Implicit tream operators
Hi,
Again a basic question:
Just to want to understand if there is any performance penalty (timing,..)
in taking advantage of kernel auto up-sampling/down-sampling...
for instance in this vector %2 downsampling :
kernel move_to(float src<>, out float dest<> {
dest = src;
}
main :
float A_str<100>;
float B_str<50>;
move_to(A,B); // B gets all values at even positions in A
So is upsampling/downsampling a pure Brook software feature, or is it de facto supported by the GPU hardware?
In any case what is the performance price to pay versus using aligned-sized vectors??
Thanks
Jean-Claude