On Fri, May 12, 2017 at 9:21 AM, Alexander Rabtchevich <alexander.v.rabtchev...@gmx.net> wrote: > Hello > > I've installed a new graphics card with Radeon 580 chip (8Gb) and so > examined darktable performance. Two opearations during regular jpg export do > not use OpenCl -
>raw demosaic (or decompression) I'm guessing you mean "raw denoise" >and applying of final > gamma. Both of these module do not have opencl kernels, so they have to be processed on CPU. And for that, the data needs to be copied from GPU memory to CPU (and back)... >Their common time equals or even prewails a whole bunch of all other > operations. The OpenCl setting in GUI is - powerful graphics card. Is there > a way to increase the performance? Darktable is from current git. Basically only by implementing those respective opencl kernels The gamma kernel would be trivial, now that the "pipe format callbacks" work has landed last fall (https://github.com/darktable-org/darktable/pull/1253) I wanted to do it, and I would do it, but (*still*, sigh) i have no opencl here. > With respect, > Alexander Rabtchevich Roman. > ___________________________________________________________________________ > darktable developer mailing list to unsubscribe send a mail to > darktable-dev+unsubscr...@lists.darktable.org ___________________________________________________________________________ darktable developer mailing list to unsubscribe send a mail to darktable-dev+unsubscr...@lists.darktable.org
