Number of buffers on the GPU which can stay allocated even after the ComputeDisparityMap command is finished. This will provide an increase in performance, as the necessary buffer doesn’t have to be newly allocated each time the command is called. As a rule of thumb this value should match the number of parallel executed ComputeDisparityMap commands. If no free static buffer is available, the command will allocate a new buffer.
This setting is for CUDA. There is an equivalent setting for the CPU.
An integer specifying how many static buffers can be used by the ComputeDisparityMap command.