Though again it's not that big if a deal. I get that it's kind of more readable to see the names but if that's the case the same argument applies to workgroup_size so 🤷 😅
Size as an array also matches textureDimensions which returns a vec2u or vec3u which can be indexed with [0], [1], [2] and which is often useful directly in texture coordinate calculations texelCoord = uv * size
anyway, I'm not suggesting changing more now. Mostly voicing that I'd mildly prefer it not be changed to return a GPUExtent3DDict in the future.
Updated the TODO for making it return a GPUExtent3DDict
Note: I'm torn on which is better, 3 numbers or the object. The object seems verbose. The object is also inconsistent with
workgroup_size
Being able to compute the number of texels with
Seems nice though I admit it's hardly that different than
It's also nice to be able to pass the size. For example algos which take a size of texture passed to a compute shader
vs
Or putting it into a uniform/storage buffer with
Though again it's not that big if a deal. I get that it's kind of more readable to see the names but if that's the case the same argument applies to
workgroup_size
so 🤷 😅Size as an array also matches
textureDimensions
which returns avec2u
orvec3u
which can be indexed with[0]
,[1]
,[2]
and which is often useful directly in texture coordinate calculationstexelCoord = uv * size
anyway, I'm not suggesting changing more now. Mostly voicing that I'd mildly prefer it not be changed to return a
GPUExtent3DDict
in the future.