Archive notice: This article was originally published on July 29, 2013. Links and embedded videos are preserved as part of the historical record.
OpenCL 2.0 is intended to be refined and finalised over the next six months. The most important feature of the upcoming release is support for Shared Virtual Memory, so that unified memory architectures between GPU and CPU can then be addressed uniformly in OpenCL.
According to Khronos, the atomic operations of the C11 language extension are now also supported directly, which is intended to make programming easier. For video editing, it is interesting that several cores can write to the same OpenCL image simultaneously.
According to Khronos, software developers should be able to use the GPU for acceleration more easily. The new version primarily benefits GPU/CPU combinations such as those offered by Intel and AMD.
All further information is available here.
About the author
Redakteurin der Digital Production
Article details
- Published
- Author
- Mirja Fürst
- Reading time
- About 1 min
- Word count
- 146