mutantsushi on 24 May 2014
I believe I've read that the GPGPU can be great to use for occlusion culling,
and there's other aspects of graphics which can be more efficiently done with the broader flexibility of GPGPU,
not to mention work that may be transferrable from CPU to GPGPU, freeing capacity on the CPU side.
I'm not sure what this SDK work IS, AFAIK, the GPGPU side is already fully addressable as such,
I suppose there could be some intermediary representation that simplifies using GPGPU to augment graphics?
