> - Are other profiles also so slow? For example the standard matrix? No, they are fast. Standard matrix is applied using OpenCL: [dev_pixelpipe] took 0.036 secs (0.028 CPU) processed `input color profile' on GPU, blended on GPU [export]
> - Does it get better when you disable OpenCL? It wasn't used for the profile > step, but sometimes copying buffers back and forth is the real bottleneck and > it gets counted as module runtime. With OpenCL disabled, the converted RawStudio profile is slow, standard matrix is fast: [dev_pixelpipe] took 9.218 secs (16.960 CPU) processed `input color profile' on CPU with tiling, blended on CPU [export] vs. [dev_pixelpipe] took 9.218 secs (16.960 CPU) processed `input color profile' on CPU with tiling, blended on CPU [export] > - How much (free) memory does your computer have? It's an old box with only 4 GB of RAM. > free total used free shared buffers cached Mem: 4046984 3515600 531384 48304 237772 1266132 -/+ buffers/cache: 2011696 2035288 Swap: 67108860 898944 66209916 > top KiB Mem: 4046984 total, 3814072 used, 232912 free, 413652 buffers KiB Swap: 67108860 total, 898728 used, 66210132 free. 1265756 cached Mem This is with darktable loaded but inactive in the background. Thanks, Kofa ------------------------------------------------------------------------------ Site24x7 APM Insight: Get Deep Visibility into Application Performance APM + Mobile APM + RUM: Monitor 3 App instances at just $35/Month Monitor end-to-end web transactions and take corrective actions now Troubleshoot faster and improve end-user experience. Signup Now! http://pubads.g.doubleclick.net/gampad/clk?id=272487151&iu=/4140 _______________________________________________ Darktable-users mailing list Darktable-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/darktable-users