native morph workaround #16910
native morph workaround #16910
Conversation
|
Please make sure to label your PR with "bug", "new feature" or "breaking change" label(s). |
|
Snapshot stored with reference name: Test environment: To test a playground add it to the URL, for example: https://snapshots-cvgtc2eugrd3cgfd.z01.azurefd.net/refs/pull/16910/merge/index.html#WGZLGJ#4600 Links to test babylon tools with this snapshot: https://playground.babylonjs.com/?snapshot=refs/pull/16910/merge To test the snapshot in the playground with a playground ID add it after the snapshot query string: https://playground.babylonjs.com/?snapshot=refs/pull/16910/merge#BCU1XR#0 |
|
Visualization tests for WebGPU |
|
WebGL2 visualization test reporter: |
|
WebGL2 visualization test reporter: |
|
Visualization tests for WebGPU |
|
WebGL2 visualization test reporter: |
|
Visualization tests for WebGPU |
|
WebGL2 visualization test reporter: |
|
Visualization tests for WebGPU |
|
WebGL2 visualization test reporter: |
|
Visualization tests for WebGPU |
morphTargetCount is a uniform int; but bgfx only support vec4 as uniform. when doing the replacement, morphTargetCount.x is used and comparison between `int i` and `morphTargetCount.x` which is a float break shader compilation with GLSL. casting `morphTargetCount` to int doesn't work: casting qualifier is removed becaused it's still an int before shader traversers. Setting morphTargetCount a float in the engine and casting iterators to float seems to work better. I've also done the change for WGSL to pass the tests. Long term is to add (or change) traversers for float/int comparison.
morphTargetCount is a uniform int;
but bgfx only support vec4 as uniform.
when doing the replacement, morphTargetCount.x is used and comparison between
int iandmorphTargetCount.xwhich is a float break shader compilation with GLSL.casting
morphTargetCountto int doesn't work: casting qualifier is removed becaused it's still an int before shader traversers.Setting morphTargetCount a float in the engine and casting iterators to float seems to work better.
I've also done the change for WGSL to pass the tests.
Long term is to add (or change) traversers for float/int comparison.