Guodong Chen

Metal Backend 代码的疑问？

13

Metal Backend下面，如果写入CommandBuffer 后，GPU还没有来得及执行DC。 这个时候更新纹理，是不是会污染已经写入CommandBuffer的DC的数据？ 这块好像没有做特殊处理？

代码的一些疑惑

2

`void _gen_stack_peek_long(struct sljit_compiler *C, s32 offset, sljit_s32 dst, sljit_sw dstw) { //dst=sp[offset]->lvalue sljit_emit_op1(C, SLJIT_MOV, dst, dstw, SLJIT_MEM1(REGISTER_SP), sizeof(StackEntry) * offset + SLJIT_OFFSETOF(StackEntry, lvalue)); }` 这段代码如果运行在 32位的CPU上面，但是java中long是8位的，那么 dst寄存器只有32位这样对么？

https://github.com/knightcrawler25/GLSL-PathTracer/blob/85c5167/thirdparty/RadeonRays/split_bvh.cpp At 432 line ``` // Calc SAH float sah = m_traversal_cost + (leftbox.surface_area() * +rightbounds[i - 1].surface_area() * rightcount) * invarea; ``` Should it be like this? ``` //...

tbb on wasm always executed on the main thread.

27

On the wasm platform Both tbb::task_group tg and tbb::parallel_for are always executed on the main thread. But std::thread executes on an asynchronous thread. What causes this? and oneapi::tbb::info::default_concurrency() > 10

Support for WebGPU

9

By testing normals, the compute shaders of WebGPU are found to be about 20 times faster than WebGL's GPGPU, so rendering with WebGPU will be much quicker. I discovered the...

Animation separation

4

I’ve noticed that many users have a need to load models and animations separately—for example, loading the mesh from one GLB file and the animation from another. However, since the...

Hope to have support for IBL from HDR file