preserve-emulation/ppsspp
1
0 Fork 0
mirror of https://github.com/hrydgard/ppsspp.git synced 2026-11-14 10:09:32 +01:00
Code Issues Packages Projects Releases Wiki Activity
47,140 Commits 86 Branches 131 Tags
435033e439541b72686a8305894de44fe9efbcdb
Commit Graph
47140 Commits
This Branch
This Branch
All Branches
Author SHA1 Message Date
Unknown W. Brackets a4558a5736 samplerjit: Take texptr/bufw as arrays. 
Prep for moving mip map sampling into linear.
 2021-12-28 12:04:16 -08:00
Unknown W. Brackets 4864850b3b samplerjit: Handle mipmap width/height in S/T calc. 2021-12-28 11:29:29 -08:00
Unknown W. Brackets a84accf713 samplerjit: Move S/T calculation into jit. 
Gives a pretty decent 5-10% improvement in many places.
 2021-12-28 09:58:23 -08:00
Unknown W. Brackets 476dfdf731 samplerjit: Add more bits for S/T, skip multiply. 
For now, we're not using those other bits yet.
 2021-12-27 18:24:37 -08:00
Unknown W. Brackets 9cc0883d53 softgpu: Correct non-SSE T clamp. 2021-12-27 15:31:37 -08:00
Henrik Rydgård 1f5680e614 Merge pull request #15253 from unknownbrackets/samplerjit-vec 
Vectorize texture sampling a bit, make filtering more accurate
 2021-12-27 21:00:51 +01:00
Unknown W. Brackets 39d5b1c221 softgpu: Reduce mipmap fraction to 4 bits. 
For CONST (and SLOPE with flat w), this produces accurate values.
SLOPE is still wrong in its handling of w, and AUTO seems to calculate
using a different and less accurate ramp.  But they both produce values
with 16 steps, in any case.
 2021-12-27 11:37:33 -08:00
Unknown W. Brackets d6b6ef4cb1 softgpu: Correct nearest filtering too. 
Turns out to have the same behavior as linear, when it comes to the
subpixel offset.
 2021-12-27 11:37:33 -08:00
Unknown W. Brackets 1dfaea9062 softgpu: Remove no longer possible report. 
Also, it's known how this behaves, now.
 2021-12-27 11:37:33 -08:00
Unknown W. Brackets 75f105f84b softgpu: Make linear filtering more accurate. 
This matches tests for various u/v offsets and x/y subpixel offsets.
Mipmaps are probably still wrong.
 2021-12-27 11:37:32 -08:00
Unknown W. Brackets 3cd19b02ac samplerjit: Handle unswizzled offsets too. 2021-12-27 11:37:32 -08:00
Unknown W. Brackets 820361f34b samplerjit: Calculate texel byte offset as vector. 2021-12-27 11:37:32 -08:00
Unknown W. Brackets 4d6a2f3919 samplerjit: Blend linear using integers. 2021-12-27 11:37:32 -08:00
Unknown W. Brackets 6f4e735757 samplerjit: Accumulate results in an XMM. 2021-12-27 11:37:32 -08:00
Unknown W. Brackets b00a66e34c samplerjit: Pass u/v coords as vector. 2021-12-27 11:37:32 -08:00
Henrik Rydgård cc767622d0 Merge pull request #15247 from unknownbrackets/debugger-hash 
Debugger: Don't try to hash invalid RAM
 2021-12-21 08:29:37 +01:00
Unknown W. Brackets 9017f91253 Debugger: Don't try to hash invalid RAM. 2021-12-20 16:17:18 -08:00
Unknown W. Brackets 31d7d1e55a Merge pull request #15244 from hrydgard/vulkan-profiler-toggle 
Make the Vulkan GPU log profiler a runtime developer setting
 2021-12-19 15:32:20 -08:00
Henrik Rydgård 74f76e5ef1 Missing stdarg.h include 2021-12-20 00:11:51 +01:00
Unknown W. Brackets fadf01dc7b Vulkan: Update profiler formatting for this. 2021-12-19 14:56:50 -08:00
Henrik Rydgård df2f0df155 Make the Vulkan GPU log profiler a runtime developer setting. 
I keep forgetting to disable the define on commit, this is a better
solution.
 2021-12-19 22:50:14 +01:00
Henrik Rydgård 748b8287a6 Add format string support to Vulkan log-profiler 2021-12-19 22:21:34 +01:00
Henrik Rydgård 2c38a14ad6 Merge pull request #15243 from unknownbrackets/icache-hash 
jit: Use a hash for invalidate/clear all
 2021-12-19 20:59:56 +01:00
Unknown W. Brackets 00f9707399 jit: Use a hash for invalidate/clear all. 
This should make sceKernelICacheClearAll() more useful.
 2021-12-19 10:40:33 -08:00
Unknown W. Brackets 3a61119e77 Merge pull request #15242 from hrydgard/unknownbrackets-vk-profiler 
Vulkan: Disable profiler by default again
 2021-12-19 10:15:35 -08:00
Unknown W. Brackets 148ca08047 Vulkan: Disable profiler by default again. 2021-12-19 09:12:58 -08:00
Henrik Rydgård a9dcf482c6 Improve the check for "slow" GPUs (with regards to tex scaling). See #15238 2021-12-18 21:51:33 +01:00
Henrik Rydgård 1e9e3376be Merge pull request #15238 from hrydgard/hardware-texscale-limits 
Vulkan: Be more restrictive about hardware texture upscaling on "slow" GPUs
 2021-12-18 16:47:43 +01:00
Henrik Rydgård 2851065889 Merge pull request #15237 from unknownbrackets/path-case 
Path: Check for PSP case insensitively
 2021-12-17 09:35:19 +01:00
Unknown W. Brackets 3e73ce87ee Path: Check for PSP case insensitively. 
This allows a game to look up ms0:/psp/ eve with the PSP special case path
handling.
 2021-12-16 15:06:15 -08:00
Henrik Rydgård 80ae4b039c Vulkan: Be more restrictive about texture upscaling on "slow" devices. 
Applies the same rules as for software upscaling in this case.

Should fix the stutters seen in #15109
 2021-12-16 21:00:49 +01:00
Henrik Rydgård d294b6a1ca Merge pull request #15235 from hrydgard/unknownbrackets-cull-remove 
GPU: Remove explicit rect/line depth cull
 2021-12-14 08:57:23 +01:00
Unknown W. Brackets 001d67b711 GPU: Remove explicit rect/line depth cull. 
This appears to be breaking NFS (#15129) and isn't fully correct since
the triangles are still later checked anyway.
 2021-12-13 23:07:26 -08:00
Henrik Rydgård c07068f89b Fix text wrapping on PromptScreen by improving the layout 2021-12-13 22:42:03 +01:00
Henrik Rydgård ff887b0f11 Vulkan: Don't generate mipmaps for huge textures (probably menu backgrounds). Other minor fixes. 2021-12-13 09:13:48 +01:00
Henrik Rydgård 9755781704 Oops, accidentally enabled the logging GPU profiler by default. 2021-12-13 09:01:13 +01:00
Henrik Rydgård cd8493f470 Merge pull request #15230 from hrydgard/4xbrz-opt 
Vulkan: 4xBRZ texture upscaling shader optimizations
 2021-12-12 23:49:41 +01:00
Henrik Rydgård c1a41bef72 Merge pull request #15231 from unknownbrackets/debugger-debounce 
Debounce debugger window updates
 2021-12-12 23:26:39 +01:00
Henrik Rydgård 0e65b86e0a Address feedback (preserve alpha-ignoring equality checks) 2021-12-12 23:21:28 +01:00
Henrik Rydgård b99a5b6613 Minor fixes 2021-12-12 23:21:28 +01:00
Henrik Rydgård 5adf23280c Turn video upscaling off. 2021-12-12 23:21:28 +01:00
Henrik Rydgård fb3b7f8e23 Optimize the xBRZ texture scaling shaders by 30% on Adreno, less on Mali 2021-12-12 23:21:26 +01:00
Henrik Rydgård dc9816ceac Add hardcoded bool toggle for upscaling video, fix log, remove a comment, minor opt 2021-12-12 23:20:57 +01:00
Henrik Rydgård 628363e303 fmt 2021-12-12 23:20:56 +01:00
Henrik Rydgård 10db19efc4 Add a 2x2 downsampled version of 4xBRZ as 2xBRZ (save memory bandwidth) 
fmt
 2021-12-12 23:19:29 +01:00
Henrik Rydgård 7e097084a6 Make a copy of tex_4xbrz 2021-12-12 23:07:11 +01:00
Unknown W. Brackets 7fe9c4e519 Debugger: Debounce updates to disasm text. 
This is a bit slow, so deferring it improves update speed.
 2021-12-12 11:57:24 -08:00
Unknown W. Brackets 045fc0f2b4 Debugger: Never wait inactive on window update. 
This blocks the UI, and we always get a message when stepping is actually
active anyway.  More importantly, we PostMessage() debugger state, so we
might've already resumed.
 2021-12-12 11:36:06 -08:00
Unknown W. Brackets db2eddcf27 Debugger: Remove double debug window updates. 
Core_EnableStepping() is already marking the debug windows for update,
there's no need to call SetDebugMode() again.
 2021-12-12 11:24:03 -08:00
Unknown W. Brackets c57e5b081d Debugger: Debounce list control updates. 2021-12-12 11:03:19 -08:00