Author Topic: Difference between CUDA and OpenCL 2010  (Read 6416 times)

0 Members and 1 Guest are viewing this topic.


  • Global Moderator
  • Hero Member
  • *****
  • Posts: 1522
    • View Profile
Difference between CUDA and OpenCL 2010
« on: April 23, 2010, 11:49:46 AM »
CUDA term                                       OpenCL term
GPU                                               Device                   
Multiprocessor                               Compute Unit
Scalar core                               Processing element
Global memory                               Global memory
Shared (per-block) memory               Local memory
Local memory (automatic, or local)       Private memory
kernel                                       program
block                                            work-group
thread                                          work item

On NVidia hardware, OpenCL is up to 10% slower (see Matt Harvey’s presentation); this is mainly because OpenCL is implemented on top of CUDA-architecture (this shouldn’t be a reason, but to say NVidia has put more energy in CUDA is just a wild guess also). On ATI 4000-series OpenCL is just slow, but gives very comparable to NVidia if compared to the 5000-series.