In the latest German (PDF) online photo magazine “fotoespresso” issue, an extensive review of darktable is published. First, I appreciate that they, as a magazine owned by a publisher of subject literature, publish such a review, especially since they earn their money with how-to-adobe books. It is written by an adobe user, and he was working only two weeks with darktable. Furthermore, he tested on an apple computer, and there are many indications that darktable runs a lot less fluent on these. For these preconditions, the conclusion of the article is not too bad. You can fid it here:
Several of the missing functionalities the article mentions are availably by lua extensions or otherwise (e.g. direct export to gimp, the possibility to change colours of UI by css, …), so it could have improved the article if the author had asked our community before finalization of this article.
Disclaimer: I only read the conclusion, it’s already late here. Hope the rest of the article fulfils what I have teased above.
Btw., the German photo podcast happy shooting talked about a darktable test as well, with similar preconditions (apple computer, adobe user), and while the tester tries not to be too harsh or even friendly, his conclusion is rather extremely negative.
While I share several of the points of criticism of the two reviews, I don’t share most of them, because they are either simply not true (missing features that just were not found), show incomplete testing (for a fair comparison the tools should be compared on all operating systems where at least one of them is running, by OS and as cross comparison. What’s the adobe performance on Linux?) or incomplete research (not asked the community for resolutions).
Hm, in the issue mentioned there is a table comparing several raw converters including darktable, and there are IMHO several mistakes. I am thinking about asking for correction, since they want to publish the second half of programs in the next issue and there would be the opportunity to “print” the adapted figures, if they follow my assessment. It’s not always better for darktable, but I think one should be correct. What is your opinion?
Lens correction: 2 stars vs 4 for adobe, I want to know what is missing in darktable?
Perspective correction: 2 stars vs 4 for adobe, I wonder if they missed the respective module that automates this and only checked the options in the crop tool.
Snapshots: They rank them equal to lightroom, but I think darktable is a bit worse (no zoom for snapshots etc.). I am not an adobe user, so what do you think?
Virtual copies: They say “?” but as I understand the feature they are available.
Batch processing: They say “not available”, but what about darktable-cli?
Plug-ins: They say “not available”, but what about the Lua extension system?
Make web galleries: They say “not available”, but it’s in the export options.
The rest of the comparison I think I agree on, from what I know from lightroom. About Affinity, which is the thirs one compared, I know too little to judge.
I haven’t read the review but I will make some casual remarks, based on my brief experience with Adobe products.
Lens correction, perspective correction: ACR definitely does an excellent job and is constantly improving. That said, dt deserves more than 2 stars. Two stars out of what – five?
Virtual copies: again, Lightroom does an excellent job here. I think it is more feature rich than dt. However, most qualms with dt has to do with people being unfamiliar with it. Look no further than a few threads on discuss about virtual copies and other topics.
Batch processing, plugins: I am not as fluent on this topic but I think it is harder to navigate in dt. What I know is that Adobe products have dedicated GUIs, menus and dialogues, so I am thinking that the reviewer didn’t bother with figuring it out.
Unit and limits not defined . They even mix up “+”, “-”, “*” and more of these, “?” as well as annotations, even in the same row.
Thanks for the comments .
Hm, maybe I did not understand what they mean. I thought it was about having duplicates of one raw in the database with different processing? There is a button for that?!
Again, I am speaking casually, from the top of my mind, so I might be wrong on a number of things.
Virtual copies in Lightroom do more than hold different processing instructions. They have a lot of depth to them. I think ACR’s treatment of copies is similar to dt’s.
I agree that batch processing is easy once you know what to do. However, I think that LR and PS are easier to figure out; i.e., batch is contained in one (popup) window, accessed from an intuitive location or menu.
Also, PS (and maybe LR) has something called Actions where you can edit a photo using step recording and scripting similar to BIMP (guessing: haven’t used BIMP before), with the additional ability of doing advanced math and programming and accessing well-documented API. which I am sure could be done in LUA in dt given enough effort and motivation from the community.
This would seem the single biggest reason for the lukewarm review; having an experienced darktable user to tutor the reviewer would certainly have made sense. It has been my experience that darktable has incredible power and versatility, however it is not always easy or immediately intuitive in how to use that power and versatility.
I’ve been using darktable for more than two years (a lot less than some of you) and still haven’t come close to a thorough knowledge of each module, or each blend mode.
A raw-developer such as daktable is a complex piece of software with a steep andlong learning-curve.
The article does laud the power of darktable’s parametric masks, but it suggests that it falls short of adobe’s offering because each mask can only bebound to one correction. Surely in darktable duplicating a module and then applying a different correction is functionally identical?
darktable is not professional, commercial software, with price-tag or subscription to match; It is FOSS. The documentation doesn’t hold your hand very much. It requires patience, practice and learning from others to gain in proficiency. But I wouldn’t trade darktable and the community we have here at pixls for adobe’s product, even if it came free-of-charge.
Yes, but he specifically states his negaive review is for darktable on the Mac; he does report that Linux users express that they have hervorragend performance. Literally, standing ahead of others, so “leading”, but best translated as “excellent.”
I’m not (and have never been) a Mac user - I rebel against being fitted into a straitjacket - which is what Apple seems to me (with apologies to Apple fans). However I have had interactions with graphic designers, medical professionals, and others, who have apreciated the polished, consistent user interface on Mac, and it seems that darktable’s Mac implementation diverges from this. So the reviewer’s complaints – about menu entries (or lack thereof), the grey background, the fonts, the size of the sidebars (did he bother to resize them?), and the ubiquitous triangle to expand modules – all make sense to me from this perspective, because he is looking for the polished consistent user-interface.
Again, as with the first review, it is probably a little unfair - darktable is FOSS. But the reviewer doesn’t care: he explicitly states that he simply wants the best product for his platform, irrespective of cost.
The reviewer’s thinking is too simple here. You can choose the same parametric parameters for two modules, but since one must come after the other in the processing pipeline, you won’t be modifying the same area in the second mask iteration.
I have read the review in fotoespresso, too. On the one hand, we can be happy that there is a review and that it is not that bad. On the other hand, it is definitely true, that it is not quite fair. I don’t know. There are so many people who just love Lr and they are not even interested in having a look at something else. The are so biased. I personally do not know why many people adore Lr. I have had the Adobe Photography Plan for years, I have had a good look at Lr, though I have never really been a Lr user. What I dislike about Lr the most is the GUI. Lr is definitely a very good an powerful program but it is just so boring. In dt however, there is always something new to discover.
I think it is really a pity that the reviewer did not stress what dt can do better than Lr. I mean it is true that dt cannot do everything that Lr can do as well as Lr but there are also many things where dt is much better than Lr. E.g. noise reduction and masks.
I have seen some reviews for darktable in internet and they are more or less always the same. There are goods and bads and this is how just life is.
I just learn to take these (and any reviews anyway) with a grain of salt. I just read them to see if I find something interesting that I have overlooked.
I am not expecting that someone who (maybe) used the program for a few days to be able to know all the details about the software …
On the other hand, a possibility to link a parametric mask “result” from one module to another one on pixel level is on my imaginary wish list for a long time. That way I would not have to find the same mask over and over again, under changing preconditions (as you explained). Does not seem too difficult, but I lack programming skills …
Huh? I must be missing something glorious because I’ve been running dt on my my old MacBook Pro for a couple of years now. I’ve yet to be disappointed because I felt that the program is lacking. Everything from the basics to batch processing to running lua scripts works well for me.
The only Pain in the A** is that we poor Mac users must often wait much longer than the rest of you for darktable updates
Once again, a GUI used by the writer is compared and evaluated with another unfamiliar GUI. With a Mercedes Benz I reach my goal just like with a Porsche. If I only drove Mercedes Benz before, the Porsche will seem very hard to me.
DT is a wonderful software. It’s all a matter of taste.
The comments I made above seem to favour Adobe products. What I was getting at is imagine the average user, someone reading this magazine for example. Would they likely find dt difficult to get into or the Adobe trifecta? I don’t know. Maybe try it on your tech oblivious family member or friend and report back. Or try to find the proverbial Millennial or Bro. Like someone already pointed out and I have said in the past FLOSS can be hard to get into but when you do there is no turning back.
. They even mix up “+”, “-”, “*” and more of these, “?” as well as annotations, even in the same row.
.
:
Or try to find the proverbial Millennial or Bro. 
Like someone already pointed out and I have said in the past FLOSS can be hard to get into but when you do there is no turning back. 