Re: [darktable-user] specs

[Remco Viëtor]

On mercredi 28 février 2024 04:15:30 CET Michael wrote:
> Gimp and darktable will run nicely on 16GB ram?

Depends on the image size and what else you have running at the same time.
I'd expect 16GB to be enough until you start on large images or (for Gimp) 
complex edits with many layers.

For darktable, you also want to look at GPU memory, though (6+GB seems a  
reasonable minimum).




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: Fwd: [darktable-user] Setting resolution for an on-line photography competition

[Remco Viëtor]

On mardi 23 janvier 2024 23:52:38 CET Bruce Williams wrote:
> Are you submitting a digital file, or a physical print?
> If it's a digital file, I'd avoid that competition, because the organisers
> don't even know what they're asking for!
> Let's take a random example:
> A digital file from a 24mp sensor is (roughly speaking) 6000 pixels x 4000
> pixels.
> This file can be said to be 83.3 inches x 55.5 inches at 72 ppi.
> Or it can be referred to as being 20 inches by 13.3 inches at 300 ppi.
> So, unless the organisers have specified a PRINT SIZE @ 72 ppi, they have
> no idea of what they are on about.
> 
> 
> -- Forwarded message -
> From: Andrew Greig 
> 
> Hi
> I have recently decided to jump into photographic competition, and one
> of the requirements for submission is to adjust my file to 72 PPI what
> would be the best settings for export please?
> 

And even if they require a physical print, I'd hesitate: "normal" printing 
resolution is about 300 DPI or higher. 72 DPI is low even for screen display 
(standard there is 96 DPI). 

As to darktable export settings: go to the export module, and select for "set 
size" either "in cm" or "in inch". That will give you the option to set a size 
in cm (or inch), and a DPI, and will display the resulting size in pixels. 


Note that a file cannot have a resolution (in DPI), it's just a bunch of pixel 
data. Until you are displaying or printing, any notion of resolution is 
irrelevant. And for a screen, its size in pixels is fixed (and usually known), 
so you don't need (or want) the DPI here either.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Installing darktable 4.6 in Ubuntu 22.04.3 why is it so hard?

[Remco Viëtor]

On mercredi 27 décembre 2023 14:04:30 CET Andrew Greig wrote:
> I have backed up my ~config/darktable of 4.4.2 and
> 
> I have issues with the installation, what is best? apt snap or flatpack?
> why do we have such a shit fight?
> In 2002 I was using Mandrake and it was as smooth as.  Best detection
> and configuration of hardware because of the outstanding installer, but
> here we are 21 years later and things are really shit!!
(...)

Darktable team can only provide so many binary packages. If they don't provide 
a package for your distribution, you'll have to wait until someone else 
prepares one.

The latest version was published dec. 21, that's about a week ago, with 
christmas in that week. Some of the packagers might have had other obligations 
during that time?

And of course, if you are in a hurry, you can always download the source code 
and compile dt yourself. Apart from a possible hurdle concerning compiler 
versions(*), it's not that difficult. (it's what I did...)

 (*: you need at least GCC 12, and then best use "export CC=gcc-12;export 
CXX=g++-12" before starting the build)



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: Fwd: Fwd: [darktable-user] Need help compiling

[Remco Viëtor]

On mardi 19 décembre 2023 11:32:06 CET Bruce Williams wrote:
> [ 21%] Built target raw_r
> make: *** [Makefile:156: all] Error 2
> 
> I'm guessing this means that the script tried to build something called
> "raw_r", and that that process failed. Is that correct?
> How do I find out why it failed?
> And the final line? I don't understand the syntax there at all.
Sorry, I'm getting out of my depth here, and when I tried to build from Git, 
there were errors as well (I don't know enough about cmake to tell it to link 
in libstdc(++) ).

To your situation:
- the *target* for that part of the build is called raw_r, which is a label 
for one part of the full build, not related to the name of an executable or 
library.
- it looks like that part of the build actually succeeded ("Built target xxx")
- that "Error 2" indicates an error in the next step, which error isn't clear 
from the log.

I hope someone who knows the build system for darktable can chime in.

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: Fwd: [darktable-user] Need help compiling

[Remco Viëtor]

On mardi 19 décembre 2023 07:59:41 CET Bruce Williams wrote:
> Thanks for the replies so far.
> After pamac search intltool, I found a version number, and promptly tried
> this:
> 
> "sudo pamac intltool  0.51.0-6"
...✔
> Available actions:
>  pamac --version
>  pamac --help, -h [action]
>  pamac search [options] 
>  pamac list   [options] 
>  pamac info   [options] 
>  pamac install[options] 
>  pamac reinstall  [options] 
>  pamac remove [options] [package(s)]
>  pamac checkupdates   [options]
>  pamac update,upgrade [options]
>  pamac clone  [options] 
>  pamac build  [options] [package(s)]
>  pamac clean  [options]
> 
> What did I do wrong?
Perhaps pamac needs to be told *what* to do with the package?
Also, I doubt the package name has spaces (they complicate shell scripts)

From your post, I gather you either never installed a package manually before,
or you used a GUI vesion of the package manager. If the latter, why not use it 
here as well?



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Need help compiling

[Remco Viëtor]

On mardi 19 décembre 2023 07:12:45 CET Bruce Williams wrote:
(...)
> -- Missing intltool-merge
(...)
>  Some external programs couldn't be found

Install the missing external programs? For me (OpenSUse) that would be the 
package 'intltool'

Note: the cited line is the only missing tool you showed, I have a feeling the 
start of the output is missing, so there could be other missing tools...




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Darktable database

[Remco Viëtor]

On jeudi 5 octobre 2023 13:28:21 CEST dboland9 wrote:
> I know from reading the docs that DT uses SQLite to store some information,
> but how are the images and the image meta data stored? The reason for the
> question is that I'm uncomfortable with putting everything in a database in
> case something goes wrong. Thanks for any helpful responses.
> 
> Dave,
> 
> Sent with [Proton Mail](https://proton.me/) secure email.
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.or

Your original images stay were you put them, darktable won't move or change 
them. Metadata from the original files and metadata you add (tags, titles, 
etc, and editing information) are stored in the database (but *never* removed 
from the original files, those are not changed in any way). Images are never 
stored in the dt database.

 (There's a cache for thumbnails, in order to speed up operations, but those 
are not criticial, and you can remove them at any moment without loss of 
data).

Metadata can also be stored in so-called sidecar files (xmp files that will 
contain the metadata and the editing operations you performed on the original 
image). You can recreate the image database from those xmp files. (dt has 
another database for e.g. styles and presets you can define, and some other 
general information).

You can also use dt without the image database (see option "-library 
:memory:") but then you better use sidecar files, as that would be the only 
way to save edits between sessions.

This also means that to get an edited image as a separate file, you need to 
explicitly export that edit to e.g. a jpg file.

Remco






darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: Fwd: [darktable-user] darktable suddenly refuses to launch on Manjaro

[Remco Viëtor]

On vendredi 25 août 2023 14:31:53 CEST Bruce Williams wrote:
> No, I had not!
> This was the error thrown by the terminal:
> 
> an error occurred while trying to execute gdb. please check if gdb is
> installed on your system.
> backtrace written to /tmp/darktable_bt_XHGFA2.txt
> zsh: segmentation fault (core dumped)  darktable
> 
> I tried "nano /tmp/darktable_bt_XHGFA2.txt" but it appears to have nothing
> in the text file.
> 

Is that just a "plain" run of darktable, which gave the gbd error?

You should be able to just start darktable from a terminal emulator like 
"Konsole" by typing "darktable" at the prompt. And there's no reason gbd 
should be called in that case.

Also, is it possible you had done a system update without immediately 
rebooting the system, and in that case, did darktable stop working after a 
reboot (or restart, logout/login cycle)? Programs don't just stop working on a 
system where nothing changed...

> 
> -- Forwarded message -
> From: Jesus Arocho 
> Date: Fri, Aug 25, 2023 at 10:19 PM
> Subject: Re: [darktable-user] darktable suddenly refuses to launch on
> Manjaro
> To: Bruce Williams 
> 
> 
> 
> On 8/25/23 07:45, Bruce Williams wrote:
> 
> Hi all,
> I'm running Manjaro, and have had darktable 4.4 running just fine in recent
> days.
> But tonight, I tried to launch it, the icon appears for a second, and then
> disappears.
> System monitor does not show it amongst the running applications.
> 
> System specs:
> 
> Operating System: Manjaro Linux
> KDE Plasma Version: 5.27.7
> KDE Frameworks Version: 5.108.0
> Qt Version: 5.15.10
> Kernel Version: 6.1.44-1-MANJARO (64-bit)
> Graphics Platform: X11
> Processors: 16 × AMD Ryzen 7 2700X Eight-Core Processor
> Memory: 31.3 GiB of RAM
> Graphics Processor: NVIDIA GeForce RTX 3070/PCIe/SSE2
> Manufacturer: Micro-Star International Co., Ltd.
> Product Name: MS-7B79
> System Version: 2.0
> 
> I do not have auto updating of the system running due to some issues with
> the GUI version of PAMAC. I do manual system updates via the command line,
> and I have not done any since I last ran dt.
> If anyone can point me in the right direction to get it launching, I would
> be most thankful.
> 
> Have you tried to launch darktable from a terminal window to see if an
> error message is displayed?
> 
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] A different build error with 4.4.2 under Mint 21.2

[Remco Viëtor]

On mardi 25 juillet 2023 10:57:27 CEST tony Hamilton wrote:
(...)Given my level of skill and understanding
> , coupled with the non-specific nature of the error messages generated
> by CMake, it is impossible for somebody like me to determine precisely
> what package dependencies are missing from my instance of Linux. Result:
> compile is bound to fail, until somebody like me gets help from somebody
> like you.
> 
> But behind all this frustration there is somebody (somebodies) who know
> EXACTLY what the dependencies are. It would transform the compile
> success rate for somebody like me if those dependencies were completely
> documented somewhere, or, better still, if there was a command which I
> could copy/paste that would act to meet all those dependencies. 

But the dependencies *are* listed, just before the (perhaps outdated) commands 
to install them. The list gives you the library name and minimum required 
version (if any). It cannot give the package names to install, as they vary 
between distributions (and perhaps even versions). Then again, I think most 
distributions have a graphical package manager, with search function (Yast for 
OpenSUSe, Synaptic(?) for Mint)... And make sure you install the required 
development packages as well..

There is a learning curve for compiling programs, but once you master it, you 
are a lot less dependant on packagers.

When you run into issues, remember to look at the first error message listed, 
it's usually the most important (note: *error* message, information and 
warning messages are often less helpful).




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] darktable flatpak opencl support

[Remco Viëtor]

On mardi 25 juillet 2023 08:21:57 CEST Michael Staats wrote:
> Hi
> I'm on Debian 11, and it seems that there is no dt 4.4.2 package on OBS.
> Same for Debian 12, so an upgrade will not help.
> 
> So I thought I'd give the flatpak version a try. Installs nicely, and
> after copying from ~/.config/darktable to
> ~/.var/app/org.darktable.Darktable/config/darktable, 4.4.2 starts and
> all looks good.
> 
> Except: There is no openCL support. Which is bad, the only reason I have
> a GTX 1660 is darktable...
> 
> I also installed org.freedesktop.Platform.ClInfo (as flatpak), but this
> did not change anything.
Did you also install the NVidia drivers as flatpak? I understood that flatpak 
won't use the system drivers (sandboxing...).

An alternative would be to compile dt yourself, it's not all that difficult if 
you follow the instructions provided with the source.

Or just wait a bit longer, the announcement of version 4.4.2 on darktable.org 
is dated 2023-07-22... Packagers may need a few days to package a new version 
(more if they are away for holidays)

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Build error with 4.4.2

[Remco Viëtor]

On lundi 24 juillet 2023 15:02:24 CEST Patrick Shanahan wrote:
(...)
(concerning build instructions)
> the directions in dt's github repository are quite good.
> 
However, what I did miss in them was how to tell the build system which 
compiler to use. Not all distributions have the latest GCC as "standard" 
version, even when it is available in their repositories.
Hence the need for something like: "export CC=gcc-12; export CXX=g++-12;".

(Would this be something to include in the build instructions, or is it too 
much "required knowledge" for those who insist on building from source?)

Remco





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Build error with 4.4.2

[Remco Viëtor]

On lundi 24 juillet 2023 10:01:59 CEST tony Hamilton wrote:
(...)
> Working in this mode of unbounded ignorance is not very satisfying, so I
> would be grateful if you can direct me to the source of your
> understanding about cmake expecting /usr/bin/c++ to be an alias of
> /usr/bin/g++. That way I can understand better the reason for this
> common compile failure.

It's not so much that c++ needs to be an alias for g++, but that you have to 
tell cmake which version to use if that is different from the "system" 
version. On my system (OpenSUSe Leap 15.5), both /usr/bin/c++ and /usr/bin/g++ 
are symlinks to another file. Both link to version 7.x. 
So to get darktable to compile, I have to tell cmake to use a different 
version (which of course needs to be installed).

To do that, there is the "export" command to set environment variables CC and 
CXX. The build script will use those two variables to decide which command to 
use for the C- and C++ compiler (resp.).

For me, the following works (after removing o possible ./build dir):
# export CC=gcc-12; export CXX=g++-12; ./build.sh

That will build the program, but not install it. It tells you the command to 
install darktable (but you will have to precede that command with "sudo", as 
it needs to write to a system directory /opt 

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] DT 4.4.1 xmp madness

[Remco Viëtor]

On mardi 18 juillet 2023 08:52:39 CEST GianLuca Sarto wrote:
> No, it doesn't work like that.
> 
> Edits are always copied to xmp.
> 
> That flag is for people who do not want the folder populated with xmp
> for images they do not edit.
Provided you have *not* set "write sidecar file for each image" to "never".
Perhaps self-evident, but still...

But I can confirm that edits are written to the sidecar when that option is 
set to "on import".

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Optimising GPU memory usage in dt 4.4.1 on Windows ?

[Remco Viëtor]

On mercredi 12 juillet 2023 22:35:10 CEST marc.ca...@gmail.com wrote:
>  MAX MEM ALLOC:2048 MB

That's not the maximum darktable can use, it's the maximum opencl can get in 
one allocation request (a quarter of the total GPU memory). See e.g.
https://registry.khronos.org/OpenCL/sdk/3.0/docs/man/html/clGetDeviceInfo.html

And of course, there usually are other programs also wanting GPU memory 
(desktop manager, Firefox, ...)

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Darktable 4.4 on Ubuntu 22.04

[Remco Viëtor]

On mardi 4 juillet 2023 14:38:09 CEST Remco Viëtor wrote:

> Perhaps not in the official/base repositories, but:
>  https://launchpad.net/~ubuntu-toolchain-r/+archive/ubuntu/ppa
> has gcc-12 available.
or perhaps this repository is better:
https://ubuntu.pkgs.org/22.04/ubuntu-updates-main-amd64/




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Darktable 4.4 on Ubuntu 22.04

[Remco Viëtor]

On mardi 4 juillet 2023 13:52:17 CEST Guillermo Rozas wrote:
> The OBS packages are having problems because Ubuntu 22.04 lacks the
> required dependencies to compile darktable 4.4 (in particular, gcc12). As
> there is a single maintainer, and he doesn't use Debian/Ubuntu in "normal
> life", it could take a while or could be solved tomorrow, there is no ETA
> currently.

Perhaps not in the official/base repositories, but:
 https://launchpad.net/~ubuntu-toolchain-r/+archive/ubuntu/ppa
has gcc-12 available.
Once you have a binary, you no longer need gcc, so then it's not too difficult 
to create a package for others on the distro. 

However, if there are other missing dependencies (as is the case for OpenSUSe 
Leap), it gets trickier: if you have to hunt for libraries dt depends on, 
other would also have to hunt for those. _Not_ nice for a distributed 
package...

As for flatpak, snap and other self-contained packages, some seem to cause 
their own troubles as soon as you want more than vanilla use, due to the 
sandboxing they do (which is a good thing, until it isn't).

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Darktable 4.4 on Ubuntu 22.04

[Remco Viëtor]

On mardi 4 juillet 2023 12:49:17 CEST Bernhard wrote:
> Remco Viëtor schrieb am 04.07.23 um 10:29:
> 
> > On mardi 4 juillet 2023 08:15:36 CEST Bernhard wrote:
> > 
> >> here is a list of what works and what not ...
> >> https://build.opensuse.org/project/show/graphics:darktable
> >> (might be possible to get to work with a lot of hacks though - obviously
> >> like William did)
> >> William Ferguson schrieb am 04.07.23 um 08:05:
> >> 
> >>> I build it
> > 
> > (...)
> >
> >
> >
> > Perhaps not all that many hacks.
> >
> >
> >
> > You'll need GCC-12 (or newer?), and tell your build system to use it, and
> > you  may have to disable some optional parts.
> >
> > e.g. on OpenSUSe I needed to tell the system to use GCC-12
> > ("export CC=gcc-12; export CXX=g++-12),
> > and I removed openEXR (which I don't use, so removing it was easier than
> > trying to get the proper version...)
> > In case of such problems, have a look at the DefineOptions.cmake file, to
> > see what could be disabled.
> >
> > Of course, if you can't get a proper version for a *required* component,
> > it  gets more difficult...
> >
> > Remco
> >
> >
> 
> depends on the viewpoint - aka base knowledge.
> I tried to build master 2 months ago - and learned I have do upgrade my
> debian.  I did that - then I learned I have to do this and that recommended
> in pixls.us, github and here. I tried but never succeeded. But I'm not a SW
> developer. I'm happy if my scripts (where I don't have the necessity to
> compile) work for me. 
> So something like
> 
> > tell the system to use GCC-12
> 
> is difficult (ok, you give two commands here - but: as a user? as root? And:
> what are the side-affects?)
 
> So I decided to wait till "normal compile" in a "standard system" will work
> again.  Btw: using the "Testing" binary in "Debian Bookworm" (manual
> download) seems to work in the meantime as a workaround. 
> -- 
So you have a working dt. Good.

If you want to be sure you can go back to stable 4.4.1, make sure you have a 
backup of the databases: a change in the database scheme may make it 
impossible to go back to an older version (iirc, a version that updates the 
databases saves a backup copy, so make sure it is there...) 

btw, the full sequence I used to compile darktable 4.4.1 was:
#> export CC=gcc-12
#> export CXX=g++-12
#> ./build.sh --prefix /opt/darktable --build-type Release --install --sudo
All as user, you'll be asked for the root password just before the install to 
/opt/darktable (which is not user-writable).

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Darktable 4.4 on Ubuntu 22.04

[Remco Viëtor]

On mardi 4 juillet 2023 08:15:36 CEST Bernhard wrote:
> here is a list of what works and what not ...
> https://build.opensuse.org/project/show/graphics:darktable
> (might be possible to get to work with a lot of hacks though - obviously
> like William did)
> William Ferguson schrieb am 04.07.23 um 08:05:
> > I build it
(...)

Perhaps not all that many hacks.

You'll need GCC-12 (or newer?), and tell your build system to use it, and you 
may have to disable some optional parts.

e.g. on OpenSUSe I needed to tell the system to use GCC-12 
("export CC=gcc-12; export CXX=g++-12),
and I removed openEXR (which I don't use, so removing it was easier than 
trying to get the proper version...)
In case of such problems, have a look at the DefineOptions.cmake file, to see 
what could be disabled.

Of course, if you can't get a proper version for a *required* component, it 
gets more difficult...

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] darktable 4.4.1 released

[Remco Viëtor]

On dimanche 2 juillet 2023 17:29:28 CEST Jean-Luc wrote:
> Le 01/07/2023 à 08:19, Pascal Obry a écrit :
> > We're proud to announce the new corrective release of darktable, 4.4.1!
> > 
> > The github release is here:
> > [https://github.com/darktable-org/darktable/releases/tag/release-4.4.1](h
> > ttps://github.com/darktable-org/darktable/releases/tag/release-4.4.1).
> > 
> > As always, please don't use the autogenerated tarball provided by
> 
> > github, but only our tar.xz file. The checksums are:
> .../...
> 
> Hello,
> 
> A few days after 4.4 release announce, I tried to compile from sources,
> following the steps I used for 4.2 - and that was still up-to-date at
> this time.
> I ended up with abort due to many missing dependancies and gcc-12+
> mandatory.
> After some time spent to satisfy most of the dependancies, I still could
> not override the gcc problem.
> I finally thought I had solved it, after some more browsing, but the
> only result was a fatal system crash and I had to reinstall from scratch.
> I am now in the previous state of dt 4.2.1 running, but not willing to
> try again the compiling adventure without being sure I will not run into
> the same issue again.
> Have the compiling instructions been updated, so I will be able to reach
> the end of the process ?
> I am currently running Ubuntu 22.04-2, and the steps for getting dt from
> sources are those :
> 
(..)
Once you have the source code and all dependencies, what worked for me was:

export CC=gcc-12; export CXX=g++-12; ./build.sh --prefix /opt/darktable --
build-type Release --install --sudo

That forces use of the GCC version 12 C and C++ compilers before starting the 
build process. But make sure you have removed a "build" directory that may 
have been created in your source tree by earlier attempts: cmake caches some 
results there and will reuse them whenever it finds them. But if you changed 
anything, those cached results will be wrong...

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] dt 4.4 on Debian 11.7, apt update issue

[Remco Viëtor]

On vendredi 30 juin 2023 11:22:01 CEST Dieter Faulbaum wrote:
> Hello Michael,
> 
> if you look here:
> https://download.opensuse.org/repositories/graphics:/darktable/
> you can see, there is no Debian_11 directory (anymore?) only a
> Debian_12
> (and Debian_Testing and Debian_Unstable) directory.
> 
> Maybe download.opensuse.org doesn't support a oldstable version
> (of
> Debian)?

There *may* be another repository on OBS with such older distributions. Those 
will be maintained by individual users. 

However, it's quite possible no one maintaining such repositories uses older 
distributions, or it may become too difficult to get the code to compile (due 
to required versions of external libraries). E.g. latest darktable needs 
GCC12, which may not be available on Debian 11 (or Ubuntu 20.04)

Remco

(Even fairly recent distributions can run into problems: there's no build for 
OpenSuse LEAP 15.4 or 15.5, due to an issue with OpenEXR, I disabled openEXR 
for my private build, but that's not a good option for a repository, IMO)



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] obs has moved or something

[Remco Viëtor]

On lundi 26 juin 2023 14:27:22 CEST Michael wrote:
> I went here  and then to the OBS page
> to make sure I had the correct source. But instead of getting the options
> of operating systems I got this . If anyone
> has the correct source could I have it? And if it isn't too much trouble
> could I have the instructions to put the source in as well?

Why not just use the search option top tight of the OBS page? It gives you 
access to all repositories providing darktable (there are quite a few, some 
for many distributions, some only for a few)

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: Aw: [darktable-user] darktable 4.4.0 released

[Remco Viëtor]

On jeudi 22 juin 2023 10:40:42 CEST Remco Viëtor wrote:
> On jeudi 22 juin 2023 09:52:42 CEST Achim Dittrich wrote:
> > Hi!
> > 
> > To upgrade from my 4.2.1 installation on macOS Ventura 13.4.1, I
> > downloaded
> > the darktable-4.4.0-arm64 dmg file, opened and dragged it over to the
> > application folder - the standard procedure, that always worked fine in
> > the
> > past.
> > This time however, when I try to start darktable, I get the error message
> > that the application would be damaged and can not be opened and that I
> > shall move it to the bin.
> > After proper deinstallation I tried to install it again but that failed.
> > 
> > 
> > What can I do now?
> 
> Latest versions of MacOS require applications to be signed. The key needed
> for that needs to be purchased from Apple (at 100$/year iirc)... So
> darktable isn't signed.
> 
> There are ways to make such programs run anyway. A Google search should
> help. (Sorry, I'm not an Apple user, so I can't give the details)
Follow-up: Actually the darktable site has the instructions here:
https://www.darktable.org/install/#macos






darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: Aw: [darktable-user] darktable 4.4.0 released

[Remco Viëtor]

On jeudi 22 juin 2023 09:52:42 CEST Achim Dittrich wrote:
> Hi!
>  
> To upgrade from my 4.2.1 installation on macOS Ventura 13.4.1, I downloaded
> the darktable-4.4.0-arm64 dmg file, opened and dragged it over to the
> application folder - the standard procedure, that always worked fine in the
> past. 
> This time however, when I try to start darktable, I get the error message
> that the application would be damaged and can not be opened and that I
> shall move it to the bin. 
> After proper deinstallation I tried to install it again but that failed.
>  
>  
> What can I do now?
Latest versions of MacOS require applications to be signed. The key needed for 
that needs to be purchased from Apple (at 100$/year iirc)... So darktable 
isn't signed.

There are ways to make such programs run anyway. A Google search should help. 
(Sorry, I'm not an Apple user, so I can't give the details)

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] workflow hints

[Remco Viëtor]

On mardi 13 juin 2023 17:58:47 CEST Germano Massullo wrote:
> Good day, I would like to ask what is your darktable workflow when you
> have an album made of RAWs + JPEGs (camera generated), and you want to
> keep certain camera generated JPEGs and generate some JPEGs with darktable.
> I import only RAWs in darktable, and last time I ended up being confused
> because I could not recognize in lighttable which photos I retouched and
> which not.
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org
Photo's you edited have a small icon top right in the light table view.




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Module that still applies even if removed

[Remco Viëtor]

On dimanche 14 mai 2023 00:36:10 CEST Jean-Luc wrote:
> Le 13/05/2023 à 18:45, Remco Viëtor a écrit :
> > On samedi 13 mai 2023 18:41:52 CEST Remco Viëtor wrote:
> >> On samedi 13 mai 2023 18:00:17 CEST Jean-Luc wrote:
> >>> Le 13/05/2023 à 14:03, Remco Viëtor a écrit :
> >>>> On samedi 13 mai 2023 12:23:12 CEST Jean-Luc wrote:
(..)
> In my case, there is no preset defined in the module.
> By the way, I would expect that removing a module from a preset would
> prevent it from being applied and activated.
> There is definitely something I have trouble in understanding.
> I think Aurélien P. or Luc V. might have done videos where this could be
> explained, I just have to get ahold of the one.

In version 4.2.1, the *only* place where you can remove an individual module 
from a preset is in the module layout. And that layout has no influence on 
whether a module is auto-applied, auto-activated or not.

We are dealing with at least three different "presets" here:
- presets in individual modules, can be autoapplied or not.
- workflow presets, selected in the global settings dialog. These will 
determine which modules are initially activated, and cannot be changed by the 
user
- module layout presets. These *only* determine which modules will be visible 
in the module section of the right sidebar. This has *no* influence on the 
modules that will be activated for a new edit or after fully clearing the 
history.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Module that still applies even if removed

[Remco Viëtor]

On samedi 13 mai 2023 18:41:52 CEST Remco Viëtor wrote:
> On samedi 13 mai 2023 18:00:17 CEST Jean-Luc wrote:
> > Le 13/05/2023 à 14:03, Remco Viëtor a écrit :
> > > On samedi 13 mai 2023 12:23:12 CEST Jean-Luc wrote:
> > >> Hello,
> > >> 
> > >> Maybe I am doing things really wrong, but I am running into a problem
> > >> that the manual could not help me to solve.
> > >> If I add a module to a workflow, it is easy to have it autoapplied.
> > >> But, if I then remove it from the workflow, it still gets applied and
> > >> activated upon compress history or new import - whatever I do.
> > >> I am getting nuts - what on earth am I doing wrong ?
> > > 
> > > Which module?
> > > 
> > > And what exactly do you mean by "adding a module to a workflow"? Afaik
> > > (version 4.2.1) it's not possible to add a module to one of the
> > > predefined
> > > workflows.
> > > 
> > > What you can do is define auto-applied *presets*. But removal of the
> > > preset
> > > should stop them from being autoapplied...
> > > 
> > > Remco
> > 
> > Hello Remco,
> > 
> > Yes, you are right, wrong terminology from my side. What I was calling
> > workflow is actually a preset from the modules menu.
> > The machine running dt is three roms away from the one I am writing from.
> > I cloned the scene referred preset and rearranged the modules within,
> > tweaking some of them - filmic RGB is one.
> > Now, I would like to remove the filmic RGB module, but even then it
> > still gets applied at new import or history compression.
> > I tried with another module - sigmoid, that I just added - it behaves
> > the same : no way to have it forgotten once it has been added.
> 
> It sounds like you tried to change a built-in preset and store it under the
> same name. I don't think you can change the built-in presets.
> 
> But the solution is simple: select the layout you want to use as a basis,
> then go to "manage presets" and "duplicate" the preset. You'll be asked to
> give it a new name, and then you can adjust it to your liking.
> 
> (Rapidement: il faut choisir un "preset" comme base, le *dupliquer* et
> modifier le copie)
Oh, and a quick follow-up: module layout set through those presets doesn't 
influence which modules will be auto-applied. That's determined by your 
workflow settings in the "settings" dialog and possible presets in individual 
modules.



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Module that still applies even if removed

[Remco Viëtor]

On samedi 13 mai 2023 18:00:17 CEST Jean-Luc wrote:
> Le 13/05/2023 à 14:03, Remco Viëtor a écrit :
> > On samedi 13 mai 2023 12:23:12 CEST Jean-Luc wrote:
> >> Hello,
> >> 
> >> Maybe I am doing things really wrong, but I am running into a problem
> >> that the manual could not help me to solve.
> >> If I add a module to a workflow, it is easy to have it autoapplied.
> >> But, if I then remove it from the workflow, it still gets applied and
> >> activated upon compress history or new import - whatever I do.
> >> I am getting nuts - what on earth am I doing wrong ?
> > 
> > Which module?
> > 
> > And what exactly do you mean by "adding a module to a workflow"? Afaik
> > (version 4.2.1) it's not possible to add a module to one of the predefined
> > workflows.
> > 
> > What you can do is define auto-applied *presets*. But removal of the
> > preset
> > should stop them from being autoapplied...
> > 
> > Remco
> 
> Hello Remco,
> 
> Yes, you are right, wrong terminology from my side. What I was calling
> workflow is actually a preset from the modules menu.
> The machine running dt is three roms away from the one I am writing from.
> I cloned the scene referred preset and rearranged the modules within,
> tweaking some of them - filmic RGB is one.
> Now, I would like to remove the filmic RGB module, but even then it
> still gets applied at new import or history compression.
> I tried with another module - sigmoid, that I just added - it behaves
> the same : no way to have it forgotten once it has been added.

It sounds like you tried to change a built-in preset and store it under the 
same name. I don't think you can change the built-in presets.

But the solution is simple: select the layout you want to use as a basis, then 
go to "manage presets" and "duplicate" the preset. You'll be asked to give it 
a new name, and then you can adjust it to your liking.

(Rapidement: il faut choisir un "preset" comme base, le *dupliquer* et 
modifier le copie)

Remco





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Module that still applies even if removed

[Remco Viëtor]

On samedi 13 mai 2023 12:23:12 CEST Jean-Luc wrote:
> Hello,
> 
> Maybe I am doing things really wrong, but I am running into a problem
> that the manual could not help me to solve.
> If I add a module to a workflow, it is easy to have it autoapplied.
> But, if I then remove it from the workflow, it still gets applied and
> activated upon compress history or new import - whatever I do.
> I am getting nuts - what on earth am I doing wrong ?
Which module?

And what exactly do you mean by "adding a module to a workflow"? Afaik 
(version 4.2.1) it's not possible to add a module to one of the predefined 
workflows.

What you can do is define auto-applied *presets*. But removal of the preset 
should stop them from being autoapplied...

Remco







darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Filtering combining multiple dual conditions

[Remco Viëtor]

On dimanche 7 mai 2023 17:50:56 CEST Jean-Luc wrote:
> Le 07/05/2023 à 15:37, Remco Viëtor a écrit :
> > On dimanche 7 mai 2023 14:24:03 CEST Jean-Luc wrote:
> >> Hello,
> >> 
> >> I need to export pictures with eight different tags but belonging to the
> >> same roll.
> >> The only way to achieve this seems to combine (rollname and tag1) or
> >> (rollname and tag2) ... or (rollname and tag8) in the filter fields.
> >> The problem is : it seems limited to ten lines, which means five
> >> conditions.> 
> > Perhaps "collection filters" with "search" option can work? I.e. select
> > the
> > filmroll with the images you need, then set up a series of collection
> > filter rules with "search" for a (partial) tag name. Don't forget to
> > switch the combination option to "or" for the second and following rules!
> > 
> > This works well for simple searches (combining all rules with the same
> > boolean operator), not sure how easy to use it is with complex filtering.
> 
> Hello Remco,
> 
> I think - I hope - this is the way I did.
> Unfortunately, as you can see on the screenshot, I can only set five
> conditions, where I need much more.
> But maybe this is not what you meant.
Indeed, I was talking about the "collections filter", which is just below the 
"collections". Basically, pick the film roll you need in the "collections" 
section, then go to the section *below* that.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Filtering combining multiple dual conditions

[Remco Viëtor]

On dimanche 7 mai 2023 14:24:03 CEST Jean-Luc wrote:
> Hello,
> 
> I need to export pictures with eight different tags but belonging to the
> same roll.
> The only way to achieve this seems to combine (rollname and tag1) or
> (rollname and tag2) ... or (rollname and tag8) in the filter fields.
> The problem is : it seems limited to ten lines, which means five conditions.
> I could not find anything in the manual that would help me, but maybe did I
> not read correctly. Might I have missed sommething that could solve my
> problem ?
Perhaps "collection filters" with "search" option can work? I.e. select the 
filmroll with the images you need, then set up a series of collection filter 
rules with "search" for a (partial) tag name. Don't forget to switch the 
combination option to "or" for the second and following rules!

This works well for simple searches (combining all rules with the same boolean 
operator), not sure how easy to use it is with complex filtering.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] dt 4.21 - missing support?

[Remco Viëtor]

On jeudi 4 mai 2023 08:09:52 CEST Bernhard wrote:
> Remco Viëtor schrieb am 04.05.23 um 08:04:
> 
> > On jeudi 4 mai 2023 07:43:52 CEST Dr. A. Krebs wrote:

> >> Is there support for NIKON D500 and NIKON D850?
> >> If handling with pics, the models are indicated as unknown?

> >>
> > What do you mean by "handling with pics"? Do you see thumbnails in the
> > lighttable view? Do they open in the darktoom? Does demosaicing work?
> > Does
> > lens correction work?

> >
> > According to https://www.darktable.org/resources/camera-support/ those
> > cameras
 are supported.
> >
> >> Do I make a mistake?
> >> What to do to overcome the situation?
> > 
> > You may check https://raw.pixls.us/ if perhaps a raw mode was missed that
> > you
 are using?
> >
> > Alternatively, you could post one of the troublesome pictures on the
> > pixls.us
 forum to check if others also have problems opening it.
> >
> 
> I own(ed) both cameras - no problems, full support. Samples are available.
> If lenses are a problem: lensfun-update-data in terminal.

I noticed OP had posted a very similar question about the D500 (only) last 
June. While he got some replies, there was no follow-up on his part, so we 
don't know if the problem was solved at that time...

Given the replies, though, it really pointed to a problem local to his system.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Only 1 CPU (or better core) is used during export

[Remco Viëtor]

On jeudi 4 mai 2023 08:11:41 CEST you wrote:
> Remco Viëtor schrieb am 04.05.23 um 07:27:
> 
> > On mercredi 3 mai 2023 22:39:06 CEST Bernhard wrote:
> > 
> >> Remco Viëtor schrieb am 03.05.23 um 17:57:
> > (...)
> > 
> >> I see 4 cores running at 100% during export on my system ...
> >>
> >>
> >>
> > Are you using openCL?

> 
> no, not available on my machine since darktable does not support Intel GPU
> and with external GPU opencl always crashes after suspend-to-ram.
Then it makes sense to use multiple cores. OP did have openGL working, in 
which case using multiple cores doesn't really make sense. And handling 
multiple images in parallel is never faster than handling the same number in 
sequence.

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] dt 4.21 - missing support?

[Remco Viëtor]

On jeudi 4 mai 2023 07:43:52 CEST Dr. A. Krebs wrote:
> Dear all:
> 
> Is there support for NIKON D500 and NIKON D850?
> If handling with pics, the models are indicated as unknown?
> 

What do you mean by "handling with pics"? Do you see thumbnails in the 
lighttable view? Do they open in the darktoom? Does demosaicing work? Does 
lens correction work?

According to https://www.darktable.org/resources/camera-support/ those cameras 
are supported.


> Do I make a mistake?
> What to do to overcome the situation?

You may check https://raw.pixls.us/ if perhaps a raw mode was missed that you 
are using?

Alternatively, you could post one of the troublesome pictures on the pixls.us 
forum to check if others also have problems opening it.

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Only 1 CPU (or better core) is used during export

[Remco Viëtor]

On mercredi 3 mai 2023 22:39:06 CEST Bernhard wrote:
> Remco Viëtor schrieb am 03.05.23 um 17:57:
> 
(...)
> 
> I see 4 cores running at 100% during export on my system ...
> 
Are you using openCL?



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Only 1 CPU (or better core) is used during export

[Remco Viëtor]

On mercredi 3 mai 2023 17:16:03 CEST Jörn Rink wrote:
> Hello,
> i have one question. I recognize, that in general dt used more than one
> core. But especially when i export images to jpeg, only one core is
> used, image AFTER image is exported. Is there a chance to corce dt to
> use more than one core? When i use high quality, the process also uses
> only one core and disc io is not very high.
> 
> Is dt running better on more cores, when i compile it myself with g++
> or gcc switches? Does dt use openMP?
> thanks and greetings

Darktable is geared towards using the GPU whenever possible; the CPU is mostly 
used for "bookkeeping".

While it might theoretically be possible to handle multiple images in 
parallel, it wouldn't get you much (if anything): treating several images in 
parallel on the GPU cannot be faster than treating them sequentially.
In addition, you'd need to have the RAM to deal with multiple images.

Remco






darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] OpenCL not working, Nvidia driver - could not get platforms: -1001

[Remco Viëtor]

On mardi 2 mai 2023 16:26:06 CEST Marco wrote:
> Hi,
> 
> I've recently acquired a new GPU and fail to get it to work with
> darktable. Here are some infos:
> 
> System: Debian stable 5.10.0-21-amd64
> GPU: NVIDIA GeForce GTX 1650
> 
> I installed the driver using
> 
>   apt install nvidia-driver
> 
> as suggested by nvidia-detect.
> 
> The modules are loaded:
(...)
> But darktable doesn't use the GPU:
(...)
> The manual states:
> 
>   “… There is a structure of files in /etc/OpenCL on your system
>   that libOpenCL.so consults to find these libraries. See if you can
>   find something fishy in there and try to fix it…”
> 
> I don't have the directory /etc/OpenCL on my system. What do do? Any
> pointers appreciated on how to get it working. If you need more
> infos, please let me know.

You have the graphics drivers installed to use openGL, i.e. graphics display.
On OpenSuse, I need an extra package to use openCL: "nvidia-compute...". This 
package sets up the /etc/OpenCL directory (with a bunch of other stuff).

No idea how that's arranged on Debian, but look for packages related to 
nvidia, opencl and CUDA (another system to use the GPU for calculations).

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] darktable-users list, fewer postings?

[Remco Viëtor]

On lundi 24 avril 2023 03:06:45 CEST Terry Pinfold wrote:
> "that you follow m$ *only* shows your age, not your
> wisdom."
> 
> Maybe this passive aggressiveness on your part also chased me away.
> 
> On Mon, 24 Apr 2023 at 11:01, Patrick Shanahan  wrote:
(snip)

Maybe, but Patrick is right in that bottom posting is an accepted standard, 
and not just in email (usenet had the same rules).
But snipping of irrelevant material is also part of the netiquette...

And the aim of those "rules" is to make understanding of the message easier 
(not necessarily the reading, there's a difference).

And, related to Michael's parallel answer, MS has a habit of implementing 
standards in its own way, (often) in a way that's not compatible with other 
implementations.

Remco





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Suspended Support of camera models

[Remco Viëtor]

On vendredi 17 mars 2023 10:53:11 CET Jörn Rink wrote:
> Hello together,
> to my surprise i read about suspended support of old camaera models in
> darktable. First time i have noticed it in the changelog of 4.2.1.
> 
> For me, also one reason to use dark table, was to work with very old
> fotos, Bruce williams mentioned this argument in one of his darktable
> youtube videos, and i really like this argument.
> 
> So, is there a definition "(of done ;-)" when camera models are not
> supported any longer? Must there be raw sample once a year or once
> every two years?
> 

It's not a matter of age, but of available sample images (which is stated 
under the heading "Suspended Support").  See 
https://discuss.pixls.us/t/raw-samples-wanted/ and https://raw.pixls.us/ for 
the details, and what is missing.

So, if you have raw files for such cameras, which you can share under the 
required license, please upload them according to the instructions at 
raw.pixls.us .

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] DT Metadata via EXIF to wordpress

[Remco Viëtor]

On mercredi 15 mars 2023 12:27:24 CET Frank J. wrote:
> Am 15.03.23 um 07:45 schrieb Remco Viëtor:
> > On mardi 14 mars 2023 18:29:23 CET Frank J. wrote:
> >> Hi,
> >> I want to use the “description” metadata from darktable [1] as caption
> >> in WordPress.
> >> In german language there are some characters which are not in ASCII but
> >> in Unicode/UTF-8.
> >> Umlaute: ä ö ü ß  ->  HTML:
> >> 
> >> When a raw-picture is exported as JPG from darktable, then the
> >> “description” metadata is embedded into the exif-data:
> >> 
> >> - description whithout “umlaute” is written to EXIF ImageDescription
> >> field (ascii).
> >> 
> >> - description WHITH “umlaute” is written to EXIF UserComment field
> >> (unicode).
> >> 
> >> I try the Plugins "Exif Caption" und "Exif Details" for import into
> >> wordpress.
> >> 
> >> As written in [2] the UserComment should be imported to wordpress:
> >>   [“caption”]
> >>   (string) Set to a non-empty value of one of the following fields:
> >>   …
> >>   EXIF UserComment field if [“title”] is unset AND
> >> 
> >> EXIF:ImageDescription is less than 80 characters
> >> 
> >> But for pictures with “umlaute” the caption in wordpress stay empty. The
> >> description is lost.
> > 
> > Iirc, EXIF can only store ASCII characters, so umlauts (and things like
> > éèà) are not allowed there. That may make the library used for reading
> > the data discard the text as invalid.
> > 
> >   But that concerns the "real" EXIF metadata, not the equivalent fields in
> > 
> > XMP(*). IPTC also allows non-ASCII. So if at all possible, you may be
> > better off using XMP or IPTC metadata.
> > 
> > Remco
> > 
> > (*: There are three "types" of image metadata: EXIF, IPTC and XMP, which
> > differ in the way they are stored in the file, the character sets used,
> > and
> > the maximum field lengths. The fields in the EXIF and IPTC sections can
> > also be stored as XMP fields, but that doesn't mean they will be read
> > when you ask for an EXIF or IPTC field).
> 
> /> you may be better off using XMP or IPTC metadata./
> 
> I'm not a darktable-developer but just a user. So I only can use, what
> darktable supports.
(...)

All I can say is that it is possible to get captions and titles exported in a 
jpeg in the XMP metadata section.
Whether your wordpress plugin is able to read that section is another 
question, better asked on a wordpress list or forum...

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] DT Metadata via EXIF to wordpress

[Remco Viëtor]

On mardi 14 mars 2023 18:29:23 CET Frank J. wrote:
> Hi,
> I want to use the “description” metadata from darktable [1] as caption
> in WordPress.
> In german language there are some characters which are not in ASCII but
> in Unicode/UTF-8.
> Umlaute: ä ö ü ß  ->  HTML:
> 
> When a raw-picture is exported as JPG from darktable, then the
> “description” metadata is embedded into the exif-data:
> 
> - description whithout “umlaute” is written to EXIF ImageDescription
> field (ascii).
> 
> - description WHITH “umlaute” is written to EXIF UserComment field
> (unicode).
> 
> I try the Plugins "Exif Caption" und "Exif Details" for import into
> wordpress.
> 
> As written in [2] the UserComment should be imported to wordpress:
> 
>  [“caption”]
>  (string) Set to a non-empty value of one of the following fields:
>  …
>  EXIF UserComment field if [“title”] is unset AND
> EXIF:ImageDescription is less than 80 characters
> 
> But for pictures with “umlaute” the caption in wordpress stay empty. The
> description is lost.
Iirc, EXIF can only store ASCII characters, so umlauts (and things like éèà) 
are not allowed there. That may make the library used for reading the data 
discard the text as invalid.

 But that concerns the "real" EXIF metadata, not the equivalent fields in 
XMP(*). IPTC also allows non-ASCII. So if at all possible, you may be better 
off using XMP or IPTC metadata.

Remco

(*: There are three "types" of image metadata: EXIF, IPTC and XMP, which 
differ in the way they are stored in the file, the character sets used, and 
the maximum field lengths. The fields in the EXIF and IPTC sections can also 
be stored as XMP fields, but that doesn't mean they will be read when you ask 
for an EXIF or IPTC field). 



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] HowTo reset a module (exposure) within darkroom to defaults?

[Remco Viëtor]

On mardi 14 mars 2023 11:19:55 CET Thomas Schweikle wrote:
> Hi!
> 
> How can I reset a module, here exposure, to defaults?
> Whatever I did this module keeps "compensate camera exposure" active, sets
> exposure to +0.7ev and black level correction to -0.0002.
> I'd like to have it:
> * "compensate camera exposure" inactive,
> * "exposure" set to 0.0ev, and
> * "black level correction" to 0.0
> 
> Any idea how to set defaults without creating a new setting?

If you want exactly those values, best use a auto-applied preset. 

I *think* the black level correction is always set to -0.0002 to avoid 
introducing negative values. "Compensate camera exposure" and the +0.7EV 
exposuse are the defaults when you use the "scene-referred" for "auto-apply 
pixel workflow default". If you use "none" there, those should be inactive (?) 
(but you get *no* default activated modules)

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] darktable 4.2.1 released

[Remco Viëtor]

On dimanche 26 février 2023 13:04:39 CET tony Hamilton wrote:
(...)
> The whole process is most unsatisfactory. It can best be described
> through an analogy: it was what I expect would be the experience of a
> blind person driving a busload of passengers along the narrow track a
> couple of hundred metres up in the gorge of the Jinshan River near Hu
> tiao xia in Yunnan Province - with verbal instructions (in local Yunnan
> dialect) being given by a highly experienced master co-driver who was
> having his lunch at the same time. 'Ah, yes, I should have warned you
> about that fragile part of the track before you drove into it - it's
> just that I don't have to think about it now after so many trips. Not to
> worry; only 95% fatalities. We have enough salvageable bits to rebuild
> the bus'.

No need to be overly dramatic, if you can't get your head around the 
compilation process, there will be ready-made packages available.

Just not necessarily the day (or week) a new version is announced, the 
packagers are volunteers with other demands on their time.

But if you do not want to wait, you'll have to *learn* how to compile from 
source. Yes, that process takes time, the first time around. No, there are no 
shortcuts, complaining it's too complicated doesn't help (if anything, it will 
get you *less* help).

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] missing database

[Remco Viëtor]

On dimanche 12 février 2023 14:55:30 CET Jan Minekus wrote:
> Hello,
> 
> I try to install darktable again in my freshly installed computer.
> Problem is the standard database is missing. Local directory does not
> have a darktable directory which I can overwrite with a backup.
> I tried to start darktable without the database but that window is
> rubbish. What is wrong?
> What can I do??

Normally, on a fresh install darktable should start and create the necessary 
files, including the database. But you will have to re-import all your images, 
styles and presets...

Remco.

P.S. If you copied the old darktablerc file, darktable would start, expecting 
a database, in principle in the same directory as darktablerc 
(~/.config/darktable for me under OpenSuse linux)...




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Disaster with history stack

[Remco Viëtor]

On vendredi 3 février 2023 07:26:59 CET Andrew Greig wrote:
> Hi all,
> 
> Another consideration is that I have a set of images at Full Size jpg in
> a separate folder, unaffected by my carelessness. Can they be any help?
If you exported them with editing instructions in the jpeg metadata, I think 
you can use them as sidecars to get the edits back. 
Not sure if you can do that in batch, though.




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] darktable 4.2.0 released

[Remco Viëtor]

On vendredi 23 décembre 2022 08:48:01 CET Jan Ingwer Baer wrote:
> I have tried to look into the source, but i am lost and could not find
> where the information is extracted/decoded from the RAW-file EXIF data
> to get a guess for which cameras the RAW-file EXIF lens-distortion data
> are supported.
You may have to look at Exiv2 information to find that: afaik, all metadata 
handling is done through Exiv2. And that kind of metadata is usually in the 
makernote section, which is "rarely" documented...

So if Exiv2 isn't capable of decoding the relevant part of the makernote, 
darktable will have no information to work with. As exiv2 is an external 
library, the set of supported cameras can change w/o warning. 

I haven't found a list of "supported cameras" on the Exiv2 site. Not 
surprising: when is a camera supported? Very few makernotes are fully 
understood. And the makernote  support can vary between models of the same 
brand (and perhaps even between firmware versions of a given model?).
But perhaps try this page: https://exiv2.org/metadata.html





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Corrupted installation of darktable on Linux Mint 21

[Remco Viëtor]

On vendredi 16 décembre 2022 00:10:00 CET Willy Williams wrote:
> I've run into a challenge on my one Linux Mint 21 computer. Since
> yesterday, when I start darktable, I get the message below. I've deleted
> the lock files repeatedly, to no avail.  I've uninstalled and
> reinstalled darktable, but get the same message every time.  I am NOT a
> Linux wizard and haven't a good idea of how to resolve this issue. 
> Anyone got some good advice?  I'm hoping to come to a successful
> resolution before darktable 4.2 springs out later this month.
> 
> Willy Williams

I get such a message when I try to restart darktable too quickly after I 
closed it. After a few seconds (10-20? at most), the problem solves itself (as 
the previous run will have finished its clean-up).
Or you may have another copy running, e.g. on another virtual desktop.

 You can use several tools to check that (one you certainly have is "ps", so 
try in a terminal window (no need to use root for this):
ps -aux | grep -i "darktable"
That should show you any copy still running with an extra line for the "grep" 
program, so if you see two or more lines, there is a copy running...
(what that command does: 'ps -aux' lists all running processes on your 
computer, 'grep -i "darktable" ' then retains only those lines that have the 
word "darktable" in them.)

Another cause might be an unexpected shutdown of your computer, after e.g. a 
power cut; in such a case, darktable won't have had the time to clean up after 
itself. (although that wouldn't explain why you' have to remove lock files 
repeatedly).

Uninstalling and reinstall darktable won't remove the lock files, so it won't 
help.

Also, if you need to manually remove the lock files, it might be a good idea 
to verify the integrity of the databases before restarting darktable... The 
locks are there to prevent database corruption by another copy of darktable, 
so if they are there while there's no copy of darktable running, there's a 
good chance the last shutdown wasn't done correctly. That means the database 
may not have been closed correctly either.





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Mouse hover in darktable 4.0.1 under Linux Mint 21 apparently not working

[Remco Viëtor]

On lundi 5 décembre 2022 16:00:21 CET tony Hamilton wrote:
> Apologies to those who have already seen my poorly expressed request for
> help in discuss.pixls.us, but this problem remains with no method of
> resolution.

As it seems to be a problem concerning your Linux distribution, you might have 
a better chance of finding a solution on a forum or mailing list for your 
distribution.



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Time for a new system

[Remco Viëtor]

On mardi 6 décembre 2022 01:01:59 CET David Vincent-Jones wrote:
> OK, ... understood. But I an just a photographer and not responsible to
> shareholders.
Keep in mind that time spent on side activities costs a professional, and 
doesn't bring in any money. So the time cost of maintaining on-site storage 
has to be factored in comparing with off-site storage.

Nothing to do with shareholders, just maximising income (which is valid for a 
one-person business as well!).




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] How To View My Photo At Actual Print Size in DT

[Remco Viëtor]

On lundi 21 novembre 2022 15:04:35 CET Gray Card wrote:
> If I'm understanding your question right, look in the top left of the
> darkroom screen.  There's a small view of the photo.  In the bottom
> right corner of the small view, there is an icon that resembles four
> corner arrows. Click on that icon and you'll see how to select 100%.  I
> hope this is the answer you're seeking.
> 
That "100%" is a view in which 1 image pixel corresponds to 1 screen pixel. 

I don't think dt has an option to show the image at the actual print size: 
there is not really an option to **set** the print size, so how can you 
display at that size? Keep in mind that the whole notion of resolution in ppi 
is not really relevant for digital image files, as they do not have an 
inherent physical size. 

Take screen sizes: screens with a native size of 1980×1080 pixels exist in 
different physical sizes, an image of 1980×1080 pixels will display full 
screen on all of those...

> 
> On 11/21/22 08:35, Victor wrote:
> > I know that in photoshop you can view your photos at actual print size
> > via the view menu item "Actual Size" which, in a nutshell, considers
> > both the size of your print and the resolution of your monitor.
> > 
> > Is there any way to do the same in Darktable?
> > 
> > If yes, how?
> > 
> > Ciao
> > 
> > Vittorio
> > 
> > 
> > 
> > __
> > __ darktable user mailing list to unsubscribe send a mail to
> > darktable-user+unsubscr...@lists.darktable.org
> 
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Pale blue in specular highlights how do I eleminate it?

[Remco Viëtor]

On lundi 21 novembre 2022 05:04:53 CET Andrew Greig wrote:
> 
> Well. strange things have happened. After shutting down dt and trying
> again, I was confronted by large wait times for loading an image, and it
> seemed to me that if I was going to wait for each frame, I may as well
> load an image and wait, crop an image, wait, tweak exposure, wait etc.,
> and then apply to several images, maybe I will only have to wait as long
> as for one image. So I applied the changes using "Selective Copy" and
> "Selective Paste". Lo and behold (gadzooks, wtf?) the problem was
> miraculously cleared and I was able to return to normal activity, 3
> chairs - no waiting! I have no idea what went on, but all is fine now.
> 

If you are using calculation-intensive modules like "diffuse or sharpen" or 
"highlight reconstruction" with "guided laplacians", you can always disable 
them while doing the other corrections like cropping, exposure, and to a 
lesser extend filmic, then at the end re-enable them.

Remember that the order of activation of a module doesn't change the order of 
calculation, so when you (re-)activate a module doesn't change the final 
result.

And while the application of a history copy may be much faster, when you open 
the image in the darkroom, you will get all the processing with the associated 
waiting. Although minutes per image sounds excessive, unless you really went 
overboard with the modules mentioned, or don't have openCL.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Pale blue in specular highlights how do I eleminate it?

[Remco Viëtor]

On dimanche 20 novembre 2022 06:33:51 CET Andrew Greig wrote:
> Hi all,
> 
> My first beach shoot on a sunny day, has left me with pale blue
> highlights. Previous beach shoots have been overcast. I thought it may
> have disappeared if I reduced the exposure, nope. Link to image with the
> problem.
> 
> 
> https://drive.google.com/drive/folders/15y-pZv55Gy9VPe5R04D9CmUXr-vcx-MF
> 
> 
> I would appreciate any guidance on this, I am a studio photographer who
> has strayed into the "blue room" (outdoors).
> 
> Many thanks
> 
> Andrew Greig
Assuming you use filmic, and color calibration as recommended, I think what 
happened is:
- your image was overexposed, to the point the sensor clipped in the 
highlights;
- then "whitebalance" (set to "camera reference") and "highlight recovery" 
made sure those areas were pure white (which is supposed to happen);
- after that, "color calibration" set the "real" white balance to something 
below 5000K, which means all colours get shifted towards the cooler side, 
*including* the pure white in the over-exposed areas.

To get it back to something more acceptable, you can use filmic 
reconstruction, or in "highlight reconstruction" try the "guided laplacians" 
method (can be very slow: seconds per update), or both.

Or drop the "color calibration" white balancing, and only use the whitebalance 
module. You can still use the other "color calibration" functions when you set 
adaptation to "none (bypass)" in the CAT tab.

Hope this helps,

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Moving DT between OS and computers?

[Remco Viëtor]

On dimanche 13 novembre 2022 20:49:29 CET Bengt Falke wrote:
> Thank you for your tips!
> 
> I still don't understand what is in the database? I am a bit worried of
> loosing all the tags
> and hope they are in the XMP-files?
> 
> Regards/Bengt

You can open an xmp file in a text editor and check for yourself which tags 
are present. Look for the XML tags  (keyword 
hierarchy) and  ("flat" keywords).
Darktable can read its own XMP files, so tags that are in them will be read in 
the new version/installation.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] darktable: symbol lookup error: /usr/bin/../lib64/darktable/libdarktable.so: undefined symbol: _ZN7Imf_3_114ChromaticitiesC1ERKN9Imath_3_14Vec2IfEES5_S5_S5_

[Remco Viëtor]

On samedi 12 novembre 2022 17:07:27 CET Knut Krüger wrote:
> Hi to all
> 
> I am using opensuse tumbleweed and I am not able to use darktable since
> a long timen now. I expected that any update will solve this error but
> it does not.
> darktable: symbol lookup error:
> /usr/bin/../lib64/darktable/libdarktable.so: undefined symbol:
> _ZN7Imf_3_114ChromaticitiesC1ERKN9Imath_3_14Vec2IfEES5_S5_S5_
> 
> Any hints for me?
> 
> Kind Regards Knut
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org
That usually means you have a different version of the library than expected 
by the executable.

Not something I'd expect to happen in that particular library with an install 
through a repository.

Which version of darktable did you try to install, and from which repository?

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Updating the Database

[Remco Viëtor]

On vendredi 21 octobre 2022 10:27:55 CEST Paul Marfell wrote:
> I don't want my system slowed down just because there might sometime be a
> crash but it would seem sensible to set a flag in the dB when dt starts and
> unset it when it shuts down normally. If the flag is set when dt starts
> then it can run some checks or just warn the user.

Such a flag exists already. It's needed to avoid database problems when you 
try to start a second instance of darktable. Not a good idea to have two 
programs writing to the same file without good synchronisation. Hard to 
implement correctly, and not really needed for a program like darktable.

I've had several occasions where dt refused to start immediately after closing 
the previous session (a few seconds later, no problems) with a message 
indicating another copy was running (as that copy was still in shutdown and 
hadn't cleared the flag).



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] how to correct lens distortion when no camera / lens defined

[Remco Viëtor]

On dimanche 2 octobre 2022 20:36:52 CEST Jan Ingwer Baer wrote:
> I would like to see an option to do fully manually distortion
> correction, at least at the level of the simple pincushion/barrel
> correction slider in RawTherapee. Additional to this first-order
> correction, possible higher order corrections.
> 
>  From my simple view it should be possible to add an option 'Manual'. If
> this option is selected there should be additional sliders to manually
> set the correction parameters that are normally come from the
> lensfun-database.

Then you have two options:
- post a feature request on github
- code it yourself and post a pull request

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] how to correct lens distortion when no camera / lens defined

[Remco Viëtor]

On samedi 1 octobre 2022 11:28:23 CEST marc.ca...@gmail.com wrote:
> Hello fellow dt users,
> 
> Is there a way to correct lens distortion when no camera or lens is defined
> in the EXIF of the TIFF file?
> 
> I am experimenting with a few old camera's (60's and 70's ) and have the b/w
> film developed and scanned by a lab. One of my pictures is showing cushion
> distortion but I can't use the 'lens correction' module and try to enhance
> the picture. The lens correction module doesn't allow me to change settings
> because the lens is undefined (in the lensfun database). Should I use
> another module ? Or, since it's about a TIFF file, perhaps another open
> source program can deal with this ?
> 
> I am using dt 4.0.1 on Windows 10.

I think Hugin can create lens distortion models, and use them (and those 
models can be used in lensfun, so you could add the lenses to your private 
lensfun database, and/or upload them).
Do keep in mind that your images must contain straight lines for this to work.

See e.g. https://hugin.sourceforge.io/tutorials/calibration/en.shtml

Remco





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Export to piwigo

[Remco Viëtor]

On mardi 27 septembre 2022 09:36:30 CEST jordi carreras sanahuja wrote:
> I don't understand because I can login in darktable correctly. And
> export the image, too
>

Darktable is on your computer, and doesn't require a specific login. Once you 
are in darktable, export to local disk doesn't require a password either (as 
far as darktable is concerned).

 Piwigo is an online server, which will require a password to allow uploading 
to your account (or anybody could put all they want on your account)... Not 
using any of those services, I have no idea how dt handles such password 
requests.

> El 27/9/22 a les 8:32, Remco Viëtor ha escrit:
> > On mardi 27 septembre 2022 08:17:01 CEST Jordi Carreras Sanahuja wrote:
> >> Thanks Pascal. This the answer of piwigo
> >> 
> >> Hi Jordi,
> >> 
> >> We have just tested the export from Darktable to Piwigo and it works
> >> fine,
> >> on another Piwigo.com account.
> >> 
> >> Now on your account, I see that your Darktable discusses with Piwigo to
> >> get
> >> the list of albums and create new albums, but no photo upload. I see 56
> >> HTTP requests with error code 401 (unauthorized), but nothing more
> >> detailed
> >> 
> >> :-/
> >> 
> >> (...)
> > 
> > That 401 error code looks like it might hint at a mispelled or forgotten
> > password.
> > 
> > Remco
> > 
> > 
> > __
> > __ darktable user mailing list
> > to unsubscribe send a mail to
> > darktable-user+unsubscr...@lists.darktable.org
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Export to piwigo

[Remco Viëtor]

On mardi 27 septembre 2022 08:17:01 CEST Jordi Carreras Sanahuja wrote:
> Thanks Pascal. This the answer of piwigo
> 
> Hi Jordi,
> 
> We have just tested the export from Darktable to Piwigo and it works fine,
> on another Piwigo.com account.
> 
> Now on your account, I see that your Darktable discusses with Piwigo to get
> the list of albums and create new albums, but no photo upload. I see 56
> HTTP requests with error code 401 (unauthorized), but nothing more detailed
> :-/
> (...)

That 401 error code looks like it might hint at a mispelled or forgotten 
password.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Canon R5 R6 RP not available dt 4 or 4.0.1

[Remco Viëtor]

On lundi 12 septembre 2022 12:34:23 CEST Alex Delaforce wrote:
> Hi,
> 
> The Canon R series cameras are not listed in my darktable for lens
> correction etc. though I can  import the files and process them (apart
> from lens correction).
> 
> R5 R6 etc not listed on the page
> https://www.darktable.org/resources/camera-support/
If you can process the files (apart from lens correction), the camera is de 
facto supported...

Lens correction is a different matter: it is handled by a third-party library, 
called lens-fun. Lens recognition is done through another library, Exiv2. The 
latter tries to find the name of the lens used from the metadata associated 
with the raw file.

For newer lenses, it's possible that the lens isn't in the lensfun library: 
either the data aren't available yet, or the version you have isn't quite up-
to-date. You can use the command-line tool 'lensfun-update-data' to download 
the latest version of the lensfun data base. Note that the tool may be 
packaged seperately from the lensfun library and data (your package manager 
should show that on a search for "lensfun").

Another possibility is that the lens name returned by Exiv2 doesn't match the 
name in the lensfun database (*). But in that case, you should be able to find 
the lens in the drop-down list, where you can then check the spelling 
(ignoring white-space)

(* : different versions of  a lens can have a slightly different name.)



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Question about DT Color-Science + Bayer-Filters

[Remco Viëtor]

On dimanche 4 septembre 2022 08:33:02 CEST Jan Ingwer Baer wrote:
> Thank you for your answer, but i am a little bit confused. In the input
> color profile i can choose between standard color matrix, enhanced color
> matrix and different fixed input profiles (sRGB, AdobeRGB, Rec2020,
> etc). But i cant see any camera specific color profile. Is the camera
> specific profile 'under-the-hood'?
The dt manual says:
"Where an image has been captured in a raw file, the input color profile 
module will normally apply either a standard or enhanced color matrix 
*specific for that camera model*, ..." (emphasis added).
(https://docs.darktable.org/usermanual/3.8/en/module-reference/processing-modules/input-color-profile/)

Unless you know what you are doing, stay away from the fixed colour spaces 
with raw files.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] dt 4.0 corrupts disk

[Remco Viëtor]

On mercredi 17 août 2022 15:12:35 CEST Jean-Luc wrote:
> Le 17/08/2022 à 13:45, Patrick Shanahan a écrit :
> > * Jean-Luc  [08-17-22 06:12]:
> >> Le 16/08/2022 à 23:17, I. Ivanov a écrit :
> >>> On 8/16/22 13:59, Patrick Shanahan wrote:
>  * Jean-Luc  [08-16-22 16:42]:
>  I would definitely try another external drive and maybe try to store
>  on a
>  local drive and move the files after finishing.
> >>> 
> >>> Agree with Patrick.
> >>> 
> >>> It is more likely the DD responds with corrupting the data because of
> >>> the operation (writing xml I expect) than Darktable actually doing
> >>> something wrong.
> >>> 
> >>> If it is a spin disk – you may want to check it for bad sectors. But
> >>> what Patrick suggest should take the disk out of the equation
> >>> completely.
> >>> 
> >>> Regards,
> >>> 
> >>> B
> >> 
> >> Hello,
> >> 
> >> First, it is Testdisk, not memtest - I do not know why I always mention
> >> the
> >> second instead of the first... 
> >> And I do not know as well why I wrote the scan lasts two hours, since it
> >> takes more than twelve. 
> >> And, yes, does not really reflect the reality : as I have still not
> >> upgraded to 22.04, the problem might be caused by anything else than dt
> >> itself, so a log would have been a good starting point.
> >> My interrogation was whether dt coul have saved a crash dump though
> >> launched as usual - ie with no commuter, aso. The answear is obviously
> >> no,
> >> unfortunately.
> >> The last time it happened I ran a full test after restoration (another
> >> 12+
> >> hours) and no bad sector was detected.
> >> In any case, I have a backup of the files, I just need to get the
> >> sidecars
> >> then I can copy to another disk and restart clean.
> > 
> > you are still not eliminating possibilities with your current approach.
> 
> Yes, I do, since I intend to copy the files to another disk. I just need
> to retrieve my sidecars first.
> 
> > try first to export to a local disk.
> 
> That might not be really relevant, the problem does not occur
> systematically. Only two times since dt upgrade and heavy work
> (reworking my whole images collection, 24k+ photos).
> 
> > also try to another removable drive.
> 
> See above, 1st reply.
> 
> > I doubt "22.04" is the problem.
> 
> It is not, since I still have not upgraded. So, rather *my* 20.04 install
> with dt4 - if even. I have installed lots of stuff, and restarting from a
> fresh 22.04 clean install might be the solution. However, restoration is
> still in progress, so I have to wait a little bit more.
> 
> Rgrds,
> 
> J.-Luc
While you  are troubleshooting, perhaps also do a memtest on the RAM. A faulty 
RAM chip once gave me a lot of "bad sectors" on a HDD (and as they were marked 
as such, replacing the RAM stopped it getting worse, but didn't recover the 
"bad" sectors...)

In any case, I'd rule out hardware problems before upgrading to a new OS 
version...

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Ubuntu 22.04.1 and OpenCL

[Remco Viëtor]

On mardi 16 août 2022 21:51:35 CEST Mikael Ståldal wrote:
> So which OpenCL library do I need? I tried to install
> libopencl-clang-dev, but it does not help.

On opensuse that would be "nvidia-computeG0x" ("x" indicating the series of 
cards handled by the driver/library, should be the same as your main driver 
package "nvidia-gfxG0x..."). That package is part of the NVidia binary driver 
(closed source). Afaik, there's no associated -devel package?

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Ubuntu 22.04.1 and OpenCL

[Remco Viëtor]

On mardi 16 août 2022 20:58:53 CEST Mikael Ståldal wrote:
> After upgrading to Ubuntu 22.04.1, OpenCL does not work any longer. It
> worked on same machine with Ubuntu 20.04. I have compiled Darktable
> 4.0.0 from source.
> 
> Any clues?
> 
> 
> $ lsb_release -a
> No LSB modules are available.
> Distributor ID: Ubuntu
> Description:Ubuntu 22.04.1 LTS
> Release:22.04
> Codename:   jammy
> 
> $ uname -a
> Linux heimdall 5.15.0-46-generic #49-Ubuntu SMP Thu Aug 4 18:03:25 UTC
> 2022 x86_64 x86_64 x86_64 GNU/Linux
> 
> $ nvidia-smi
> Tue Aug 16 20:55:49 2022
> +---
> --+
> | NVIDIA-SMI 515.65.01Driver Version: 515.65.01CUDA Version:
> 11.7 |
> 
> |---+--+
> |--+|
> | GPU  NamePersistence-M| Bus-IdDisp.A | Volatile
> 
> Uncorr. ECC |
> 
> | Fan  Temp  Perf  Pwr:Usage/Cap| Memory-Usage | GPU-Util
> 
> Compute M. |
> 
> MIG M. |
> 
> |===+==+
> |==||
> |   0  NVIDIA T600 Off  | :08:00.0  On |
> 
>   N/A |
> 
> | 39%   50CP8N/A /  41W |   1506MiB /  4096MiB |  1%
> 
> Default |
> 
>   N/A |
> +---+--+
> --+
> 
> $ echo $XDG_SESSION_TYPE
> x11
> 
> $ /opt/darktable/bin/darktable-cltest
> [opencl_init] opencl related configuration options:
> [opencl_init] opencl: ON
> [opencl_init] opencl_scheduling_profile: 'default'
> [opencl_init] opencl_library: 'default path'
> [opencl_init] opencl_device_priority: '*/!0,*/*/*/!0,*'
> [opencl_init] opencl_mandatory_timeout: 200
> [opencl_init] opencl_synch_cache: active module
> [opencl_init] no working opencl library found. Continue with opencl disabled
That looks like a big clue to what the problem is...

> [opencl_init] FINALLY: opencl is NOT AVAILABLE on this system.
> [opencl_init] initial status of opencl enabled flag is OFF.
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Highlights recovery / Filmic v6

[Remco Viëtor]

On mardi 16 août 2022 15:11:06 CEST Jean-Luc wrote:
> Hello,
> 
> Running dt 4.0 from ppa on Ubuntu 20.04 LTS (still not 22.04 yet).
> I am trying to understand whether it makes sense or not to use
> Highlights reconstruction together with filmic v6, since it seems it
> brings its own highlight reconstruction tools.
> The french online help seems no longer available, and maybe I did not
> understand correctly.
> Could someone please confirm ?
I think the two tools work differently.

The highlight reconstruction module (at least for "clip" and "reconstruct in 
Lch" modes) works on individual pixels, where filmic uses the surrounding area 
as well to get some detail back in clipped areas.

The "laplacian" reconstruction in the "highlight reconstruction" uses the 
surrounding area as well, but is horrendously slow on larger areas.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] ETTR and under-expose?

[Remco Viëtor]

On mardi 9 août 2022 10:27:39 CEST Terry Pinfold wrote:
> " worked fine for me with darktable and ETTR (or almost R, leaving some
> space on the right side, before risking overexposure)."
> 
> I taught one of my photography students who was a painter and therefore
> very sensitive to subtle colour changes in images about ETTR. She later
> taught me that pushing this too far caused a loss of saturation. Having
> listened to some of AP's description of filmic I realise why this occurred.
> Saturation drops off towards the extremities. (...)

While that's true for the normal colour spaces we use in photography (and in 
what we see), I'm not sure it's true for a photographic sensor.

The main difference between a sensor and film is that the sensor response is 
*linear* with the light energy over the full usable range of the sensor. So 
increasing or decreasing exposure doesn't change the ratio between the 
colours, as long as you avoid clipping (or rather, signals over the raw white 
point*)

That does not mean the *in-camera jpeg* won't show desaturation if you push 
exposure to the right. But in-between the raw signal and the jpeg, the camera 
has at least applied a tone curve (cf. dt's basecurves) and pushed the colours 
into sRGB (or AdobeRGB, if set).

(*: the raw white point can be set a bit lower than the theoretical clipping 
point, to avoid non-linearities.)



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Recommended processing order

[Remco Viëtor]

On lundi 8 août 2022 20:35:52 CEST Terry Pinfold wrote:
> Just looking at the DT pipeline and I see cropping is placed before filmic.
> For most images I doubt that cropping would have a significant effect on
> tone mapping, but if it is obvious that it would then do it early in the
> processing steps.

No, cropping has no effect on the tone mapping (other than that you may want 
to adjust the black and white reference; depending on what you cropped out).

So if you want, you can start your editing with cropping. Just be aware that 
certain other operations may require you to disable the module temporarily or 
force you to review your crop.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Recommended processing order

[Remco Viëtor]

On lundi 8 août 2022 18:16:55 CEST David Vincent-Jones wrote:
> My first objective with raw data is to establish the outline and tones
> of the final image. I use perspective control and cropping prior to any
> other function. Where these modules come in the 'pipeline' is really of
> little importance to me. The 'frame' of the image is key to all further
> processing..
 As I said elsewhere, some modules are better applied before cropping.
Among these are "lens correction" (it often deforms the image, which van 
require small corrections to the crop) and "retouch" (where you can only use 
source regions you can see...)

Apart from such exceptions, you can indeed use crop early in your workflow.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] ETTR and under-expose?

[Remco Viëtor]

On lundi 8 août 2022 17:58:38 CEST David Vincent-Jones wrote:
> My understanding:
> 
> The 'L' histogram is based on the potential JPEG output  however the
> R, G, B histograms are based on the RAW data. This was information
> offered, I think correctly, from a Fujifilm source.
And may very well be correct *for Fuji* (my camera doesn't have an 'L' 
histogram, for starters.) Even for Fuji, it may depend on the exact model.

> I wonder what others think that their camera exposure meter is doing?
> What do people think that 'auto-exposure' achieves? Well, here is my
> take: left to its own devices the camera believes that it is looking at
> a 18% gray card and will establish an exposure at a specific level below
> this value to account for highlight values. (in my case the exposure
> appears to be precisely 1 stop below, but other cameras are definitely
> different.) If I have a situation that involves back-lighting or other
> anomalies, then I use the 'compensation dial' to further make such
> adjustments.
While indeed the exposure to 18% gray seems standard, all other settings are 
camera-dependent. As are the tone curves applied to get the in-camera jpeg 
(preview).

> Remember ... filmic tone-mapping is based on a 'pivot point' of an 18%
> gray and not on an ETTR theory.
Of course filmic has nothing to do with ETTR, which is a method use at the 
moment you take the picture.

> When I open an image (RAW) for processing my 1 stop Fujifilm allowance
> is automatically reversed and so also is the compensation dial setting.
> Using this simple system I find that 95% of my images require absolutely
> zero further adjustment of the basic exposure in dt.
And it's the 5% remaining that requires 95% of the effort, and that's where 
you have to pay a bit more attention to exposure etc.
For easy situations, you can be a bit less precise with exposure and post-
processing. When you get into tricky backlit situations, theatre and such, you 
either work very carefully (if possible!) or accept a loss in image quality 
(more noise, some burned-out regions, ...), or, often enough, both...

> Why make it so hard when our tools make it so easy?
Two reasons:
- Our tools make it easy for the common situations
- The ease of use often means accepting a (possibly small) loss of quality 
(e.g. shadow noise is still present, so properly using ETTR can gain you a bit 
there).

Whether all this is important depends on the kind of photography you do, and 
the subjects you deal with. A fine-art nature photographer who loves backlit 
scenes will not work the same way as a studio photographer or a photo reporter 
(who might not even bother with raw).

But even then, it never hurts to understand the theory behind common terms, 
and know how to apply the methods involved. Or, as in this case, get a 
clarification for the manual (for me, the text looks correct, but the author 
glossed over a few steps to get from A to B).

> On 2022-08-08 01:38, Terry Pinfold wrote:
> > This seems  to be conflicting info. I would read it as  ettr is
> > recommended, however, even under exposing by up to one stop is not
> > detrimental.
> > 
> > On Mon, 8 Aug 2022, 18:17 Mikael Ståldal,  wrote:
> >> I don't quite understand this section in the manual:
> >> 
> >> 
> >> https://docs.darktable.org/usermanual/development/en/overview/workflow/pr
> >> ocess/#take-a-well-exposed-photograph
> >> 
> >> It recommends exposure to-the-right (ETTR), but then talks about
> >> under-expose images by 0.5 to 1 EV. Isn't ETTR about over-exposing
> >> rather than under-exposing?
> >> 
> >> _
> >> ___ darktable user mailing list
> >> to unsubscribe send a mail to
> >> darktable-user+unsubscr...@lists.darktable.org
> > 
> > __
> > __ darktable user mailing list
> > to unsubscribe send a mail
> > todarktable-user+unsubscr...@lists.darktable.org
> 
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Recommended processing order

[Remco Viëtor]

On lundi 8 août 2022 11:41:15 CEST Mikael Ståldal wrote:
> The
> https://docs.darktable.org/usermanual/development/en/overview/workflow/proce
> ss/ section in the user manual outlines the recommended processing order. It
> places cropping towards the end along with creative adjustments.
> Wouldn't it make more sense to do cropping first, before tone
> adjustments etc?

Well, the method in the section you refer to ("introduction to darktable's 
workflow") starts with doing the required "technical" corrections with the 
minimum number of modules needed: "exposure", "filmic RGB" and "color balance 
RGB". The aim there is to get a decent starting point for further "creative" 
corrections in as simple a way as possible.

But that section is an **introduction**, i.e. kept as simple as possible to 
get the principles across. Once you are a bit more familiar with darktable and 
its tools, nothing stops you from using a different working order. 

And certain operations are easier before cropping:
- "lens correction" can move parts of the image a bit
- "rotate and perspective" will move and/or deform parts of the image
- in "retouch"  you can use 'source' areas outside the final cropped image, 
provided you can see them while retouching.

Remember that the order in which you use modules doesn't influence the order 
in which they are applied to the image. So even if you activate "rotate and 
perspective" after cropping, it will still be applied before the crop. Which 
means your crop is probably no longer exactly what you intended.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] ETTR and under-expose?

[Remco Viëtor]

On lundi 8 août 2022 10:15:47 CEST Mikael Ståldal wrote:
> I don't quite understand this section in the manual:
> 
> https://docs.darktable.org/usermanual/development/en/overview/workflow/proce
> ss/#take-a-well-exposed-photograph
> 
> It recommends exposure to-the-right (ETTR), but then talks about
> under-expose images by 0.5 to 1 EV. Isn't ETTR about over-exposing
> rather than under-exposing?
The idea behind "Expose to the right" is that you expose  to get as much 
usable signal in the shadows, without over-exposing the highlights where you 
want to keep detail. So in a low-contrast scene that probably means "over-
exposing" compared to the meter reading. But when you have a high-contrast (or 
high dynamic range) scene, with part in full sunlight and part in shadow, you 
may have to *under-expose* wrt to the meter reading. 

That is based on the sensors we had 10-15 years ago, which had a much more 
limited dynamic range than modern sensors. That's also why exposure bracketing 
and HDR work were discussed so much.
Modern sensors have a much larger dynamic range (up to 12-13EV), so you have a  
lot more data in the shadow parts. That means that ETTR is less important 
("less", not "not").

But you **still** have to prevent over-exposure of any area of interest in 
your image. Any pixel that gets saturated means you lose data. If you have 
zones where one colour is over-exposed (i.e. the pixels are saturated and 
signal is clipped), you lose colour information for that zone, but details 
should be recoverable. If all three colours are ever-exposed, you have neither 
colour nor details in that zone, and you end up with flat white (or magenta 
due to the whitebalance correction, which pushes red and blue).

And the in-camera histogram is of remarkably little value in judging over 
exposure, as it is based on the in-camera jpeg, not on the raw data. Hence the 
remark about under-exposing 0.5-1 EV. Avoiding clipping in the highlights is 
really that important (and very easy to do: take a sunset, unless you under-
expose a lot, the sky and clouds near the sun will be clipped. The sun as 
well, but that's really hard to avoid...)

The above is for working with raw files, if you only take jpeg images, the 
situation is easier: expose to get the best possible image in camera, large 
edits are not easy to do correctly (if at all possible) on jpeg files.



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] start raw in gimp

[Remco Viëtor]

On mardi 19 juillet 2022 19:25:03 CEST Michael wrote:
> there was a slight typo:
(...) 
So you sent the whole 25k again...
 
spam folder...





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Color calibration overloaded?

[Remco Viëtor]

On jeudi 14 juillet 2022 17:15:00 CEST Mikael Ståldal wrote:
> Is it only me who thinks that the color calibration module is quite
> overloaded with functionality, at least from a UI perspective?
> 
> The user manual says
> 
> "This simple yet powerful module can be used in the following ways:"
> 
> then lists five ways for it to be used, and then follows a very long
> description of all the options.
> 
> Not sure I would agree this is a "simple" module. Maybe from a
> technical/mathematical standpoint, but not from usage.
> 
> To me, it would make more sense if the CAT part was its own module, and
> the channel mixing was its own.

Short answer: all the various ways of using the color calibration module use 
the same math, and all the settings can be combined into one 3×3 matrix, which 
makes the module very efficient.

Splitting the module means more calculations,  more conversions, and more 
exchanges between main memory and GPU memory.



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] "while balance applied twice" warning in DT 4.0

[Remco Viëtor]

On mercredi 13 juillet 2022 11:41:40 CEST Mikael Ståldal wrote:
> I'm having a hard time finding stuff on pixls.us. Can you give some
> pointers?
> 
> On 2022-07-13 10:42, Remco Viëtor wrote:
> > For more precise information, you may want to check Aurelien Pierre's
> > posts
> > and articles (on pixls.us iirc).
> 
> 
> darktable user mailing list
> to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

https://discuss.pixls.us/t/introducing-color-calibration-module-formerly-known-as-channel-mixer-rgb/21227



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] "while balance applied twice" warning in DT 4.0

[Remco Viëtor]

On mercredi 13 juillet 2022 10:20:50 CEST Mikael Ståldal wrote:
> Now I get the same problem when importing a new photo in 4.0. It's fine
> right after import, but as soon as I change something in the White
> Balance module, I get the "while balance applied twice" warning, and in
> Color Calibration it says "white balance module error".
> 
> I get this in the console:
> -
> [color calibration] white balance error
> [color calibration] white balance error
> [color calibration] white balance error
> [white balance] double application of white balance
> [white balance] double application of white balance
> [color calibration] white balance error
That's exactly the planned behaviour: if you want to use the color calibration 
module, the white balance module should be set to "camera reference".
The reasoning behind that is that the demosaic module needs some form of white 
balancing/camera correction to give a decent result, but the proper place to 
correct for illuminant (aka colour of scene lighting) is later in the 
pipeline. 

For more precise information, you may want to check Aurelien Pierre's posts 
and articles (on pixls.us iirc).

So, either leave WB module in "camera reference", correct white balance in 
color calibration module, 
or don't use the color calibration module for setting the white balance. You  
can set the adaptation to "none (bypass)" if you want to use the other 
possibilities of this module.

Remco

Remco. 



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] An issue with watermarks under Linux Mint

[Remco Viëtor]

On dimanche 10 juillet 2022 18:12:19 CEST Willy Williams wrote:
(...)
> As a related note, I've not been able to locate where the darktable
> executables reside under Linux Mint.  They seem to be carefully
> concealed, despite the fact that darktable seems to be working properly,
> as do all the other applications that I've managed to load.
> 
> Willy Williams

If you can easily start dt from the command line, the executable must be in a 
directory specified in $PATH (try "echo $PATH" in a terminal).

Or your package manager might be able to tell you were the various files are 
installed.

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Missing Flip

[Remco Viëtor]

On samedi 11 juin 2022 18:16:19 CEST David Vincent-Jones wrote:
> I notice that 'flip' is no longer available in the crop module. Will it
> reappear at some point?

Module "orientation", "transform" section, two buttons on the right...



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Exposure module, what happens here?

[Remco Viëtor]

On dimanche 15 mai 2022 17:33:10 CEST Niranjan Rao wrote:
> Greetings,
> 
> darkatable version 3.8.1, on Ubuntu 20.04
> 
> 
> I'm trying to understand what happens when I change the exposure module
> mode from automatic to manual. I've noticed it many times, if I change
> the mode from automatic to manual, application shows "working" message
> and does something to image. I've not changed any other settings of
> exposure module yet, just switching the mode.

If I switch the module from manual to automatic (or back) I see a small change 
in the applied exposure. It looks like the two sets of parameters are 
completely independant.
That means that the whole pipeline has to be recalculated, including possibly 
'expensive' modules like "diffuse or sharpen".

Remco





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Question regarding .svg watermarks comparing Windows and Ubuntu Linux

[Remco Viëtor]

On mercredi 4 mai 2022 02:38:36 CEST Top Rock Photography wrote:
> The issue has nothing to do with Windows, nor fonts, and everything to do
> with darktable.
> I have only used Linux for the last umpteen years, and I had created some
> watermarks —in Linux— to use with darktable. Everything was fine until the
> 3.x branch. (may have been earlier).
> 
> All my text, which looks great in any other SVG viewer, and used to look
> great in darktable, now has overlapping characters. Converting the text to
> a curve does not work, since the majority of my text is dynamic. $Title,
> $Description, $date, $sequence-number, etc.
> 
> I should have made a bug report, but, due to the excellent team at
> darktable, I kept telling myself, “no problem, boss, dem soon fix dis.”
> 
> Well, I think that it is time for a bug report.

Then what's stopping you?

*We* cannot do it for you, as we don't know exactly what you did/didn't do, 
nor do we know your system.

And your mail certainly doesn't give enough info for a decent bug report...

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] The configuration and operation files of darktable V4.2

[Remco Viëtor]

On mercredi 6 avril 2022 23:00:19 CEST tony Hamilton wrote:
> On 06/04/2022 20:48, Patrick Shanahan wrote:
> > * tony Hamilton  [04-06-22 14:30]:
> >> Is this most useful of documents, by Jean-Pierre Verrue, dated
> >> 06/03/2019,
> >> the latest version for dt 3.8 ?
> > 
> > what document would that be?
> 
> The document with the name as given in the subject line of my original
> post. Easily found in Pixls,Us under the topic
> 
> 
>   "How to tell where darktable’s data folder is
>  
>  28>"
A link which could have been given in the original post, avoiding the need for 
many to waste time before they can even start answering the question...

It's simple, really: providing the information needed up front makes it easier 
to get an answer (and usually quicker). No protocol, just basic self-interest.
The more as those that are best placed to answer, are usually also the most 
solicited.

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Life cycle of darktable tag data (in Windows)

[Remco Viëtor]

On mercredi 6 avril 2022 09:48:05 CEST tony Hamilton wrote:
> If an image is removed from the dt catalog in the windows version, is
> the tag and metadata, which had been applied to the image, automatically
> lost? Or is it still available in some way such that if the image is
> added back into the catalog, that tag and metadata will be re-associated
> with the image?
> 
> As a secondary question, where is this data held, in the Windows version?
I'll let the devs answer the part about where data is stored and whether re-
associating old data with a new image is possible...

My question is: would it be *acceptable* to have a newly imported image get 
the data from an older, removed image? What criterium would you use to 
associate the new image with the old data? "name" isn't particularly reliable, 
"id" isn't possible (the 'new' images hasn't one yet...), so that would leave 
a checksum or fingerprint (relatively slow operation).

Also, I'd expect the tags and added metadata to be stored in the database, and 
remain there after just removing an image (just marked as deleted).
But: that database can be cleaned/purged/compressed every now and then. In 
that case, the data of removed images *is* deleted from the database. And that 
would make re-using existing data unpredictable...

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Recovering xmp data from Exported jpg files

[Remco Viëtor]

On dimanche 20 mars 2022 08:28:07 CET Andrew Greig wrote:
> On 20/3/22 17:59, Remco Viëtor wrote:
> > On dimanche 20 mars 2022 00:20:09 CET Andrew Greig wrote:
> >> I have been told that I can recover the xmp files from my exported (and
> >> correct) jpgs. So I need to discover the process. I imagine that I would
> >> need to load my jpg files into Darktable instead of my RAW files with
> >> their ovewritten xmp files. But what is next, please?
> > 
> > No, you should *not* inport your jpegs in darktable.
> > 
> > What you can do, is read the jpeg as a sidecar file:
> > select the required image in the lightroom, and in the right sidebar, the
> > module "history stack" has an option "load sidecar file". Click that and
> > select the jpeg you want to use the edit from.
> > Afaik, there's no easy way to automate that for a group of images.
> > 
> > And may I suggest regular backups of at least the database (ideally also
> > of
> > the images and sidecar files)?
> > 
> > Remco
> > 
> > P.S. your previous problems were with the *"export"* module, here you seem
> > to have been using the "history stack module". Different modules, so the
> > "append/ overwrite" switches are independent!
> > 
> > 
> > __
> > __ darktable user mailing list
> > to unsubscribe send a mail to
> > darktable-user+unsubscr...@lists.darktable.org
> Hi Remco,
> 
> I have done exactly as instructed, Light-table view with the selected
> file highlighted
> 
> In the History stack module I clicked on the Load Sidecar file and
> navigated to the Folder containing my full size edited jpgs, choice was
> xmp files or all files. Nothing showed with xmp files button but when I
> selected all files, I had access to the jpgs but selecting the correct
> jpg did nothing but return me to the Light-table view.  Is there any way
> to just extract the xmp data without Darktable running?
> 
> Kind regards
> 
> Andrew
Exiftool or Exiv2 can do that, as far as I know. But I've never tried to do 
that.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Recovering xmp data from Exported jpg files

[Remco Viëtor]

On dimanche 20 mars 2022 00:20:09 CET Andrew Greig wrote:
> I have been told that I can recover the xmp files from my exported (and
> correct) jpgs. So I need to discover the process. I imagine that I would
> need to load my jpg files into Darktable instead of my RAW files with
> their ovewritten xmp files. But what is next, please?

No, you should *not* inport your jpegs in darktable.

What you can do, is read the jpeg as a sidecar file:
select the required image in the lightroom, and in the right sidebar, the 
module "history stack" has an option "load sidecar file". Click that and 
select the jpeg you want to use the edit from.
Afaik, there's no easy way to automate that for a group of images.

And may I suggest regular backups of at least the database (ideally also of 
the images and sidecar files)?

Remco

P.S. your previous problems were with the *"export"* module, here you seem to 
have been using the "history stack module". Different modules, so the "append/
overwrite" switches are independent!



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] 3.8.1 Under expose / black level warning error ? and FILMIC !

[Remco Viëtor]

On mercredi 16 février 2022 21:42:41 CET Terry Pinfold wrote:
> I agree with Bruce that Filmic tries to reduce clipping of the blacks and
> the whites and this is the desired behaviour of filmic. It is really worth
> watching Aurélien's video on this. Even if some of it gets very technical
> it gives a good understanding of the what and why of filmic.
> https://www.youtube.com/watch?v=leZVK2s68QA
> 
Also, a right-click on the over/underexposure button (the divided square) 
gives you the option to indicate what you consider under-exposure in EV below 
100% white. Useful if you know how the final image is going to be used 
(screen, print...). 



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Mask won't apply

[Remco Viëtor]

On lundi 7 février 2022 20:01:16 CET Patrick Shanahan wrote:
> * Hieke van Hoogdalem  [02-07-22 13:13]:
> > ? I don't see new text __
> > 
> > Op 07-02-2022 18:51 heeft Patrick Shanahan  
geschreven:
> > * Hieke van Hoogdalem  [02-07-22 12:47]:
> > > Tnx but like I explained: Safari has huge problems. I can’t visit
> > > tutorials now. I can send and receive emails and use Whatsapp,
> > > luckily, but my browser refuses often. Like now.
> > > 
> > > Hartelijke groet,
> > > 
> > > Hieke van Hoogdalem
> > > 
> > > Van: Ludger Bolmerg 
> > > Datum: maandag 7 februari 2022 om 18:35
> > > Aan: 
> > > Onderwerp: Re: [darktable-user] Mask won't apply
> > > 
> > > Here you can find some excellent videos on masking in DT
> > > 
> > > https://www.youtube.com/channel/UCxHYygok15XQ6bqu9FK-oCw/videos
> > > 
> > > On 2/7/22 18:19, Hieke van Hoogdalem wrote:
> > > 
> > > Hi there,
> > > 
> > > I'm trying to apply masks. The idea is simple, but I can't get it
> > > working. I searched for YouTube tutorials, but they don't cover my
> > > problem. Also: my new M1 Macbook is malfunctioning and often stops
> > > internet all of a sudden without restarting properly. So my online
> > > resources are really small. Watching on my phone is an option, but
> > > the content doesn't cover my question, like I said.
> > > 
> > > So here's the situation. I have this image that I had to brighten up
> > > quite a bit, but now a small part is overexposed. I want to use a
> > > mask to bring that down. But… when I draw a mask and change the
> > > exposure, it's being applied to the complete image. (Yes, I did
> > > create a new instance in the expose module before I started to
> > > create the mask.)
> > > 
> > > I found this tutorial that showed something about 'blend mode' and
> > > telling that 'normal' applies to the complete image. You have to
> > > chose 'mask'. But when I click on 'blend mode' I get a much larger
> > > list without the option 'mask'.
> > > 
> > > What can I do to make this work?

> maybe you should actually read postings.
And ccould you please remove excessive signatures?




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: ****SPAM(2.9)**** Re: ****SPAM(2.9)**** Re: [darktable-user] Mask won't apply

[Remco Viëtor]

On mardi 8 février 2022 21:41:07 CET Terry Pinfold wrote:
(removed about *2000* lines... RV)
> --Hi Hieke,
>I am posting my reply again from my gmail server rather than my
> University server. I apologise that I can not control if my post appears at
> the top or bottom of the list. The server does whatever the server wants to
> do.  I have also tried to attach a screenshot for clarity.

That's... not true in my experience: the editor may put the cursor at the 
bottom or the top of the quoted text, but they all allow you to move that 
cursor. 

But in either case, that's no reason to leave *over a 1000 lines* of 
irrelevant material, including signatures. Not all Terry's fault in this 
case...

> With respect I say that you seem confused about working with masks. I hope
> the following explanation may help you. Masks for me are the whole reason I
> love using DT compared to alternative programs.
> 
>  I often want to brighten or darken a section of an image, similar to how I
> would have dodged or burned in a traditional film darkroom. To achieve this
> with a digital image I make a new instance of exposure and adjust the
> exposure slider (sometimes I deliberately overdo the adjustment and will
> return to it later). This adjustment moves the whole of the image's
> exposure, but then I use the drawn mask option which looks like a crayon. I
> then often use the add path option to very roughly select the area I want
> to apply the effect to. The transition will be harsh and obvious. To fix
> this problem I then move the feather radius and blurring radius sliders
> about a third of the way to the right (amount may vary between images).
> This feathering and blurring then creates an invisible transition. I return
> to the exposure slider and fine turn the strength of the effect.

But that's after you have a basic mask selecting the area you want to modify. 
Problem here seems to be in that first step...

> Forget the blend mode options until you have mastered basic masking
> techniques.
I don't agree. You select a blend mode only *after* you have a mask that 
affects the area you want. 






darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Tokina lens not detected

[Remco Viëtor]

On lundi 7 février 2022 18:29:08 CET Ludger Bolmerg wrote:
> Hi
> 
> I am on DT 3.8.0 on FreeBSD with lensfun 0.3.95. Although this may not
> be a DT issue I would appreciate any hint how to solve my problem.
> DT does not recognize my Tokina 11-20mm F2.8 IF DX lens and only shows
> "lens 139" instead. I double-checked the lensfun database and found an
> entry for this lens.  I also ran lensfun-update-data to upddate the
> lensfun database.
As you are using a third-party lens, the information available to lensfun 
might cover several lenses. As your lens is in the database, you can use the 
drop-down menu in the lens correction module to manually select the lens, or 
create a .exiv2 file in your home directory linking the lens ID to the lens 
(look on the web for details on how to do that)



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Mask won't apply

[Remco Viëtor]

On lundi 7 février 2022 18:19:57 CET Hieke van Hoogdalem wrote:
> Hi there,
> 
> I'm trying to apply masks. The idea is simple, but I can't get it working. I
> searched for YouTube tutorials, but they don't cover my problem.
>(...)
> So here's the situation. I have this image that I had to brighten up quite a
> bit, but now a small part is overexposed. I want to use a mask to bring
> that down. But… when I draw a mask and change the exposure, it's being
> applied to the complete image. (Yes, I did create a new instance in the
> expose module before I started to create the mask.)
If you are using the scene-referred workflow, you could use the tone equaliser 
for this (or perhaps adjust filmic).

As for the mask, did you visualise it ("circle in square" icon next to "mask 
refinement) and did it show the area you wanted to mask (yellow) ?

> I found this tutorial that showed something about 'blend mode' and telling
> that 'normal' applies to the complete image. You have to chose 'mask'. But
> when I click on 'blend mode' I get a much larger list without the option
> 'mask'.
I haven't seen a "mask" option in blendmode ever (that is, the last few 
years). I found that I always have to check the date on any kind of tutorial, 
as changes are rather fast (not only for darktable) .
> 
> What can I do to make this work?
First we have to know what went wrong...
Keep in mind that in your case, masking and reducing exposition can lead to  
strange tone transitions at the edges. That's why I suggested tone equaliser 
or filmic adjustments instead of masking.




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Style with automatic selection

[Remco Viëtor]

On dimanche 30 janvier 2022 08:38:07 CET Terry Pinfold wrote:
> I imagine it would be difficult for the developers to program DT so that a
> style will use the dropper with the new image. I can see advantages if it
> was practical to implement.

Well, there's also the little detail of the area or point used with the 
dropper: for e.g. exposure, the default is to use the whole image, but you can 
use a small area to determine auto-exposure. 

Of course, storing that in the style is possible, but (imo) useless for a 
style: you rarely want to re-use the exact same pixel locations. And if you 
have to select the area "by hand", the usefulness of a style disappears.




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Style with automatic selection

[Remco Viëtor]

On dimanche 30 janvier 2022 04:39:37 CET Niranjan Rao wrote:
> On darktable 3.8.0
> 
> 
> Many modules have a dropper or equivalent to pickup the optimized
> values, examples are filmic and color calibration modules. Question is
> when I create style from this, will it use the actual value or for the
> new style it will treat as if dropper is clicked.
The style uses the values at the moment of the creation of the style, and 
won't use the dropper. 

That is something that you can easily test for yourself: create a style (with 
only the module you want to test) and apply it to another (ideally completely 
different) image, check the value after application of the style.



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Re: Files suddenly invisable and unimportable

[Remco Viëtor]

On dimanche 23 janvier 2022 16:45:14 CET Hieke van Hoogdalem wrote:
> (...)

Could you at least remove the attached images when you reply? Twice the same 
image isn't adding anything to your post, just using up disk space and 
bandwidth.

(one image isn't much, but you are not the only one sending mail)



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Timestamp oddity

[Remco Viëtor]

On samedi 22 janvier 2022 07:28:57 CET Top Rock Photography wrote:
> IF one's card is 32GB or less, it will not store time zone data, and that
> may cause problems. If one does not have local time being kept correctly
> by the OS, that may cause problems. If one's camera does not save local
> time correctly, that may cause problems.

Two different things going on here:
We have the file timestamps as handled by the different OSs. Those can change 
for different reasons, mostly access and file modification (but there's the 
"touch" command).

We also have metadata timestamps, generated by the camera. Most of these 
shouldn't change. And afaik, Exif metadata couldn't store timezone as a 
standard tag before 2016 (makernote is a different issue.

I'm getting a bit confused which of these groups of timestamps are being 
discussed here...

Remco




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: Fwd: Fwd: [darktable-user] Timestamp oddity

[Remco Viëtor]

On jeudi 20 janvier 2022 14:41:47 CET Bernhard wrote:
> Remco Viëtor schrieb am 20.01.22 um 14:13:
> > I think the only solution for now would be to set your camera to UTC
> > (which
> > has disadvantages as well).
> 
> That's what I am doing for years now.
> Imagine you are travelling along the border Spain/Portugal or
> Greece/Macedonia/Albania and go back and forth - ok, 2 time zones. But if
> you do that in spring or autumn in the time of switch from standard time to
> daylight saving time or back you end up with total mess.
Oh, I agree that that's getting very messy. But as long as the camera metadata 
don't retain time zone information, every solution I can think of will be 
messy. 

Using UTC has at least the merit that it's a fairly neutral point, with no 
messing with DST and such. And if you ever need the local time, you can get 
the information needed for the correction. 

Needs differ of course, but I've rarely needed the exact time when an image 
was taken. Even when linking images to a GPS track, what I use are time 
differences (to get a correlation between the times recorded in the GPS track 
and the timestamps in the images). I don't keep my camera synchronised to any 
particular clock anyway (more like "set and forget"...).





darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: Fwd: Fwd: [darktable-user] Timestamp oddity

[Remco Viëtor]

On mercredi 19 janvier 2022 23:56:49 CET Bruce Williams wrote:
> OK, I think I understand what you said Nick. :)
> I also tried some of those ls switches on my own memory card and was able
> to replicate your results. Of course, mine are all +11, not -5, but that's
> to be expected.
> But where does that leave us?
> Is this a dt issue that the devs can fix?
> Or is it a linux issue?
That will depend on whether the camera stores the time zone somewhere in the 
metadata when it writes "local time" there. The Exif standard has fields for 
storing time offsets (time zone info) since 2016 or so, but that's not 
necessarily implemented even in newer cameras. Older cameras *can't* use such 
fields (other than in the makernote section).

As said earlier, any program reading times from the metadata without time zone 
info will have to make an assumption, either use the "computer time zone" or 
UTC. And there is no way to know which is correct, so not a Linux issue.

And how would you handle import of e.g. older images taken on a trip abroad, 
in a completely different timezone?

I think the only solution for now would be to set your camera to UTC (which 
has disadvantages as well).
The dt devs could perhaps add an option to specify a timezone on import, but I 
have no idea how complicated that would be, nor what other implications that 
would have.



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] DT can't access files on external hdd with special software

[Remco Viëtor]

On mercredi 12 janvier 2022 14:17:47 CET Hieke van Hoogdalem wrote:
(...)
> 
> 
> 
> I didn’t download the latest version yet. Will it work on M1 Macs? The
> installation notes say that Intel is required – can that be it? I can
> hardly believe but I just needed to ask.
> 

Well, better believe it. Darktable is developed by unpaid volunteers, who work 
on what they want to work on *within the possibilities of available hardware*.

In order to get dt to work on M1 Macs, a developer has to have access to one 
of those machines, and be willing to make the effort to port dt to it. Apple 
is particularly unpleasant for this, as programs need to be signed and 
notarised by Apple, which requires at least one of the developers to register 
for 100$/year. (Strictly speaking, you can get an unsigned application to 
work, but it's not fun for the end user...)

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Imported images are always way too dark

[Remco Viëtor]

On mercredi 12 janvier 2022 12:08:08 CET Frank J. wrote:
> Am 10.01.22 um 22:12 schrieb bwh vanBeest:
> > Hi,
> > When I import a  series of images straight from the camera (Nikon
> > D850, Nikon nef format) in Darktable (scene-referred workflow), the
> > images invariably show up as way too dark, as if they are
> > underexposed. The histogram shows a big unexposed part on the
> > highlights. I am pretty sure though that when I took the pictures the
> > exposure is about right (according to the camera histogram). The
> > camera-generated jpg is also ok.
> > What can be the cause of this and what can I do about it?
> > 
> > Kind Regards,
> > Bertwim
> 
> Same effect here.
> 
> I use Olympus-cameras, RAW only, correction +/- 0 EV in the Camera.
> 
> After importing in DT the default in module "exposure" is "+0.5 EV" but
> this is too dark for nearly all pictures.
> 
> I need to use a correction between +1.5 and +2.5 EV for the best look.
> 
> This is not new in DT 3.8 but in 3.6 (or 3.4?) with the change to
> scene-referred workflow.
> 
> I never compared with camera JPG.
> 
> Frank
You have two effects here: one is the exposure of the middle gray parts of 
your image (the one that "exposure" works on), the other is the tone curve 
applied by the camera. That tonecurve can add quite a bit of midtone contrast, 
limiting the room available for shadows and highlights, and also increase the 
apparent exposure a bit (to "protect" the shadows). That tone curve is 
replaced by filmic, which tends to be softer/more delicate with the shadows 
and the highlights.

The correction you need to apply in exposure seems to depend on the camera 
brand, I find I need to add about 1-1.5 EV for my Sony.

Remco



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Broken images after upgrade to 3.8.0

[Remco Viëtor]

On mardi 4 janvier 2022 18:33:36 CET Michael Rasmussen wrote:
> On Tue, 4 Jan 2022 18:13:13 +0100 (CET)
> 
> Jean-Luc CECCOLI  wrote:
> > Many softwares that can open NEF files, XNview for instance, free and
> > easy to use.
> 
> What you see is the embedded jpeg file. Since you have darktable
> installed you will also have exiftool installed so a quick and dirty
> solution: exiftool -b -JpgFromRaw -w .jpg -ext NEF some.nef > some.jpg

Or use it to check for the image size and orientation first?
It might be useful to have a look at the rawspeed database as well 
(/usr/share/darktable/rawspeed/cameras.xml for me). Although if you start 
messing with that file, I'm not responsible for any results.




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Fresh git install.

[Remco Viëtor]

On samedi 18 décembre 2021 07:05:59 CET David Vincent-Jones wrote:
> Hi Patrick ... thanks for that, hopefully somebody with Manjaro may have
> an idea.
> 
Not using Manjaro, but perhaps you can check if the required header files are 
present under /usr/include. Under OpenSUse they are under /usr/include/OpenEXR
and installed from package openexr-devel



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] ICC profiles / SoftProof / Export

[Remco Viëtor]

On jeudi 9 décembre 2021 14:21:22 CET KueWW wrote:
> Hello,
> 
> I've downloaded ICC profiles from a printing laboratory.
> I've placed them in the ".config/darktable/color/out" and dt detects
> them. Everything's fine so far.
> 
> Selecting the profile to SoftProof works and the result is shown well, I
> think.
> 
> Even when I select the ICC profile to export the image using this
> profile dt says "the profile is replaced bay sRGB because the selected
> ICC profile is not supported". There are just some profiles this message
> appears. Some others seems to work fine.
> 
> My question are:
> Why does the SoftProof preview seems to work and the export don't?
> What can I do to check / solve this issue?
Those ICC profiles you downloaded are "device profiles", not "colour spaces" 
(yes, very confusing that they look the same...).

Device profiles are specific to a device, and describe the differences between 
the colours a device actually shows, and those it is supposed to show 
according to the colour space it uses. For that reason, you never (need to) 
apply them on export. But when using such a profile in "soft proof" mode, you 
can check if the colours in your image fit within what the device can 
represent. 

In your case, softproof will show if the print more or less corresponds to 
what you expect. As paper and screen are completely different media, the 
correspondence will always be only approximate (for starters, your screen 
emits light, often too much, and paper only reflects light).

On export, you select a "colour space", which describes what observed colour a 
given RGB triplet (for sRGB) represents (under defined lighting conditions). 
And again, device profiles are not colour spaces, so cannot be used there.




darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Filmic .. Auto-Tune

[Remco Viëtor]

On dimanche 5 décembre 2021 22:58:44 CET Terry Pinfold wrote:
> Thanks for your reply. So is it your opinion that if I adjust the image in
> filmic to get the 'starting' look that pleases me and then use the RGB
> curves module to further tweak the image you see no problems in that within
> the pipeline it comes before filmic but in my workflow it comes after? I
> really do appreciate the feedback because I would always adjust filmic
> before RGB curves in the workflow. I will also read the manual sections to
> understand the design of the RGB curves module.

So do I. So if I know I'm going to do things that would require re-adjustment 
of filmic, I tend to leave a bit of wiggle room in filmic (esp. the white 
point about 1 EV or so higher than strictly necessary). 

On a first pass, I mostly do "technical" corrections: exposure, white balance, 
lens and perspective corrections, and cropping. This includes a first 
adjustment of filmic. If I want to do more creative editing, I tend to leave 
at least 1-2 EV headroom in filmic (depending on what I plan to do).
Then, after the creative editing is done, I do a final adjustment in filmic.

That's something I try to plan for early on, not always easy, but it makes the 
work  a lot easier (basically, try to previsualise where you want to go with 
the image).



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org

Re: [darktable-user] Filmic .. Auto-Tune

[Remco Viëtor]

On dimanche 5 décembre 2021 09:21:12 CET Terry Pinfold wrote:
> We are discouraged from changing the carefully thought out order of the
> modules. I would really appreciate some feedback on if there are any
> perceived issues with changing the order of the following modules.  

There can be issues with moving a module to the other side of filmic, see 
below

> I would
> like to put the contrast equalizer after filmic for the reasons identified
> by David at the start of this thread. I would also like to move the RGB
> levels and RGB curves modules after filmic since I often want to use these
> for contrast or lighting adjustments after filmic. I also do not understand
> why tone equalizer is placed before filmic, 

The modules you mention are all supposed to work in unbound linear scene-
referred mode. That means that the pixel values are linear with the light 
energy (if you multiply the energy by 2, you multiply the pixel value by 2). 
And the pixel values are not restricted to the interval 0.0 .. 1.0. So larger 
corrections that push pixel values beyond 1.0 are not a problem.
You may have to adjust the settings in filmic to get a proper mapping, but 
there will be no loss of information or detail.

Filmic transforms the linear scene-referred mode to display mode. That means 
that after filmic, the pixel values are proportional to the log of the light 
energy (bluntly said, if you multiply the light energy by 2, you increase the 
pixel value by 1). In addition, filmic performs a tone mapping to pull the 
pixel values back into the display range of 0..1 (for current LDR displays, 
not sure what's going to happen for HDR displays)

That means that using the modules meant for linear scene-referred images after 
filmic can cause colour shifts and other strange behaviour, especially when 
they  push pixel values outside the 0..1 range, where you get hard clipping, 
i.e. loss of information and detail.
Also a given correction will look different between light and dark regions.

> but I blindly trust that
> Aurelein is responsible for this order and he has the wisdom to know where
> to place his modules. I have heard him say that some modules are placed in
> the wrong order, but I don't know which ones. Advice would be appreciated.

The LUT module is a bit problematic, as it uses external files that can assume 
either scene space (i.e. the LUT has to go before filmic) or display space (in 
which case the LUT has to come after filmic). So for that module, there isn't 
one correct place in the pipeline.

Apart from that one module, I don't know about modules being in the wrong 
order. Only that in *some* cases, moving a module to a different place gave a 
better result. 



darktable user mailing list
to unsubscribe send a mail to darktable-user+unsubscr...@lists.darktable.org