ppsspp

preserve-emulation/ppsspp

0 Fork 0

mirror of https://github.com/hrydgard/ppsspp.git synced 2026-10-14 15:00:30 +02:00

Files

T

History

Henrik RydgårdandClaude Opus 5.5 63a05db8d2 SoftGPU: Skin the x, y and z components together in vector lanes

Skinning did each component's weighted bone entries and its chain of
GEAdds separately in scalar code. The components now go in lanes, with
GEMulFloat24x4 for the weighted entries, one-term row sums for the
products and GEAdd4 for the sums, in the same order. Checked against the
scalar version on every vertex of 143 skinned frame dumps.

GEMulFloat24x4 matches ProductToFloat24, so inf and NaN bone entries
behave as before. GEAdd4 handles two zero operands without its fallback.

Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>

2026-10-09 09:07:12 +02:00

BinManager.cpp

SoftGPU: Store depth in VRAM as the GE does

2026-10-08 18:35:48 +02:00

BinManager.h

SoftGPU: Store depth in VRAM as the GE does

2026-10-08 18:35:48 +02:00

Clipper.cpp

SoftGPU: Cull primitives entirely outside an x or y plane

2026-10-02 15:09:43 -06:00

Clipper.h

SoftGPU: Cull primitives entirely outside an x or y plane

2026-10-02 15:09:43 -06:00

DrawPixel.cpp

SoftGPU: Don't flush for JIT compiles where there's no JIT

2026-10-05 14:55:17 -06:00

DrawPixel.h

SoftGPU: Store depth in VRAM as the GE does

2026-10-08 18:35:48 +02:00

DrawPixelX86.cpp

SoftGPU: Store depth in VRAM as the GE does

2026-10-08 18:35:48 +02:00

FuncId.cpp

SoftGPU: Keep a not-black color test when dithering, fog or a logic op is on

2026-10-08 18:37:32 +02:00

FuncId.h

SoftGPU: Store depth in VRAM as the GE does

2026-10-08 18:35:48 +02:00

GEMath.cpp

SoftGPU: Skin the x, y and z components together in vector lanes

2026-10-09 09:07:12 +02:00

GEMath.h

SoftGPU: Skin the x, y and z components together in vector lanes

2026-10-09 09:07:12 +02:00

Lighting.cpp

GEMath: GERowSum4 takes its left operands as one vector

2026-10-09 09:07:12 +02:00

Lighting.h

SoftGPU: Light vectors as one matrix row from the model position

2026-10-02 15:09:43 -06:00

Rasterizer.cpp

SoftGPU: Store depth in VRAM as the GE does

2026-10-08 18:35:48 +02:00

Rasterizer.h

SoftGPU: Don't change a state the threads may be drawing with

2026-10-05 14:55:17 -06:00

RasterizerRectangle.cpp

SoftGPU: Only draw 1:1 textured triangle rectangles as sprites

2026-10-02 15:09:43 -06:00

RasterizerRectangle.h

SoftGPU: Keep textured rectangles as triangles unless they'd match

2026-10-02 15:09:42 -06:00

RasterizerRegCache.cpp

GPU: modernize use C++17 constexpr for precalculate compilation

2026-05-19 21:57:45 +03:00

RasterizerRegCache.h

Delete remains of support for the MIPS architecture as host

2026-02-17 10:20:51 +01:00

README.txt

Add snapshot of the whole source code.

2012-11-01 16:19:01 +01:00

Sampler.cpp

SoftGPU: Don't flush for JIT compiles where there's no JIT

2026-10-05 14:55:17 -06:00

Sampler.h

SoftGPU: Sprite UVs and texel addressing like the GE

2026-10-02 15:09:42 -06:00

SamplerX86.cpp

SoftGPU: Leave textures with an invalid level address to the generic samplers

2026-10-05 13:05:46 -06:00

SoftGpu.cpp

Core: Let 16-bit CPU accesses through the depth mirrors see the GE's layout

2026-10-08 18:37:23 +02:00

SoftGpu.h

Core: Let 16-bit CPU accesses through the depth mirrors see the GE's layout

2026-10-08 18:37:23 +02:00

TransformUnit.cpp

SoftGPU: Skin the x, y and z components together in vector lanes

2026-10-09 09:07:12 +02:00

TransformUnit.h

SoftGPU: Carry UVs and normals into formats without them as the GE does

2026-10-09 02:03:31 +02:00

trirast.txt

Add snapshot of the whole source code.

2012-11-01 16:19:01 +01:00

README.txt

To get to 100% compatibility, we will need a software renderer as there are games out there that do tricks
that can't really be faked in a sensible way. Useful for homebrew too that mix sw and accel rendering.