The steps skipped bones with a zero weight. With weights that vary from vertex to vertex that
branch mispredicts a lot, and it saves only a few multiplies. Now every weight is applied, with
fused multiply-adds, as the vertex JIT does. Without the vertex JIT (as on iOS): Tekken 4.5% fewer
cycles, God Eater 2.6%, God of War 1.7%.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
saveIndexed = true in [options] saves the CLUT4/CLUT8 textures listed in
[indexed] as palette PNGs, saveIndexed = all saves every one of them.
The file is named <hash>_indexed.png, with the CLUT hash zeroed so one
file serves every palette of the texture. The other textures are dumped
as before.
The indices are decoded from PSP memory and the palette comes from the
current CLUT, so the dump looks like the game's texture. Add the file to
[hashes] and [indexed] to use it as a replacement.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
For where there's no vertex JIT. It skins the same way as the steps:
the skin matrix is 0 + bone * weight, kept as the columns that
Vec3ByMatrix43() and Norm3ByMatrix43() multiply and add unfused in the
same order (by lane on ARM64, like those), and only rebuilt when the
weight changes. 2.2 ns a vertex in God of War against the JIT's 1.95
and the steps' 12-18.
Without the vertex JIT, God of War's instructions drop by 6.8% per 10
emulated seconds.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The decoders clear gstate_c.vertexFullAlpha when a vertex has less than
full alpha, and a batch decodes each of its draws in turn. The God of War
and GTA decoders assigned it instead, so a draw with full alpha set it
back after an earlier one had cleared it. God of War's also read the
alpha from the normal: its struct had color and normal swapped (the
output came out right, as each went to the matching slot), so the flag
never got set for its most used format.
The vertex decoder test now decodes a vertex with full alpha after a
draw that cleared the flag and after one that didn't.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
For where there's no vertex JIT (iOS), as they're a few percent slower
than the JIT's but four to eleven times faster than the steps. The
formats carry most of the vertices of ATV Offroad Fury: Blazin' Trails,
Wipeout Pure and Killzone: Liberation, and a lot of Gods Eater Burst's.
They convert like the steps (PrescaleUV moves to the header for that),
and the vertex decoder test now also compares each format's decoder
without the JIT against the steps.
Without the vertex JIT, 10 emulated seconds on an M1, instructions:
ATV -9.9%, Wipeout -6.0%, Killzone -5.4%, Gods Eater -2.2%.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
They were only used for UV from texture coordinates. With the texture
matrix or environment map, UV isn't prescaled, which is the same as a
scale of 1 and no offset, so NoPrescale versions pass that. God of War
draws 5 million of its vertices per 10 seconds this way, which without
the vertex JIT took 9.6 ns each through the steps and now 2.4.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
An [indexed] entry set to gpu uploads the indices as an R8 image, and
the CLUT8_INDEXED depal shader applies the palette. A texture whose
palette changes a lot then costs no CPU time per palette, and keeps one
texture instead of one per palette.
Once such a replacement is found, the texture is looked up without the
CLUT hash, like with the TextureCLUTInShader compat flag. Only the first
mip level is used, and the replacer's filtering overrides are skipped
since filtering indices mixes up palette entries. Without shader bitwise
ops it falls back to coloring on the CPU.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
The triangle is convex, so once a row has been entered and left, the
rest of it is outside. This cut cycles by 5% in Wipeout Pure, where depth
raster is on for the lens flares.
The test checks the covered pixels against a per-pixel edge test.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
TestBoundingBoxFast transformed and tested one vertex per iteration,
with a scalar divide and a min/max chain each vertex had to wait on.
It now transposes four vertices into lanes and does it all as vectors,
with the same operations in the same order, so the results don't change.
It was 16.6% of the emulator thread in Wipeout Pure (with depth raster
off) and 5% in GTA: Vice City Stories.
Per vertex, it's 1.8x faster on an M1, 2x on x86-64, 1.7x on a Galaxy S8+
and 10% faster in 32-bit mode there.
The test compares it with the old version on random draws in every
position and index format.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Indexed replacements can now also be KTX2 files in the R8_UNORM format,
uncompressed or zstd supercompressed, with the mip levels stored in the
file. Lossy formats (Basis, BC) are not usable, since an error in an index
picks the wrong color.
The template textures.ini documents the supported files.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
FakeJit was the native JIT for architectures without one, but the config
already forces those onto the IR interpreter, so CreateNativeJit() now
returns that instead. The vertex decoder fallback no longer needs
FakeEmitter.
JitBlockCache was the block cache of the pre-IR JITs. Its header keeps
the debug interface types the IR block caches use. Also removes the
JitInterface link hooks only it called, and the JitState fields tied to
it. This fixes a null dereference in the ImDebugger memory section,
which read GetBlockCache(), always null with the IR JITs.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Resistance: Retribution sets its scissor to end at 480,272, one past its
drawing region. The CPU depth rasterizer (SoftwareRasterDepth) only
clipped to the scissor, so it wrote row 272 of the depth buffer, which is
where the game keeps its logo textures. The startup logos got a garbage
strip on hardware backends and failed to fade out.
SoftGPU already clips to the minimum of the two.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Since the sequential copies heuristic went (50696c5b1f), every partial readback
waits for the GPU. Motorstorm reads parts of the screen about 18 times a frame,
and ran about 75% slower than 1.20.4: 5403 blocking readbacks in 10 emulated
seconds, against 156.
With the flag, the first partial color readback of a buffer reads all of it into
a copy outside PSP memory, and the pieces until the next render to the buffer
are copied from there. Unlike the old heuristic, only what the game asked for is
written to PSP memory. Motorstorm is back to 153 blocking readbacks and 1.20.4's
speed, with identical output. Grand Knights History, which the heuristic was
added for, gets the flag too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Fixes#15145
A replacement listed in a new [indexed] section of textures.ini holds
palette indices instead of colors: a palette PNG (the file's palette index)
or an 8-bit grayscale PNG (the gray value), at any resolution. When the
texture is built, the indices go through the game's current CLUT, including
the CLUT index shift, mask and start position, and the result is uploaded as
an ordinary RGBA texture. A game that swaps or customizes a palette keeps
working with a high-resolution replacement.
[indexed] keys use the [hashes] format and are matched with the same
wildcards, so a key with a zero CLUT hash covers every palette of the
texture. The contents hash only covers the indices, so one image is shared
by all palettes (the loaded data is shared, each palette still has its own
texture cache entry).
Saving does not dump indexed textures yet.
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
A draw that blends with the source alpha while setting the stencil (the
alpha on the PSP) needs dual source blending to write both. Without it the
alpha came out wrong, and later draws that blend with the destination alpha
differed: WipEout Pure, Juiced 2 and 16131 UCES00001 on a Mali-G71 and a
PowerVR GE8320. With framebuffer fetch, blend those in the shader instead,
which matches the Mac's Vulkan render now. Without fetch it would take a
framebuffer copy per draw, too slow for something this common.
A draw that writes no color can't need the source alpha, so it writes the
stencil as alpha directly.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
ResizeFramebufFBO copied color and depth into the new FBO but left its
stencil cleared. Need for Speed Shift grows its framebuffer from 480 to 512
wide just before stencil testing, and the frame came out far too dark on
Android (112k px off the PSP on an Adreno 840, now 4k). MoltenVK happened
to keep it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
PerformWriteStencilFromMemory bound the framebuffer it wrote the stencil to
and, unless it blitted, left it bound. When that wasn't the current render
target, the following draws landed in it. Draws that copy the target for
shader blending rebind it themselves, which hid this; with framebuffer fetch
they don't, and OutRun 2006's masked water draws went into the old 5551
buffer at the address instead of the new 565 one.
Over the framedumps set on Vulkan, 9 dumps get closer to the PSP (OutRun
ULUS10064_0002 12295 -> 483 px, F1 2009, God of War) and one a little further
(16131 ULUS10193, a slight brightness shift).
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Without clip distances, the fragment shader's min/max Z discard only came on
when both minz and maxz were set, while the clip planes take each side on
its own. Pursuit Force's sky has minz 0 and maxz 300 and lies past 300, so
the PSP drops it (its replays of 11399 UCES00019 and 11216 UCUS98703 show
that), but GPUs without clip distances drew it over the scene: 13.9k px off
the PSP on a Mali-G71, now 8 as with clip distances. The discard can now
come with the fragment depth clamp, for the side that's at the limit.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The fragment shader's min/max Z discard declared it with the default
precision, the vertex shader with highp, and PowerVR refuses to link such a
program, so every draw that needed the discard was dropped. On a PowerVR
GE8320 that was Street Riders, Valkyria Chronicles, Flatout and Frontier
Gate, and with the discard taking one-sided depth ranges too (next commit),
most of Pursuit Force.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Picking a byte out of a 16-bit framebuffer needs more than the defaults. The
texcoord reached the shader as a mediump varying, which Adreno keeps in fp16:
its byte coordinate (u times up to 2048) then picks the wrong byte for every
other one past 512, and SOCOM Tactical Strike's night vision came out striped
in the right half. On a PowerVR GE8320 the default lowp in the rest of the
shader made SOCOM's night vision and full screen effect far too bright (53k
and 17k px off the Mac's Vulkan render, now 0). Vulkan already uses highp.
Where the fragment shader has full precision, the vertex shader now writes the
texcoord in highp, and CLUT8 depal reads it that way and computes in highp.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
GL_TEXTURE_LOD_BIAS is desktop GL only, and the queue runner skipped it on
GLES, but GLES still reported sampler LOD bias as supported, so the bias was
dropped and textures used the wrong mip levels: Super Monkey Ball Adventure's
intro came out blurred on the Adreno 840. Report it unsupported, so the shader
applies it (GPU_USE_SHADER_LOD_BIAS), and give the GL backend the uniform that
Vulkan has in its uniform block.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
PosFloat loaded its clamp constants into XMM4/XMM5 but clamped with XMM2/XMM3,
and the S8/S16 morphs overwrote the morph weight pointer with the constant's
address. These only run when the JIT cache lands more than 2GB from the
constants, which CPUEquivalence's allocations made happen on Linux CI
(SplineTessellation). Checked by forcing every RipAccessible() false, which
SplineTessellation and VertexJit now pass.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The fallback set the rasterizer's Z to a constant so nothing would clip, which
lost the clipping at the PSP's near plane too. Hardware depth clamp loses it
the same way, which the vertex shader can only make up for with a clip plane.
Give the rasterizer (z + w) / 65536 instead: 0 at the near plane, so it clips
there, and past w only beyond 65535 * w. And without clip distances, use the
fragment fallback instead of hardware depth clamp.
With --disable-gpu-features for all three on the Mac, the framedumps set now
differs from the full render by 15570 px instead of 184229 (WALL-E, Hokuto no
Ken, Frontier Gate, FF Zero). The full render doesn't change.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
For GPUs without depth clamp (old Mali), found by comparing with
--disable-gpu-features=depth-clamp on the Mac:
- Write Z / 65536 like the vertex shader, not / 65535.
- Pass the fragment shader Z after the vertex depth rounding, which hardware
clamp sees too (Puyo Puyo Fever 2's ground at Z = 0).
- Also clamp draws whose Z only touches 0 or 65535.
- Draws that skip the bounds check (skinned, morphed) get the fragment clamp
and min/max discard wherever the hardware versions would be on, since those
don't depend on the check either.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Curves skipped the per-draw bounding box check and relied on the vertex
shader's near clip plane. On old Mali (no clip distances), that left the
geometry behind the PSP's near plane to the GPU: Test Drive's hills reached
over the sky.
A patch lies within the convex hull of its control points, so run the PRIM
check on those: a patch entirely outside is skipped, and only one with points
outside the clip volume in Z goes through software transform. Missing clip
distances are now also a reason to software-clip, for PRIM too.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
VertexDecoderRiscV.cpp got <cfloat> through SmallDataConvert.h, and three
headers used malloc/free without <cstdlib>, which libc++ no longer brings in.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Config.h is edited about 100 times a year and each edit recompiled 198
files; now 172. SmallDataConvert.h pulls in arm_neon.h and was included by
headers that don't use it.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
- Fixes#21391
Makes the hardware backends tessellate Bezier patches and splines with the
GE-exact evaluation the software renderer uses. Fixes the z-fighting glow
under the floor grates in Coded Arms.
AGENTS.md: Put "- Fixes #N" first in commit messages
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Depth readback and the GLES stencil readback converted whatever their
buffer last held into PSP memory when the copy failed. The generic stencil
readback asked for depth, which asserts on Vulkan and D3D11. And
ReadFramebufferToMemory clamped the width to the buffer but not the height,
which a block transfer can exceed.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The crosshair turns red, and lock-on works, when stencil bit 3 is set at
the centre of the frame, which the game tests by reading the pixel from
VRAM with the CPU (#17936). Hook that and read back the stencil of that
row into the alpha bytes in RAM. The alpha the hardware backends keep
doesn't follow stencil everywhere, so reading color isn't enough.
Adds PerformReadbackStencilToMemory for this (8888 only for now).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
Buzz! Master Quiz converts its question videos (sceMpegBaseCscAvcRange)
straight into the middle of the frame it has drawn, at 84,69. The
framebuffer upload only matched writes starting at a framebuffer's
address, so the hardware backends never showed the video. Pass the
written width along and draw just that rectangle into the framebuffer.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
A memcpy of part of one row into a framebuffer was drawn at the buffer's full
width, pulling in whatever followed the source. Everybody's Golf copies its
loading screen spinner back to the screen 64 pixels at a time, so its sprite's
next rows were drawn alongside, as a row of discs across the bottom.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
After more than three partial readbacks of one buffer in a frame,
ReadFramebufferToMemory read the whole buffer instead for the rest of the
session, writing rows the game never asked for.
Everybody's Golf 2 reads back four portraits' color and depth per frame, which
turned it on. The whole 480x273 depth buffer then got written to VRAM, and its
last row is where the game keeps its texture heap. A texture header was
overwritten: the portrait showed the scene behind it, and the next time the
portraits were built the game crashed writing through a null pointer.
It was added for speed in Grand Knights History (c42d04829f), which copies many
slivers of a framebuffer, so that game may now do more readbacks.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The game builds each 64x64 portrait by memcpying a rect of the framebuffer one
partial row at a time, then setting each texel's alpha bit from the depth
buffer, which it reads directly from VRAM. Neither got read back: the row
copies don't line up with a framebuffer, and depth is never downloaded.
Hook the copy function to read back both, whole rows at a time. That needed two
fixes to requested depth copies: they were dropped when the depth buffer lies
above all color buffers, and only an exact full-buffer size matched.
The portrait texture now matches the software renderer, though it still
doesn't show up on screen with Vulkan.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
ApplyGESkinning computes skinned positions and normals the way the GE
does and overwrites what the decoder skinned in float, so that was
wasted, 4% of the emulation thread in God of War. A new decoder option,
callerSkins, leaves out the weights, position and normal steps for
weighted vertices outside through mode and only reserves the float
position and normal. The software renderer's transform unit gets its
own decoders with it; bounding boxes, splines and the debugger still
use the ones that skin. Checked to give the same output with the vertex
decoder JIT as without the option.
God of War takes 4.0 s for five emulated seconds (from 4.1 s).
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
ReadVertex took the GE's viewport arithmetic one coordinate at a time,
with a reciprocal of w for each, and stored the position, the UVs and
the whole vertex in pieces that were then reloaded with wider loads,
each a store forwarding stall. x, y and z now go through one GERecip and
the four-lane multiply and add (the scalar path stays for non-finite
input and zero or denormal w), the position is loaded as a vector, and
the vertex is written straight into its destination.
Co-Authored-By: Claude Opus 5.5 (1M context) <noreply@anthropic.com>
The Tantalus games (Cars Race-O-Rama, MX vs ATV Reflex and Untamed, SpongeBob)
turn depth into fog with a CLUT16 read of the swizzled depth mirror, stretched
2x into a 565 view of an 8888 buffer. A following 8888 copy writes the scene's
RGB and keeps the alpha, so each pixel's fog factor is the upper 565 half.
At a higher render scale, the depth X swizzle snapped every render pixel to its
PSP pixel's left edge, so the fog came out in PSP-pixel blocks. It now keeps the
position within the PSP pixel. That leaves the upper half sampling the right
half of the depth texel, half a pixel off the scene, which drew magenta halos on
the right of objects in Cars. With the new DepthCLUTIntoHalves compat flag, the
depal pass outputs twice as wide so each render pixel reads its own 8888 pixel's
depth.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
At a higher render scale each PSP pixel spans several render pixels, and the
reinterpret mapped them all to the PSP pixel's first one, which pixel-doubled
the result horizontally. A render pixel k into its PSP pixel now reads the
render pixels k into the PSP pixels it's made of. Cars Race-O-Rama renders at
the full render scale again, without the magenta stripes along edges that the
doubling caused. 1x is unchanged.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Reading a 16-bit framebuffer as CLUT8 with u = 1 + 2x puts every pixel center
exactly on a byte edge. The GE lands on the edge only when the sprite's u
gradient is exact (gpu/probe exp256), otherwise just below it. The software
transform now nudges the short sprites' u down a little, sprite by sprite, and
the shader truncates to the GE's 1/16 texel. SOCOM keeps reading its pixels'
high bytes; Syphon Filter and Resistance: Retribution's 64x128 bright pass
now reads the next pixels' low bytes, as on the PSP.
Syphon Filter's warehouse and dark room dumps get much closer to the PSP
(18310 -> 7192, 9885 -> 3925 px on Vulkan), and Killzone: Liberation's improve.
Resistance's and Syphon Filter mission 5's get further away, from the 5551
precision difference their bright pass also depends on.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The GE samples at the pixel center like the GPU; the comment saying it uses the
corner was wrong. SOCOM's u = 1 + 2x reads the high byte because a sprite's
fixed-point UV gradient falls just short of a texel edge (gpu/probe exp87,
exp163). Move to the PSP pixel's center and 1/16 byte below it, the GE's step,
instead of to its left edge. That's the same byte for one or two bytes per
pixel, but not for four, where the edge could pick the component before. No
frame dump changes, at 1x or 2x.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
int(t * 254.99) gave each byte value minus one: c / 255 * 254.99 lands just
below c. With 255.99 it's c for every byte. The texel also gets the best
available precision: in fp16 the product rounds, and 16 of the 256 bytes
came out one low. Burnout Dominator's car select lens flare gets closer to the
PSP on Vulkan (57 -> 48 px); nothing else in the frame dump set changes.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
They reach the thousands (bytes across a wide framebuffer), past what fp16
holds to a fraction, and NVIDIA's Vulkan driver runs the default lowp as fp16
(as in #22481). The byte picked out of each pixel could then land on the wrong
one. No change on the M1.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>