preserve-emulation/ppsspp
1
0
Fork 0
mirror of https://github.com/hrydgard/ppsspp.git synced 2026-07-25 00:04:49 +02:00
Code Issues Packages Projects Releases Wiki Activity
11,573 Commits 76 Branches 131 Tags
cb4e72058aaa7165dd9da170cb1cca74b89ae281
Commit Graph
1536 Commits
Author SHA1 Message Date
Henrik Rydgard 9c6e7e2958 Fix bug in vtxdec jit when hardware skinning 2013-11-30 16:26:50 +01:00
Henrik Rydgard e3e9408b87 Resurrect the vtx decoder jit option for debugging 2013-11-30 16:11:54 +01:00
Henrik Rydgard eea8413a82 Make the texcoord speedhack a bit less aggressive at eliminating uvscaleoffset uniform updates. 
Should fix #4688 and the new problem in #2919 while keeping Frontier Gate Boost working.
 2013-11-30 15:54:20 +01:00
Henrik Rydgård 562582320a In LINEARFMV mode, let the presence of a video override the alpha and color test checks. 2013-11-29 18:08:29 +01:00
Unknown W. Brackets 41a7157aa6 Eat some cycles when changing thread priority. 
Some games call this a lot.
 2013-11-28 01:01:22 -08:00
Sacha c71186b35a Partial Stretch: This setting is needed on other devices too. 2013-11-28 14:59:17 +10:00
Henrik Rydgård 9aa0c9a489 Minor x86 optimizations 2013-11-26 18:23:08 +01:00
Henrik Rydgard 17c9e0d172 x86 vtxjit: Use MOVZX to avoid loading too many bits when loading 16-bit colors. 
Probably doesn't fix anything but more correct.
 2013-11-26 14:30:32 +01:00
raven02 4d14f88455 Alternative consistent fix 2013-11-26 11:34:33 +08:00
raven02 a8d9796010 Fix texture cord in HW T&L 2013-11-25 21:07:30 +08:00
Henrik Rydgard 866ed28f64 Use VFMA in the one place where it's convenient. Some more optimization. 2013-11-24 19:50:56 +01:00
Henrik Rydgard e5e23f3ce1 ARM: Fix vsgn. Some vertex decoder tweaks. 2013-11-24 18:21:47 +01:00
Henrik Rydgard 030e6460cc ARM: NEON-optimize software skinning 2013-11-24 18:03:42 +01:00
Henrik Rydgard acb3994172 Split VertexDecoder into files for ARM and x86, got too large. 2013-11-24 15:58:15 +01:00
Henrik Rydgard dfea160491 ARM: Use PLD (cache preload) in vertex decoder loop. 2013-11-24 15:08:47 +01:00
Henrik Rydgard f650b23c90 ARM: Add NEON widening and narrowing moves, and float/int convert. 
Experiment a little in the vertex decoder.
 2013-11-24 13:30:28 +01:00
Sacha 756f651eb2 Buildfix for older compilers, such as GCC 4.6.3 2013-11-22 16:22:41 +10:00
Unknown W. Brackets 70b6a4a796 Small optimization to vertex preview. 2013-11-20 22:36:48 -08:00
Henrik Rydgard e3d471f590 Fix issues with texturing from render targets with prescaled UV (texcoord speed hack) 2013-11-19 23:38:36 +01:00
Unknown W. Brackets d2bccfeba5 Omit a few commands from "DL Unknown" in debug. 2013-11-18 07:51:10 -08:00
Unknown W. Brackets ed16e42ca8 Cleanup, don't even need z here. 2013-11-17 14:49:51 -08:00
Unknown W. Brackets 018d95180a Fix 3d vertex preview. 2013-11-17 14:31:21 -08:00
Unknown W. Brackets a3bd2f1365 Fix Vec3ByMatrix44() and use it for matrix math. 2013-11-17 14:10:57 -08:00
Unknown W. Brackets b541c81ba3 Clean up Mat3x3 etc. constness. 2013-11-17 13:27:51 -08:00
Unknown W. Brackets fcc77f525f Implement some basic vertex previews on prim. 
3D doesn't work correctly (sometimes it does...)  2D should be working.
 2013-11-17 13:27:50 -08:00
Unknown W. Brackets ac8edb4282 Support reversed buffer formats in the ge debugger. 
So that it doesn't show garbage under the softgpu.
 2013-11-17 02:14:34 -08:00
Henrik Rydgård 4cf2b99316 Merge pull request #4529 from raven02/patch-19 
Fix Summon Night 5 shadow breaking
 2013-11-17 02:09:22 -08:00
Henrik Rydgård cd511cc43c In Freeze-Frame mode, don't wipe GPU resources when loading save states. 2013-11-15 16:30:55 +01:00
Henrik Rydgård f1ed1a9a17 Flush the lit pool before flushing the icache. 2013-11-15 11:14:24 +01:00
Henrik Rydgård bd23f91e0c Flush the lit pool in the vtx dec 2013-11-15 10:34:27 +01:00
raven02 312f290bde Fix typo 2013-11-15 10:24:32 +08:00
Henrik Rydgard adde016338 vtxdec sw skin arm: Simplify code, try a slightly different approach. 2013-11-14 23:07:23 +01:00
Henrik Rydgard b0ccf5981c Don't bother with glDrawRangeElements, seems to not improve perf. 2013-11-14 17:33:43 +01:00
Henrik Rydgard d6d54c8bb4 Fix a couple of bugs in vtxdec 2013-11-14 15:06:51 +01:00
Henrik Rydgard 9521d0e0a6 NormalizeVertices should not try to skin if we've already software-skinned. 2013-11-14 14:34:47 +01:00
Henrik Rydgard 3b63ef7005 Remove the SubmitPrim param forceIndexType, optimize BBOX more. 2013-11-14 14:03:03 +01:00
raven02 8af2e66bd5 This seems to be better 2013-11-14 19:33:20 +08:00
Henrik Rydgard 35ae239eb9 Optimize bbox some more 2013-11-14 12:25:53 +01:00
Henrik Rydgard 4f93654a88 Oops, accidentally enabled some bbox debugging code 2013-11-14 11:49:06 +01:00
Henrik Rydgard 8a69543ec4 BBOX: Transform the planes by the matrix so we don't need to transform the box 2013-11-14 11:44:13 +01:00
Henrik Rydgard 871d8b28ab Assorted vtxdec x86 optimizations 2013-11-14 11:44:13 +01:00
Sacha eaa52c4a61 VertexDecoder: Aligned code seems to help Symbian (armv6) performance. It is also used for CPU JIT so there's likely a reason. 2013-11-14 15:30:07 +10:00
Henrik Rydgard ff1d53c9c7 vtxdec x86: Don't load bones unless software skinning is enabled 2013-11-14 00:56:24 +01:00
Henrik Rydgard a9b365aa71 Fix for possible misalignment crash on x86 2013-11-14 00:37:04 +01:00
Henrik Rydgård a6ee2648d0 Merge pull request #4531 from hrydgard/software-skinning 
Software skinning
 2013-11-13 12:54:41 -08:00
Henrik Rydgård 9bbdd1907d Kind-of optimized ARM software skinning (non-NEON) 2013-11-13 18:11:35 +01:00
Henrik Rydgård 821a2f10f8 Delete obsolete code 2013-11-13 18:10:58 +01:00
Henrik Rydgard 6221dbaf5d Optimize software skinning for x86. 
Can't seem to get a win on Windows vs hardware skinning though, even though
draw calls drop by 2/3rd...
 2013-11-13 18:10:58 +01:00
Henrik Rydgard 9333d3ea76 Vtx dec jit: Combine the scale and offset registers to save 1 more xmm register. 2013-11-13 18:10:58 +01:00
Henrik Rydgard f0cacf46d0 No reason to involve the FPU when loading matrices 2013-11-13 18:10:58 +01:00