I've recently sucessfully used it with Ubuntu 20.04 and an old NVidia card. Perhaps you want to give Ubuntu a try?
Btw: Are you using the proprietary drivers or are you using nouveau? Perhaps switching the drivers might help?
Search found 9 matches
- Mon Jun 22, 2020 7:52 am
- Forum: Bug reports
- Topic: createScreenShot GLES2
- Replies: 4
- Views: 2789
- Thu Jun 18, 2020 10:16 am
- Forum: Bug reports
- Topic: createScreenShot GLES2
- Replies: 4
- Views: 2789
createScreenShot GLES2
Hi, When using createScreenShot red and blue channels are mixed up in GLES2 (tested on Linux). I currently use a workaround in COGLES2Driver::createScreenShot by switching these channels back afterwards: for(u32 x=0; x<ScreenSize.Width; x++){ for(u32 y=0; y<ScreenSize.Height; y...
- Wed Jun 10, 2020 2:15 pm
- Forum: Bug reports
- Topic: [fixed]CPLYMeshFileLoader
- Replies: 3
- Views: 2550
Re: CPLYMeshFileLoader
Yes it's possible that I fixed this locally a long time ago and forgot about it.
Here's the ply: https://we.tl/t-Eb0I5DQLvr
Here's the ply: https://we.tl/t-Eb0I5DQLvr
- Mon Jun 08, 2020 3:10 pm
- Forum: Bug reports
- Topic: [fixed]CPLYMeshFileLoader
- Replies: 3
- Views: 2550
[fixed]CPLYMeshFileLoader
Hi, since I updated my Irrlicht fork to the latest Irrlicht from the ogles branch I get an "invalid read" when testing with valgrind in the CPLYMeshFileLoader: Here's the diff: @@ -571,7 +572,7 @@ c8* CPLYMeshFileLoader::getNextLine() StartPointer = LineEndPointer + 1; //...
- Fri May 29, 2020 11:22 am
- Forum: Bug reports
- Topic: [fixed]COpenGLCoreTexture.h: glGenerateMipMap in GLES2
- Replies: 3
- Views: 2646
Re: COpenGLCoreTexture.h: glGenerateMipMap in GLES2
Yes it's with the newest ogles-branch. I think it makes sense to add it to Irrlicht. However I don't have an explanation for this. It's just the result of trial and error. There might also be a problem with the current emulator from Android Studio. Btw: IRR_OPENGL_HAS_glGenerateMipmap is defined in ...
- Thu May 28, 2020 3:13 pm
- Forum: Bug reports
- Topic: [fixed]COpenGLCoreTexture.h: glGenerateMipMap in GLES2
- Replies: 3
- Views: 2646
[fixed]COpenGLCoreTexture.h: glGenerateMipMap in GLES2
Hi, there seems to be something wrong with glGenerateMipMap in COpenGLCoreTexture.h: In this line I get an GL_INVALID_ENUM error when I run this on linux or a real Android device. In the Android emulator I get an GL_INVALID_OPERATION . However the enum parameter is correct (GL_TEXTURE_2D). I noticed...
- Thu Feb 27, 2020 4:31 pm
- Forum: Code Snippets
- Topic: My Irrlicht Extensions: SDFF, 2D functions, GUI Elements,etc
- Replies: 1
- Views: 2981
My Irrlicht Extensions: SDFF, 2D functions, GUI Elements,etc
Hi, I've made my extension available at github: https://github.com/wolfgang1991/CommonLibraries Most interesting part for most people might be the SDFF (Signed Distance Field Font) implementation with some effects and support for arbitrary font transformations. New GUI Elements optimized for touch i...
- Mon Jul 01, 2019 10:01 am
- Forum: Advanced Help
- Topic: GPU side memory leak?
- Replies: 2
- Views: 950
Re: GPU side memory leak?
I believe I found the solution: IVideoDriver::removeHardwareBuffer must be called to avoid the memory leak when deleting the meshbuffer. Btw: After looking at the code of CMeshCache::removeMesh I believe I found a bug. This function does not remove the hardware buffers when deleting a mesh: void CMe...
- Mon Jul 01, 2019 8:24 am
- Forum: Advanced Help
- Topic: GPU side memory leak?
- Replies: 2
- Views: 950
GPU side memory leak?
Hi, I'm wondering if the following steps would create a memory leak on GPU side: 1. Create an SMeshBuffer with some content 2. Use setHardwareMappingHint(EHM_STATIC) 3. Render it many times via IVideoDriver::drawMeshBuffer 4. Delete the SMeshBuffer I believe the Vertex Buffer Objects still persist a...