[GH-ISSUE #4179] Everybody Golf.... Graphics problem. #1694

Open
opened 2026-03-18 02:42:35 +03:00 by kerem · 59 comments
Owner

Originally created by @SimonPPC on GitHub (Oct 14, 2013).
Original GitHub issue: https://github.com/hrydgard/ppsspp/issues/4179

Hi

this is my first report.
Sorry for my bad English.

In the videogame Everybody Golf there are some remaining errors.

The characters and the ball appear higher than they should.

http://s23.postimg.org/966b6bc3v/Logerror.jpg

http://s10.postimg.org/l1dmbut49/screen00000.jpg

http://s10.postimg.org/s5vfkw0dl/screen00001.jpg

Originally created by @SimonPPC on GitHub (Oct 14, 2013). Original GitHub issue: https://github.com/hrydgard/ppsspp/issues/4179 Hi this is my first report. Sorry for my bad English. In the videogame Everybody Golf there are some remaining errors. The characters and the ball appear higher than they should. http://s23.postimg.org/966b6bc3v/Logerror.jpg http://s10.postimg.org/l1dmbut49/screen00000.jpg http://s10.postimg.org/s5vfkw0dl/screen00001.jpg
Author
Owner

@hrydgard commented on GitHub (Oct 14, 2013):

Do you have pictures from a real PSP of how it should look like?

<!-- gh-comment-id:26251093 --> @hrydgard commented on GitHub (Oct 14, 2013): Do you have pictures from a real PSP of how it should look like?
Author
Owner
<!-- gh-comment-id:26252447 --> @SimonPPC commented on GitHub (Oct 14, 2013): From a real psp this is similar http://s9.postimg.org/4911ru4b3/everybody_s_golf_psp_review_1.jpg http://s2.postimg.org/n3i3m3imx/everygolf2psp_001_large.jpg
Author
Owner

@xsacha commented on GitHub (Oct 17, 2013):

Does it look different when you disable/enable buffered rendering?

<!-- gh-comment-id:26485176 --> @xsacha commented on GitHub (Oct 17, 2013): Does it look different when you disable/enable buffered rendering?
Author
Owner

@SimonPPC commented on GitHub (Oct 17, 2013):

Yes, I've tried various options and systematically cross and the problem remains.

<!-- gh-comment-id:26492947 --> @SimonPPC commented on GitHub (Oct 17, 2013): Yes, I've tried various options and systematically cross and the problem remains.
Author
Owner

@unknownbrackets commented on GitHub (Jan 11, 2014):

Has this improved or changed at all? How does it look with "software rendering" enabled (it will be very slow)?

-[Unknown]

<!-- gh-comment-id:32105684 --> @unknownbrackets commented on GitHub (Jan 11, 2014): Has this improved or changed at all? How does it look with "software rendering" enabled (it will be very slow)? -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (Jan 12, 2014):

I have test with the last version v0.9.6-409...

Download at SpeedyShare

The error remains the same.

The speed is excellent in all modes in my system

<!-- gh-comment-id:32111402 --> @SimonPPC commented on GitHub (Jan 12, 2014): I have test with the last version v0.9.6-409... <a href="http://speedy.sh/KyYhW/ppsspplog.txt">Download at SpeedyShare</a> The error remains the same. The speed is excellent in all modes in my system
Author
Owner

@unknownbrackets commented on GitHub (Jan 12, 2014):

Hmm, "Vertex decoder JIT failed! fmt = 000801bc".

Software rendering is the last option in the graphics section of the settings. So even with that the graphics problems are the same?

-[Unknown]

<!-- gh-comment-id:32111680 --> @unknownbrackets commented on GitHub (Jan 12, 2014): Hmm, "Vertex decoder JIT failed! fmt = 000801bc". Software rendering is the last option in the graphics section of the settings. So even with that the graphics problems are the same? -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (Jan 12, 2014):

With software rendering the problem seems to disappear, unfortunately I can not get up to play the emulator crashes ..
But where you see the character for a change of clothes is in the right position.

New log
Download at SpeedyShare

<!-- gh-comment-id:32111966 --> @SimonPPC commented on GitHub (Jan 12, 2014): With software rendering the problem seems to disappear, unfortunately I can not get up to play the emulator crashes .. But where you see the character for a change of clothes is in the right position. New log <a href="http://speedy.sh/DT9PQ/Log.zip">Download at SpeedyShare</a>
Author
Owner

@unknownbrackets commented on GitHub (Jun 9, 2014):

Has this improved in the latest git build?

-[Unknown]

<!-- gh-comment-id:45454934 --> @unknownbrackets commented on GitHub (Jun 9, 2014): Has this improved in the latest git build? -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (Jun 9, 2014):

In the latest version (1088) the problem is the same.
In addition, if they have added another that was not there before.

When I press the X button to activate the indicator as in the screen goes grab ...
UCES00767 00000

<!-- gh-comment-id:45504520 --> @SimonPPC commented on GitHub (Jun 9, 2014): In the latest version (1088) the problem is the same. In addition, if they have added another that was not there before. When I press the X button to activate the indicator as in the screen goes grab ... <a href='http://postimg.org/image/c9y21bmk5/' target='_blank'><img src='http://s18.postimg.org/c9y21bmk5/UCES00767_00000.jpg' border='0' alt="UCES00767 00000" /></a>
Author
Owner

@hrydgard commented on GitHub (Jun 9, 2014):

That looks like a stride mismatch (512 vs 480) of some sort... Is that with plain buffered rendering + block transfer?

<!-- gh-comment-id:45504877 --> @hrydgard commented on GitHub (Jun 9, 2014): That looks like a stride mismatch (512 vs 480) of some sort... Is that with plain buffered rendering + block transfer?
Author
Owner

@SimonPPC commented on GitHub (Jun 9, 2014):

With Buffer Rendering or Not-Buffer Rendering the problem appears in any graphics mode.
With Framebuffer in memory (CPU) or (GPU) the problem does not appear

<!-- gh-comment-id:45506303 --> @SimonPPC commented on GitHub (Jun 9, 2014): With Buffer Rendering or Not-Buffer Rendering the problem appears in any graphics mode. With Framebuffer in memory (CPU) or (GPU) the problem does not appear
Author
Owner

@unknownbrackets commented on GitHub (Jun 9, 2014):

Does it happen with "simulate block transfer" unchecked?

-[Unknown]

<!-- gh-comment-id:45518146 --> @unknownbrackets commented on GitHub (Jun 9, 2014): Does it happen with "simulate block transfer" unchecked? -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (Jun 9, 2014):

I do not know where that option using the Italian version ...

<!-- gh-comment-id:45522982 --> @SimonPPC commented on GitHub (Jun 9, 2014): I do not know where that option using the Italian version ...
Author
Owner

@unknownbrackets commented on GitHub (Jun 9, 2014):

It's right under "Rendering mode" "buffered rendering".

-[Unknown]

<!-- gh-comment-id:45527987 --> @unknownbrackets commented on GitHub (Jun 9, 2014): It's right under "Rendering mode" "buffered rendering". -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (Jun 10, 2014):

If I remove the option "simulated transfer block" is a still image of a scene previous in the game ...

<!-- gh-comment-id:45578046 --> @SimonPPC commented on GitHub (Jun 10, 2014): If I remove the option "simulated transfer block" is a still image of a scene previous in the game ...
Author
Owner

@SimonPPC commented on GitHub (Jun 10, 2014):

or a black screen if I restart the emulator...

<!-- gh-comment-id:45578286 --> @SimonPPC commented on GitHub (Jun 10, 2014): or a black screen if I restart the emulator...
Author
Owner

@unknownbrackets commented on GitHub (Jul 20, 2014):

Has this improved at all in the latest git build with "simulate block transfer" enabled?

-[Unknown]

<!-- gh-comment-id:49537854 --> @unknownbrackets commented on GitHub (Jul 20, 2014): Has this improved at all in the latest git build with "simulate block transfer" enabled? -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (Jul 20, 2014):

The game suffers from slowdown already different versions.
In addition you added a graphical issue at a particular time of the game
Playing with the ball on the green.

Thank you that you are continuing to develop this emulator which I think is really great!

<!-- gh-comment-id:49538880 --> @SimonPPC commented on GitHub (Jul 20, 2014): The game suffers from slowdown already different versions. In addition you added a graphical issue at a particular time of the game Playing with the ball on the green. Thank you that you are continuing to develop this emulator which I think is really great!
Author
Owner

@ppmeis commented on GitHub (Jul 25, 2015):

Tested with latest build. Same issues. It also affects Everybody's Golf 2:
image
image

Proper:
01285-03

<!-- gh-comment-id:124901771 --> @ppmeis commented on GitHub (Jul 25, 2015): Tested with latest build. Same issues. It also affects Everybody's Golf 2: ![image](https://cloud.githubusercontent.com/assets/4381277/8891405/65011be6-3328-11e5-8c96-5dc7ff42923c.png) ![image](https://cloud.githubusercontent.com/assets/4381277/8891406/6af2489a-3328-11e5-9aaf-341227678f08.png) Proper: ![01285-03](https://cloud.githubusercontent.com/assets/4381277/8891408/8c468f1a-3328-11e5-93ea-b8c55415a523.jpg)
Author
Owner

@FlatOutPS2 commented on GitHub (Apr 13, 2016):

This issue has been fixed. I did encounter it in 1.1.1, but as of 1.2.1 both the characters in the introductions and ball in the bottom right are shown as it does on the PSP.

Only the character portraits are still empty, as can be seen in the screenshots posted above in the empty frames near the top left and/or right.

<!-- gh-comment-id:209403802 --> @FlatOutPS2 commented on GitHub (Apr 13, 2016): This issue has been fixed. I did encounter it in 1.1.1, but as of 1.2.1 both the characters in the introductions and ball in the bottom right are shown as it does on the PSP. Only the character portraits are still empty, as can be seen in the screenshots posted above in the empty frames near the top left and/or right.
Author
Owner

@unknownbrackets commented on GitHub (Apr 14, 2016):

I guess that's probably a block transfer we don't detect. Can you try to find it using the GE debugger?

https://github.com/hrydgard/ppsspp/wiki/How-to-find-a-graphic-issue-with-the-GE-debugger

Glad to hear the first issue is fixed.

-[Unknown]

<!-- gh-comment-id:209739344 --> @unknownbrackets commented on GitHub (Apr 14, 2016): I guess that's probably a block transfer we don't detect. Can you try to find it using the GE debugger? https://github.com/hrydgard/ppsspp/wiki/How-to-find-a-graphic-issue-with-the-GE-debugger Glad to hear the first issue is fixed. -[Unknown]
Author
Owner

@FlatOutPS2 commented on GitHub (Apr 14, 2016):

I found two possible candidates with GE debugger. One with Texture shown as disabled and one with a transparent, empty texture:
EG Tex disabled.txt
EG Tex transparent.txt

<!-- gh-comment-id:209874758 --> @FlatOutPS2 commented on GitHub (Apr 14, 2016): I found two possible candidates with GE debugger. One with Texture shown as disabled and one with a transparent, empty texture: [EG Tex disabled.txt](https://github.com/hrydgard/ppsspp/files/218921/EG.Tex.disabled.txt) [EG Tex transparent.txt](https://github.com/hrydgard/ppsspp/files/218922/EG.Tex.transparent.txt)
Author
Owner

@unknownbrackets commented on GitHub (Apr 17, 2016):

Yeah, 040cc040 seems likely to be where it's trying to draw the portrait from.

If you get to that area, and then open Debug -> Disassembly, and hit Stop, it'll pause the emulator. This is the CPU debugger.

From there, hit "Breakpoint", and enter "040cc040" in for the Address. Under Size, put 0x100. Uncheck "Read", and then hit OK. Now, click the "Go" button (which is where the "Stop" button was before.)

If the game is writing to that area from the CPU, it'll pause again (and change back to "Go".) If that happens, can you take a screenshot of the Disassembly window? I want to see the code around where it's happening.

It may also not trip. If it doesn't, that means it's not drawing the portrait on the CPU, so it must be something else.

-[Unknown]

<!-- gh-comment-id:210940224 --> @unknownbrackets commented on GitHub (Apr 17, 2016): Yeah, 040cc040 seems likely to be where it's trying to draw the portrait from. If you get to that area, and then open Debug -> Disassembly, and hit Stop, it'll pause the emulator. This is the CPU debugger. From there, hit "Breakpoint", and enter "040cc040" in for the Address. Under Size, put 0x100. Uncheck "Read", and then hit OK. Now, click the "Go" button (which is where the "Stop" button was before.) If the game is writing to that area from the CPU, it'll pause again (and change back to "Go".) If that happens, can you take a screenshot of the Disassembly window? I want to see the code around where it's happening. It may also not trip. If it doesn't, that means it's not drawing the portrait on the CPU, so it must be something else. -[Unknown]
Author
Owner

@FlatOutPS2 commented on GitHub (Apr 17, 2016):

I didn't need to press Go after clicking OK on setting up the breakpoint, it resumes automatically. But it doesn't pause afterwards.

<!-- gh-comment-id:210981158 --> @FlatOutPS2 commented on GitHub (Apr 17, 2016): I didn't need to press Go after clicking OK on setting up the breakpoint, it resumes automatically. But it doesn't pause afterwards.
Author
Owner

@FlatOutPS2 commented on GitHub (Apr 19, 2016):

I finally did manage to trip it. It only seems to trigger it when starting a new round, it doesn't trip on course or while loading the next hole.

eg disassembly

<!-- gh-comment-id:211947595 --> @FlatOutPS2 commented on GitHub (Apr 19, 2016): I finally did manage to trip it. It only seems to trigger it when starting a new round, it doesn't trip on course or while loading the next hole. ![eg disassembly](https://cloud.githubusercontent.com/assets/15942346/14642808/2c1575dc-064c-11e6-8c89-76b95c805b5c.png)
Author
Owner

@unknownbrackets commented on GitHub (Apr 20, 2016):

Okay, well, that looks like it may be some form of memcpy, seems to be copying a1 (being a2 originally) 16-bit values. We aren't detecting that it's trying to write to a render surface there, so I guess we could try hooking that function.

-[Unknown]

<!-- gh-comment-id:212246305 --> @unknownbrackets commented on GitHub (Apr 20, 2016): Okay, well, that looks like it may be some form of memcpy, seems to be copying a1 (being a2 originally) 16-bit values. We aren't detecting that it's trying to write to a render surface there, so I guess we could try hooking that function. -[Unknown]
Author
Owner

@FlatOutPS2 commented on GitHub (Jul 7, 2016):

Okay, well, that looks like it may be some form of memcpy, seems to be copying a1 (being a2 originally) 16-bit values. We aren't detecting that it's trying to write to a render surface there, so I guess we could try hooking that function.

Has anyone looked into this yet?

<!-- gh-comment-id:231069981 --> @FlatOutPS2 commented on GitHub (Jul 7, 2016): > Okay, well, that looks like it may be some form of memcpy, seems to be copying a1 (being a2 originally) 16-bit values. We aren't detecting that it's trying to write to a render surface there, so I guess we could try hooking that function. Has anyone looked into this yet?
Author
Owner

@TechLucky commented on GitHub (Mar 17, 2017):

This will do the work...
http://ppssppbestconfig.blogspot.se/
[adfly removed]

<!-- gh-comment-id:287304880 --> @TechLucky commented on GitHub (Mar 17, 2017): This will do the work... http://ppssppbestconfig.blogspot.se/ [adfly removed]
Author
Owner

@hrydgard commented on GitHub (Mar 17, 2017):

@TechLucky Please stop spamming adfly links.

<!-- gh-comment-id:287307414 --> @hrydgard commented on GitHub (Mar 17, 2017): @TechLucky Please stop spamming adfly links.
Author
Owner

@unknownbrackets commented on GitHub (May 15, 2017):

Does it render the portrait in the software renderer? It's slow, but that'll help confirm the problem.

-[Unknown]

<!-- gh-comment-id:301362929 --> @unknownbrackets commented on GitHub (May 15, 2017): Does it render the portrait in the software renderer? It's slow, but that'll help confirm the problem. -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (May 15, 2017):

I use 1.4-200 with the Software Renderer... Confirm the problem.
immagine

<!-- gh-comment-id:301521341 --> @SimonPPC commented on GitHub (May 15, 2017): I use 1.4-200 with the Software Renderer... Confirm the problem. ![immagine](https://cloud.githubusercontent.com/assets/5682097/26066859/a195ca66-3998-11e7-9730-f3f82b407091.png)
Author
Owner

@unknownbrackets commented on GitHub (May 15, 2017):

Aha - if that's not with a savestate, it probably means we're mis-emulating something (maybe even HLE jpeg funcs.) Software rendering doesn't have memcpy issues, so it theoretically can't be that.

-[Unknown]

<!-- gh-comment-id:301523502 --> @unknownbrackets commented on GitHub (May 15, 2017): Aha - if that's not with a savestate, it probably means we're mis-emulating something (maybe even HLE jpeg funcs.) Software rendering doesn't have memcpy issues, so it theoretically can't be that. -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (May 15, 2017):

Not with savestate... I have load the game ;)

<!-- gh-comment-id:301557815 --> @SimonPPC commented on GitHub (May 15, 2017): Not with savestate... I have load the game ;)
Author
Owner

@unknownbrackets commented on GitHub (May 16, 2017):

Thanks for confirming, and for baring with the slowness of the software renderer. So that means the memcpy won't help...

There's no reported functions though:
http://report.ppsspp.org/logs/game/UCES00012_1.00

Hm.

-[Unknown]

<!-- gh-comment-id:301643669 --> @unknownbrackets commented on GitHub (May 16, 2017): Thanks for confirming, and for baring with the slowness of the software renderer. So that means the memcpy won't help... There's no reported functions though: http://report.ppsspp.org/logs/game/UCES00012_1.00 Hm. -[Unknown]
Author
Owner

@SimonPPC commented on GitHub (May 16, 2017):

Thanks to all of you that you continue to develop this emulator!

<!-- gh-comment-id:301851104 --> @SimonPPC commented on GitHub (May 16, 2017): Thanks to all of you that you continue to develop this emulator!
Author
Owner

@FlatOutPS2 commented on GitHub (Dec 7, 2017):

The portraits are now visible with the software renderer in v1.5.0 builds. No change for D3D9/11 and OGL.

<!-- gh-comment-id:349817917 --> @FlatOutPS2 commented on GitHub (Dec 7, 2017): The portraits are now visible with the software renderer in v1.5.0 builds. No change for D3D9/11 and OGL.
Author
Owner

@FlatOutPS2 commented on GitHub (Dec 7, 2017):

eg

A screenshot of the character portrait being visible at the top left with the software renderer. Though the software renderer does have other issue like missing grass textures (the grey/blue section at the bottom) and a line of garbage at the very bottom and right of the screen. These two issues are not there with any of the hardware renderers.

<!-- gh-comment-id:349986467 --> @FlatOutPS2 commented on GitHub (Dec 7, 2017): ![eg](https://user-images.githubusercontent.com/15942346/33720627-cb0fde2c-db64-11e7-9b9d-7d0461368550.jpg) A screenshot of the character portrait being visible at the top left with the software renderer. Though the software renderer does have other issue like missing grass textures (the grey/blue section at the bottom) and a line of garbage at the very bottom and right of the screen. These two issues are not there with any of the hardware renderers.
Author
Owner

@unknownbrackets commented on GitHub (Dec 9, 2017):

The missing grass is a long standing bug in the software renderer. The garbage though may be new, oops...

Anyway, if it's possible to find the git build that fixed the portrait (you can go by halves, starting with the one halfway between 1.4.2 and 1.5.0, for example, it would only take about 8-10 tests), it would help significantly. If we can find what change fixed the bug in software, it should point very suggestively to the bug in Vulkan/OpenGL/etc.

-[Unknown]

<!-- gh-comment-id:350498997 --> @unknownbrackets commented on GitHub (Dec 9, 2017): The missing grass is a long standing bug in the software renderer. The garbage though may be new, oops... Anyway, if it's possible to find the git build that fixed the portrait (you can go by halves, starting with the one halfway between 1.4.2 and 1.5.0, for example, it would only take about 8-10 tests), it would help significantly. If we can find what change fixed the bug in software, it should point very suggestively to the bug in Vulkan/OpenGL/etc. -[Unknown]
Author
Owner

@FlatOutPS2 commented on GitHub (Dec 9, 2017):

I just noticed that Everybody's Golf 2/ HSG Open Tee 2(the version I used for the above screenshot) used to crash with the software renderer. And applying the fix for that crash to PPSSPP 1.4.0 showed the portraits would have been visible back then(the software renderer garbage does indeed seem to be a new/worsened regression introduced between 1.4.0 and 1.5.0). So there hasn't been an improvement to portraits in EG2, but there's a difference between the two PSP games when using the software renderer(both games lack the portraits with the other renderers). I removed the ISO of EG1 a while ago, so I can't tell if there has been an improvement for that game right now. I'll have to look for the UMD.

<!-- gh-comment-id:350504869 --> @FlatOutPS2 commented on GitHub (Dec 9, 2017): I just noticed that Everybody's Golf 2/ HSG Open Tee 2(the version I used for the above screenshot) used to crash with the software renderer. And applying the fix for that crash to PPSSPP 1.4.0 showed the portraits would have been visible back then(the software renderer garbage does indeed seem to be a new/worsened regression introduced between 1.4.0 and 1.5.0). So there hasn't been an improvement to portraits in EG2, but there's a difference between the two PSP games when using the software renderer(both games lack the portraits with the other renderers). I removed the ISO of EG1 a while ago, so I can't tell if there has been an improvement for that game right now. I'll have to look for the UMD.
Author
Owner

@nassau-tk commented on GitHub (Aug 10, 2021):

v1.11.3-1101-gf233c7726 still missing the portrait.
★OpenGL/Vulkan/D3D 11 all
image

And, If use software rendering then it will happpen Invalid Memory Access error in some scene.
Example : Character select / Tee shot scene
image

<!-- gh-comment-id:896377604 --> @nassau-tk commented on GitHub (Aug 10, 2021): v1.11.3-1101-gf233c7726 still missing the portrait. ★OpenGL/Vulkan/D3D 11 all ![image](https://user-images.githubusercontent.com/48179091/128946531-11ab8afa-b827-45b1-a12a-a5c696168ae6.png) And, If use software rendering then it will happpen Invalid Memory Access error in some scene. Example : Character select / Tee shot scene ![image](https://user-images.githubusercontent.com/48179091/128946704-73523ec1-40bc-42e5-8bdd-b81575cd7f8e.png)
Author
Owner

@unknownbrackets commented on GitHub (Aug 11, 2021):

Can you make a frame dump (from OpenGL/similar) of one of the scenes where the software renderer would crash? I'm hoping the frame dump will reproduce it.

-[Unknown]

<!-- gh-comment-id:896472896 --> @unknownbrackets commented on GitHub (Aug 11, 2021): Can you make a frame dump (from OpenGL/similar) of one of the scenes where the software renderer would crash? I'm hoping the frame dump will reproduce it. -[Unknown]
Author
Owner

@nassau-tk commented on GitHub (Aug 11, 2021):

@unknownbrackets

Um.....
I tried getting the GEdump.
But, OpnGL/Vulkan/SoftwareRendering all backends cant appear any picture.
Why? Did I wrong way???

UCJS10001_DUMP.zip

<!-- gh-comment-id:896508309 --> @nassau-tk commented on GitHub (Aug 11, 2021): @unknownbrackets Um..... I tried getting the GEdump. But, OpnGL/Vulkan/SoftwareRendering all backends cant appear any picture. Why? Did I wrong way??? [UCJS10001_DUMP.zip](https://github.com/hrydgard/ppsspp/files/6966016/UCJS10001_DUMP.zip)
Author
Owner

@ghost commented on GitHub (Sep 17, 2021):

SW Renderer doesn't work anymore on this game, Im using the latest git build commit github.com/hrydgard/ppsspp@cd05f3ad29
Please solve this issue on sw renderer.

Edit: Commit github.com/hrydgard/ppsspp@e6a2da69c5 sw renderer is working again on this game.

<!-- gh-comment-id:921881485 --> @ghost commented on GitHub (Sep 17, 2021): SW Renderer doesn't work anymore on this game, Im using the latest git build commit https://github.com/hrydgard/ppsspp/commit/cd05f3ad29dd80ef1f559e42b24491f27030a823 Please solve this issue on sw renderer. Edit: Commit https://github.com/hrydgard/ppsspp/commit/e6a2da69c5dde977fecb468cc7906addfca1fb3e sw renderer is working again on this game.
Author
Owner

@ghost commented on GitHub (Sep 17, 2021):

SW Renderer
Screenshot_2021-09-17-23-38-58-031_org ppsspp ppsspp
VK & OGL Renderer
Screenshot_2021-09-17-23-42-05-601_org ppsspp ppsspp
using savestate from sw rendering.

DUMP

ppsspp dump.zip

<!-- gh-comment-id:921901822 --> @ghost commented on GitHub (Sep 17, 2021): SW Renderer ![Screenshot_2021-09-17-23-38-58-031_org ppsspp ppsspp](https://user-images.githubusercontent.com/90336524/133817289-954c434c-f862-418e-9d5d-91dd769a7d2f.jpg) VK & OGL Renderer ![Screenshot_2021-09-17-23-42-05-601_org ppsspp ppsspp](https://user-images.githubusercontent.com/90336524/133817349-3ea7572f-5abc-4d41-b0b7-897afd3c8538.jpg) using savestate from sw rendering. ### DUMP [ppsspp dump.zip](https://github.com/hrydgard/ppsspp/files/7186873/ppsspp.dump.zip)
Author
Owner

@ghost commented on GitHub (Sep 19, 2021):

@SimonPPC outdated no feedback?

<!-- gh-comment-id:922461981 --> @ghost commented on GitHub (Sep 19, 2021): @SimonPPC outdated no feedback?
Author
Owner

@SimonPPC commented on GitHub (Sep 19, 2021):

@SimonPPC outdated no feedback?

Hi, I haven't used the emulator for a long time.

<!-- gh-comment-id:922479766 --> @SimonPPC commented on GitHub (Sep 19, 2021): > > > @SimonPPC outdated no feedback? Hi, I haven't used the emulator for a long time.
Author
Owner

@ghost commented on GitHub (Oct 10, 2022):

Can still reproduce this issue using ppsspp build github.com/hrydgard/ppsspp@bc9215ca93
Simulate Block Transfer Effects is enable I try also to disable/enable hw transform no luck.

Vulkan/OpenGL
Screenshot_2022-10-10-13-52-54-50

Software
Screenshot_2022-10-10-13-52-03-82

REAL HW
Screenshot_2022-10-10-13-50-57-55

Dump:
UCES00012.ppdmp.zip

<!-- gh-comment-id:1272817384 --> @ghost commented on GitHub (Oct 10, 2022): Can still reproduce this issue using ppsspp build https://github.com/hrydgard/ppsspp/commit/bc9215ca93f139ed80e4d47747086040a8d6608d Simulate Block Transfer Effects is enable I try also to disable/enable hw transform no luck. Vulkan/OpenGL ![Screenshot_2022-10-10-13-52-54-50](https://user-images.githubusercontent.com/37603562/194805618-9319156a-3d53-43a1-86f0-5313bb6f4870.png) Software ![Screenshot_2022-10-10-13-52-03-82](https://user-images.githubusercontent.com/37603562/194805637-cd9df9ee-202b-4193-8046-215022e861de.png) REAL HW ![Screenshot_2022-10-10-13-50-57-55](https://user-images.githubusercontent.com/37603562/194805667-59839e5a-e918-4e02-b4ec-fd2ae1ee8069.png) Dump: [UCES00012.ppdmp.zip](https://github.com/hrydgard/ppsspp/files/9743588/UCES00012.ppdmp.zip)
Author
Owner

@sum2012 commented on GitHub (Nov 21, 2023):

v1.16.6-333-gecab50346 still missing portrait

<!-- gh-comment-id:1820930781 --> @sum2012 commented on GitHub (Nov 21, 2023): v1.16.6-333-gecab50346 still missing portrait
Author
Owner

@sum2012 commented on GitHub (Nov 21, 2023):

v1.4.2 soft gpu crash\hang in-game

<!-- gh-comment-id:1820995074 --> @sum2012 commented on GitHub (Nov 21, 2023): v1.4.2 soft gpu crash\hang in-game
Author
Owner

@hrydgard commented on GitHub (Nov 22, 2023):

Yup, portraits still need investigation. But, hang in v1.4.2? That's a very old version, so not that interesting?

<!-- gh-comment-id:1822293110 --> @hrydgard commented on GitHub (Nov 22, 2023): Yup, portraits still need investigation. But, hang in v1.4.2? That's a very old version, so not that interesting?
Author
Owner

@sum2012 commented on GitHub (Nov 22, 2023):

@hrydgard Because @unknownbrackets asked it

The missing grass is a long standing bug in the software renderer. The garbage though may be new, oops...

Anyway, if it's possible to find the git build that fixed the portrait (you can go by halves, starting with the one halfway between 1.4.2 and 1.5.0, for example, it would only take about 8-10 tests), it would help significantly. If we can find what change fixed the bug in software, it should point very suggestively to the bug in Vulkan/OpenGL/etc.

-[Unknown]

<!-- gh-comment-id:1823186936 --> @sum2012 commented on GitHub (Nov 22, 2023): @hrydgard Because @unknownbrackets asked it > The missing grass is a long standing bug in the software renderer. The garbage though may be new, oops... > > Anyway, if it's possible to find the git build that fixed the portrait (you can go by halves, starting with the one halfway between 1.4.2 and 1.5.0, for example, it would only take about 8-10 tests), it would help significantly. If we can find what change fixed the bug in software, it should point very suggestively to the bug in Vulkan/OpenGL/etc. > > -[Unknown]
Author
Owner

@sum2012 commented on GitHub (Dec 2, 2023):

portraits maybe saved in the memory from player select screen
1

<!-- gh-comment-id:1837230120 --> @sum2012 commented on GitHub (Dec 2, 2023): portraits maybe saved in the memory from player select screen ![1](https://github.com/hrydgard/ppsspp/assets/2177532/0c83584c-0612-46a6-8521-4a376781d6f5)
Author
Owner

@sum2012 commented on GitHub (Feb 8, 2026):

Can @hrydgard or anyone give instruction to find hash value (better write in wiki) to use in ReplaceTables.cpp ?

Image
<!-- gh-comment-id:3867193672 --> @sum2012 commented on GitHub (Feb 8, 2026): Can @hrydgard or anyone give instruction to find hash value (better write in wiki) to use in ReplaceTables.cpp ? <img width="572" height="395" alt="Image" src="https://github.com/user-attachments/assets/ad92e17a-1591-45b8-89da-de0ea38189da" />
Author
Owner

@hrydgard commented on GitHub (Feb 8, 2026):

you can rename the function then do "save symbol map" to whatever format is text.. I know, we need a much better way :)

<!-- gh-comment-id:3868546336 --> @hrydgard commented on GitHub (Feb 8, 2026): you can rename the function then do "save symbol map" to whatever format is text.. I know, we need a much better way :)
Author
Owner

@sum2012 commented on GitHub (Feb 9, 2026):

@hrydgard
"save symbol map" do not have hash value
08804000 z_un_08804000,0024
08804040 z_un_08804040,0088
088040C8 z_un_088040c8,0060

<!-- gh-comment-id:3871768140 --> @sum2012 commented on GitHub (Feb 9, 2026): @hrydgard "save symbol map" do not have hash value 08804000 z_un_08804000,0024 08804040 z_un_08804040,0088 088040C8 z_un_088040c8,0060
Author
Owner

@hrydgard commented on GitHub (Feb 9, 2026):

Hm, yeah :/

I used to have some method I can't remember, but I'm gonna make a nice way instead soon.

<!-- gh-comment-id:3872044625 --> @hrydgard commented on GitHub (Feb 9, 2026): Hm, yeah :/ I used to have some method I can't remember, but I'm gonna make a nice way instead soon.
Author
Owner

@hrydgard commented on GitHub (Feb 9, 2026):

@sum2012 see #21229

<!-- gh-comment-id:3873276137 --> @hrydgard commented on GitHub (Feb 9, 2026): @sum2012 see #21229
Author
Owner

@sum2012 commented on GitHub (Feb 12, 2026):

I tired 3 days still don't get fixed. There is no Hook tutorial in wiki.

<!-- gh-comment-id:3890878721 --> @sum2012 commented on GitHub (Feb 12, 2026): I tired 3 days still don't get fixed. There is no Hook tutorial in wiki.
Sign in to join this conversation.
No milestone
No project
No assignees
1 participant
Notifications
Due date
The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference
starred/ppsspp#1694
No description provided.