libretro/scummvm
1
0
Fork 0
mirror of https://github.com/libretro/scummvm.git synced 2024-12-14 13:50:13 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity
0182b46679
scummvm/engines/kyra/sound_adlib.cpp

2309 lines
64 KiB
C++
Raw Normal View History

Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
/* ScummVM - Scumm Interpreter
* Copyright (C) 2006 The ScummVM project
*
* This program is free software; you can redistribute it and/or
* modify it under the terms of the GNU General Public License
* as published by the Free Software Foundation; either version 2
* of the License, or (at your option) any later version.
* This program is distributed in the hope that it will be useful,
* but WITHOUT ANY WARRANTY; without even the implied warranty of
* MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
* GNU General Public License for more details.
* You should have received a copy of the GNU General Public License
* along with this program; if not, write to the Free Software
* Foundation, Inc., 51 Franklin Street, Fifth Floor, Boston, MA 02110-1301, USA.
*
* $URL$
* $Id$
*
*/
#include "common/stdafx.h"
#include "common/system.h"
#include "common/mutex.h"
#include "common/timer.h"
#include "kyra/resource.h"
#include "kyra/sound.h"
#include "sound/mixer.h"
#include "sound/fmopl.h"
#include "sound/audiostream.h"
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Basic Adlib Programming:
// http://www.gamedev.net/reference/articles/article446.asp
Compensate for rounding errors in _samplesPerCallback by having readBuffer() generate "leap samples". I doubt this will make any noticeable difference, but at least now we can probably rule out timer inaccuracies from the list of possible suspects for the remaining bugs. Also removed unnecessary mutex locking/unlocking from readBuffer(), spotted by LordHoto. svn-id: r21239
2006-03-12 17:50:36 +00:00
#define CALLBACKS_PER_SECOND 72
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
namespace Kyra {
class AdlibDriver : public AudioStream {
public:
AdlibDriver(Audio::Mixer *mixer);
~AdlibDriver();
int callback(int opcode, ...);
void callback();
// AudioStream API
int readBuffer(int16 *buffer, const int numSamples) {
Changed type for samplesleft in order for MIN macro to work properly on all platforms svn-id: r21264
2006-03-13 21:38:20 +00:00
int32 samplesLeft = numSamples;
After several days, telling LordHoto over and over that "no, I don't think the timer inaccuracies can cause that kind of problems, or we'd have noticed it in the other game engines as well", guess what? The other game engines do not necessary use a timer for their Adlib music. So now Kyra doesn't either. Fortunately for my dignity, the music is still a bit uneven at times, but the situation does seem to have improved a bit, and the sound effects sound better to me now. svn-id: r21237
2006-03-12 17:05:59 +00:00
memset(buffer, 0, sizeof(int16) * numSamples);
while (samplesLeft) {
if (!_samplesTillCallback) {
callback();
_samplesTillCallback = _samplesPerCallback;
Compensate for rounding errors in _samplesPerCallback by having readBuffer() generate "leap samples". I doubt this will make any noticeable difference, but at least now we can probably rule out timer inaccuracies from the list of possible suspects for the remaining bugs. Also removed unnecessary mutex locking/unlocking from readBuffer(), spotted by LordHoto. svn-id: r21239
2006-03-12 17:50:36 +00:00
_samplesTillCallbackRemainder += _samplesPerCallbackRemainder;
if (_samplesTillCallbackRemainder >= CALLBACKS_PER_SECOND) {
_samplesTillCallback++;
_samplesTillCallbackRemainder -= CALLBACKS_PER_SECOND;
}
After several days, telling LordHoto over and over that "no, I don't think the timer inaccuracies can cause that kind of problems, or we'd have noticed it in the other game engines as well", guess what? The other game engines do not necessary use a timer for their Adlib music. So now Kyra doesn't either. Fortunately for my dignity, the music is still a bit uneven at times, but the situation does seem to have improved a bit, and the sound effects sound better to me now. svn-id: r21237
2006-03-12 17:05:59 +00:00
}
int32 render = MIN(samplesLeft, _samplesTillCallback);
samplesLeft -= render;
_samplesTillCallback -= render;
YM3812UpdateOne(_adlib, buffer, render);
buffer += render;
}
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return numSamples;
}
bool isStereo() const { return false; }
bool endOfData() const { return false; }
Added debug calls to the adlib functions, also uses mixer samplerate now. Fixed a bug in updateCallback41. svn-id: r20992
2006-03-01 16:44:32 +00:00
int getRate() const { return _mixer->getOutputRate(); }
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
private:
struct OpcodeEntry {
typedef int (AdlibDriver::*DriverOpcode)(va_list &list);
DriverOpcode function;
const char *name;
};
static const OpcodeEntry _opcodeList[];
static const int _opcodesEntries;
int snd_ret0x100(va_list &list);
int snd_ret0x1983(va_list &list);
int snd_initDriver(va_list &list);
int snd_deinitDriver(va_list &list);
int snd_setSoundData(va_list &list);
int snd_unkOpcode1(va_list &list);
int snd_startSong(va_list &list);
int snd_unkOpcode2(va_list &list);
int snd_unkOpcode3(va_list &list);
int snd_readByte(va_list &list);
int snd_writeByte(va_list &list);
int snd_setUnk5(va_list &list);
int snd_unkOpcode4(va_list &list);
int snd_dummy(va_list &list);
int snd_getNullvar4(va_list &list);
int snd_setNullvar3(va_list &list);
int snd_setFlag(va_list &list);
int snd_clearFlag(va_list &list);
Added comment detailing what I believe I know about the remaining "unk" variables in OutputState. svn-id: r21207
2006-03-10 14:53:48 +00:00
// These variables have not yet been named, but some of them are partly
// known nevertheless:
//
// unk4 - Unknown. Related to sound timing?
Cleaned up comment and did some more renaming: unk4 -> position (measured on a very short ruler :-) unk5 -> duration svn-id: r21245
2006-03-12 22:33:24 +00:00
// unk11 - Unknown. Used for updating random durations.
Added comment detailing what I believe I know about the remaining "unk" variables in OutputState. svn-id: r21207
2006-03-10 14:53:48 +00:00
// unk16 - Sound-related. Possibly some sort of pitch bend.
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
// unk18 - Sound-effect. Used for secondaryEffect1()
// unk19 - Sound-effect. Used for secondaryEffect1()
// unk20 - Sound-effect. Used for secondaryEffect1()
// unk21 - Sound-effect. Used for secondaryEffect1()
// unk22 - Sound-effect. Used for secondaryEffect1()
// unk29 - Sound-effect. Used for primaryEffect1()
// unk30 - Sound-effect. Used for primaryEffect1()
// unk31 - Sound-effect. Used for primaryEffect1()
// unk32 - Sound-effect. Used for primaryEffect2()
// unk33 - Sound-effect. Used for primaryEffect2()
// unk34 - Sound-effect. Used for primaryEffect2()
// unk35 - Sound-effect. Used for primaryEffect2()
// unk36 - Sound-effect. Used for primaryEffect2()
// unk37 - Sound-effect. Used for primaryEffect2()
// unk38 - Sound-effect. Used for primaryEffect2()
Added comment detailing what I believe I know about the remaining "unk" variables in OutputState. svn-id: r21207
2006-03-10 14:53:48 +00:00
// unk39 - Currently unused, except for updateCallback56()
// unk40 - Currently unused, except for updateCallback56()
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
// unk41 - Sound-effect. Used for primaryEffect2()
Added comment detailing what I believe I know about the remaining "unk" variables in OutputState. svn-id: r21207
2006-03-10 14:53:48 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
struct Channel {
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
uint8 opExtraLevel2;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 *dataptr;
Cleaned up comment and did some more renaming: unk4 -> position (measured on a very short ruler :-) unk5 -> duration svn-id: r21245
2006-03-12 22:33:24 +00:00
uint8 duration;
Renamed 'unk9' to a slightly more descriptive 'repeatCounter'. (I should have done that in the previous commit.) svn-id: r20987
2006-03-01 08:13:38 +00:00
uint8 repeatCounter;
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
int8 baseOctave;
Further guesswork: 'unk2' appears to be the priority of a sound, since the callbackOutput() and updateCallback3() functions won't change the channel data pointer unless the new sound has a higher unk2 and the current one. Since it's set to 0 when the data pointer is nulled, I've changed the priority to be unsigned. The updateCallback22() function is now update_setPriority(). svn-id: r21206
2006-03-10 14:22:27 +00:00
uint8 priority;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 dataptrStackPos;
uint8 *dataptrStack[4];
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
int8 baseNote;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 unk29;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 unk31;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint16 unk30;
uint16 unk37;
uint8 unk33;
uint8 unk34;
uint8 unk35;
uint8 unk36;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 unk32;
uint8 unk41;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 unk38;
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
uint8 opExtraLevel1;
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
uint8 spacing2;
- Removes deleted waitTicks prototype in KyraEngine - Adds sound fading support to the adlib player - Adds support for stopping running tracks - Changes baseFreq type back to uint8 - corrects octave handling in updateAndOutput1 (Thanks to eriktorbjorn for doing that) svn-id: r21185
2006-03-09 20:50:19 +00:00
uint8 baseFreq;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 tempo;
uint8 position;
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
uint8 regAx;
uint8 regBx;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
typedef void (AdlibDriver::*Callback)(Channel&);
Renamed callback1 to primaryEffect and callback2 to secondaryEffect. svn-id: r21246
2006-03-12 22:47:51 +00:00
Callback primaryEffect;
Callback secondaryEffect;
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
uint8 fractionalSpacing;
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
uint8 opLevel1;
uint8 opLevel2;
uint8 opExtraLevel3;
Renamed "algorithm" "twoChan" to be more consistent with our MidiDriver_ADLIB class, where the corresponding variable is called _twochan. svn-id: r21198
2006-03-10 08:55:19 +00:00
uint8 twoChan;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 unk39;
uint8 unk40;
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
uint8 spacing1;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 unk11;
uint8 unk19;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 unk18;
Added guesswork comments for stateCallback2_1() and its related functions and variables, and cleaned it up slightly. (Some of its variables are now signed.) svn-id: r21203
2006-03-10 11:18:40 +00:00
int8 unk20;
int8 unk21;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 unk22;
uint16 offset;
- Renames some functions: updateCallback31 -> update_setChannelTempo updateCallback42 -> update_resetToGlobalTempo updateCallback55 -> update_setTempoReset - Renames some vars: unk1 -> tempo unk6 -> tempoReset - Fixes tempo bugs svn-id: r21244
2006-03-12 21:54:06 +00:00
uint8 tempoReset;
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
uint8 rawNote;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
int8 unk16;
};
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void primaryEffect1(Channel &channel);
void primaryEffect2(Channel &channel);
void secondaryEffect1(Channel &channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
void resetAdlibState();
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
void writeOPL(byte reg, byte val);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void initChannel(Channel &channel);
void noteOff(Channel &channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
void unkOutput2(uint8 num);
Note to self: compile before commit. svn-id: r21103
2006-03-05 20:11:43 +00:00
uint16 getRandomNr();
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void setupDuration(uint8 duration, Channel &channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void setupNote(uint8 rawNote, Channel &channel, bool flag = false);
void setupInstrument(uint8 regOffset, uint8 *dataptr, Channel &channel);
void noteOn(Channel &channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void adjustVolume(Channel &channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint8 calculateOpLevel1(Channel &channel);
uint8 calculateOpLevel2(Channel &channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint16 checkValue(int16 val) {
if (val < 0)
val = 0;
if (val > 0x3F)
val = 0x3F;
return val;
}
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
void setupPrograms();
void executePrograms();
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
struct ParserOpcode {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
typedef int (AdlibDriver::*POpcode)(uint8 *&dataptr, Channel &channel, uint8 value);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
POpcode function;
const char *name;
};
static const ParserOpcode _parserOpcodeTable[];
static const int _parserOpcodeTableSize;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int update_setRepeat(uint8 *&dataptr, Channel &channel, uint8 value);
int update_checkRepeat(uint8 *&dataptr, Channel &channel, uint8 value);
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
int update_setupProgram(uint8 *&dataptr, Channel &channel, uint8 value);
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int update_setNoteSpacing(uint8 *&dataptr, Channel &channel, uint8 value);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int update_jump(uint8 *&dataptr, Channel &channel, uint8 value);
int update_jumpToSubroutine(uint8 *&dataptr, Channel &channel, uint8 value);
int update_returnFromSubroutine(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setBaseOctave(uint8 *&dataptr, Channel &channel, uint8 value);
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int update_stopChannel(uint8 *&dataptr, Channel &channel, uint8 value);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int update_playRest(uint8 *&dataptr, Channel &channel, uint8 value);
int update_writeAdlib(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback12(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setBaseNote(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setupSecondaryEffect1(uint8 *&dataptr, Channel &channel, uint8 value);
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int update_stopOtherChannel(uint8 *&dataptr, Channel &channel, uint8 value);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int updateCallback16(uint8 *&dataptr, Channel &channel, uint8 value);
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int update_setupInstrument(uint8 *&dataptr, Channel &channel, uint8 value);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int update_setupPrimaryEffect1(uint8 *&dataptr, Channel &channel, uint8 value);
int update_removePrimaryEffect1(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setBaseFreq(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setupPrimaryEffect2(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setPriority(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback23(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback24(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setExtraLevel1(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback26(uint8 *&dataptr, Channel &channel, uint8 value);
int update_playNote(uint8 *&dataptr, Channel &channel, uint8 value);
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int update_setFractionalNoteSpacing(uint8 *&dataptr, Channel &channel, uint8 value);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int update_setTempo(uint8 *&dataptr, Channel &channel, uint8 value);
int update_removeSecondaryEffect1(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setChannelTempo(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setExtraLevel3(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setExtraLevel2(uint8 *&dataptr, Channel &channel, uint8 value);
int update_changeExtraLevel2(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setAMDepth(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setVibratoDepth(uint8 *&dataptr, Channel &channel, uint8 value);
int update_changeExtraLevel1(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback38(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback39(uint8 *&dataptr, Channel &channel, uint8 value);
int update_removePrimaryEffect2(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback41(uint8 *&dataptr, Channel &channel, uint8 value);
int update_resetToGlobalTempo(uint8 *&dataptr, Channel &channel, uint8 value);
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int update_nop1(uint8 *&dataptr, Channel &channel, uint8 value);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int updateCallback44(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback45(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback46(uint8 *&dataptr, Channel &channel, uint8 value);
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int update_nop2(uint8 *&dataptr, Channel &channel, uint8 value);
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
int update_setupRhythmSection(uint8 *&dataptr, Channel &channel, uint8 value);
int update_playRhythmSection(uint8 *&dataptr, Channel &channel, uint8 value);
int update_removeRhythmSection(uint8 *&dataptr, Channel &channel, uint8 value);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int updateCallback51(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback52(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback53(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback54(uint8 *&dataptr, Channel &channel, uint8 value);
int update_setTempoReset(uint8 *&dataptr, Channel &channel, uint8 value);
int updateCallback56(uint8 *&dataptr, Channel &channel, uint8 value);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
private:
More comments, and some very minor cleanup. svn-id: r21208
2006-03-10 15:30:22 +00:00
// These variables have not yet been named, but some of them are partly
// known nevertheless:
//
// _unk5 - Currently unused, except for updateCallback54()
// _unkValue1 - Unknown. Used for updating _unkValue2
// _unkValue2 - Unknown. Used for updating _unkValue4
// _unkValue3 - Unknown. Used for updating _unkValue2
// _unkValue4 - Unknown. Used for updating _unkValue5
// _unkValue5 - Unknown. Used for controlling updateCallback24().
// _unkValue6 - Unknown. Something to do with channel 1 volume?
// _unkValue7 - Unknown. Something to do with channel 2 volume?
// _unkValue8 - Unknown. Something to do with channel 2 volume?
// _unkValue9 - Unknown. Something to do with channel 3 volume?
// _unkValue10 - Unknown. Something to do with channel 3 volume?
// _unkValue11 - Unknown. Something to do with channel 2 volume?
// _unkValue12 - Unknown. Something to do with channel 2 volume?
// _unkValue13 - Unknown. Something to do with channel 3 volume?
// _unkValue14 - Unknown. Something to do with channel 3 volume?
// _unkValue15 - Unknown. Something to do with channel 3 volume?
// _unkValue16 - Unknown. Something to do with channel 3 volume?
// _unkValue17 - Unknown. Something to do with channel 2 volume?
// _unkValue18 - Unknown. Something to do with channel 2 volume?
// _unkValue19 - Unknown. Something to do with channel 1 volume?
// _unkValue20 - Unknown. Something to do with channel 1 volume?
// _unkOutputByte2 - Unknown. Something to do with the BD register.
// _unkTable[] - Probably frequences for the 12-tone scale.
// _unkTable2[] - Unknown. Currently only used by updateCallback46()
// _unkTable2_1[] - One of the tables in _unkTable2[]
// _unkTable2_2[] - One of the tables in _unkTable2[]
// _unkTable2_3[] - One of the tables in _unkTable2[]
After several days, telling LordHoto over and over that "no, I don't think the timer inaccuracies can cause that kind of problems, or we'd have noticed it in the other game engines as well", guess what? The other game engines do not necessary use a timer for their Adlib music. So now Kyra doesn't either. Fortunately for my dignity, the music is still a bit uneven at times, but the situation does seem to have improved a bit, and the sound effects sound better to me now. svn-id: r21237
2006-03-12 17:05:59 +00:00
int32 _samplesPerCallback;
Compensate for rounding errors in _samplesPerCallback by having readBuffer() generate "leap samples". I doubt this will make any noticeable difference, but at least now we can probably rule out timer inaccuracies from the list of possible suspects for the remaining bugs. Also removed unnecessary mutex locking/unlocking from readBuffer(), spotted by LordHoto. svn-id: r21239
2006-03-12 17:50:36 +00:00
int32 _samplesPerCallbackRemainder;
After several days, telling LordHoto over and over that "no, I don't think the timer inaccuracies can cause that kind of problems, or we'd have noticed it in the other game engines as well", guess what? The other game engines do not necessary use a timer for their Adlib music. So now Kyra doesn't either. Fortunately for my dignity, the music is still a bit uneven at times, but the situation does seem to have improved a bit, and the sound effects sound better to me now. svn-id: r21237
2006-03-12 17:05:59 +00:00
int32 _samplesTillCallback;
Compensate for rounding errors in _samplesPerCallback by having readBuffer() generate "leap samples". I doubt this will make any noticeable difference, but at least now we can probably rule out timer inaccuracies from the list of possible suspects for the remaining bugs. Also removed unnecessary mutex locking/unlocking from readBuffer(), spotted by LordHoto. svn-id: r21239
2006-03-12 17:50:36 +00:00
int32 _samplesTillCallbackRemainder;
After several days, telling LordHoto over and over that "no, I don't think the timer inaccuracies can cause that kind of problems, or we'd have noticed it in the other game engines as well", guess what? The other game engines do not necessary use a timer for their Adlib music. So now Kyra doesn't either. Fortunately for my dignity, the music is still a bit uneven at times, but the situation does seem to have improved a bit, and the sound effects sound better to me now. svn-id: r21237
2006-03-12 17:05:59 +00:00
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
int _lastProcessed;
Renamed setTempo to update_setTempo, also fixed tempo handling (intro is too fast now though, but seems to be another problem). Also changed some unsigned values to signed. svn-id: r21111
2006-03-06 16:37:03 +00:00
int8 _flagTrigger;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int _curChannel;
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
uint8 _rhythmSection;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 _unk5;
int _soundsPlaying;
The updateUnk6Value() appears to be some sort of random number generator, so I have renamed it getRandomNr(). The _unk6 variable has been renamed _rnd. (Our MidiDriver_ADLIB class also has its own random number generator, so it's not such a far-fetched guess as it may seem.) svn-id: r21102
2006-03-05 20:09:02 +00:00
uint16 _rnd;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 _unkValue1;
uint8 _unkValue2;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 _unkValue3;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 _unkValue4;
uint8 _unkValue5;
uint8 _unkValue6;
uint8 _unkValue7;
uint8 _unkValue8;
uint8 _unkValue9;
uint8 _unkValue10;
uint8 _unkValue11;
uint8 _unkValue12;
uint8 _unkValue13;
uint8 _unkValue14;
uint8 _unkValue15;
uint8 _unkValue16;
uint8 _unkValue17;
uint8 _unkValue18;
uint8 _unkValue19;
uint8 _unkValue20;
int _flags;
FM_OPL *_adlib;
uint8 *_soundData;
uint8 _soundIdTable[0x10];
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
Channel _channels[10];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 _unkOutputByte2;
Renamed _unkOutputByte1 _curRegOffset. It doesn't always correspond to the _curTable channel, but it probably does so whenever both of them are used together. svn-id: r21204
2006-03-10 11:26:32 +00:00
uint8 _curRegOffset;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 _tempo;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
const uint8 *_tablePtr1;
const uint8 *_tablePtr2;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
static const uint8 _regOffset[];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
static const uint16 _unkTable[];
static const uint8 *_unkTable2[];
static const uint8 _unkTable2_1[];
static const uint8 _unkTable2_2[];
static const uint8 _unkTable2_3[];
static const uint8 _unkTables[][32];
Common::Mutex _mutex;
Audio::Mixer *_mixer;
void lock() { _mutex.lock(); }
void unlock() { _mutex.unlock(); }
};
AdlibDriver::AdlibDriver(Audio::Mixer *mixer) {
_mixer = mixer;
_flags = 0;
Added debug calls to the adlib functions, also uses mixer samplerate now. Fixed a bug in updateCallback41. svn-id: r20992
2006-03-01 16:44:32 +00:00
_adlib = makeAdlibOPL(getRate());
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
assert(_adlib);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
memset(_channels, 0, sizeof(_channels));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_soundData = 0;
Renamed _unkOutputByte1 _curRegOffset. It doesn't always correspond to the _curTable channel, but it probably does so whenever both of them are used together. svn-id: r21204
2006-03-10 11:26:32 +00:00
_unkOutputByte2 = _curRegOffset = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
_lastProcessed = _flagTrigger = _curChannel = _rhythmSection = 0;
The updateUnk6Value() appears to be some sort of random number generator, so I have renamed it getRandomNr(). The _unk6 variable has been renamed _rnd. (Our MidiDriver_ADLIB class also has its own random number generator, so it's not such a far-fetched guess as it may seem.) svn-id: r21102
2006-03-05 20:09:02 +00:00
_rnd = 0x1234;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
More renamings: _unkTableByte1 is now _tempo, and updateCallback29() is setTempo(). svn-id: r21110
2006-03-06 16:31:19 +00:00
_tempo = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
_unkValue3 = 0xFF;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_unkValue1 = _unkValue2 = _unkValue4 = _unkValue5 = 0;
_unkValue6 = _unkValue7 = _unkValue8 = _unkValue9 = _unkValue10 = 0;
_unkValue11 = _unkValue12 = _unkValue13 = _unkValue14 = _unkValue15 =
_unkValue16 = _unkValue17 = _unkValue18 = _unkValue19 = _unkValue20 = 0;
_tablePtr1 = _tablePtr2 = 0;
_mixer->setupPremix(this);
Compensate for rounding errors in _samplesPerCallback by having readBuffer() generate "leap samples". I doubt this will make any noticeable difference, but at least now we can probably rule out timer inaccuracies from the list of possible suspects for the remaining bugs. Also removed unnecessary mutex locking/unlocking from readBuffer(), spotted by LordHoto. svn-id: r21239
2006-03-12 17:50:36 +00:00
_samplesPerCallback = getRate() / CALLBACKS_PER_SECOND;
_samplesPerCallbackRemainder = getRate() % CALLBACKS_PER_SECOND;
After several days, telling LordHoto over and over that "no, I don't think the timer inaccuracies can cause that kind of problems, or we'd have noticed it in the other game engines as well", guess what? The other game engines do not necessary use a timer for their Adlib music. So now Kyra doesn't either. Fortunately for my dignity, the music is still a bit uneven at times, but the situation does seem to have improved a bit, and the sound effects sound better to me now. svn-id: r21237
2006-03-12 17:05:59 +00:00
_samplesTillCallback = 0;
Compensate for rounding errors in _samplesPerCallback by having readBuffer() generate "leap samples". I doubt this will make any noticeable difference, but at least now we can probably rule out timer inaccuracies from the list of possible suspects for the remaining bugs. Also removed unnecessary mutex locking/unlocking from readBuffer(), spotted by LordHoto. svn-id: r21239
2006-03-12 17:50:36 +00:00
_samplesTillCallbackRemainder = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
AdlibDriver::~AdlibDriver() {
_mixer->setupPremix(0);
OPLDestroy(_adlib);
_adlib = 0;
}
int AdlibDriver::callback(int opcode, ...) {
lock();
if (opcode >= _opcodesEntries || opcode < 0) {
warning("AdlibDriver: calling unknown opcode '%d'", opcode);
return 0;
}
debugC(9, kDebugLevelSound, "Calling opcode '%s' (%d)", _opcodeList[opcode].name, opcode);
va_list args;
va_start(args, opcode);
int returnValue = (this->*(_opcodeList[opcode].function))(args);
va_end(args);
unlock();
return returnValue;
}
// Opcodes
int AdlibDriver::snd_ret0x100(va_list &list) {
return 0x100;
}
int AdlibDriver::snd_ret0x1983(va_list &list) {
return 0x1983;
}
int AdlibDriver::snd_initDriver(va_list &list) {
_lastProcessed = _soundsPlaying = 0;
resetAdlibState();
return 0;
}
int AdlibDriver::snd_deinitDriver(va_list &list) {
resetAdlibState();
return 0;
}
int AdlibDriver::snd_setSoundData(va_list &list) {
if (_soundData) {
delete [] _soundData;
_soundData = 0;
}
_soundData = va_arg(list, uint8*);
return 0;
}
int AdlibDriver::snd_unkOpcode1(va_list &list) {
warning("unimplemented snd_unkOpcode1");
return 0;
}
int AdlibDriver::snd_startSong(va_list &list) {
int songId = va_arg(list, int);
_flags |= 8;
Fixes hangs in the "Pool of Sorrow" scene (at least I don't get them anymore). svn-id: r20964
2006-02-27 23:41:27 +00:00
_flagTrigger = 1;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint16 offset = READ_LE_UINT16(&_soundData[songId << 1]);
uint8 firstByte = *(_soundData + offset);
if ((songId << 1) != 0) {
if (firstByte == 9) {
if (_flags & 2)
return 0;
} else {
if (_flags & 1)
return 0;
}
}
_soundIdTable[_soundsPlaying] = songId;
++_soundsPlaying;
_soundsPlaying &= 0x0F;
return 0;
}
int AdlibDriver::snd_unkOpcode2(va_list &list) {
warning("unimplemented snd_unkOpcode2");
return 0;
}
int AdlibDriver::snd_unkOpcode3(va_list &list) {
int value = va_arg(list, int);
int loop = value;
if (value < 0) {
value = 0;
loop = 9;
}
loop -= value;
++loop;
while (loop--) {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = value;
Channel &channel = _channels[_curChannel];
channel.priority = 0;
channel.dataptr = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
if (value != 9) {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
noteOff(channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
++value;
}
return 0;
}
int AdlibDriver::snd_readByte(va_list &list) {
This may be needless paranoia, but I always get nervous about order of evaluation when there are multiple side-effects on the same variable (in this case the internal state modified by va_arg()) in the same expression. So let's just humor me and do the va_arg() calls one at a time. svn-id: r20970
2006-02-28 08:49:51 +00:00
int a = va_arg(list, int);
int b = va_arg(list, int);
uint8 *ptr = _soundData + READ_LE_UINT16(&_soundData[a << 1]) + b;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return *ptr;
}
int AdlibDriver::snd_writeByte(va_list &list) {
This may be needless paranoia, but I always get nervous about order of evaluation when there are multiple side-effects on the same variable (in this case the internal state modified by va_arg()) in the same expression. So let's just humor me and do the va_arg() calls one at a time. svn-id: r20970
2006-02-28 08:49:51 +00:00
int a = va_arg(list, int);
int b = va_arg(list, int);
int c = va_arg(list, int);
uint8 *ptr = _soundData + READ_LE_UINT16(&_soundData[a << 1]) + b;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 oldValue = *ptr;
This may be needless paranoia, but I always get nervous about order of evaluation when there are multiple side-effects on the same variable (in this case the internal state modified by va_arg()) in the same expression. So let's just humor me and do the va_arg() calls one at a time. svn-id: r20970
2006-02-28 08:49:51 +00:00
*ptr = (uint8)c;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return oldValue;
}
int AdlibDriver::snd_setUnk5(va_list &list) {
warning("unimplemented snd_setUnk5");
return 0;
}
int AdlibDriver::snd_unkOpcode4(va_list &list) {
warning("unimplemented snd_unkOpcode4");
return 0;
}
int AdlibDriver::snd_dummy(va_list &list) {
return 0;
}
int AdlibDriver::snd_getNullvar4(va_list &list) {
warning("unimplemented snd_getNullvar4");
return 0;
}
int AdlibDriver::snd_setNullvar3(va_list &list) {
warning("unimplemented snd_setNullvar3");
return 0;
}
int AdlibDriver::snd_setFlag(va_list &list) {
int oldFlags = _flags;
_flags |= va_arg(list, int);
return oldFlags;
}
int AdlibDriver::snd_clearFlag(va_list &list) {
int oldFlags = _flags;
_flags &= ~(va_arg(list, int));
return oldFlags;
}
// timer callback
void AdlibDriver::callback() {
lock();
Fixes hangs in the "Pool of Sorrow" scene (at least I don't get them anymore). svn-id: r20964
2006-02-27 23:41:27 +00:00
--_flagTrigger;
Renamed setTempo to update_setTempo, also fixed tempo handling (intro is too fast now though, but seems to be another problem). Also changed some unsigned values to signed. svn-id: r21111
2006-03-06 16:37:03 +00:00
if (_flagTrigger < 0)
Minor cleanup and comments. svn-id: r20983
2006-03-01 06:56:53 +00:00
_flags &= ~8;
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
setupPrograms();
executePrograms();
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 temp = _unkValue3;
More renamings: _unkTableByte1 is now _tempo, and updateCallback29() is setTempo(). svn-id: r21110
2006-03-06 16:31:19 +00:00
_unkValue3 += _tempo;
- Renames some functions: updateCallback31 -> update_setChannelTempo updateCallback42 -> update_resetToGlobalTempo updateCallback55 -> update_setTempoReset - Renames some vars: unk1 -> tempo unk6 -> tempoReset - Fixes tempo bugs svn-id: r21244
2006-03-12 21:54:06 +00:00
if (_unkValue3 < temp) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
if (!(--_unkValue2)) {
_unkValue2 = _unkValue1;
++_unkValue4;
}
}
unlock();
}
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
void AdlibDriver::setupPrograms() {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
while (_lastProcessed != _soundsPlaying) {
Fixed some comments and made a couple of cleanups to make it a bit more clear just how similar callbackOutput() and updateCallback3() are. svn-id: r21333
2006-03-16 17:27:52 +00:00
uint16 add = _soundIdTable[_lastProcessed] << 1;
uint8 *ptr = _soundData + READ_LE_UINT16(_soundData + add);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint8 chan = *ptr++;
Further guesswork: 'unk2' appears to be the priority of a sound, since the callbackOutput() and updateCallback3() functions won't change the channel data pointer unless the new sound has a higher unk2 and the current one. Since it's set to 0 when the data pointer is nulled, I've changed the priority to be unsigned. The updateCallback22() function is now update_setPriority(). svn-id: r21206
2006-03-10 14:22:27 +00:00
uint8 priority = *ptr++;
// Only start this sound if its priority is higher than the one
// already playing.
Fixed some comments and made a couple of cleanups to make it a bit more clear just how similar callbackOutput() and updateCallback3() are. svn-id: r21333
2006-03-16 17:27:52 +00:00
Channel &channel = _channels[chan];
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
if (priority >= channel.priority) {
initChannel(channel);
channel.priority = priority;
channel.dataptr = ptr;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
channel.tempo = 0xFF;
channel.position = 0xFF;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.duration = 1;
I don't know if it's our bug, or a bug in the original Kyra music driver, but updateCallback3() would call unkOutput2(9) in at least one case. This is obviously wrong because it a) reads outside _regOffsets[], and b) writes to invalid Adlib registers. Now unkOutput2() has the same safeguards as noteOff() already had, making callbackOutput() and updateCallback3() even more similar. svn-id: r21335
2006-03-16 17:39:23 +00:00
unkOutput2(chan);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
++_lastProcessed;
_lastProcessed &= 0x0F;
}
}
Added a block of comments about command parsing and timing. Perhaps it will be of use in tracking down the tempo bugs. svn-id: r21230
2006-03-12 10:47:20 +00:00
// A few words on opcode parsing and timing:
//
Cleaned up comment and did some more renaming: unk4 -> position (measured on a very short ruler :-) unk5 -> duration svn-id: r21245
2006-03-12 22:33:24 +00:00
// First of all, We simulate a timer callback 72 times per second. Each timeout
// we update each channel that has something to play.
Added a block of comments about command parsing and timing. Perhaps it will be of use in tracking down the tempo bugs. svn-id: r21230
2006-03-12 10:47:20 +00:00
//
Cleaned up comment and did some more renaming: unk4 -> position (measured on a very short ruler :-) unk5 -> duration svn-id: r21245
2006-03-12 22:33:24 +00:00
// Each channel has its own individual tempo, which is added to its position.
// This will frequently cause the position to "wrap around" but that is
// intentional. In fact, it's the signal to go ahead and do more stuff with
// that channel.
Added a block of comments about command parsing and timing. Perhaps it will be of use in tracking down the tempo bugs. svn-id: r21230
2006-03-12 10:47:20 +00:00
//
Cleaned up comment and did some more renaming: unk4 -> position (measured on a very short ruler :-) unk5 -> duration svn-id: r21245
2006-03-12 22:33:24 +00:00
// Each channel also has a duration, indicating how much time is left on the
// its current task. This duration is decreased by one. As long as it still has
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
// not reached zero, the only thing that can happen is that the note is turned
// off depending on manual or automatic note spacing. Once the duration reaches
// zero, a new set of musical opcodes are executed.
Added a block of comments about command parsing and timing. Perhaps it will be of use in tracking down the tempo bugs. svn-id: r21230
2006-03-12 10:47:20 +00:00
//
// An opcode is one byte, followed by a variable number of parameters. Since
// most opcodes have at least one one-byte parameter, we read that as well. Any
Cleaned up comment and did some more renaming: unk4 -> position (measured on a very short ruler :-) unk5 -> duration svn-id: r21245
2006-03-12 22:33:24 +00:00
// opcode that doesn't have that one parameter is responsible for moving the
// data pointer back again.
Added a block of comments about command parsing and timing. Perhaps it will be of use in tracking down the tempo bugs. svn-id: r21230
2006-03-12 10:47:20 +00:00
//
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
// If the most significant bit of the opcode is 1, it's a function; call it.
// The opcode functions return either 0 (continue), 1 (stop) or 2 (stop, and do
// not run the effects callbacks).
Added a block of comments about command parsing and timing. Perhaps it will be of use in tracking down the tempo bugs. svn-id: r21230
2006-03-12 10:47:20 +00:00
//
// If the most significant bit of the opcode is 0, it's a note, and the first
// parameter is its duration. (There are cases where the duration is modified
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
// but that's an exception.) The note opcode is assumed to return 1, and is the
// last opcode unless its duration is zero.
Added a block of comments about command parsing and timing. Perhaps it will be of use in tracking down the tempo bugs. svn-id: r21230
2006-03-12 10:47:20 +00:00
//
// Finally, most of the times that the callback is called, it will invoke the
// effects callbacks. The final opcode in a set can prevent this, if it's a
// function and it returns anything other than 1.
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
void AdlibDriver::executePrograms() {
Fixed some comments and made a couple of cleanups to make it a bit more clear just how similar callbackOutput() and updateCallback3() are. svn-id: r21333
2006-03-16 17:27:52 +00:00
// Each channel runs its own program. There are ten channels: One for
// each Adlib channel (0-8), plus one "control channel" (9) which is
// the one that tells the other channels what to do.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
for (_curChannel = 9; _curChannel >= 0; --_curChannel) {
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
int result = 1;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
if (!_channels[_curChannel].dataptr) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
continue;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
Channel &channel = _channels[_curChannel];
_curRegOffset = _regOffset[_curChannel];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
if (channel.tempoReset) {
channel.tempo = _tempo;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 backup = channel.position;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.position += channel.tempo;
if (channel.position < backup) {
if (--channel.duration) {
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
if (channel.duration == channel.spacing2)
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
noteOff(channel);
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
if (channel.duration == channel.spacing1 && _curChannel != 9)
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
noteOff(channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
} else {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
while (channel.dataptr) {
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
uint8 opcode = *channel.dataptr++;
uint8 param = *channel.dataptr++;
if (opcode & 0x80) {
opcode &= 0x7F;
if (opcode >= _parserOpcodeTableSize)
opcode = _parserOpcodeTableSize - 1;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
debugC(9, kDebugLevelSound, "Calling opcode '%s' (%d) (channel: %d)", _parserOpcodeTable[opcode].name, opcode, _curChannel);
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
result = (this->*(_parserOpcodeTable[opcode].function))(channel.dataptr, channel, param);
if (result)
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
break;
} else {
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
debugC(9, kDebugLevelSound, "Note on opcode 0x%02X (duration: %d) (channel: %d)", opcode, param, _curChannel);
setupNote(opcode, channel);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
noteOn(channel);
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
setupDuration(param, channel);
if (param)
break;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
}
}
}
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
if (result == 1) {
if (channel.primaryEffect)
(this->*(channel.primaryEffect))(channel);
if (channel.secondaryEffect)
(this->*(channel.secondaryEffect))(channel);
}
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
}
//
void AdlibDriver::resetAdlibState() {
Added debug calls to the adlib functions, also uses mixer samplerate now. Fixed a bug in updateCallback41. svn-id: r20992
2006-03-01 16:44:32 +00:00
debugC(9, kDebugLevelSound, "resetAdlibState()");
The updateUnk6Value() appears to be some sort of random number generator, so I have renamed it getRandomNr(). The _unk6 variable has been renamed _rnd. (Our MidiDriver_ADLIB class also has its own random number generator, so it's not such a far-fetched guess as it may seem.) svn-id: r21102
2006-03-05 20:09:02 +00:00
_rnd = 0x1234;
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
// Authorize the control of the waveforms
writeOPL(0x01, 0x20);
// Select FM music mode
writeOPL(0x08, 0x00);
// I would guess the main purpose of this is to turn off the rhythm,
// thus allowing us to use 9 melodic voices instead of 6.
writeOPL(0xBD, 0x00);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
int loop = 10;
while (loop--) {
if (loop != 9) {
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
// Silence the channel
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0x40 + _regOffset[loop], 0x3F);
writeOPL(0x43 + _regOffset[loop], 0x3F);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
initChannel(_channels[loop]);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
}
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
// Old calling style: output0x388(0xABCD)
// New calling style: writeOPL(0xAB, 0xCD)
Minor cleanup and comments. svn-id: r20983
2006-03-01 06:56:53 +00:00
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
void AdlibDriver::writeOPL(byte reg, byte val) {
OPLWriteReg(_adlib, reg, val);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::initChannel(Channel &channel) {
debugC(9, kDebugLevelSound, "initChannel(%d)", &channel - _channels);
memset(&channel.dataptr, 0, sizeof(Channel) - ((char*)&channel.dataptr - (char*)&channel));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
channel.tempo = 0xFF;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.priority = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
// normally here are nullfuncs but we set 0 for now
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.primaryEffect = 0;
channel.secondaryEffect = 0;
Cleaned up the comments in unkOutput2(), and added a big note about the bug where low-frequent noises are playing at the beginning of some new sounds. svn-id: r21265
2006-03-13 23:05:11 +00:00
channel.spacing1 = 1;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::noteOff(Channel &channel) {
debugC(9, kDebugLevelSound, "noteOff(%d)", &channel - _channels);
More comments, and some very minor cleanup. svn-id: r21208
2006-03-10 15:30:22 +00:00
Fixed some comments and made a couple of cleanups to make it a bit more clear just how similar callbackOutput() and updateCallback3() are. svn-id: r21333
2006-03-16 17:27:52 +00:00
// The control channel has no corresponding Adlib channel
if (_curChannel >= 9)
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return;
More comments, and some very minor cleanup. svn-id: r21208
2006-03-10 15:30:22 +00:00
Fixed some comments and made a couple of cleanups to make it a bit more clear just how similar callbackOutput() and updateCallback3() are. svn-id: r21333
2006-03-16 17:27:52 +00:00
// When the rhythm section is enabled, channels 6, 7 and 8 are special.
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
if (_rhythmSection && _curChannel >= 6)
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// This means the "Key On" bit will always be 0
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.regBx &= 0xDF;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
// Octave / F-Number / Key-On
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xB0 + _curChannel, channel.regBx);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::unkOutput2(uint8 chan) {
debugC(9, kDebugLevelSound, "unkOutput2(%d)", chan);
More comments, and some very minor cleanup. svn-id: r21208
2006-03-10 15:30:22 +00:00
I don't know if it's our bug, or a bug in the original Kyra music driver, but updateCallback3() would call unkOutput2(9) in at least one case. This is obviously wrong because it a) reads outside _regOffsets[], and b) writes to invalid Adlib registers. Now unkOutput2() has the same safeguards as noteOff() already had, making callbackOutput() and updateCallback3() even more similar. svn-id: r21335
2006-03-16 17:39:23 +00:00
// The control channel has no corresponding Adlib channel
if (chan >= 9)
return;
More comments, and some very minor cleanup. svn-id: r21208
2006-03-10 15:30:22 +00:00
// I believe this has to do with channels 6, 7, and 8 being special
// when Adlib's rhythm section is enabled.
Cleaned up the comments in unkOutput2(), and added a big note about the bug where low-frequent noises are playing at the beginning of some new sounds. svn-id: r21265
2006-03-13 23:05:11 +00:00
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
if (_rhythmSection && chan >= 6)
More comments, and some very minor cleanup. svn-id: r21208
2006-03-10 15:30:22 +00:00
return;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint8 offset = _regOffset[chan];
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Cleaned up the comments in unkOutput2(), and added a big note about the bug where low-frequent noises are playing at the beginning of some new sounds. svn-id: r21265
2006-03-13 23:05:11 +00:00
// The channel is cleared: First the attack/delay rate, then the
Fix typo in comment. svn-id: r21298
2006-03-14 19:08:09 +00:00
// sustain level/release rate, and finally the note is turned off.
Cleaned up the comments in unkOutput2(), and added a big note about the bug where low-frequent noises are playing at the beginning of some new sounds. svn-id: r21265
2006-03-13 23:05:11 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0x60 + offset, 0xFF);
writeOPL(0x63 + offset, 0xFF);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0x80 + offset, 0xFF);
writeOPL(0x83 + offset, 0xFF);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Cleaned up the comments in unkOutput2(), and added a big note about the bug where low-frequent noises are playing at the beginning of some new sounds. svn-id: r21265
2006-03-13 23:05:11 +00:00
writeOPL(0xB0 + chan, 0x00);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Cleaned up the comments in unkOutput2(), and added a big note about the bug where low-frequent noises are playing at the beginning of some new sounds. svn-id: r21265
2006-03-13 23:05:11 +00:00
// ...and then the note is turned on again, with whatever value is
// still lurking in the A0 + chan register, but everything else -
// including the two most significant frequency bit, and the octave -
// set to zero.
//
Added workaround in FMOPL for the pathological case where a note was turned off while still at the very beginning of the "attack" phase. This is the very lowest point on the attack curve, yet it would continue from the beginning of the release curve, i.e. its very highest point. This is what caused Kyra to often play low-frequency notes at the very beginning of a new song. (That, and a truly bizarre function for initialising the channels.) The proper fix would be to locate the correct point on the release curve and continue from there. For now, though, only handle the trivial case. svn-id: r21302
2006-03-14 23:01:44 +00:00
// This is very strange behaviour, and causes problems with the ancient
// FMOPL code we borrowed from AdPlug. I've added a workaround. See
// fmopl.cpp for more details.
Further clarifications to the comment about the "low-frequency notes at beginning of music" bugs. As an experiment, I tried hooking up ScummVM with the most recent version of the FMOPL code, and the bug went away. Unfortunately, I believe we are stuck with the ancient version for license (and performance?) reasons. svn-id: r21267
2006-03-13 23:53:15 +00:00
//
Added workaround in FMOPL for the pathological case where a note was turned off while still at the very beginning of the "attack" phase. This is the very lowest point on the attack curve, yet it would continue from the beginning of the release curve, i.e. its very highest point. This is what caused Kyra to often play low-frequency notes at the very beginning of a new song. (That, and a truly bizarre function for initialising the channels.) The proper fix would be to locate the correct point on the release curve and continue from there. For now, though, only handle the trivial case. svn-id: r21302
2006-03-14 23:01:44 +00:00
// More recent versions of the MAME FMOPL don't seem to have this
// problem, but cannot currently be used because of licensing and
// performance issues.
Updated comment with more circumstantial evidence that the low-frequency note bug is a bug in our fmopl.cpp. Ken Silverman's Adlib emulator doesn't seem to exhibit the bug either. (But is apparently less feature complete than MAME's.) svn-id: r21275
2006-03-14 09:55:41 +00:00
//
// Ken Silverman's Adlib emulator (which can be found on his Web page -
// http://www.advsys.net/ken - and as part of AdPlug) also seems to be
Added workaround in FMOPL for the pathological case where a note was turned off while still at the very beginning of the "attack" phase. This is the very lowest point on the attack curve, yet it would continue from the beginning of the release curve, i.e. its very highest point. This is what caused Kyra to often play low-frequency notes at the very beginning of a new song. (That, and a truly bizarre function for initialising the channels.) The proper fix would be to locate the correct point on the release curve and continue from there. For now, though, only handle the trivial case. svn-id: r21302
2006-03-14 23:01:44 +00:00
// immune, but is apparently not as feature complete as MAME's.
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xB0 + chan, 0x20);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
The updateUnk6Value() appears to be some sort of random number generator, so I have renamed it getRandomNr(). The _unk6 variable has been renamed _rnd. (Our MidiDriver_ADLIB class also has its own random number generator, so it's not such a far-fetched guess as it may seem.) svn-id: r21102
2006-03-05 20:09:02 +00:00
// I believe this is a random number generator. It actually does seem to
// generate an even distribution of almost all numbers from 0 through 65535,
// though in my tests some numbers were never generated.
uint16 AdlibDriver::getRandomNr() {
_rnd += 0x9248;
uint16 lowBits = _rnd & 7;
_rnd >>= 3;
_rnd |= (lowBits << 13);
return _rnd;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::setupDuration(uint8 duration, Channel &channel) {
debugC(9, kDebugLevelSound, "setupDuration(%d, %d)", duration, &channel - _channels);
if (channel.unk11) {
channel.duration = duration + (getRandomNr() & channel.unk11);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return;
}
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
if (channel.fractionalSpacing) {
channel.spacing2 = (duration >> 3) * channel.fractionalSpacing;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.duration = duration;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
// This function may or may not play the note. It's usually followed by a call
// to noteOn(), which will always play the current note.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::setupNote(uint8 rawNote, Channel &channel, bool flag) {
debugC(9, kDebugLevelSound, "setupNote(%d, %d)", rawNote, &channel - _channels);
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.rawNote = rawNote;
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int8 note = (rawNote & 0x0F) + channel.baseNote;
int8 octave = ((rawNote + channel.baseOctave) >> 4) & 0x0F;
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
// There are only twelve notes. If we go outside that, we have to
// adjust the note and octave.
if (note >= 12) {
note -= 12;
octave++;
} else if (note < 0) {
note += 12;
octave--;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
As LordHoto pointed out to me, updateCallback41() was slightly different after all: it adjusts the frequency even when state.unk16 is zero. Fixed that, and added some comments to hopefully explain the differences between the cleaned up code and the original disassembly. svn-id: r21184
2006-03-09 18:42:05 +00:00
// The calculation of frequency looks quite different from the original
// disassembly at a first glance, but when you consider that the
// largest possible value would be 0x0246 + 0xFF + 0x47 (and that's if
Some more/updated comments. svn-id: r21196
2006-03-10 07:56:22 +00:00
// baseFreq is unsigned), freq is still a 10-bit value, just as it
// should be to fit in the Ax and Bx registers.
As LordHoto pointed out to me, updateCallback41() was slightly different after all: it adjusts the frequency even when state.unk16 is zero. Fixed that, and added some comments to hopefully explain the differences between the cleaned up code and the original disassembly. svn-id: r21184
2006-03-09 18:42:05 +00:00
//
// If it were larger than that, it could have overflowed into the
// octave bits, and that could possibly have been used in some sound.
// But as it is now, I can't see any way it would happen.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint16 freq = _unkTable[note] + channel.baseFreq;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
As LordHoto pointed out to me, updateCallback41() was slightly different after all: it adjusts the frequency even when state.unk16 is zero. Fixed that, and added some comments to hopefully explain the differences between the cleaned up code and the original disassembly. svn-id: r21184
2006-03-09 18:42:05 +00:00
// When called from callback 41, the behaviour is slightly different:
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
// We adjust the frequency, even when channel.unk16 is 0.
As LordHoto pointed out to me, updateCallback41() was slightly different after all: it adjusts the frequency even when state.unk16 is zero. Fixed that, and added some comments to hopefully explain the differences between the cleaned up code and the original disassembly. svn-id: r21184
2006-03-09 18:42:05 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
if (channel.unk16 || flag) {
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
const uint8 *table;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
if (channel.unk16 >= 0) {
table = _unkTables[(channel.rawNote & 0x0F) + 2];
freq += table[channel.unk16];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
} else {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
table = _unkTables[channel.rawNote & 0x0F];
freq -= table[-channel.unk16];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.regAx = freq & 0xFF;
channel.regBx = (channel.regBx & 0x20) | (octave << 2) | ((freq >> 8) & 0x03);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
// Keep the note on or off
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xA0 + _curChannel, channel.regAx);
writeOPL(0xB0 + _curChannel, channel.regBx);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::setupInstrument(uint8 regOffset, uint8 *dataptr, Channel &channel) {
debugC(9, kDebugLevelSound, "setupInstrument(%d, %p, %d)", regOffset, (const void *)dataptr, &channel - _channels);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Amplitude Modulation / Vibrato / Envelope Generator Type /
// Keyboard Scaling Rate / Modulator Frequency Multiple
Since updateAndOutput2() is the only function that sets up wave forms and stuff, it pretty much has to be our "set instrument" function. Also, while I'm not entirely sure of the implications, "unk23" is the Algorithm bit for the Feedback / Algorithm register, so I've renamed it "algorithm". svn-id: r21197
2006-03-10 08:19:53 +00:00
writeOPL(0x20 + regOffset, *dataptr++);
writeOPL(0x23 + regOffset, *dataptr++);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 temp = *dataptr++;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Feedback / Algorithm
Since updateAndOutput2() is the only function that sets up wave forms and stuff, it pretty much has to be our "set instrument" function. Also, while I'm not entirely sure of the implications, "unk23" is the Algorithm bit for the Feedback / Algorithm register, so I've renamed it "algorithm". svn-id: r21197
2006-03-10 08:19:53 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
// It is very likely that _curChannel really does refer to the same
Since updateAndOutput2() is the only function that sets up wave forms and stuff, it pretty much has to be our "set instrument" function. Also, while I'm not entirely sure of the implications, "unk23" is the Algorithm bit for the Feedback / Algorithm register, so I've renamed it "algorithm". svn-id: r21197
2006-03-10 08:19:53 +00:00
// channel as regOffset, but there's only one Cx register per channel.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xC0 + _curChannel, temp);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Since updateAndOutput2() is the only function that sets up wave forms and stuff, it pretty much has to be our "set instrument" function. Also, while I'm not entirely sure of the implications, "unk23" is the Algorithm bit for the Feedback / Algorithm register, so I've renamed it "algorithm". svn-id: r21197
2006-03-10 08:19:53 +00:00
// The algorithm bit. I don't pretend to understand this fully, but
// "If set to 0, operator 1 modulates operator 2. In this case,
// operator 2 is the only one producing sound. If set to 1, both
// operators produce sound directly. Complex sounds are more easily
// created if the algorithm is set to 0."
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.twoChan = temp & 1;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Waveform Select
Since updateAndOutput2() is the only function that sets up wave forms and stuff, it pretty much has to be our "set instrument" function. Also, while I'm not entirely sure of the implications, "unk23" is the Algorithm bit for the Feedback / Algorithm register, so I've renamed it "algorithm". svn-id: r21197
2006-03-10 08:19:53 +00:00
writeOPL(0xE0 + regOffset, *dataptr++);
writeOPL(0xE3 + regOffset, *dataptr++);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.opLevel1 = *dataptr++;
channel.opLevel2 = *dataptr++;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Level Key Scaling / Total Level
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0x40 + regOffset, calculateOpLevel1(channel));
writeOPL(0x43 + regOffset, calculateOpLevel2(channel));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Attack Rate / Decay Rate
Since updateAndOutput2() is the only function that sets up wave forms and stuff, it pretty much has to be our "set instrument" function. Also, while I'm not entirely sure of the implications, "unk23" is the Algorithm bit for the Feedback / Algorithm register, so I've renamed it "algorithm". svn-id: r21197
2006-03-10 08:19:53 +00:00
writeOPL(0x60 + regOffset, *dataptr++);
writeOPL(0x63 + regOffset, *dataptr++);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Sustain Level / Release Rate
Since updateAndOutput2() is the only function that sets up wave forms and stuff, it pretty much has to be our "set instrument" function. Also, while I'm not entirely sure of the implications, "unk23" is the Algorithm bit for the Feedback / Algorithm register, so I've renamed it "algorithm". svn-id: r21197
2006-03-10 08:19:53 +00:00
writeOPL(0x80 + regOffset, *dataptr++);
writeOPL(0x83 + regOffset, *dataptr++);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
// Apart from playing the note, this function also updates the variables for
// primary effect 2.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::noteOn(Channel &channel) {
debugC(9, kDebugLevelSound, "noteOn(%d)", &channel - _channels);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Added some guesswork comments about stateCallback1_2() and related functions and variables. svn-id: r21200
2006-03-10 10:18:05 +00:00
// The "note on" bit is set, and the current note is played.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.regBx |= 0x20;
writeOPL(0xB0 + _curChannel, channel.regBx);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int8 shift = 9 - channel.unk33;
uint16 temp = channel.regAx | (channel.regBx << 8);
channel.unk37 = ((temp & 0x3FF) >> shift) & 0xFF;
channel.unk38 = channel.unk36;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::adjustVolume(Channel &channel) {
debugC(9, kDebugLevelSound, "adjustVolume(%d)", &channel - _channels);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Level Key Scaling / Total Level
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0x43 + _regOffset[_curChannel], calculateOpLevel2(channel));
if (channel.twoChan)
writeOPL(0x40 + _regOffset[_curChannel], calculateOpLevel1(channel));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// This is presumably only used for some sound effects, e.g. Malcolm blowing up
// the trees in the intro (but not the effect where he "booby-traps" the big
// tree) and turning Kallak to stone. Related functions and variables:
//
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
// update_setupPrimaryEffect1()
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// - Initialises unk29, unk30 and unk31
// - unk29 is not further modified
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
// - unk30 is not further modified, except by update_removePrimaryEffect1()
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
//
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
// update_removePrimaryEffect1()
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// - Deinitialises unk30
//
// unk29 - determines how often the notes are played
// unk30 - modifies the frequency
// unk31 - determines how often the notes are played
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::primaryEffect1(Channel &channel) {
debugC(9, kDebugLevelSound, "Calling primaryEffect1 (channel: %d)", _curChannel);
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 temp = channel.unk31;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.unk31 += channel.unk29;
if (channel.unk31 >= temp)
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return;
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// Initialise unk1 to the current frequency
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint16 unk1 = ((channel.regBx & 3) << 8) | channel.regAx;
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// This is presumably to shift the "note on" bit so far to the left
// that it won't be affected by any of the calculations below.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint16 unk2 = ((channel.regBx & 0x20) << 8) | (channel.regBx & 0x1C);
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int16 unk3 = (int16)channel.unk30;
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
if (unk3 >= 0) {
unk1 += unk3;
if (unk1 >= 734) {
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// The new frequency is too high. Shift it down and go
// up one octave.
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
unk1 >>= 1;
if (!(unk1 & 0x3FF))
++unk1;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
unk2 = (unk2 & 0xFF00) | ((unk2 + 4) & 0xFF);
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
unk2 &= 0xFF1C;
}
} else {
unk1 += unk3;
if (unk1 < 388) {
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// The new frequency is too low. Shift it up and go
// down one octave.
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
unk1 <<= 1;
if (!(unk1 & 0x3FF))
--unk1;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
unk2 = (unk2 & 0xFF00) | ((unk2 - 4) & 0xFF);
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
unk2 &= 0xFF1C;
}
}
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// Make sure that the new frequency is still a 10-bit value.
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
unk1 &= 0x3FF;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xA0 + _curChannel, unk1 & 0xFF);
channel.regAx = unk1 & 0xFF;
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
Added guesswork comments for stateCallback1_1() and its related functions and variables. svn-id: r21202
2006-03-10 10:55:36 +00:00
// Shift down the "note on" bit again.
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
uint8 value = unk1 >> 8;
value |= (unk2 >> 8) & 0xFF;
value |= unk2 & 0xFF;
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xB0 + _curChannel, value);
channel.regBx = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Fixed some obvious errors in the comments I added before. svn-id: r21201
2006-03-10 10:30:31 +00:00
// This is presumably only used for some sound effects, e.g. Malcolm entering
// and leaving Kallak's hut. Related functions and variables:
Added some guesswork comments about stateCallback1_2() and related functions and variables. svn-id: r21200
2006-03-10 10:18:05 +00:00
//
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
// update_setupPrimaryEffect2()
Added some guesswork comments about stateCallback1_2() and related functions and variables. svn-id: r21200
2006-03-10 10:18:05 +00:00
// - Initialises unk32, unk33, unk34, unk35 and unk36
// - unk32 is not further modified
// - unk33 is not further modified
// - unk34 is a countdown that gets reinitialised to unk35 on zero
// - unk35 is based on unk34 and not further modified
// - unk36 is not further modified
//
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
// noteOn()
Added some guesswork comments about stateCallback1_2() and related functions and variables. svn-id: r21200
2006-03-10 10:18:05 +00:00
// - Plays the current note
// - Updates unk37 with a new (lower?) frequency
// - Copies unk36 to unk38. The unk38 variable is a countdown.
//
// unk32 - determines how often the notes are played
// unk33 - modifies the frequency
// unk34 - countdown, updates frequency on zero
// unk35 - initialiser for unk34 countdown
// unk36 - initialiser for unk38 countdown
// unk37 - frequency
Fixed some obvious errors in the comments I added before. svn-id: r21201
2006-03-10 10:30:31 +00:00
// unk38 - countdown, begins playing on zero
Added some guesswork comments about stateCallback1_2() and related functions and variables. svn-id: r21200
2006-03-10 10:18:05 +00:00
// unk41 - determines how often the notes are played
//
// Note that unk41 is never initialised. Not that it should matter much, but it
// is a bit sloppy.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::primaryEffect2(Channel &channel) {
debugC(9, kDebugLevelSound, "Calling primaryEffect2 (channel: %d)", _curChannel);
if (channel.unk38) {
--channel.unk38;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return;
}
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 temp = channel.unk41;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.unk41 += channel.unk32;
if (channel.unk41 < temp) {
uint16 unk1 = channel.unk37;
if (!(--channel.unk34)) {
- Changes usage of regBx to regAx in primaryEffect2 (seems to be a bug introduced when renaming). - Changes unk41 and unk32 to signed. - Fixes wrong sound when dropping items (that was also a little bug in primaryEffect2) svn-id: r21240
2006-03-12 17:58:04 +00:00
unk1 ^= 0xFFFF;
++unk1;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.unk37 = unk1;
channel.unk34 = channel.unk35;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint16 unk2 = (channel.regAx | (channel.regBx << 8)) & 0x3FF;
- Changes usage of regBx to regAx in primaryEffect2 (seems to be a bug introduced when renaming). - Changes unk41 and unk32 to signed. - Fixes wrong sound when dropping items (that was also a little bug in primaryEffect2) svn-id: r21240
2006-03-12 17:58:04 +00:00
unk2 += unk1;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.regAx = unk2 & 0xFF;
channel.regBx = (channel.regBx & 0xFC) | (unk2 >> 8);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Octave / F-Number / Key-On
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xA0 + _curChannel, channel.regAx);
writeOPL(0xB0 + _curChannel, channel.regBx);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
}
Added guesswork comments for stateCallback2_1() and its related functions and variables, and cleaned it up slightly. (Some of its variables are now signed.) svn-id: r21203
2006-03-10 11:18:40 +00:00
// I don't know where this is used. The same operation is performed several
// times on the current channel, using a chunk of the _soundData[] buffer for
// parameters. The parameters are used starting at the end of the chunk.
Fleshed out some guesswork comments. svn-id: r21205
2006-03-10 13:29:34 +00:00
//
// Since we use _curRegOffset to specify the final register, it's quite
// unlikely that this function is ever used to play notes. It's probably only
// used to modify the sound. Another thing that supports this idea is that it
// can be combined with any of the effects callbacks above.
//
Added guesswork comments for stateCallback2_1() and its related functions and variables, and cleaned it up slightly. (Some of its variables are now signed.) svn-id: r21203
2006-03-10 11:18:40 +00:00
// Related functions and variables:
//
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
// update_setupSecondaryEffect1()
Added guesswork comments for stateCallback2_1() and its related functions and variables, and cleaned it up slightly. (Some of its variables are now signed.) svn-id: r21203
2006-03-10 11:18:40 +00:00
// - Initialies unk18, unk19, unk20, unk21, unk22 and offset
// - unk19 is not further modified
// - unk20 is not further modified
// - unk22 is not further modified
// - offset is not further modified
//
// unk18 - determines how often the operation is performed
// unk19 - determines how often the operation is performed
// unk20 - the start index into the data chunk
// unk21 - the current index into the data chunk
// unk22 - the operation to perform
// offset - the offset to the data chunk
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
void AdlibDriver::secondaryEffect1(Channel &channel) {
debugC(9, kDebugLevelSound, "Calling secondaryEffect1 (channel: %d)", _curChannel);
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 temp = channel.unk18;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.unk18 += channel.unk19;
if (channel.unk18 < temp) {
if (--channel.unk21 < 0) {
channel.unk21 = channel.unk20;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(channel.unk22 + _curRegOffset, _soundData[channel.offset + channel.unk21]);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint8 AdlibDriver::calculateOpLevel1(Channel &channel) {
int8 value = channel.opLevel1 & 0x3F;
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
if (channel.twoChan) {
value += channel.opExtraLevel1;
value += channel.opExtraLevel2;
value += channel.opExtraLevel3;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
// Don't allow the total level to overflow into the scaling level bits.
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
if (value > 0x3F) {
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
value = 0x3F;
} else if (value < 0)
value = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
// Preserve the scaling level bits from opLevel1
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
return value | (channel.opLevel1 & 0xC0);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint8 AdlibDriver::calculateOpLevel2(Channel &channel) {
int8 value = channel.opLevel2 & 0x3F;
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
value += channel.opExtraLevel1;
value += channel.opExtraLevel2;
value += channel.opExtraLevel3;
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
// Don't allow the total level to overflow into the scaling level bits.
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
if (value > 0x3F) {
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
value = 0x3F;
} else if (value < 0)
value = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
// Preserve the scaling level bits from opLevel2
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
return value | (channel.opLevel2 & 0xC0);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
// parser opcodes
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setRepeat(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.repeatCounter = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_checkRepeat(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
++dataptr;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
if (--channel.repeatCounter) {
Use signed integers for jump offsets, so that the jump opcodes can jump both forward and backward. This seems to fix the looping of the Black forest music (and probably others as well), and might fix the invalid pointer crashes and wrong music that would sometimes play after a song had finished. Changed the names of the jump opcode functions, and added "update_" prefixes to the previously cleaned-up opcode functions. svn-id: r20986
2006-03-01 08:06:54 +00:00
int16 add = READ_LE_UINT16(dataptr - 2);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
dataptr += add;
}
return 0;
}
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
int AdlibDriver::update_setupProgram(uint8 *&dataptr, Channel &channel, uint8 value) {
- Renames some functions: updateCallback31 -> update_setChannelTempo updateCallback42 -> update_resetToGlobalTempo updateCallback55 -> update_setTempoReset - Renames some vars: unk1 -> tempo unk6 -> tempoReset - Fixes tempo bugs svn-id: r21244
2006-03-12 21:54:06 +00:00
if (value == 0xFF)
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
uint16 add = value << 1;
uint8 *ptr = _soundData + READ_LE_UINT16(_soundData + add);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint8 chan = *ptr++;
Further guesswork: 'unk2' appears to be the priority of a sound, since the callbackOutput() and updateCallback3() functions won't change the channel data pointer unless the new sound has a higher unk2 and the current one. Since it's set to 0 when the data pointer is nulled, I've changed the priority to be unsigned. The updateCallback22() function is now update_setPriority(). svn-id: r21206
2006-03-10 14:22:27 +00:00
uint8 priority = *ptr++;
Fixed some comments and made a couple of cleanups to make it a bit more clear just how similar callbackOutput() and updateCallback3() are. svn-id: r21333
2006-03-16 17:27:52 +00:00
Channel &channel2 = _channels[chan];
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
if (priority >= channel2.priority) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_flagTrigger = 1;
_flags |= 8;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
initChannel(channel2);
channel2.priority = priority;
channel2.dataptr = ptr;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
channel2.tempo = 0xFF;
channel2.position = 0xFF;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel2.duration = 1;
unkOutput2(chan);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
return 0;
}
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int AdlibDriver::update_setNoteSpacing(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.spacing1 = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_jump(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
Use signed integers for jump offsets, so that the jump opcodes can jump both forward and backward. This seems to fix the looping of the Black forest music (and probably others as well), and might fix the invalid pointer crashes and wrong music that would sometimes play after a song had finished. Changed the names of the jump opcode functions, and added "update_" prefixes to the previously cleaned-up opcode functions. svn-id: r20986
2006-03-01 08:06:54 +00:00
int16 add = READ_LE_UINT16(dataptr); dataptr += 2;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
dataptr += add;
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_jumpToSubroutine(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
Use signed integers for jump offsets, so that the jump opcodes can jump both forward and backward. This seems to fix the looping of the Black forest music (and probably others as well), and might fix the invalid pointer crashes and wrong music that would sometimes play after a song had finished. Changed the names of the jump opcode functions, and added "update_" prefixes to the previously cleaned-up opcode functions. svn-id: r20986
2006-03-01 08:06:54 +00:00
int16 add = READ_LE_UINT16(dataptr); dataptr += 2;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.dataptrStack[channel.dataptrStackPos++] = dataptr;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
dataptr += add;
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_returnFromSubroutine(uint8 *&dataptr, Channel &channel, uint8 value) {
dataptr = channel.dataptrStack[--channel.dataptrStackPos];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setBaseOctave(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.baseOctave = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int AdlibDriver::update_stopChannel(uint8 *&dataptr, Channel &channel, uint8 value) {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.priority = 0;
if (_curChannel != 9) {
noteOff(channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
dataptr = 0;
return 2;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_playRest(uint8 *&dataptr, Channel &channel, uint8 value) {
setupDuration(value, channel);
noteOff(channel);
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
return (value != 0);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_writeAdlib(uint8 *&dataptr, Channel &channel, uint8 value) {
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(value, *dataptr++);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback12(uint8 *&dataptr, Channel &channel, uint8 value) {
setupNote(value, channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
value = *dataptr++;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
setupDuration(value, channel);
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
return (value != 0);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setBaseNote(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.baseNote = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setupSecondaryEffect1(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.unk18 = value;
channel.unk19 = value;
channel.unk20 = channel.unk21 = *dataptr++;
channel.unk22 = *dataptr++;
channel.offset = READ_LE_UINT16(dataptr); dataptr += 2;
channel.secondaryEffect = &AdlibDriver::secondaryEffect1;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int AdlibDriver::update_stopOtherChannel(uint8 *&dataptr, Channel &channel, uint8 value) {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
Channel &channel2 = _channels[value];
channel2.duration = 0;
channel2.priority = 0;
channel2.dataptr = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback16(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 *ptr = _soundData;
ptr += READ_LE_UINT16(&_soundData[value << 1]);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
Channel &channel2 = _channels[*ptr];
if (!channel2.dataptr) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
dataptr -= 2;
return 2;
}
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int AdlibDriver::update_setupInstrument(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 *ptr = _soundData;
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
ptr += READ_LE_UINT16(_soundData + (value << 1) + 0x1F4);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
setupInstrument(_curRegOffset, ptr, channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setupPrimaryEffect1(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.unk29 = value;
channel.unk30 = READ_BE_UINT16(dataptr);
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
dataptr += 2;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.primaryEffect = &AdlibDriver::primaryEffect1;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
channel.unk31 = 0xFF;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_removePrimaryEffect1(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.primaryEffect = 0;
channel.unk30 = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setBaseFreq(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.baseFreq = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setupPrimaryEffect2(uint8 *&dataptr, Channel &channel, uint8 value) {
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
channel.unk32 = value;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.unk33 = *dataptr++;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 temp = *dataptr++;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.unk34 = temp + 1;
channel.unk35 = temp << 1;
channel.unk36 = *dataptr++;
channel.primaryEffect = &AdlibDriver::primaryEffect2;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setPriority(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.priority = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback23(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
value >>= 1;
_unkValue1 = _unkValue2 = value;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
_unkValue3 = 0xFF;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_unkValue4 = _unkValue5 = 0;
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback24(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
if (_unkValue5) {
if (_unkValue4 & value) {
_unkValue5 = 0;
return 0;
}
}
if (!(value & _unkValue4)) {
++_unkValue5;
}
dataptr -= 2;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.duration = 1;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 2;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setExtraLevel1(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.opExtraLevel1 = value;
adjustVolume(channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback26(uint8 *&dataptr, Channel &channel, uint8 value) {
setupDuration(value, channel);
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
return (value != 0);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_playNote(uint8 *&dataptr, Channel &channel, uint8 value) {
setupDuration(value, channel);
noteOn(channel);
I was going to hold off on this until the "Pool of Sorrow" music had been fixed but it seems that it drifts out of sync after a few repeats even with the original interpreter. It's a feature! This cleanup clears up some of the more confusing break/continues in the callbackProcess() function, and gets rid of _continueFlag. svn-id: r21259
2006-03-13 11:57:09 +00:00
return (value != 0);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int AdlibDriver::update_setFractionalNoteSpacing(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.fractionalSpacing = value & 7;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setTempo(uint8 *&dataptr, Channel &channel, uint8 value) {
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
_tempo = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_removeSecondaryEffect1(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.secondaryEffect = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setChannelTempo(uint8 *&dataptr, Channel &channel, uint8 value) {
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
channel.tempo = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setExtraLevel3(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.opExtraLevel3 = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setExtraLevel2(uint8 *&dataptr, Channel &channel, uint8 value) {
int channelBackUp = _curChannel;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = value;
Channel &channel2 = _channels[value];
channel2.opExtraLevel2 = *dataptr++;
adjustVolume(channel2);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = channelBackUp;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_changeExtraLevel2(uint8 *&dataptr, Channel &channel, uint8 value) {
int channelBackUp = _curChannel;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = value;
Channel &channel2 = _channels[value];
channel2.opExtraLevel2 += *dataptr++;
adjustVolume(channel2);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = channelBackUp;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setAMDepth(uint8 *&dataptr, Channel &channel, uint8 value) {
Cleaned up updateCallback35() and renamed it setAMDepth(). It looks like the _unkOutputByte2 variable is there to keep track of the current value of the BD register (AM Depth / Vibrato Depth / Rhythm) svn-id: r20984
2006-03-01 07:21:51 +00:00
if (value & 1)
_unkOutputByte2 |= 0x80;
else
_unkOutputByte2 &= 0x7F;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Cleaned up updateCallback35() and renamed it setAMDepth(). It looks like the _unkOutputByte2 variable is there to keep track of the current value of the BD register (AM Depth / Vibrato Depth / Rhythm) svn-id: r20984
2006-03-01 07:21:51 +00:00
// The AM Depth bit is set or cleared, the others remain unchanged
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0xBD, _unkOutputByte2);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setVibratoDepth(uint8 *&dataptr, Channel &channel, uint8 value) {
Cleaned up updateCallback36() and renamed it setVibratoDepth(). svn-id: r20985
2006-03-01 07:27:35 +00:00
if (value & 1)
_unkOutputByte2 |= 0x40;
else
_unkOutputByte2 &= 0xBF;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Cleaned up updateCallback36() and renamed it setVibratoDepth(). svn-id: r20985
2006-03-01 07:27:35 +00:00
// The Vibrato Depth bit is set or cleared, the others remain unchanged
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0xBD, _unkOutputByte2);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_changeExtraLevel1(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.opExtraLevel1 += value;
adjustVolume(channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback38(uint8 *&dataptr, Channel &channel, uint8 value) {
int channelBackUp = _curChannel;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = value;
Channel &channel2 = _channels[value];
channel2.duration = channel2.priority = 0;
channel2.dataptr = 0;
channel2.opExtraLevel2 = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
if (value != 9) {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint8 outValue = _regOffset[value];
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Feedback strength / Connection type
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xC0 + _curChannel, 0x00);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Key scaling level / Operator output level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x43 + outValue, 0x3F);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Sustain Level / Release Rate
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x83 + outValue, 0xFF);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Key On / Octave / Frequency
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xB0 + _curChannel, 0x00);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = channelBackUp;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback39(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint16 unk = *dataptr++;
unk |= value << 8;
The updateUnk6Value() appears to be some sort of random number generator, so I have renamed it getRandomNr(). The _unk6 variable has been renamed _rnd. (Our MidiDriver_ADLIB class also has its own random number generator, so it's not such a far-fetched guess as it may seem.) svn-id: r21102
2006-03-05 20:09:02 +00:00
unk &= getRandomNr();
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
uint16 unk2 = ((channel.regBx & 0x1F) << 8) | channel.regAx;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
unk2 += unk;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
unk2 |= ((channel.regBx & 0x20) << 8);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Frequency
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xA0 + _curChannel, unk2 & 0xFF);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Key On / Octave / Frequency
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
writeOPL(0xB0 + _curChannel, (unk2 & 0xFF00) >> 8);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_removePrimaryEffect2(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.primaryEffect = 0;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback41(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.unk16 = value;
setupNote(channel.rawNote, channel, true);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_resetToGlobalTempo(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
channel.tempo = _tempo;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
int AdlibDriver::update_nop1(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback44(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.unk11 = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback45(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
if (value & 0x80) {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
value += channel.tempo;
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
if (value >= channel.tempo)
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
value = 1;
} else {
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
uint8 temp = value;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
value += channel.tempo;
- Renames some functions: updateCallback31 -> update_setChannelTempo updateCallback42 -> update_resetToGlobalTempo updateCallback55 -> update_setTempoReset - Renames some vars: unk1 -> tempo unk6 -> tempoReset - Fixes tempo bugs svn-id: r21244
2006-03-12 21:54:06 +00:00
if (value < temp)
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
value = 0xFF;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
- Fixes a bug in update_primaryEffect (wrong sound from the birds outside brandons home is fixed with this) - Changes _tempo, tempo, position, unk41, unk32, unk18, unk19, _unkValue3 to unsigned (should be correct this time) svn-id: r21293
2006-03-14 15:15:04 +00:00
channel.tempo = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback46(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 entry = *dataptr++;
_tablePtr1 = _unkTable2[entry++];
_tablePtr2 = _unkTable2[entry];
if (value == 2) {
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Frequency
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0xA0, _tablePtr2[0]);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
return 0;
}
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
// TODO: This is really the same as update_nop1(), so they should be combined
// into one single update_nop().
int AdlibDriver::update_nop2(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
return 0;
}
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
int AdlibDriver::update_setupRhythmSection(uint8 *&dataptr, Channel &channel, uint8 value) {
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int channelBackUp = _curChannel;
Renamed _unkOutputByte1 _curRegOffset. It doesn't always correspond to the _curTable channel, but it probably does so whenever both of them are used together. svn-id: r21204
2006-03-10 11:26:32 +00:00
int regOffsetBackUp = _curRegOffset;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Corrects implementation of updateCallback48. svn-id: r21105
2006-03-05 21:48:55 +00:00
uint8 entry = value << 1;
uint8 *ptr = _soundData + READ_LE_UINT16(_soundData + entry + 0x1F4);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = 6;
_curRegOffset = _regOffset[6];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_unkValue6 = *(ptr + 6);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
setupInstrument(_curRegOffset, ptr, channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Corrects implementation of updateCallback48. svn-id: r21105
2006-03-05 21:48:55 +00:00
entry = *dataptr++ << 1;
ptr = _soundData + READ_LE_UINT16(_soundData + entry + 0x1F4);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = 7;
_curRegOffset = _regOffset[7];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_unkValue7 = entry = *(ptr + 5);
_unkValue8 = entry = *(ptr + 6);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
setupInstrument(_curRegOffset, ptr, channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Corrects implementation of updateCallback48. svn-id: r21105
2006-03-05 21:48:55 +00:00
entry = *dataptr++ << 1;
ptr = _soundData + READ_LE_UINT16(_soundData + entry + 0x1F4);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = 8;
_curRegOffset = _regOffset[8];
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_unkValue9 = entry = *(ptr + 5);
_unkValue10 = entry = *(ptr + 6);
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
setupInstrument(_curRegOffset, ptr, channel);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Octave / F-Number / Key-On for channels 6, 7 and 8
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_channels[6].regBx = *dataptr++ & 0x2F;
writeOPL(0xB6, _channels[6].regBx);
Corrects implementation of updateCallback48. svn-id: r21105
2006-03-05 21:48:55 +00:00
writeOPL(0xA6, *dataptr++);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_channels[7].regBx = *dataptr++ & 0x2F;
writeOPL(0xB7, _channels[7].regBx);
Corrects implementation of updateCallback48. svn-id: r21105
2006-03-05 21:48:55 +00:00
writeOPL(0xA7, *dataptr++);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_channels[8].regBx = *dataptr++ & 0x2F;
writeOPL(0xB8, _channels[8].regBx);
Corrects implementation of updateCallback48. svn-id: r21105
2006-03-05 21:48:55 +00:00
writeOPL(0xA8, *dataptr++);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
_rhythmSection = 0x20;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Renamed _unkOutputByte1 _curRegOffset. It doesn't always correspond to the _curTable channel, but it probably does so whenever both of them are used together. svn-id: r21204
2006-03-10 11:26:32 +00:00
_curRegOffset = regOffsetBackUp;
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
_curChannel = channelBackUp;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
int AdlibDriver::update_playRhythmSection(uint8 *&dataptr, Channel &channel, uint8 value) {
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Amplitude Modulation Depth / Vibrato Depth / Rhythm
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
writeOPL(0xBD, (((value & 0x1F) ^ 0xFF) & _rhythmSection) | 0x20);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
value |= _rhythmSection;
_rhythmSection = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
value |= _unkOutputByte2;
value |= 0x20;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Cleanup. svn-id: r21304
2006-03-14 23:39:42 +00:00
// The original only wrote to the data port here, but that shouldn't
// make any real difference.
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
Cleanup. svn-id: r21304
2006-03-14 23:39:42 +00:00
writeOPL(0xBD, value);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
int AdlibDriver::update_removeRhythmSection(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
--dataptr;
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
_rhythmSection = 0;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Amplitude Modulation Depth / Vibrato Depth / Rhythm
Corrects former commit which deleted sound file loading after the scene with malcolm and kallak. Also corrects callback50() which seem to improve the sound a bit in the kallak writing scene. svn-id: r21238
2006-03-12 17:08:46 +00:00
writeOPL(0xBD, _unkOutputByte2 & 0xC0);
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback51(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint16 temp = (value << 8) | *dataptr++;
if (value & 1) {
uint8 val = temp & 0xFF;
_unkValue12 = val;
val += _unkValue7;
val += _unkValue11;
val += _unkValue12;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 2: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x51, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 2) {
uint8 val = temp & 0xFF;
_unkValue14 = val;
val += _unkValue10;
val += _unkValue13;
val += _unkValue14;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 3: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x55, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 4) {
uint8 val = temp & 0xFF;
_unkValue15 = val;
val += _unkValue9;
val += _unkValue16;
val += _unkValue15;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 3: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x52, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 8) {
uint8 val = temp & 0xFF;
_unkValue18 = val;
val += _unkValue8;
val += _unkValue17;
val += _unkValue18;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 2: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x54, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 16) {
uint8 val = temp & 0xFF;
_unkValue20 = val;
val += _unkValue6;
val += _unkValue19;
val += _unkValue20;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 1: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x53, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback52(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint16 temp = (value << 8) | *dataptr++;
if (value & 1) {
uint8 val = temp & 0xFF;
val += _unkValue7;
val += _unkValue11;
val += _unkValue12;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 2: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x51, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 2) {
uint8 val = temp & 0xFF;
val += _unkValue10;
val += _unkValue13;
val += _unkValue14;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 3: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x55, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 4) {
uint8 val = temp & 0xFF;
val += _unkValue9;
val += _unkValue16;
val += _unkValue15;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 3: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x52, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 8) {
uint8 val = temp & 0xFF;
val += _unkValue8;
val += _unkValue17;
val += _unkValue18;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 2: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x54, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 16) {
uint8 val = temp & 0xFF;
val += _unkValue6;
val += _unkValue19;
val += _unkValue20;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 1: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x53, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback53(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint16 temp = (value << 8) | *dataptr++;
if (value & 1) {
uint8 val = temp & 0xFF;
_unkValue11 = val;
val += _unkValue7;
val += _unkValue12;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 2: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x51, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 2) {
uint8 val = temp & 0xFF;
_unkValue13 = val;
val += _unkValue10;
val += _unkValue14;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 3: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x55, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 4) {
uint8 val = temp & 0xFF;
_unkValue16 = val;
val += _unkValue9;
val += _unkValue15;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 3: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x52, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 8) {
uint8 val = temp & 0xFF;
_unkValue17 = val;
val += _unkValue8;
val += _unkValue18;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 2: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x54, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (value & 16) {
uint8 val = temp & 0xFF;
_unkValue19 = val;
val += _unkValue6;
val += _unkValue20;
Added some basic comments, along with some very, very minor cleanup. svn-id: r20972
2006-02-28 15:44:39 +00:00
// Channel 1: Level Key Scaling / Total Level
Cleanup: output0x388(0xABCD) is now writeOPL(0xAB, 0xCD), which is slightly more easy to read. svn-id: r21081
2006-03-04 20:27:51 +00:00
writeOPL(0x53, checkValue(val));
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback54(uint8 *&dataptr, Channel &channel, uint8 value) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_unk5 = value;
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::update_setTempoReset(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.tempoReset = value;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
int AdlibDriver::updateCallback56(uint8 *&dataptr, Channel &channel, uint8 value) {
channel.unk39 = value;
channel.unk40 = *dataptr++;
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
return 0;
}
// static res
#define COMMAND(x) { &AdlibDriver::x, #x }
const AdlibDriver::OpcodeEntry AdlibDriver::_opcodeList[] = {
COMMAND(snd_ret0x100),
COMMAND(snd_ret0x1983),
COMMAND(snd_initDriver),
COMMAND(snd_deinitDriver),
COMMAND(snd_setSoundData),
COMMAND(snd_unkOpcode1),
COMMAND(snd_startSong),
COMMAND(snd_unkOpcode2),
COMMAND(snd_unkOpcode3),
COMMAND(snd_readByte),
COMMAND(snd_writeByte),
COMMAND(snd_setUnk5),
COMMAND(snd_unkOpcode4),
COMMAND(snd_dummy),
COMMAND(snd_getNullvar4),
COMMAND(snd_setNullvar3),
COMMAND(snd_setFlag),
COMMAND(snd_clearFlag)
};
const AdlibDriver::ParserOpcode AdlibDriver::_parserOpcodeTable[] = {
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 0
Use signed integers for jump offsets, so that the jump opcodes can jump both forward and backward. This seems to fix the looping of the Black forest music (and probably others as well), and might fix the invalid pointer crashes and wrong music that would sometimes play after a song had finished. Changed the names of the jump opcode functions, and added "update_" prefixes to the previously cleaned-up opcode functions. svn-id: r20986
2006-03-01 08:06:54 +00:00
COMMAND(update_setRepeat),
COMMAND(update_checkRepeat),
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
COMMAND(update_setupProgram),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_setNoteSpacing),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 4
Use signed integers for jump offsets, so that the jump opcodes can jump both forward and backward. This seems to fix the looping of the Black forest music (and probably others as well), and might fix the invalid pointer crashes and wrong music that would sometimes play after a song had finished. Changed the names of the jump opcode functions, and added "update_" prefixes to the previously cleaned-up opcode functions. svn-id: r20986
2006-03-01 08:06:54 +00:00
COMMAND(update_jump),
COMMAND(update_jumpToSubroutine),
COMMAND(update_returnFromSubroutine),
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
COMMAND(update_setBaseOctave),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 8
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
COMMAND(update_playRest),
Fixed updateCallback11() according to LordHoto's directions, and renamed it update_writeAdlib(). (The "update_" prefix will probably be changed later, but for now I want it to remain obvious that the functions belong to the other updateCallback*() functions.) svn-id: r20990
2006-03-01 13:57:32 +00:00
COMMAND(update_writeAdlib),
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback12),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 12
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
COMMAND(update_setBaseNote),
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
COMMAND(update_setupSecondaryEffect1),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopOtherChannel),
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback16),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 16
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_setupInstrument),
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
COMMAND(update_setupPrimaryEffect1),
COMMAND(update_removePrimaryEffect1),
More cleanups and renamings: * Rewrote updateCallback41() in terms of updateAndOutput1(), since they were almost identical to begin with. * Rewrote updateAndOutput1() for better readability. * Made unk10 signed, and renamed it baseOctave. * Made unk14 signed, and renamed it baseNote. * Made unk15 signed, and renamed it baseFreq. * Renamed unk17 regAx. It appears to be a cache for that register. * Renamed unkOutputValue1 regBx. It appears to be a cache for that register. * Renamed updateCallback8() update_setBaseOctave(). * Renamed updateCallback13() update_setBaseNote(). * Renamed updateCallback20() update_setBaseFreq(). Which still leaves the mystery of unk16 and _unkTables[]. Perhaps *this* is the pitch bend? svn-id: r21183
2006-03-09 17:04:00 +00:00
COMMAND(update_setBaseFreq),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 20
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
COMMAND(update_setupPrimaryEffect2),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
COMMAND(update_stopChannel),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 24
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
COMMAND(update_stopChannel),
Further guesswork: 'unk2' appears to be the priority of a sound, since the callbackOutput() and updateCallback3() functions won't change the channel data pointer unless the new sound has a higher unk2 and the current one. Since it's set to 0 when the data pointer is nulled, I've changed the priority to be unsigned. The updateCallback22() function is now update_setPriority(). svn-id: r21206
2006-03-10 14:22:27 +00:00
COMMAND(update_setPriority),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 28
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback23),
COMMAND(updateCallback24),
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
COMMAND(update_setExtraLevel1),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 32
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback26),
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
COMMAND(update_playNote),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
COMMAND(update_stopChannel),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 36
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_setFractionalNoteSpacing),
COMMAND(update_stopChannel),
Renamed setTempo to update_setTempo, also fixed tempo handling (intro is too fast now though, but seems to be another problem). Also changed some unsigned values to signed. svn-id: r21111
2006-03-06 16:37:03 +00:00
COMMAND(update_setTempo),
And now, for a change, some more renamings: update1() -> setupDuration() updateAndOutput1() -> setupNote() setInstrument() -> setupInstrument() (just for the symmetry) updateAndOutput3() -> noteOn() That made the following two renamings seem natural: updateCallback10() -> update_playRest() updateCallback27() -> update_playNote() I decided to number the secondary effects, even though there's only one: update_setupSecondaryEffect() -> update_setupSecondaryEffect1() update_removeSecondaryEffect() -> update_removeSecondaryEffect1() And finally, I renamed the effects callbacks: stateCallback1_1() -> primaryEffect1() stateCallback1_2() -> primaryEffect2() stateCallback2_1() -> secondaryEffect1() svn-id: r21232
2006-03-12 12:40:14 +00:00
COMMAND(update_removeSecondaryEffect1),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 40
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
- Renames some functions: updateCallback31 -> update_setChannelTempo updateCallback42 -> update_resetToGlobalTempo updateCallback55 -> update_setTempoReset - Renames some vars: unk1 -> tempo unk6 -> tempoReset - Fixes tempo bugs svn-id: r21244
2006-03-12 21:54:06 +00:00
COMMAND(update_setChannelTempo),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
COMMAND(update_setExtraLevel3),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 44
The calculateLowByte1() and 2() functions are used for calculating the "total level", which I believe is the individual channel volume. (The functions also return the "scaling level" bits, but they are preserved, not calculated.) I have renamed these functions calculateOpLevel1() and 2(). The unk25 and unk26 variables have been renamed opLevel1 and opLevel2. These are called oplvl_1 and oplvl_2 in our MidiDriver_ADLIB class. The unk26, unk27 and unk28 variables are potentially added to both operator levels, and have been renamed opExtraLevel1, opExtralevel2 and opExtraLevel3. The updateCallback25() function has been remamed update_setExtraLevel1(). The updateCallback33() function has been renamed update_setExtraLevel2(). The updateCallback32() function has been renamed update_setExtraLevel3(). Note that these callbacks aren't quite as similar as their names would seem to indicate: They differ in whether or not the volume is updated and/or how the parameters are passed. svn-id: r21199
2006-03-10 09:24:50 +00:00
COMMAND(update_setExtraLevel2),
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
COMMAND(update_changeExtraLevel2),
Use signed integers for jump offsets, so that the jump opcodes can jump both forward and backward. This seems to fix the looping of the Black forest music (and probably others as well), and might fix the invalid pointer crashes and wrong music that would sometimes play after a song had finished. Changed the names of the jump opcode functions, and added "update_" prefixes to the previously cleaned-up opcode functions. svn-id: r20986
2006-03-01 08:06:54 +00:00
COMMAND(update_setAMDepth),
COMMAND(update_setVibratoDepth),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 48
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
COMMAND(update_changeExtraLevel1),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
COMMAND(update_stopChannel),
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback38),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 52
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback39),
Some more renaming: updateCallback18() -> update_setupPrimaryEffect1() updateCallback19() -> update_removePrimaryEffect1() updateCallback21() -> update_setupPrimaryEffect2() updateCallback40() -> update_removePrimaryEffect2() updateCallback14() -> update_setupSecondaryEffect() updateCallback30() -> update_removeSecondaryEffect() updateCallback37() -> update_changeExtraLevel1() updateCallback34() -> update_changeExtraLevel2() svn-id: r21217
2006-03-11 17:25:12 +00:00
COMMAND(update_removePrimaryEffect2),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 56
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback41),
- Renames some functions: updateCallback31 -> update_setChannelTempo updateCallback42 -> update_resetToGlobalTempo updateCallback55 -> update_setTempoReset - Renames some vars: unk1 -> tempo unk6 -> tempoReset - Fixes tempo bugs svn-id: r21244
2006-03-12 21:54:06 +00:00
COMMAND(update_resetToGlobalTempo),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_nop1),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 60
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback44),
COMMAND(updateCallback45),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel),
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback46),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 64
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_nop2),
More renamings: callbackOutput() -> setupPrograms() callbackProcess() -> executePrograms() updateCallback3() -> update_setupProgram() _unk4 -> _rhythmSection updateCallback48() -> update_setupRhythmSection() updateCallback49() -> update_playRhythmSection() updateCallback50() -> update_removeRhythmSection() svn-id: r21336
2006-03-16 18:01:14 +00:00
COMMAND(update_setupRhythmSection),
COMMAND(update_playRhythmSection),
COMMAND(update_removeRhythmSection),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 68
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback51),
COMMAND(updateCallback52),
COMMAND(updateCallback53),
COMMAND(updateCallback54),
Added lock() and unlock() again to readBuffer. Implemented stateCallback1_1. Fix some small bugs here and there (can't say what the fixed but it should be correct now) Also set's the correct timer call value (with the other the music was too fast) svn-id: r20973
2006-02-28 17:02:14 +00:00
// 72
- Renames some functions: updateCallback31 -> update_setChannelTempo updateCallback42 -> update_resetToGlobalTempo updateCallback55 -> update_setTempoReset - Renames some vars: unk1 -> tempo unk6 -> tempoReset - Fixes tempo bugs svn-id: r21244
2006-03-12 21:54:06 +00:00
COMMAND(update_setTempoReset),
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
COMMAND(updateCallback56),
Things continue to fall into place. In other words, more renamings: unk3 -> spacing1 unk7 -> spacing2 unk12 -> fractionalSpacing updateCallback4() -> update_setNoteSpacing() updateCallback9() -> update_stopChannel() updateCallback15() -> update_stopOtherChannel() updateCallback17() -> update_setupInstrument() updateCallback28() -> update_setFractionalNoteSpacing() updateCallback43() -> update_nop1() updateCallback47() -> update_nop2() svn-id: r21261
2006-03-13 13:56:19 +00:00
COMMAND(update_stopChannel)
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
};
#undef COMMAND
const int AdlibDriver::_opcodesEntries = ARRAYSIZE(AdlibDriver::_opcodeList);
const int AdlibDriver::_parserOpcodeTableSize = ARRAYSIZE(AdlibDriver::_parserOpcodeTable);
Set down in comments what I know about _outputTable[], and what I suspect about _unkTable[]. Investigating the latter further might help unravelling several of the "unk" variables. (Right now, I'm suspecting that unk15 is the pitch bend.) svn-id: r21164
2006-03-09 12:36:11 +00:00
// This table holds the register offset for operator 1 for each of the nine
// channels. To get the register offset for operator 2, simply add 3.
Until now, we have confusingly used the terms "state" or "table" where we really mean "channel". Fixed that: OutputState -> Channel _curTable -> _curChannel _outputTables[] -> _channels[] The biggest change was modifying all the "state" and "table" local variables and function parameters. Ahd, at no extra charge: _outputTable[] -> _regOffset[] svn-id: r21247
2006-03-12 23:21:01 +00:00
const uint8 AdlibDriver::_regOffset[] = {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
0x00, 0x01, 0x02, 0x08, 0x09, 0x0A, 0x10, 0x11,
0x12
};
Set down in comments what I know about _outputTable[], and what I suspect about _unkTable[]. Investigating the latter further might help unravelling several of the "unk" variables. (Right now, I'm suspecting that unk15 is the pitch bend.) svn-id: r21164
2006-03-09 12:36:11 +00:00
// Given the size of this table, and the range of its values, it's probably the
// F-Numbers (10 bits) for the notes of the 12-tone scale. However, it does not
// match the table in the Adlib documentation I've seen.
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
const uint16 AdlibDriver::_unkTable[] = {
0x0134, 0x0147, 0x015A, 0x016F, 0x0184, 0x019C, 0x01B4, 0x01CE, 0x01E9,
0x0207, 0x0225, 0x0246
};
More comments, and some very minor cleanup. svn-id: r21208
2006-03-10 15:30:22 +00:00
// These tables are currently only used by updateCallback46(), which only ever
// uses the first element of one of the sub-tables.
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
const uint8 *AdlibDriver::_unkTable2[] = {
AdlibDriver::_unkTable2_1,
AdlibDriver::_unkTable2_2,
AdlibDriver::_unkTable2_1,
AdlibDriver::_unkTable2_2,
AdlibDriver::_unkTable2_3,
AdlibDriver::_unkTable2_2
};
const uint8 AdlibDriver::_unkTable2_1[] = {
0x50, 0x50, 0x4F, 0x4F, 0x4E, 0x4E, 0x4D, 0x4D,
0x4C, 0x4C, 0x4B, 0x4B, 0x4A, 0x4A, 0x49, 0x49,
0x48, 0x48, 0x47, 0x47, 0x46, 0x46, 0x45, 0x45,
0x44, 0x44, 0x43, 0x43, 0x42, 0x42, 0x41, 0x41,
0x40, 0x40, 0x3F, 0x3F, 0x3E, 0x3E, 0x3D, 0x3D,
0x3C, 0x3C, 0x3B, 0x3B, 0x3A, 0x3A, 0x39, 0x39,
0x38, 0x38, 0x37, 0x37, 0x36, 0x36, 0x35, 0x35,
0x34, 0x34, 0x33, 0x33, 0x32, 0x32, 0x31, 0x31,
0x30, 0x30, 0x2F, 0x2F, 0x2E, 0x2E, 0x2D, 0x2D,
0x2C, 0x2C, 0x2B, 0x2B, 0x2A, 0x2A, 0x29, 0x29,
0x28, 0x28, 0x27, 0x27, 0x26, 0x26, 0x25, 0x25,
0x24, 0x24, 0x23, 0x23, 0x22, 0x22, 0x21, 0x21,
0x20, 0x20, 0x1F, 0x1F, 0x1E, 0x1E, 0x1D, 0x1D,
0x1C, 0x1C, 0x1B, 0x1B, 0x1A, 0x1A, 0x19, 0x19,
0x18, 0x18, 0x17, 0x17, 0x16, 0x16, 0x15, 0x15,
0x14, 0x14, 0x13, 0x13, 0x12, 0x12, 0x11, 0x11,
0x10, 0x10
};
// no don't ask me WHY this table exsits!
const uint8 AdlibDriver::_unkTable2_2[] = {
0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x07,
0x08, 0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F,
0x10, 0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17,
0x18, 0x19, 0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x1F,
0x20, 0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x27,
0x28, 0x29, 0x2A, 0x2B, 0x2C, 0x2D, 0x2E, 0x2F,
0x30, 0x31, 0x32, 0x33, 0x34, 0x35, 0x36, 0x37,
0x38, 0x39, 0x3A, 0x3B, 0x3C, 0x3D, 0x3E, 0x3F,
0x40, 0x41, 0x42, 0x43, 0x44, 0x45, 0x46, 0x47,
0x48, 0x49, 0x4A, 0x4B, 0x4C, 0x4D, 0x4E, 0x4F,
0x50, 0x51, 0x52, 0x53, 0x54, 0x55, 0x56, 0x57,
0x58, 0x59, 0x5A, 0x5B, 0x5C, 0x5D, 0x5E, 0x6F,
0x60, 0x61, 0x62, 0x63, 0x64, 0x65, 0x66, 0x67,
0x68, 0x69, 0x6A, 0x6B, 0x6C, 0x6D, 0x6E, 0x6F,
0x70, 0x71, 0x72, 0x73, 0x74, 0x75, 0x76, 0x77,
0x78, 0x79, 0x7A, 0x7B, 0x7C, 0x7D, 0x7E, 0x7F
};
const uint8 AdlibDriver::_unkTable2_3[] = {
0x40, 0x40, 0x40, 0x3F, 0x3F, 0x3F, 0x3E, 0x3E,
0x3E, 0x3D, 0x3D, 0x3D, 0x3C, 0x3C, 0x3C, 0x3B,
0x3B, 0x3B, 0x3A, 0x3A, 0x3A, 0x39, 0x39, 0x39,
0x38, 0x38, 0x38, 0x37, 0x37, 0x37, 0x36, 0x36,
0x36, 0x35, 0x35, 0x35, 0x34, 0x34, 0x34, 0x33,
0x33, 0x33, 0x32, 0x32, 0x32, 0x31, 0x31, 0x31,
0x30, 0x30, 0x30, 0x2F, 0x2F, 0x2F, 0x2E, 0x2E,
0x2E, 0x2D, 0x2D, 0x2D, 0x2C, 0x2C, 0x2C, 0x2B,
0x2B, 0x2B, 0x2A, 0x2A, 0x2A, 0x29, 0x29, 0x29,
0x28, 0x28, 0x28, 0x27, 0x27, 0x27, 0x26, 0x26,
0x26, 0x25, 0x25, 0x25, 0x24, 0x24, 0x24, 0x23,
0x23, 0x23, 0x22, 0x22, 0x22, 0x21, 0x21, 0x21,
0x20, 0x20, 0x20, 0x1F, 0x1F, 0x1F, 0x1E, 0x1E,
0x1E, 0x1D, 0x1D, 0x1D, 0x1C, 0x1C, 0x1C, 0x1B,
0x1B, 0x1B, 0x1A, 0x1A, 0x1A, 0x19, 0x19, 0x19,
0x18, 0x18, 0x18, 0x17, 0x17, 0x17, 0x16, 0x16,
0x16, 0x15
};
Some more/updated comments. svn-id: r21196
2006-03-10 07:56:22 +00:00
// This table is used to modify the frequency of the notes, depending on the
// note value and unk16. In theory, we could very well try to access memory
// outside this table, but in reality that probably won't happen.
//
// This could be some sort of pitch bend, but I have yet to see it used for
// anything so it's hard to say.
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
const uint8 AdlibDriver::_unkTables[][32] = {
Fixes _unkTables, I made a mistake while extracting those last time it seems. svn-id: r21287
2006-03-14 13:37:15 +00:00
// 0
{ 0x00, 0x01, 0x02, 0x03, 0x04, 0x05, 0x06, 0x08,
0x09, 0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, 0x10,
0x11, 0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x19,
0x1A, 0x1B, 0x1C, 0x1D, 0x1E, 0x1F, 0x20, 0x21 },
// 1
{ 0x00, 0x01, 0x02, 0x03, 0x04, 0x06, 0x07, 0x09,
0x0A, 0x0B, 0x0C, 0x0D, 0x0E, 0x0F, 0x10, 0x11,
0x12, 0x13, 0x14, 0x15, 0x16, 0x17, 0x18, 0x1A,
0x1B, 0x1C, 0x1D, 0x1E, 0x1F, 0x20, 0x22, 0x24 },
// 2
{ 0x00, 0x01, 0x02, 0x03, 0x04, 0x06, 0x08, 0x09,
0x0A, 0x0C, 0x0D, 0x0E, 0x0F, 0x11, 0x12, 0x13,
0x14, 0x15, 0x16, 0x17, 0x19, 0x1A, 0x1C, 0x1D,
0x1E, 0x1F, 0x20, 0x21, 0x22, 0x24, 0x25, 0x26 },
// 3
{ 0x00, 0x01, 0x02, 0x03, 0x04, 0x06, 0x08, 0x0A,
0x0B, 0x0C, 0x0D, 0x0E, 0x0F, 0x11, 0x12, 0x13,
0x14, 0x15, 0x16, 0x17, 0x18, 0x1A, 0x1C, 0x1D,
0x1E, 0x1F, 0x20, 0x21, 0x23, 0x25, 0x27, 0x28 },
// 4
{ 0x00, 0x01, 0x02, 0x03, 0x04, 0x06, 0x08, 0x0A,
0x0B, 0x0C, 0x0D, 0x0E, 0x0F, 0x11, 0x13, 0x15,
0x16, 0x17, 0x18, 0x19, 0x1B, 0x1D, 0x1F, 0x20,
0x21, 0x22, 0x23, 0x24, 0x25, 0x26, 0x28, 0x2A },
// 5
{ 0x00, 0x01, 0x02, 0x03, 0x05, 0x07, 0x09, 0x0B,
0x0C, 0x0D, 0x0E, 0x0F, 0x10, 0x11, 0x13, 0x15,
0x16, 0x17, 0x18, 0x19, 0x1B, 0x1D, 0x1F, 0x20,
0x21, 0x22, 0x23, 0x25, 0x27, 0x29, 0x2B, 0x2D },
// 6
{ 0x00, 0x01, 0x02, 0x03, 0x05, 0x07, 0x09, 0x0B,
0x0C, 0x0D, 0x0E, 0x0F, 0x10, 0x11, 0x13, 0x15,
0x16, 0x17, 0x18, 0x1A, 0x1C, 0x1E, 0x21, 0x24,
0x25, 0x26, 0x27, 0x29, 0x2B, 0x2D, 0x2F, 0x30 },
// 7
{ 0x00, 0x01, 0x02, 0x04, 0x06, 0x08, 0x0A, 0x0C,
0x0D, 0x0E, 0x0F, 0x10, 0x11, 0x13, 0x15, 0x18,
0x19, 0x1A, 0x1C, 0x1D, 0x1F, 0x21, 0x23, 0x25,
0x26, 0x27, 0x29, 0x2B, 0x2D, 0x2F, 0x30, 0x32 },
// 8
{ 0x00, 0x01, 0x02, 0x04, 0x06, 0x08, 0x0A, 0x0D,
0x0E, 0x0F, 0x10, 0x11, 0x12, 0x14, 0x17, 0x1A,
0x19, 0x1A, 0x1C, 0x1E, 0x20, 0x22, 0x25, 0x28,
0x29, 0x2A, 0x2B, 0x2D, 0x2F, 0x31, 0x33, 0x35 },
// 9
{ 0x00, 0x01, 0x03, 0x05, 0x07, 0x09, 0x0B, 0x0E,
0x0F, 0x10, 0x12, 0x14, 0x16, 0x18, 0x1A, 0x1B,
0x1C, 0x1D, 0x1E, 0x20, 0x22, 0x24, 0x26, 0x29,
0x2A, 0x2C, 0x2E, 0x30, 0x32, 0x34, 0x36, 0x39 },
// 10
{ 0x00, 0x01, 0x03, 0x05, 0x07, 0x09, 0x0B, 0x0E,
0x0F, 0x10, 0x12, 0x14, 0x16, 0x19, 0x1B, 0x1E,
0x1F, 0x21, 0x23, 0x25, 0x27, 0x29, 0x2B, 0x2D,
0x2E, 0x2F, 0x31, 0x32, 0x34, 0x36, 0x39, 0x3C },
// 11
{ 0x00, 0x01, 0x03, 0x05, 0x07, 0x0A, 0x0C, 0x0F,
0x10, 0x11, 0x13, 0x15, 0x17, 0x19, 0x1B, 0x1E,
0x1F, 0x20, 0x22, 0x24, 0x26, 0x28, 0x2B, 0x2E,
0x2F, 0x30, 0x32, 0x34, 0x36, 0x39, 0x3C, 0x3F },
// 12
{ 0x00, 0x02, 0x04, 0x06, 0x08, 0x0B, 0x0D, 0x10,
0x11, 0x12, 0x14, 0x16, 0x18, 0x1B, 0x1E, 0x21,
0x22, 0x23, 0x25, 0x27, 0x29, 0x2C, 0x2F, 0x32,
0x33, 0x34, 0x36, 0x38, 0x3B, 0x34, 0x41, 0x44 },
// 13
{ 0x00, 0x02, 0x04, 0x06, 0x08, 0x0B, 0x0D, 0x11,
0x12, 0x13, 0x15, 0x17, 0x1A, 0x1D, 0x20, 0x23,
0x24, 0x25, 0x27, 0x29, 0x2C, 0x2F, 0x32, 0x35,
0x36, 0x37, 0x39, 0x3B, 0x3E, 0x41, 0x44, 0x47 }
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
};
#pragma mark -
SoundAdlibPC::SoundAdlibPC(Audio::Mixer *mixer, KyraEngine *engine)
: Sound(engine, mixer), _driver(0), _trackEntries(), _soundDataPtr(0) {
memset(_trackEntries, 0, sizeof(_trackEntries));
_driver = new AdlibDriver(mixer);
assert(_driver);
_sfxPlayingSound = -1;
_soundFileLoaded = "";
}
SoundAdlibPC::~SoundAdlibPC() {
delete [] _soundDataPtr;
delete _driver;
}
bool SoundAdlibPC::init() {
_driver->callback(2);
_driver->callback(16, int(4));
return true;
}
void SoundAdlibPC::setVolume(int volume) {
}
int SoundAdlibPC::getVolume() {
return 0;
}
Music in the last two intro scenes plays now again with adlib. Cleaned up the sound code a bit (removed some unneeded functions). svn-id: r21234
2006-03-12 14:22:27 +00:00
void SoundAdlibPC::loadMusicFile(const char *file) {
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
loadSoundFile(file);
}
Music in the last two intro scenes plays now again with adlib. Cleaned up the sound code a bit (removed some unneeded functions). svn-id: r21234
2006-03-12 14:22:27 +00:00
void SoundAdlibPC::playTrack(uint8 track) {
Oops forgot to enable music again. svn-id: r20993
2006-03-01 18:18:16 +00:00
playSoundEffect(track);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
void SoundAdlibPC::haltTrack() {
- Removes deleted waitTicks prototype in KyraEngine - Adds sound fading support to the adlib player - Adds support for stopping running tracks - Changes baseFreq type back to uint8 - corrects octave handling in updateAndOutput1 (Thanks to eriktorbjorn for doing that) svn-id: r21185
2006-03-09 20:50:19 +00:00
unk1();
unk2();
Music in the last two intro scenes plays now again with adlib. Cleaned up the sound code a bit (removed some unneeded functions). svn-id: r21234
2006-03-12 14:22:27 +00:00
//_engine->_system->delayMillis(3 * 60);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
void SoundAdlibPC::playSoundEffect(uint8 track) {
uint8 soundId = _trackEntries[track];
if ((int8)soundId == -1 || !_soundDataPtr)
return;
soundId &= 0xFF;
while ((_driver->callback(16, 0) & 8)) {
The adlib code was calling the Kyra delay() function instead of the system delay, which was causing concurrency problems. svn-id: r21136
2006-03-08 01:38:27 +00:00
// We call the system delay and not the game delay to avoid concurrency issues.
_engine->_system->delayMillis(10);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
if (_sfxPlayingSound != -1) {
_driver->callback(10, _sfxPlayingSound, int(1), int(_sfxSecondByteOfSong));
_driver->callback(10, _sfxPlayingSound, int(3), int(_sfxFourthByteOfSong));
_sfxPlayingSound = -1;
}
int firstByteOfSong = _driver->callback(9, soundId, int(0));
if (firstByteOfSong != 9) {
_sfxPlayingSound = soundId;
_sfxSecondByteOfSong = _driver->callback(9, soundId, int(1));
_sfxFourthByteOfSong = _driver->callback(9, soundId, int(3));
int newVal = ((((-_sfxFourthByteOfSong) + 63) * 0xFF) >> 8) & 0xFF;
newVal = -newVal + 63;
_driver->callback(10, soundId, int(3), newVal);
newVal = ((_sfxSecondByteOfSong * 0xFF) >> 8) & 0xFF;
_driver->callback(10, soundId, int(1), newVal);
}
_driver->callback(6, soundId);
}
void SoundAdlibPC::beginFadeOut() {
- Removes deleted waitTicks prototype in KyraEngine - Adds sound fading support to the adlib player - Adds support for stopping running tracks - Changes baseFreq type back to uint8 - corrects octave handling in updateAndOutput1 (Thanks to eriktorbjorn for doing that) svn-id: r21185
2006-03-09 20:50:19 +00:00
playSoundEffect(1);
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
}
void SoundAdlibPC::loadSoundFile(const char *file) {
if (_soundFileLoaded == file)
return;
- Removes deleted waitTicks prototype in KyraEngine - Adds sound fading support to the adlib player - Adds support for stopping running tracks - Changes baseFreq type back to uint8 - corrects octave handling in updateAndOutput1 (Thanks to eriktorbjorn for doing that) svn-id: r21185
2006-03-09 20:50:19 +00:00
if (_soundDataPtr) {
haltTrack();
}
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
uint8 *file_data = 0; uint32 file_size = 0;
char filename[25];
sprintf(filename, "%s.ADL", file);
file_data = _engine->resource()->fileData(filename, &file_size);
if (!file_data) {
warning("Couldn't find music file: '%s'", filename);
return;
}
- Removes deleted waitTicks prototype in KyraEngine - Adds sound fading support to the adlib player - Adds support for stopping running tracks - Changes baseFreq type back to uint8 - corrects octave handling in updateAndOutput1 (Thanks to eriktorbjorn for doing that) svn-id: r21185
2006-03-09 20:50:19 +00:00
unk2();
unk1();
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
_driver->callback(8, int(-1));
_soundDataPtr = 0;
uint8 *p = file_data;
memcpy(_trackEntries, p, 120*sizeof(uint8));
p += 120;
int soundDataSize = file_size - 120;
_soundDataPtr = new uint8[soundDataSize];
assert(_soundDataPtr);
memcpy(_soundDataPtr, p, soundDataSize*sizeof(uint8));
delete [] file_data;
file_data = p = 0;
file_size = 0;
_driver->callback(4, _soundDataPtr);
_soundFileLoaded = file;
}
- Removes deleted waitTicks prototype in KyraEngine - Adds sound fading support to the adlib player - Adds support for stopping running tracks - Changes baseFreq type back to uint8 - corrects octave handling in updateAndOutput1 (Thanks to eriktorbjorn for doing that) svn-id: r21185
2006-03-09 20:50:19 +00:00
void SoundAdlibPC::unk1() {
playSoundEffect(0);
//_engine->_system->delayMillis(5 * 60);
}
void SoundAdlibPC::unk2() {
playSoundEffect(0);
}
Added WIP Adlib sound playing code to kyra. It needs some checks since it doesn't work correctly in every case at the moment. svn-id: r20960
2006-02-27 22:39:55 +00:00
} // end of namespace Kyra
Reference in New Issue Copy Permalink