I raised the possibility of using an Open Source program to produce something like a “tin type” photograph. A list of characteristics was provided by @hatsnp. Here is my initial attempt.
This file is licensed Creative Commons, By-Attribution, Share-Alike.
EDIT: The one thing I forgot to do was add some grain
Sorry @epeeist I may have mislead you on the grain bit. Did some research and it seems like ferrotypes didn’t have grain due to the silver particles being extremely small.
What’s the title of this picture?
Chameleon-ruins?
Ruins in disguise?
I did not try to do the tin thing… Just my postcard pic’
History: https://www.hmdb.org/m.asp?m=127938
_CTW5427_01.NEF.xmp (99.8 KB)
Note: only load up this XMP with the current nightly; It may crash darktable 5.6.1 and below.
_CTW5427.NEF.xmp (34,1 KB)
So this is pretty hard. I have only seen 1 old tintype in person that I can remember. So I pulled some basic constraints out of my ass.
First, a max of 4-4.5EV total difference between paper (iron, tin, aluminum) white and blackest black. Total conjecture based on paper printing. A rough, but imperfect, way to develop around that is by setting the clipping indicator between those values.
Two, no black lacquer on the substrate. This probably results in less contrast but also gives higher whites…pure speculation.
Three, paper white is below the ‘highlight’ line in the waveform. Total speculation and technically (at a minimum) false, as a polished tin/aluminum can reflect a great deal of light. A thin emulsion might still be very bright in the right light.
Four, between chemistry and assumptions re: lens coatings, UV would have had a significant impact on the image and its crunchiness. I am allowing myself to use a local contrast instance to add crunch…which is completely random but we do not have a way to replicate that using modern kit/software.
Finally, exposure times were slow (given cloud conditions) so the grass would have been blurry due to motion. Added motion blur at an arbitrary value with little finesse during masking.
Edit: I know most things were toned back then. I went with a barely there version to straddle the line between the un-toned and heavily toned
It looks pretty bleak.
My interpretation
_CTW5427.NEF.xmp (23.2 KB)
I am using my experimental dcp profile support. I don’t expect it to work 100% on other’s people darktable.
I didn’t try to create a tin type photograph because if I wanted that I would try and take one. I met a photographer in Hobart Tasmania that did a exhibition of his tin types and it was very interesting work.
Here is my DT 5.6 edit and may the heavens forgive me I used the retouch module to remove the metal post which intruded into the image.
_CTW5427.NEF.xmp (14.0 KB)
From what I have read, the ISO value was 1 (yes, you read it correctly).
Assynt is on the West coast of Scotland, the picture was taken in October. It was bleak.
Calda House.
It was built in the early 18th century, but the owners went in to debt and were forced to put it up for sale. Clan members burnt it down, so that an enemy clan couldn’t get hold of it.
So, I downloaded your xmp file and tried to load it as a sidecar file.
Darktable crashed. Each time I try to restart it, it tries tries to load the xmp file again, and crashes.
I have tried deleting the cache, but this makes no difference. I suspect that the problem is in one of the database files, but I am loathe to go back to a previous version.
Any suggestions as to what I should do?
This is annoying and will probably happen with my XMPs at the moment as well. I’m sorry for that and will switch back to a version where I am sure, that this will not happen.
Maybe we should think about establishing a rule that development versions are not welcome in Play Raw.
It never really made much sense to use master or dev versions in PlayRaw. Same with GIMP imo, anything that doesn’t generate a sidecar or a history of edits goes a bit against the spirit of the thing
I wouldn’t go that far to exclude Gimp as long as there are infos about editing process.
It’s one thing to offer a not completely transparent edit. I agree, that this is not as helpful as a side car file. But offering an XMP which brings trouble to the one who just want to learn is on a completely different level.
Oh I am sorry about that, that sucks.
1.) Did you try to delete the xmp from the disk?
2.) darktable keeps snapshots of its DB in it’s configuration directory - you could try restoring the most recent library.db (if that is not what you said, you “loathed” 
- should be a rather straight foward file-explorer-operation though).
3.) Could you launch darktable with -d common and provide a log?
I think this needs debugging; xmp files being incompatible is one thing, loading them making darktable unstartable is another.
Yes
In the DT gui I get a pop up with a message about a “highlights” error, but it is quickly overwritten by other messages, so I don’t see the whole of the message.
Here is the log file, I had to zip it, because the text is too big to be accepted
dtError.zip (4.7 KB)
That is the trace from the segfault, could you also launch darktable from the terminal with -d common and provide that output?
