preserve-emulation/ppsspp
1
0 Fork 0
mirror of https://github.com/hrydgard/ppsspp.git synced 2026-11-13 01:29:41 +01:00
Code Issues Packages Projects Releases Wiki Activity
45,234 Commits 83 Branches 131 Tags
9ed5dcdc27bdc8c7da3959e8b76c5efde2a73fa5
Commit Graph
9553 Commits
Author SHA1 Message Date
Chris Healy 9ed5dcdc27 GPU: Use inversesqrt for light distance and half-vector calculations 
Replace length()/normalize() with inversesqrt() pattern for better
performance on mobile GPUs. The inversesqrt instruction is a single
hardware operation, while length() requires sqrt(dot()) and normalize()
requires inversesqrt internally plus a multiply.

Changes:
1. Light distance calculation:
   Before: distance = length(toLight); toLight /= distance;
   After:  distSq = dot(toLight, toLight);
           invDist = inversesqrt(distSq);
           distance = distSq * invDist;
           toLight *= invDist;

2. Specular half-vector normalization:
   Before: ldot = dot(normalize(toLight + vec3(0,0,1)), worldnormal);
   After:  halfVec = toLight + vec3(0,0,1);
           halfInvLen = inversesqrt(dot(halfVec, halfVec));
           ldot = dot(halfVec, worldnormal) * halfInvLen;

Also reuses distSq in attenuation calculation to avoid recomputing
distance*distance.

Offline shader compiler results (vertex shader with 4-light ubershader):

Mali (malioc):
  G31 (Bifrost): 44.90 -> 40.50 cycles (-9.8%)
  G52 (Bifrost): 14.97 -> 13.50 cycles (-9.8%)
  G310 (Valhall): 13.88 -> 12.88 cycles (-7.2%)
  G57 (Valhall):  7.60 ->  6.97 cycles (-8.3%)

PowerVR (Profiling Compilers):
  Series 8 GE8300: 322 -> 302 ALU cycles (-6.2%)
  Series 9 GM9445: 296 -> 280 ALU cycles (-5.4%)
  Series B BXM:    560 -> 540 instructions (-3.6%)

Signed-off-by: Chris Healy <cphealy@gmail.com>
 2026-03-08 19:13:41 -07:00
Henrik Rydgård 5e6ec70280 Minor stuff 2026-03-01 12:09:43 +01:00
Henrik Rydgård 845f9f3d55 Add missing check to Vulkan init. Disable validation on android-x86-64 
The latter is broken in the Android SDK emulator.
 2026-02-27 23:57:50 +01:00
Henrik Rydgård bae1a7c7f9 Add "FastEmulatedGPU" compat flag. For games like Outrun that have intermittent slowdowns on hardware, it's unnecessary under emulation. 2026-02-26 10:44:16 +01:00
Henrik Rydgård 5dc169ecbe Fix more race conditions if you exit during game bootup 2026-02-24 00:56:45 +01:00
Henrik Rydgård 202259eb09 Add a quick workaround for Mahjongg Artifacts framebuffer expiration. 
Fixes #15828, well at least except for savestate support.
 2026-02-22 23:42:13 +01:00
Henrik Rydgård 5aeb38412d Fix another valgrind warning 2026-02-19 11:46:14 +01:00
Henrik Rydgård 5a5c7028b9 Assorted warning fixes and data initialization to please valgrind 2026-02-19 11:24:46 +01:00
Henrik Rydgård 12d280839c Microoptimizations in VertexDecoder interpreter 2026-02-17 11:59:30 +01:00
Henrik Rydgård c89582cfd6 Merge pull request #21267 from hrydgard/delete-mips-jit 
Delete remains of support for the MIPS architecture as host
 2026-02-17 10:55:57 +01:00
Henrik Rydgård f046857d2b Delete remains of support for the MIPS architecture as host 2026-02-17 10:20:51 +01:00
Henrik Rydgård 0c70c4f9b4 Simplify 2026-02-17 10:18:08 +01:00
Henrik Rydgård e71ffc59a6 Remove unnecessary IsStarted method 2026-02-17 10:18:08 +01:00
Henrik Rydgård ae6ba469c1 Fix some more screenshot problems 
Followup to #21249
 2026-02-13 14:56:31 +01:00
Henrik Rydgård 386fc442a6 Rework the internal screenshot API, fix bug causing screenshotting to get stuck 2026-02-13 12:34:26 +01:00
Henrik Rydgård 8ceffa08cb Ignore extreme Y offsets when matching framebuffers. 
Fixes #15828
 2026-02-12 11:03:46 +01:00
Henrik Rydgård 3ecb54e70a More tweaking of gamelist icons 2026-02-11 01:24:11 +01:00
Henrik Rydgård 5d080d1de8 Add a quick check against bad clut vs depth texture framebuffer matches 
See #15828 and #20847
 2026-02-10 15:51:49 +01:00
Henrik Rydgård 88266eee50 Warning fixes, comments 2026-02-10 13:59:32 +01:00
Henrik Rydgård d8b8a51309 Correct a bunch of edge cases, move some code around 2026-02-08 17:07:22 +01:00
Henrik Rydgård df2d00208e Improve screenshots in "Skip buffer effects" mode 2026-02-08 16:33:48 +01:00
Henrik Rydgård 1599ea7b1d SaveState: Separate screenshot operations from other ops 2026-02-08 16:11:44 +01:00
Henrik Rydgård eb234a1563 Finish the split. Greatly simplifies the render code in EmuScreen. 2026-02-08 10:46:16 +01:00
Henrik Rydgård 2ea94534e6 Split up FramebufferManagerCommon::CopyDisplayToOutput 2026-02-08 10:17:42 +01:00
Henrik Rydgård aa17599f66 Make things more consistent in the no-present case 2026-02-08 10:13:34 +01:00
Henrik Rydgård 814713e932 More refactor 2026-02-08 01:06:35 +01:00
Henrik Rydgård c139d4aaff Split up Presentation::CopyToOutput into passes and output. 2026-02-08 01:06:35 +01:00
Henrik Rydgård f01aaeafb1 Split CopyDisplayToOutput in preparation for the next step 2026-02-08 01:06:32 +01:00
Henrik Rydgård dd28f75852 Buildfixes 2026-02-07 12:03:44 +01:00
Henrik Rydgård 34108551da Fix a problem with copies to RAM - forgot to set up the copy dest rectangle 
Fixes #21187
 2026-02-05 14:04:34 +01:00
Henrik Rydgård d4896fbecd VtxDec_Tu16_C8888_Pfloat: Avoid over-reading the input 2026-02-05 13:34:34 +01:00
Henrik Rydgård 991d7bdfab VertexDecoder: Refactor away lowerbound/upperbound parameters 2026-02-05 13:26:17 +01:00
Henrik Rydgård 841e4c8564 Add various checks trying to avoid various crashes found in Google Play crash reports. 2026-02-05 11:12:53 +01:00
Henrik Rydgård 7b1ba55e72 Prevent touch controls from fading away while being held. 
Fixes #20531
 2026-02-04 00:00:38 +01:00
Henrik Rydgård 7db8709e56 Buildfix 2026-02-02 09:44:18 +01:00
Henrik Rydgård e6a2a36bb8 Touch screen controls layout: In portrait mode, move everything up a bit by default. More comfortable. 2026-02-01 14:19:51 +01:00
Henrik Rydgård 10759dc8b9 DrawEngineCommon: Avoid using member variables as loop counters, bad for perf 2026-01-30 14:10:36 +01:00
Henrik Rydgård 7d987cd78b Make fast_matrix_mul_4x4 inlineable 2026-01-30 14:10:32 +01:00
Henrik Rydgård 475390a968 Add workaround for Tales of Phantasia X. 
Fixes #21162
 2026-01-29 23:16:51 +01:00
Henrik Rydgård d6ebfed432 Fix some bugs and warnings from a pass of static analysis 2026-01-29 17:41:54 +01:00
Henrik Rydgård c3897fef4c Finish up #21151 
See #21151
 2026-01-24 10:42:43 +01:00
Henrik Rydgård 60a065b242 Make the Brave Story hack safer, hide it behind a compat flag. 2026-01-23 23:39:11 +01:00
Henrik Rydgård 732288c03a Brave Story: Hack to make the bloom effect run much more efficiently. 
Fixes #19820

Not ready for merge, it needs more checks and maybe a compat flag.
 2026-01-23 23:13:04 +01:00
Henrik Rydgård 8728be5455 Fix joining TRIANGLE_LIST draws with a vertex count non-divisible by 3 
Fixes #7503
 2026-01-23 18:54:22 +01:00
Henrik Rydgård 30e23000c0 Correct the fallback path for intra-framebuffer block copies 2026-01-22 14:25:09 +01:00
Henrik Rydgård 7bca84252e Vulkan: Add support for doing intra-buffer block copies with vkCmdCopyImage. 2026-01-22 13:47:41 +01:00
Henrik Rydgård 47f6b9975e Optimize the color alpha computation for color morph 2026-01-20 13:51:24 +01:00
Henrik Rydgård 064ad64a55 More minor optimizations 2026-01-20 13:51:24 +01:00
Henrik Rydgård 73e41750f7 More non-JIT morph optimizations 2026-01-20 13:51:24 +01:00
Henrik Rydgård 3378faca5e Optimize morph code (non-JIT vertex decoders) 2026-01-20 13:51:24 +01:00