This change introduced several character animation regressions in the
game, which may seem as a worse experience than the original issue with
the pirate dialogue, for most end users (Trac#13997).
This reverts commit eb851041ed.
Waxworks DOS has both digital and limited MIDI SFX. The game scripts use two
different opcodes for trigging digital and MIDI SFX; if a sound effect has both
a digital and a MIDI version, both opcodes are triggered. When digital SFX are
active, ScummVM would try to handle both the digital and the MIDI opcode by
playing a digital SFX. This would crash the game if the MIDI SFX ID was invalid
as a digital SFX ID. Strangely enough, this bug also seems to affect the Amiga
version. Apparently both opcodes are still in the scripts of this version, even
though it does not support MIDI at all AFAIK.
This commit fixes the issue by only playing the MIDI SFX if digital SFX are
turned off and ignoring them otherwise.
This adds support for the AdLib and MT-32 sound effects in Elvira 2 and
Waxworks. It adds an option to the UI to toggle between sampled and
synthesized SFX. It also adds the following enhancements:
- AdLib OPL3 mode for Elvira 2, Waxworks and Simon 1 floppy demo. This can be
selected using a new UI option.
- Mixed AdLib/MIDI mode for Elvira 2 and Waxworks.
- Implemented "monophonic chords", a feature of the original MIDI code which
would play only the highest note of a chord on AdLib. Most noticable in the
Waxworks music.
- Added UI option to select Simon 1 DOS music tempos.
- Rewrite of the AdLib and MT-32 drivers to remove duplication and make use of
features of the standard multisource drivers.
- Refactored MidiPlayer to standardize interface and remove code moved to the
drivers and parsers.
- Add dual layer graphics display to display the Japanese characters in their proper (double) resolution on top of the graphics. The original actually uses text mode.
- Adapt mouse handling to dual layer code.
- For consistency I got rid of all direct calls to _system->lockScreen() and _system->unlockScreen(), even in the sub engines that aren't affected. I find it easier to debug and to maintain this way.
Recent GCC versions complain if you memset() a class or struct that
contain non-POD data types. Get around that by either initializing
the object when created, or by adding a reset() method.
