Improvement Suggestions

I don’t know how much I could add given that I’m not a developer, but I can say that I really struggled when I first started with Darktable, but I also didn’t have the benefit of all the video tutorials that are out there today, plus I had a lot to learn about image processing as well (and I still do… :wink: ). However I was able to make progress and ask better questions once I got a handhold on basic processing with the program.

I’m wondering if a new discussion topic like “Getting Started” along with some pinned basic tutorials from @s7habo and @Bruce_Williams might help new users to get on their feet.

1 Like

The one I mention is from DPreview or imaging resources. I forget…it has a full variety of thing to test a module… I don’t think I can post the link here. I will pm it to you…

I think I have said this before and Dave you allude to it. You really need to have some basic understanding of color and color profiles and what a raw image really is. A lot of people think that they can bash around in a program and learn it but without that base knowledge worrying about what modules and what sliders to use is getting the cart before the horse. I wonder how many people read the sections of the manual I shared here

Embedded in that information are leads or answers to many of the common questions and complaints due to knowledge gaps…

2 Likes

There are a good dozen great raw developers out there that cater to the non-technical audience! Darktable is different, with its deep control and flexibility. I love Darktable exactly for that complexity and honesty!

For sure we should be welcoming to new users, but not at the cost of our core values. Darktable does not need to chase revenue, after all.

And let’s not forget that e.g. Lightroom’s UI is absolutely horrendous in some areas! Hold Alt on various sliders and they change behavior completely, with no indication anywhere, that this could be so. Select a mask and now some edits are localized to that mask, but others aren’t. Dragging the crop rectangle moves the image, not the crop rectangle. I could go on. It’s also instructive to have a look at their manual. Because I couldn’t find one. But they are entrenched enough that Lightroom is taught at university, and that somehow mitigates the UI/docs issues. I’m sure nobody would complain about Darktable’s UI if they had taken a university course on it!

8 Likes

I am not very familiar with Lightroom, but I used Photoshop and Gimp in the past, and experimented with other FOSS photo editors, always coming back to Darktable.

I conceptualize Darktable as a set of parametrized image transformations applied in sequence. This is not unlike a lot of other non-destructive editors out there, FOSS or commercial, but Darktable is different because the parameter space is huge (actually infinite because you can duplicate and reorder modules, but huge even with the default pipeline).

There is a style of photo editing that is prevalent in some circles, and it goes like this: there is a bunch sliders, relatively few, and you keep moving them left and right until you get the desired result. A more sophisticated approach is that there are groups of sliders, pretty well related to some single aspect of the image, and you keep fiddling with them to get what you want.

This kind of “exploratory” photo editing has very limited support in Darktable. Yes, there is the Quick access panel, but it is very basic and if one is using only that with the default setup, there is no reason to use Darktable at all.

I find that the best way to approach a photo in Darktable is to decide what I want, eg “increase the contrast in the sky”, and then figure out how to do it, where I usually have 3–5 options with different module or module chains.

But this requires an understanding of what modules actually do. Each module has to be learnt and understood, which itself is an iterative process (I revisit tutorial videos after a while to gain a deeper understanding). It takes a lot of investment, but when that is made, it is a very liberating experience because the modules are so powerful and a lot of thought went into them.

Conversely, if a new user just starts “experimenting” with the nontrivial modules without intent to learn, chances are high that they will just mess up their photo beyond redemption. One does not need to cite diffuse or sharpen as an example, color calibration is like that too: a perfect footgun if you don’t understand how it works, but if you do, extremely powerful.

What about the new users? I guess that the harsh truth of Darktable as is is that they have to decide whether they want to invest the time in learning at least the fraction of Darktable useful for their editing style, or pick another tool. It is definitely not a plug-in replacement for Lightroom, and frankly, none of the developers are very interested in evolving it in that direction so that’s probably a how it will work in the future too; which I am happy about because it is a tool that suits my purposes.

10 Likes

I, too, have used PS, as well as Lightroom and Affinity.

Frankly, while it is fairly easy to just fiddle with a set of sliders with these tools, it is much more difficult to get the same amount of control that DT gives you.

And once you try to wrestle them to get what you want, the learning curve is just as steep as DT.

Admittedly, there is more resource for them on the web than DT, and they do have professional writers producing their documentation. But there is also a plethora of dross and outdated material out there.

2 Likes

Although I’ve been using darktable for a while I’m still at a pretty basic level. One of the most important things for me is the non-destructive nature of darktable editing - I learnt very early on that I can always work back through the history stack to find where it started to get messed up, or work on a duplicate, or capture a snapshot to check I’m not making it worse! Redemption is (in darktable) always possible.

7 Likes

I get the impression that very few newbies take the time to read the manual, or at best they take a cursory glance over. Personally, the manual is the first place I go to with a question and then search past discussions on the forum and/or videos. Starting on a new topic is usually my last resort.

I understand where you’re coming from, that understanding the scene referred workflow is key to understanding DT, but I have to wonder if that’s a bit much for people who are starting out with raw processing.

Whatever assumption people want to make concerning prior knowledge of raw processing, I think that a dedicated tag for beginning or starting out might help as repository for the questions that new users are likely to ask. That’s all I have to suggest.

I think the biggest difference with DT is that you have to be more deliberative with your edits and you have to think about your end state before you begin, and I believe that is what causes problems for users coming off of commercial programs. That has its pros and cons. For example, dodging and burning is a lot easier in some programs because you can easily brush the effects on and off, where you can come close with mask brushes in DT but its harder to adjust… it’s just not as interactive. On the other hand, while LR and others have AI and other autoselect masking options I found myself spending more time refining and correcting those masks, than I do with combination of drawn and parametric masks in DT.

I think that’s exactly right. New users need to be willing to learn. And honestly, some of the very worst reviews I’ve seen of DT are from people who refused to learn the program, refused to ask for help and then trashed the program because they couldn’t understand it. Its a real pet peeve of mine. It’s one thing to say that the time required to learn isn’t worth it, but a reviewer has an obligation to provide an informed opinion and that means taking the time to do the homework on behalf of their audience.

5 Likes

I agree, and it’s also why there is the constant request for JPEG matching. Many users are conditioned to having their RAWs developed for them, and all they need to do is tweak exposure, colour, contrast, etc.

Despite its reputation as a programmer’s program, I actually think Darktable is better suited to artists. It gives more of a blank canvas and lends itself to those with a vision. Although every photographer is an artist, not all are as interested in the “arty” side of it, and maybe they just want the quickest way to get a sharp, Instagram-worthy image that pops. Commercial software excels in this area.

To be honest, I often struggle with where to take an image. It’s not always obvious what kind of final image I want. I still fall prey to processing many more images than I need to instead of realizing that some just aren’t worth it. But I don’t let it bother me too much because I enjoy processing and practising with different techniques. Sometimes a really nice image is produced as a bonus.

2 Likes

I try and discern this while I’m capturing the image. Even if its just remembering the feeling, you can work from that.

2 Likes

It seems that everyone is selling LR presets… one more thing we don’t see in DT. Sometimes I’ll try to replicate a LR preset based on some ad using Darktable and there are times when I get pretty close.

2 Likes

Yes, very good tip! But sometimes I take 500 snapshots on holiday and the feeling is always the same: “Ooh, that’s nice!” haha :smiley:

1 Like

I agree that it’s more of an artists program. Also, most users of darktable enjoy the FOSSiness (if there is such a word!) and that you can play and mess around with a photo to your hearts content. Some photographers only shoot jpeg whereas others see that is is only half the job. Darktable is a journey that I’m able to enjoy even if I fail to understand some parts of modules but I will try to learn. It keeps my ageing grey stuff sharp! I view my photos on my large TV screen these days and enjoy the end result of my efforts. It’s a very cathertic process for me.

2 Likes

As someone who develops open source simulation software for a living I have a somewhat more nuanced opinion when it comes to users, reviewers, and their requests. In particular one of my primary goals is for my software to be adopted and to be useful. This often means implementing features that I personally do not care for, but which will be of a benefit to others. Indeed, over the past couple of years a large part of my development time has gone towards making life easier for those who are coming from commercial packages and reducing the number of ways users can ‘screw up’. It isn’t easy and it is mostly thankless.

It is tempting to write users off as lazy, or as people who can’t be bothered to read documentation. This is as true for us as it is for Darktable—even though most of our users have PhD’s. I also can not tell you how many times I’ve seen a ‘bad’ review published which has found our performance to be ‘unfavourable’ compared with alternatives simply because the reviewer did not look at our appropriately named performance tuning guide in the documentation.

The standard we ended up adopting when dealing with requests and feedback is to first ascertain if the user is a professional or not: are they someone with a track record in the field (say using other pieces of software), or are they someone just starting out who likely does not currently have the resources to use our software to produce anything of merit. (As a general point at commercial rates the amount of electricity required to run a simulation of any real relevance is likely well in excess of $1,000. We’ve measured some of our larger simulations in the number of tonnes of coal which would need to be burned an hour to produce the required electricity.) Then, for people who are professionals, we regard them as being right.

Even if they have done something wrong, or failed to read the documentation, or otherwise disregarded it, the problem is with us and we should work to ensure this issue doesn’t happen again.

Thus, when people say they want the initial results to be closer to what they get out of their camera JPEG’s the question we should be asking is who are they? If the request is frequently coming from people who are professional photographers—and by professional I mean their photography is their primary source of income—then we should likely take the request seriously even if it is something we do not need, or understand the relevance of.

We should also be realistic about getting people to substantially change their workflows even if they are objectively inferior to an alternative. Now, it again is easy to say that people who don’t do this are behaving in an irrational manner, but I’m willing to bet virtually everyone here falls into this camp. Case in point: with the few exceptions (hello France!) most people use QWERTY keyboards; either physical or virtual. Yet, QWERTY is in no way optimal for English and there is good evidence to suggest that the Dvorak layout is superior. But, I’m willing to bet that virtually no one here—even though we all type—uses it.

Regards, Freddie.

6 Likes

OK I’m a French guy but I do not buy your point. At the end using QWERTY, AZERTY or Dvorak to type “hello” you’ll get the same word written (eventually faster using a specific keyboard) here we are talking about using a workflow that will give you a better result and full control. So that’s say comparing apples and oranges I’d say. But please don’t use this to start a thread about keyboards :slight_smile:

2 Likes

Why should the word of a photographer-for-money carry more weight than of a photographer-for-art or a photographer-for-fun? Especially where there already are plenty of “professional” tools for “professional” users?

You make a good point in general, that one must listen to the set of users that one wants to attract. That these users are “right”. But that just begs the question, who Darktable’s intended user actually is.

Not to put too fine a point on it, but I like Darktable because it is different from the “professional” mold.

6 Likes

And there is a third kind, those like me. I wanted to use the software to develop my images. So, I started to watch a lot of videos and edited by trial and error (and I still do). Later I tried to understand some underlying module principles and even read the manual :wink:. But I mostly still learn by trying and of course Boris wonderful videos!

No offense, but I don’t think it’s helpful to start in a new field by making suggestions for improvements. I am a little old school in this respect. First try to understand a topic, then start to make it better. I also do support in a linux forum, and we have the same kind of discussions also there. Users switch software and would like to have the advantages of the old and new programs combined with special emphasis for their personal needs. These kinds of discussions tend to be unproductive.

7 Likes

I think I’d still put you in my first category, which I probably didn’t word very well. You still put effort into learning how to use the software, even if you didn’t start out by getting into the maths and technical details. Trial and error is a very valid way of learning, but the key is not immediately giving up. I should maybe have worded it as “those who put effort into learning Darktable ultimately enjoy it”.

I doubt you were aiming that comment at me, but for what it’s worth, I fully agree with you.

3 Likes

I already agreed with your originally comment. I just wanted to add, that there are users that focus on just using the software.

No, definitely not :grinning:.

1 Like

I see your point. But:

  • If you don’t want to adapt your workflow to a program you choose to use, is that a sufficient reason for the developers of said program to change it?
  • Otoh, changing workflow requires retraining, with an initial loss of productivity. So there has to be a compensation for that for users to accept that retraining. In the keyboard case, there is no such short term compensation, and a big loss of productivity.

There’s important difference between your software and darktable:

That means, “no users” probably translates to “no income”?

3 Likes