mirror of
https://github.com/hrydgard/ppsspp.git
synced 2026-04-24 21:56:10 +03:00
[GH-ISSUE #1808] Saint Seiya Omega Missing Character (GLES2) #603
Labels
No labels
Atrac3+
Audio
CPU emulation
D3D11
D3D9 (removed)
Depth / Z
Feature Request
Font Atlas
GE emulation
Guardband / Range Culling
HLE/Kernel
I/O
Input/Controller
MP3
Multithreading
Needs hardware testing
Networking/adhoc/infrastructure
No Feedback / Outdated?
OpenGL
PGF / sceFont
PSMF / MPEG
Platform-specific (Android)
Platform-specific (Windows)
Platform-specific (iOS)
PowerVR GPU
SDL2
Saving issue
User Interface
Vulkan
arm64jit
armjit
armv6
x86jit
No milestone
No project
No assignees
1 participant
Notifications
Due date
No due date set.
Dependencies
No dependencies set.
Reference
starred/ppsspp#603
Loading…
Add table
Add a link
Reference in a new issue
No description provided.
Delete branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @kaiser20xx on GitHub (May 14, 2013).
Original GitHub issue: https://github.com/hrydgard/ppsspp/issues/1808
Saint Seiya Omega model become invicible again on this version until now (this work fine on v0.7.6 build).
@hrydgard commented on GitHub (May 16, 2013):
Yes, we know. They're gone after we fixed Ridge Racer, and I really don't understand why :(
@dbz400 commented on GitHub (Jun 6, 2013):
Just for record , that fix also broken Gundam AGE Universe as below
@unknownbrackets commented on GitHub (Jun 6, 2013):
Hmm, can you paste a link to a frame dump for that?
-[Unknown]
@dbz400 commented on GitHub (Jun 6, 2013):
Here is the link of frame dump .Generated using latest source
http://pastebin.com/3Ac6H2zS
@hrydgard commented on GitHub (Jun 6, 2013):
Looks like a typical case of the Z buffer not getting cleared... why, I don't know.
@dbz400 commented on GitHub (Jun 7, 2013):
It requires depthWrite always set to TRUE in clearmode to fix the issue propertly "glstate.depthWrite.set(GL_TRUE);" for both Saint Seiya and Gundam AGE Universe.
@hrydgard commented on GitHub (Jun 7, 2013):
That's a hack, not a proper fix. The game seems to request depth writes to be off, so it doesn't make sense to turn it on.
@papel commented on GitHub (Jul 7, 2013):
This bug is fixed.
@papel commented on GitHub (Sep 3, 2013):
I think this bug is back. :(

@brujo5, can you confirm it?
Reported in:
http://forums.ppsspp.org/showthread.php?tid=362&pid=43125#pid43125
@dbz400 commented on GitHub (Sep 4, 2013):
Yes I can confirm It .
@dbz400 commented on GitHub (Sep 9, 2013):
This issue can be fixed in hacking way by applying '"Always Depth Write' here under hack settings
@papel commented on GitHub (Sep 11, 2013):
@unknownbrackets
The issue is fixed with "Always Depth Write" hack.
http://forums.ppsspp.org/showthread.php?tid=362&pid=47567#pid47567
Since it is a hack, should the bug continue open to remember to fix without hack?
@unknownbrackets commented on GitHub (Sep 11, 2013):
Well, I personally consider "having to use a hack" a bug, so e.g. I would consider bugs that said "Game X only works with buffered rendering off" or "Game X only works with depth write forced on" or etc. valid. Thus, to me, this continues to be a bug unless someone wants to close it just to open a new one and replicate the same detail, but about having to use a hack.
That said, if there are multiple games that work only with this hack, it may be better to put them all into one new bug and close the individual bugs, unless they are just coincidences.
Again, personally, I would only close a bug for good as fixed with a hack if I did not plan/expect to ever fix the issue correctly. For example, if the thing was simply not possible using OpenGL, and we had done the best approximation possible, but had a hack to use an alternative approximation that some games require, then a bug doesn't necessarily help that situation.
-[Unknown]
@dbz400 commented on GitHub (Jan 5, 2014):
This one can be closed :) Depth issue resolved
@hrydgard commented on GitHub (Jan 5, 2014):
Not on Android though I guess... yet.
@brujo5 commented on GitHub (Jan 5, 2014):
fixed on IOS whit hack setting
@unknownbrackets commented on GitHub (Feb 15, 2014):
Does this work any better on mobile now without any hacks enabled? We recently fixed some depth issues.
Or, maybe it still requires the depth buffer to be retained, like Jeanne d'Arc.
-[Unknown]
@hrydgard commented on GitHub (Feb 16, 2014):
This one is fixed by depth buffer copy, IIRC, and is still broken without it.
@brujo5 commented on GitHub (Feb 16, 2014):
yeah this one only work whit hack setting.
@dbz400 commented on GitHub (Apr 24, 2014):
On my Androd 4.3 with OpenGL 3.0 , depth copy is working good .Both Jeanne d'Arc and Saint Seiya can see the characters.
@unknownbrackets commented on GitHub (Apr 24, 2014):
It requires GL_EXT_framebuffer_blit / GLES 3.0. We could maybe try glBlitFramebufferEXT where supported. By the way, this is the case also for the Brave Story issue.
Aside from trying to split the depth to a separately tracked buffer or manage attachments dynamically or something, we could also try blitting using a separate render pass, which is probably basically what glBlitFramebuffer does.
-[Unknown]
@dbz400 commented on GitHub (Apr 28, 2014):
I'm wondering if still any ppl got invisible character here on his mobile device with unsupported GL_EXT_framebuffer_object on ES 2.0?
(i think whenever GL_EXT_framebuffer_object presents , we should use glBlitFramebufferEXT or just simply glBlitFramebuffer?)
@dbz400 commented on GitHub (Apr 28, 2014):
Humm seen a user in forum mentioning it is not working for him and he is using Tegra Note 7 ......(Tegra 4 Soc)
@brujo5 commented on GitHub (Apr 29, 2014):
@unknownbrackets
@raven02
still need force depth write hack enabled on iphone 5S
also why show ES 2.0 in system information?
iphone 5S is ES 3.0
@hrydgard commented on GitHub (Apr 29, 2014):
Hm, I think we need to add some code to request ES 3.0 support when initializing OGL on iOS.
@brujo5 commented on GitHub (Apr 29, 2014):
sounds good.
TIA henrik
@dbz400 commented on GitHub (Apr 30, 2014):
I seen on the internet , iPhone 5S is the only one that fully support Open GL ES 3.0 .
What iOS version you are using ?
@unknownbrackets commented on GitHub (Apr 30, 2014):
Well, we don't try yet:
github.com/hrydgard/native@4d49541aa5/gfx/gl_common.h (L7)-[Unknown]
@dbz400 commented on GitHub (Apr 30, 2014):
Arr yes , probably i can send out a pull and @brujo5 can test it out on his device.
https://developer.apple.com/library/ios/documentation/3ddrawing/conceptual/opengles_programmingguide/AdoptingOpenGLES3/AdoptingOpenGLES3.html
OpenGLES/ES3/gl.h
OpenGLES/ES3/glext.h
@dbz400 commented on GitHub (Apr 30, 2014):
There is another case which uses Tegra Note 7 (Tegra 4) on Android 4.3 and should support ES 3.0 as well .
@brujo5 commented on GitHub (Apr 30, 2014):
@raven02
ios 7.0.6
i dont have a MAC,cant compile.
@angelxwind
help here,please.
@akemin-dayo commented on GitHub (Apr 30, 2014):
A new PPSSPP (Testing) build will be available from KarenBuildBot in about two hours.
Sorry for the unusual delay, it's down for maintenance because some I'm making some
modifications to the code to make it work better with the BigBoss APT repository, which some of my packages are now mirrored on.
Cydia/APT address for my repo, Karen's Pineapple Repo: http://cydia.angelxwind.net/
Download page for my PPSSPP builds (iOS, Android, OS X, Linux): http://ppsspp.angelxwind.net/?page/downloads
KarenBuildBot package index for those who are curious: http://karenbuildbot.angelxwind.net/
@brujo5 commented on GitHub (Apr 30, 2014):
@angelXwind
contain the new fixes for opengl 3.0 for iphone 5S?
github.com/hrydgard/native@4d49541aa5/gfx/gl_common.h (L7)thx
@brujo5 commented on GitHub (Apr 30, 2014):
@raven02
merge the changes and karen can compile it.
thx
@akemin-dayo commented on GitHub (Apr 30, 2014):
KarenBuildBot is back up, latest commit for the master branch was just built.
@raven02's OGLES3 commit hasn't been merged yet (see the page for pull request 207 https://github.com/hrydgard/native/pull/207)
@Some1Else46 commented on GitHub (Apr 30, 2014):
@raven02 The Tegra Note 7 (Tegra 4) is Not OpenGL ES 3 capable. Only PoverVR Series 6, Adreno 3xx, and Mali T6xx/7xx support GLES3.
@dbz400 commented on GitHub (May 1, 2014):
It is a shame that Tegra 4 is a pretty new Soc but doesn;t support ES 3.0 .Wondering if it is not currently support in 4.3 but available in 4.4
@Some1Else46 commented on GitHub (May 1, 2014):
The hardware itself is not capable. Tegra K1 should support it though when it starts getting used.
@dbz400 commented on GitHub (May 2, 2014):
I see. I have asked the user dumped the OGL extension and found a GL_NV_framebuffer_blit that we are not providing support yet . I will make a pull to see if can do something here
@dbz400 commented on GitHub (May 2, 2014):
We've biltted depth and color buffer.Wondering if bilt stencil buffer would help some games like midnight club 4? Not too sure if someone already tested but not helping.
Probably I can do some.testing here.
@brujo5 commented on GitHub (May 5, 2014):
@raven02
henrik merge https://github.com/hrydgard/native/pull/207
but iphone 5S still didnt use ES 3.0.
still needed force depth write hack
why?
@LunaMoo commented on GitHub (May 5, 2014):
@brujo5 iphone 5S doesn't have required extenshions for this fix to work anyway, so doesn't matter what ES version it'll use, you'll have to continue using the hack on this device.
As for not detecting correct version ~ as I understand PPSSPP now tries to use the highest possible first, but then drops to lower versions on failure. It generally seems that those iphones ES 3.0 support is very limited and so probably not enough for PPSSPP.
Edit: Oh well, I only looked at this short list here: https://developer.apple.com/library/ios/documentation/DeviceInformation/Reference/iOSDeviceCompatibility/OpenGLESPlatforms/OpenGLESPlatforms.html
and assumed if there's no framebuffer object nor anything with blit listed then it's not supported.
@hrydgard commented on GitHub (May 5, 2014):
I think we need to do something extra on iOS to get ES 3.0, something like CreateBestEAGLContext here:
https://developer.apple.com/library/ios/documentation/3DDrawing/Conceptual/OpenGLES_ProgrammingGuide/WorkingwithOpenGLESContexts/WorkingwithOpenGLESContexts.html#//apple_ref/doc/uid/TP40008793-CH2-SW1
iOS has full ES 3 support so glBlitFramebuffer should be available as it's core functionality, we just need to do the initialization correctly.
@dbz400 commented on GitHub (May 10, 2014):
The user who owns Tegra Note 7 reported that the character can be shown up now however missing powerbar and other stuffs
http://forums.ppsspp.org/showthread.php?tid=362&page=5
@brujo5 commented on GitHub (May 28, 2014):
fixed thx to @rock88 for add ES3.0 for ios
no needed depth write hack anymore.
@akemin-dayo commented on GitHub (May 28, 2014):
Heh. Nice to see you around again, @rock88 c:
@yourtechnogeeks commented on GitHub (Jul 24, 2014):
Good after noon every one im looking for assistance on this game it seems the issue is still around for android arm users how can i assist in getting it resolved please see attached




@brujo5 commented on GitHub (Jul 24, 2014):
For OpenGL 2.0 need to Enable always depth writte hack.
@yourtechnogeeks commented on GitHub (Jul 24, 2014):
Depth write hack is enabled and I have the same result
@xsm1997 commented on GitHub (Oct 13, 2014):
Sorry for my bad English. The same result when Depth write hack is enabled. It happens only when I'm using my cellphone, and my GPU model is Mali-450MP.
@hrydgard commented on GitHub (Feb 21, 2019):
Nothing left to do, we can't solve this in a good way for ES 2.0.