-
Notifications
You must be signed in to change notification settings - Fork 2.2k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Me and My Katamari Visual Bug Background #6411
Comments
Does that mean #1853 is fixed? -[Unknown] |
Actually yes, sorry i having problems signing in that account. So that issue for me turned out to be iso problem probably due to compressing it smaller or something There is still a bug that "crashes" the game thay still remains for all devices ive tried . After you finish the tutorial and are asked to name your island it just hangs but music is playing but there is no input. You can fast forward and theusic speeds up accordingly but you cannot procede unless you use a save state which are all 100% done and take away from playing |
That hang can be passed if you change rendering mode to read framebuffers to memory;3 |
@LunaMoo your EU save works fine in the US version. What's the easiest way for a Katamari n00b to experience this issue? I tried the first couple levels but neither was outside, which I'm guessing is the reason this wasn't evident. I have some code to probably implement depth download on desktop, although detecting the "right" depth is tricky. I'd also like to try hardcoding depth=0 on texture and see how it looks (which is what mobile may be locked to.) Would possibly use zero depth when "block transfer" is disabled OR "disable slow effects" is enabled. Though I'm not sure how slow it will be, I don't expect downloading depth to RAM to be a thing GPUs like. -[Unknown] |
Okay, confirmed (not shocking) that downloading makes it work. One trick with zero: it's actually (of course) using a CLUT16 here. CLUT=0 is 100%, so reading the texture data as zero would cause the artifacting. We'd want to set the texture as zero instead. Also, we can't just download at the end of the frame (a la "read framebuffers to memory"), because it will read zero. The game uses the depth, and then right after texturing from it, clears the depth. It seems mostly doable to detect depth. One thing I think I've seen before that WILL get in the way: turning off depth write and using the depth space as a temporary color buffer / or texture memory. I think it will be enough to prioritize color buffers, hopefully... -[Unknown] |
By the way, making depth clears write to memory (as in #8994 with color) makes it solid red. as mentioned here (because it clears to zero, I guess.) I wonder if it'd help artifacts in any other game, though... -[Unknown] |
Hi! I think I am having this Issue, is it fixed? How could one help maybe :) Thanks! |
This issue is still happening in Gold v1.5.4 with ULUS10094. When you roll the Katamari large, especially in stages where you start large (like the Cicada stage indicated above) a colored wall creeps into your vision and blocks your view Shot of real hardware in the same location for comparison None of the mitigations mentioned above seem to help. Block Transfer is disabled and Disable Slow Effects is enabled. I also tried turning off any texture scaling, but the problem still exists. A savestate is attached if it will help. |
Yeah, the game still has serious problems. There are at least two issues. First, it uses the depth buffer as a texture looking up into a palette to do custom fog, which is what you're seeing it fail to do here. PPSSPP does not support that currently, I have a rough implementation in a branch but there were some problems I couldn't explain at the time, should probably give it another shot. Secondly, for window transparency in a few stages it does something similar but textures from a texture map that's black while it really shouldn't be. I dove deep into it a while ago but failed to figure it out... |
has anyone found a fix for this? |
|
Same happens in latest build, they removed framebuffer to memory in version 1.6.0 |
Read framebuffers to memory was only passing a long fixed hang, did nothing to this glitch if not making it worse. The glitchy effect from here can be removed by my cwcheat workaround ~ https://github.com/LunaMoo/PPSSPP_workarounds |
But how the glitchy effects to drawn? It's doesn't present on a real PSP. |
It's downloading the depth buffer and using that as a texture. This is entirely separate from what that setting did, which was related to color buffers and is now essentially automatic. I haven't seen it mentioned here, but software rendering (which will be pretty slow) should render this correctly. -[Unknown] |
Thank you :-) I finally have been able to play the game fully with double stick config (remapping to second gamepad stick the cross,square,triagle,circle buttons) and in hd, thanks again! |
Hello!
Hello! Downloaded your workaround and followed the instructions and it still showing the glitch. Im currently using PPSSPP v1.9.3 |
I have no idea about switch, sorry, but the port to that platform seems to still be WIP and unfinished. Maybe cheats just doesn't work there, or something's wrong with jit invalidation as cheats which modify the game code requires it. |
Thanks for the answer! |
This will not be fixable with D3D9 and older OpenGL devices, but I hope to get in fixes for D3D11 and OpenGL 3.0 before v1.11 release. |
The last one is OpenGL ES 3.0+. but why it's not fixable on GLES2 and D3D9 Devices? |
After #14042 Here's a comparison for my PC and my android phone: All versions tested are using latest build. |
Yeah, that fix isn't working for OpenGL on Android, that's known unfortunately :( |
I'm still experiencing this on the latest PPSSPP version and the EU Version of Me & My Katamari. I tried switching to Vulkan but this still occurs and since the cheat file only seems to work on the US and JP version is there anything I can do? |
@DerGefallene what device are you using? |
Just my Windows PC |
@DerGefallene PC specs? |
That's extremely weird, this should work on any Vulkan device. What GPU do you have? And are you really running a fresh-enough PPSSPP version? |
Windows 10 21H2
My PPSSPP version is always updating itself so I figured it would be the latest version. I have been using it for some years tho so maybe I should've tried to manually update it. At first I had been using D3D11 and the above visual bug occured and when I switched to Vulkan the slightly transparent texture becomes completely solid so after reaching a certain height I can basically only see a colored texture around my Katamari and nothing else |
Are you using latest build? |
That was the issue. Like I said I always thought since PPSSPP was constantly updating itself that it would be up to date. Turns out my PPSSPP version was 1.10.3. Now I've downloaded 1.13.2 and there it is fixed. I tried copying my 1.10.3 PPSSPP savestate of a game level where the bug occured and it also occurs there. But if I load up the game normally on 1.13.2 then the bug doesn't occur. |
It's not actually updating itself at all :) Closing. |
Well, the main fix was long ago, I just extended it to work with OpenGL ES, I think... |
Hey guys, so tbis issue has been around since ive been trying the past year. I posted the issue in the forums but maybe this is a better spot to be seen. Im running 9.8 1385 and the problem remains in buffered and unbuffered rendering
Only happens when you are larger when the camera is in tight the glitch does not remain even off in the distance
Im having difficulty getting a log. Logcat apl dods not record anything when i leave and run ppsspp and come back to it. I do not have a computer at the moment either. I am rooted though.
The text was updated successfully, but these errors were encountered: