On lundi 16 décembre 2019 21:03:39 CET Jochen Keil wrote:
> Hi Michael,
>
> thanks for your reply! I'm pretty confident that darktable is using GPU
> acceleration. I've checked the output of `darktable-cltest` several times
> (like someone else suggested) and it says that OpenCL is available and in
> use.
>
> For reference, here's a RAW + XMP:
>
> https://drive.google.com/open?id=1FBunVv3reyUTXIXFFGP2dE02EAyHvkoa
> https://drive.google.com/open?id=1LEs9cIo760yplkyE57S_5Mqkm8-pVATl
>
> Attached you'll find the log of
>
> `$ time darktable-cli 2019-11-23T23:23:35+0100_7871.arw
> 2019-11-23T23:23:35+0100_7871.arw.xmp 2019-11-23T23:23:35+0100_7871.png
> --core --library :memory: -d opencl -d perf 2>&1 > dt_log.txt`
>
> I hope this will shed some light on the issue. Or maybe I just went (too)
> crazy on the modules :)
If I look at that output, I see:
53,785890 [opencl_profiling] spent 3,2569 seconds totally in command
queue (with 0 events missing)
53,785899 [dev_process_export] pixel pipeline processing took 53,032 secs
(130,516 CPU)
So by far most of the time is spend using the CPU, not the GPU.
And there are no error messages, so no cases where darktable tried to use the
GPU and failed.
That seems to indicate (to me at least) that there is no need to upgrade your
graphics card to speed up exporting.
Also, most of the time is spent in modules that are normally rather fast:
exposure1 - 17s, exposure2 - 17s and tonecurve3 - 10s. Other instances of
those same modules in your stack do execute fast, about 120ms each...
Demosaicing takes 4s, profiled denoising about 1s; both of these are described
as "slow" modules (demosaic when using Amaze)...
This makes me suspect that the cause of the long processing times is in the
masking you use. And that is CPU-bound, and not using the GPU at all.
Remco
____________________________________________________________________________
darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org
