![]() ![]() This pr introduces a decision in the opencl pixelpipe codeġ. the "overall processed memory" weighed by the tested performance in 1) is a good estimation of the resulting workload. (The results we already have in opencl.c are by far not good enough as they don't reflect real-world performance.)Ģ. ![]() we have a proper benchmarking result measuring raw processing powers cpu vs cl device. **proposed solution**: follows two assumptions:ġ. Suggested "solutions" like per-module settings (whatever the complexity in the UI might be) are not good at all as runtime defined parameters are not taken into account. Such a situation was rare, now we have as examples D&S or laplacian highlights which might behave bad and it would have been better to use cpu code instead. There is no problem using OpenCL code in modules as long as that doesn't lead to excessive tiling with large overlapping areas. **Rationale**: On many machines we find pretty fast multicore cpus and a lot of … system ram but a significantly gpu memory with somewhat resrticted number crunching power. ![]() This is a companion discussion topic for the original entry at We’re proud to announce the new feature release of darktable, 4.2.0!Īs always, please don’t use the autogenerated tarball provided by github, but only our tar.xz file. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |