Commit Graph

16752 Commits

Author SHA1 Message Date
Henrik Rydgard
a71be3f75f Start moving GPU feature compatibility decisions to startup
(and a single, cache-friendly flag field)
2015-09-05 19:58:47 +02:00
Henrik Rydgard
86a02fed61 Minor cleanup 2015-09-05 19:42:58 +02:00
Henrik Rydgård
825662d06e Update native 2015-09-02 10:23:18 +02:00
Jonas Höglund
a78d71029a Use DEVICE_ID_ANY for the NK_BACK key in default mappings 2015-09-02 10:23:18 +02:00
Jonas Höglund
06fadd5cad Set DPad keys in native for UI navigation
Complements the update to the key repeat logic in native.  Also renames
UpdateConfirmCancelKeys to UpdateNativeMenuKeys since even before this commit,
the function also set the TabLeft and TabRight keys.
2015-09-02 10:23:18 +02:00
Jonas Höglund
5bb2f20493 Use KeyDef for Confirm/Cancel/TabLeft/TabRight keys 2015-09-02 10:23:18 +02:00
Jonas Höglund
0a6ab528ea Move KeyDef and AxisPos to native 2015-09-02 10:23:18 +02:00
Henrik Rydgård
ff3bb35d15 Merge pull request #7937 from potmdehex/indentation
Fix indentation
2015-09-01 12:30:55 +02:00
potmdehex
3d7d4b090f Replace tabs with space 2015-09-01 19:13:51 -07:00
John Åkerblom
d4741c2d50 Fix indentation 2015-08-31 20:12:21 -07:00
Henrik Rydgard
08b340f423 More D3D hackery: Take inverted Z in proj matrix into account.
Also, *2 breaks it for no good reason so don't do that.
2015-08-30 22:45:24 +02:00
Henrik Rydgård
9bf5d40084 Merge pull request #7936 from unknownbrackets/arm64-minor
arm64: Avoid overflow in imm math
2015-08-30 22:13:59 +02:00
Unknown W. Brackets
9b94266c4a arm64: Avoid overflow in imm math. 2015-08-30 13:00:18 -07:00
Henrik Rydgård
037b9cfdc4 Merge pull request #7935 from unknownbrackets/ui-tweaks
Allow audio to be enabled in the menu
2015-08-30 19:08:10 +02:00
Unknown W. Brackets
58b5a74748 Allow audio to be enabled in the menu.
Previously, if it was disabled at first, enabling it wouldn't work - you
still would not get the audio.
2015-08-30 10:05:13 -07:00
Henrik Rydgård
bbde0919d8 Merge pull request #7925 from FireyFly/analog-testscreen-fix
Add axis event listener to the analog test screen.
2015-08-30 13:45:29 +02:00
Henrik Rydgard
33ff7e2075 Update submodules. Fix in native for #7931 2015-08-30 12:59:46 +02:00
Henrik Rydgård
12884eceb9 Merge pull request #7920 from hrydgard/vshader-depth-round
Round depth values to 16-bit precision to match the PSP's 16-bit depth buffer
2015-08-30 11:30:12 +02:00
Henrik Rydgård
e99ba8682a Merge pull request #7930 from aroulin/x64-emitter-rcp
x64Emitter: add RCPPS and RCPSS SSE instructions
2015-08-29 21:12:13 +02:00
Henrik Rydgard
488c62e534 This makes slightly more sense than the last commit. 2015-08-29 20:38:19 +02:00
Henrik Rydgard
7b0a31ade5 Adjust D3D depth calculations to get them to match OpenGL as close as I can get it. 2015-08-29 17:43:09 +02:00
Jonas Höglund
afb82e3998 Add axis event listener to the analog test screen. 2015-08-28 01:24:38 +02:00
Henrik Rydgard
76eacf32f4 Change the offset to 0.0 2015-08-27 18:02:45 +02:00
Henrik Rydgard
2b7d08e3af Also apply the GT fix in D3D, doh 2015-08-27 18:01:00 +02:00
Henrik Rydgard
fd18f14508 Fix GT and hopefully other games that got a black screen 2015-08-27 17:58:41 +02:00
Henrik Rydgård
c671b075d8 Merge pull request #7924 from sum2012/win32_color_depth
Add back  color depth check from last commit
2015-08-27 15:13:05 +02:00
sum2012
39d814b214 Add back color depth check from f3e0029285
But in driver error.
2015-08-27 20:26:04 +08:00
Henrik Rydgard
f3e0029285 Remove probably outdated and unnecessary color depth check 2015-08-26 23:56:38 +02:00
Henrik Rydgard
7e3292d275 Minor optimization 2015-08-26 18:46:55 +02:00
Henrik Rydgard
bf6aef44cd Oops, didn't finish the D3D code 2015-08-26 17:45:21 +02:00
Henrik Rydgard
d89da71057 Get rid of a division in the vertex shader by precomputing the inverse. 2015-08-26 17:40:06 +02:00
Henrik Rydgard
5dbc8abd94 experiment: Add a different offset that unknown found during testing, plus remove rounding of through mode depth (which should already be rounded) 2015-08-26 17:28:34 +02:00
Henrik Rydgard
dedd2b60e4 Subtracting an offset of 0.5 works on my nVidia and I have no idea why, this makes no sense. 2015-08-26 17:02:07 +02:00
Henrik Rydgard
160980ee2e Properly dirty the uniform in Execute_ViewportType 2015-08-26 16:28:10 +02:00
Henrik Rydgard
44ce0c8dee Implement vertex shader depth rounding in DX9 2015-08-26 16:07:40 +02:00
Henrik Rydgard
c91b58ac8e Fix typo. This breaks it again though, strangely. 2015-08-26 16:07:11 +02:00
Henrik Rydgard
fee776b6e0 Correct scale factors for through mode depth rounding. 2015-08-26 15:55:58 +02:00
Henrik Rydgard
9e18fb7e90 Round Z in the vertex shader to match a 16-bit Z buffer.
However this inexplicably does still not solve the Phantasy Star issue!
2015-08-26 15:25:21 +02:00
Henrik Rydgard
a5fb4cffd5 Failed quick attempt at Z rounding in the vshader 2015-08-26 15:25:20 +02:00
Henrik Rydgard
1e8a1b801f Rename viewportX1/X2 etc to scale + center, which is what they actually are. 2015-08-26 15:24:35 +02:00
Henrik Rydgard
29fc9e81a4 Use the gstate.getViewport* accessors. 2015-08-26 15:24:35 +02:00
Henrik Rydgård
6096ef30c5 Merge pull request #7919 from pal1000/work
Ignore Windows/win-version.h created since #7908
2015-08-26 14:58:21 +02:00
pal1000
d21357a20f Ignore Windows/win-version.h created since #7908 2015-08-26 12:56:54 +03:00
Unknown W. Brackets
8199b10422 Update submodules. 2015-08-25 18:16:50 -07:00
Unknown W. Brackets
e6adf48c38 Fix a glitch when PPSSPP_GIT_VERSION_NO_UPDATE = 1.
Would cause the other version not to generate properly.
2015-08-24 21:07:11 -07:00
Henrik Rydgard
4ff8c8c8e0 GEDebugger: Make the pixel info box wider 2015-08-25 01:15:02 +02:00
Henrik Rydgård
7e1eabe5e2 Merge pull request #7918 from unknownbrackets/gedebug-hover
Show pixel value on hover in GE debugger
2015-08-24 19:42:15 +02:00
Henrik Rydgård
9a4c019f08 Merge pull request #7908 from unknownbrackets/win-version-gen
Generate windows version automatically
2015-08-24 19:29:08 +02:00
Unknown W. Brackets
296173317d Let's also show 24-bit values in a 16-bit range. 2015-08-23 17:55:54 -07:00
Unknown W. Brackets
95d4f88e01 Read depth values as floats on OpenGL.
This preserves the full precision better than a 16-bit value does.
2015-08-23 17:52:15 -07:00