Size of the compute thread group that supports efficient memory sharing on the GPU (Read Only).
The most efficient thread group size for a set of compute shader calls. This value takes synchronization and shared data into account.
This value is a subset of the total workgroup size. It is also known as a "warp" or a "wavefront".
Additional resources:ComputeShader class, supportsComputeShaders.