Open CL - Comparison With CUDA

Comparison With CUDA

Two comparisons have been drawn between CUDA and OpenCL since its inception, and both arrived at the same conclusion: if the OpenCL implementation is correctly tweaked to suit the target architecture, it performs no worse than CUDA. Because the key feature of OpenCL is portability (via its abstracted memory and execution model), the programmer is not able to directly use GPU-specific technologies (such as inline PTX) unless they are willing to give up direct portability. CUDA is more directly connected to the platform upon which it will be executing because it is limited to Nvidia hardware. Compiler technology for both standards supported by Nvidia's toolkit is based upon LLVM and compiles to Nvidia's PTX instruction set abstraction.