hi,
Since darktable is actually using gphoto2 for tethered shooting you can use
gphoto2 cli for your script to take an image when the LEDs are in desired
state. cli usage docs here
http://www.gphoto.org/doc/manual/ref-gphoto2-cli.html see --capture-image
argument.
hth
mike
Am Mi., 28. Apr. 2021
Hi
unless he custom installed somehow this path to the darktable binary should
work:
"/Applications/darktable.app/Contents/MacOS/darktable -d perf"
and
"/Applications/darktable.app/Contents/MacOS/darktable-cltest"
respectively
hth
mike
Am Mo., 15. Feb. 2021 um 20:46 Uhr schrieb Bernhard <
dar
Hi Jim
I just found time to run the mentioned benchmark on darktable v3.2.1 on
macOS 10.15.6 MacBook Pro (Touchbar, 13-inch, 2018, 4xTB3), 2.7 GHz Intel
Core i7, 16GB
CPU only: 25.2secs (significantly faster than v3.0.0 on the same machine
congrats and thanks to the coders!)
with Intel Iris Graph
Hi Peter
you can start it from terminal with debugging on by typing this:
/Applications/darktable.app/Contents/MacOS/darktable -d all
for more options (limit debugging to certain things) check the --help
It'll show you on terminal stdout what problem it runs into, with that info
you can either
Hi Tom
This is a known problem, some macOS stupidity keeps the compiling of OpenCL
modules from succeeding in a random fashion, it is not in DT developers
hands to fix that. Only workaround I know is to restart darktable until it
succeeds to build all kernels. I have written a dumb shellscript to
this must be improvement of DT 3.0 over 2.6.2 as I have the same "AMD
Radeon Pro 580 Compute Engine" and am not anywhere near those numbers using
DT 2.6.2
Very impressive work from the team behind darktable, thanks in advance
since I haven't upgraded yet :)
Am Fr., 20. Dez. 2019 um 09:25 Uhr sch
g.txt`
>
> I hope this will shed some light on the issue. Or maybe I just went (too)
> crazy on the modules :)
>
> Thanks and best,
>
> Jochen
>
>
> On Mon, Dec 16, 2019 at 2:27 PM Michael Kefeder
> wrote:
>
>> Hi Jochen,
>>
>> Have you conf
Hi Jochen,
Have you confirmed darktable is actually running GPU acceleration? e.g.
looking at the logs of darktable (e.g. using arguments: -d opencl -d perf)?
I ran benchmarks on my computer with a 121MB RAW file (download of a Fuji
GFX100 DNG) with 17 modules active and that takes 34sec to export
If you feel comfortable in sqlite3 shell or some gui-tool for sqlite you
can update the filepath in the darktable DB, that's what I did when I moved
computers and my username changed. Backup the db file and do a sqlite
command like this (on purpose pseudo code, do this only if you know how a
DB wor
g would already help
tremendously, since the metadata is managed by third party app already.
Am Do., 16. Mai 2019 um 09:19 Uhr schrieb Remco Viëtor <
remco.vie...@wanadoo.fr>:
> On jeudi 16 mai 2019 08:14:54 CEST Michael Kefeder wrote:
> > Why overengineer this, just dump IPTC info into
Why overengineer this, just dump IPTC info into JSON and use that for
freestyle attributes. Sqlite even has an extension for fast parsing of that
https://sqlite.org/json1.html
William Ferguson schrieb am Do., 16. Mai 2019, 04:38:
>
>
> On Wed, May 15, 2019 at 5:50 PM wrote:
>
>> William Ferguso
Hi Florian,
Long time no see ;)
Besides that, exif data holds "default crop size" of 6000x4000 pixels. That
seems to be applied for DNG for you. I also assume those extra pixels are
there for lens correction, since it's the same on a6x00 series cameras
without IBIS they also have 6048pixel wide r
Hi darktable users
upgraded to 2.6.2 and opencl was again "not available" due to the silly
compile bug in macos.
Since restarting darktable manually multiple times until all compiles are
done is annoying I created a mini helper bash script that starts/kills
darktable until the kernels are successf
Looks like data-corruption to me (few bytes changed somewhere in the file),
maybe your harddrive starts to die? You can the smart-status for potential
errors of your disk.
Am Di., 12. März 2019 um 13:34 Uhr schrieb Michael :
> what is going. Most of the pictures I took last year with the sony hav
dirty hack that comes to my mind is you write a small tool (e.g. python
script) that removes the "done" folder from the sqlite DB of darktable
until you've finished your massive import (so they wont show in DT
anymore). After that you can add them again.
clean hack, you add a "done"-flag to the sq
There's an off chance your dedicated GPU is device 1 and your cpu is 0. By
default the config only does consider one GPU in the system = 0, you can
try (make sure the scheduler is default) to force use of device 1:
opencl_device_priority=1,!0/!0,!1/1,!0/1,!0
Am So., 2. Dez. 2018, 15:58 hat Volker
Normally you should restart "darktable -d opencl" many times until the
output looks like this, no errors compiling the kernels:
...
2,535671 [opencl_create_kernel] successfully loaded kernel
`denoiseprofile_accu' (175) for device 0
2,535708 [opencl_create_kernel] successfully loaded kernel
`denois
I am fairly sure this
is how it's best.
Am Fr., 26. Okt. 2018 um 00:50 Uhr schrieb I. Ivanov :
>
> On 2018-10-25 15:27, Michael Kefeder wrote:
> > Hi
> >
> > I have the same issue on mac, I use
> > opencl_device_priority=1,!0/!0,!1/1,!0/1,!0
> > and imp
Hi
I have the same issue on mac, I use
opencl_device_priority=1,!0/!0,!1/1,!0/1,!0
and important: opencl_scheduling_profile=default else the priority seems to
be ignored. With these it never tries to use the Intel GPU (0 on my
machine).
hth
mike
Am Fr., 26. Okt. 2018 um 00:13 Uhr schrieb I. I
I agree with the technical limitations of videos you mentioned, however I
am not looking to find technical info in those either. I am expecting to
see how others use the same tool I do, to achieve a goal - so more a
creative input in what can be done than an explanation what the
module/button/short
I run darktable on mac for many years now, and in fact I recently played
with it and benchmarked an eGPU using Mojave. Works perfectly fine for me,
migrated my image library yesterday.
For best performance, just make sure all opencl kernels are compiled. May I
suggest to get an iMac with a dedicate
Hi List
I bought a used MacBook Pro (13-inch, 2017, Two Thunderbolt 3 ports) and
wanted to compare the speed to my old laptop. And of course my new eGPU,
which is intended to be used for videoediting but why not also use it for
rawprocessing. Specs:
CPU: 2.3 GHz Intel Core i5
RAM: 16GB
GPU: Intel
22 matches
Mail list logo