Search Issue Tracker
Won't Fix
Votes
0
Found in [Package]
12.0.0
Issue ID
1334330
Regression
No
Accessing structured buffers with when using Nvidia RTX series GPUs produces visual artifacts
How to reproduce:
1. Open the attached project named "testStructuredBuffer.zip"
2. Open the SampleScene
3. Open the Game view and observe the Sphere on the screen
Expected results: The sphere is rendered in solid red "buffercorrect.png"
Actual results: The sphere has square-shaped artifacts "bufferissue.png"
Reproducible with: 2019.4.26f1(7.6.0), 2020.3.7f1(10.4.0), 2021.1.7f1(11.0.0), 2021.2.0a17(12.0.0)
Could not test with: 2018.4.35f1(HDRP was in early preview stage)
Notes:
1. Issue seems to only occur when using RTX GPUs
2. Reproduce with RTX4000, RTX 2080(Driver version - 461.72, 466.27), RTX2060(Driver version - 466.27)
3. Does not reproduce with GTX1080 (Driver version - 461.72, 466.27), GTX1060 from Asus GL702V laptop(Driver version - 466.27)
Add comment
All about bugs
View bugs we have successfully reproduced, and vote for the bugs you want to see fixed most urgently.
Latest issues
- Texture2D hash changes inside of an AssetBundle when rebuilding a SpriteAtlas bundle with an empty AssetPostprocessor Script enabled
- Aniso Level still applies when Generate MipMap is disabled in Texture Import Settings
- Mipmap Limit Groups long names are not truncated when creating a new Mipmap Limit Group with a long name
- “ArgumentException: Invalid double parameter.” error is thrown when Infinity is typed into the Fixed Timestep field
- GameObject becomes gray when using HDRP and STP together on macOS
Resolution Note:
The issue seems to originate in the repro project's custom scripts. The problem is that the stride declared for the buffer is incorrect, if the buffer is declared as a float in the shader, so will need to be the stride (and so count) on C# side (TestRWStructuredBuffer was declared as <float>, but with a count/stride for it containing <float2>). Making the mentioned changes should resolve the issue.