I have seen quite long, extensive and tyring rants about dt 4.0 and what has gone wrong and why it’s all bad and whatnot on YouTube. A lot seemed to be about usability and the general consent seems to be that darktable has become too slow and too bloated and too whatnot …
I don’t get it. I am a user and I have not found 4.0 to be any worse or slower than the previous versions. Yes, the star-rating system hasn’t been activated “my” way when I started the program after installation, but next to that … no complaints. Everything runs smooth and the way I like it.
I find those rants counterproductive. A newbie or any user looking left and right for a new converter will most likely get the impresion that dt is just shitty software, even if the video-rants talk about things that most users will never get in touch with and their epic length could be shortened to 1/10th of the time by avoiding repetitions.
I am just happy that dt exists and provides a great tool for my creative work.
(No, I will not name the authors, but if you follow dt-inclined channels here and there, you know what I am talking about.)
I agree largely - I’m almost completely happy with DT 4.0, and the change from 3.8 was nothing major at all, for me. But then I don’t pay much attention to youtubers (or anyone else…!) who point out bad bits without bothering to find the good bits! Dunno really
Fully agree. The only thing that got “worse” is highlight handling which is a thing to learn and will be improved with next release as far as I know. I can somehow agree with the rant from Aurelien Pierre about overloaded UI, but that’s one viewpoint and I’m sure many users out there will appreciate the one or the other UI feature I don’t need or click on.
I didn’t watch any other rants to be honest - but looks like ranting has become a culture.
Nothing to worry about when a single dev is going mad at others for no reason and cannot work in a group with different opinions. My point of view is that dt is not perfect and certainly will never be as perfect has no meaning when in the hand of thousand of different users with different need. We try to focus on the need and try to mitigate issues when too many users are not fond of something.
I think DT 4.0 overall is fantastic and a clear jump from 3.8, which was already great. The interface is complex and can be improved, but it’s far from being a showstopper.
I think “no reason” is a bit dismissive but unfortunately the good points that particular dev makes are lost in the torrent of abuse. He has many good reasons, none of which justify the way in which he makes his arguments. I think many of them could be addressed in the main darktable project as well.
Agreed. Let’s just say, “not a people person”.
Perfect is also a verb. We should always be perfecting.
This is a laudable goal, though I’m not sure we have enough good ways of getting user feedback to judge this. Github is a bit of an echo chamber and it can be hard to garner more general opinion from the issues. Some of the rants are about “code quality”, which is harder to gauge from user feedback (and harder to agree on a solution to).
That’s an easy rant, everyone has its own view of what is good and what is bad. I’m an Ada person, using programming contracts a lot… I can say that C is bad and find the code not well written… And what? Actually the plug-in like architecture is good, yes there is some code quality issue but I’d like to point that we have many devs doing refactoring or rewriting part to make things better. As all code, it has variance but I feel that in the long run we are converging toward something better.
I tend to have the code moving to get new features, the code is not always perfect but then in following cycles it gets better. I could block lot of PR and discuss again and again the code quality, sure, and we would have maybe one version every 2 years and maybe with half the changes we have today. Again, that’s a way of perception. I just put my view clear here. The next maintainer will maybe have another way to handle this, and I’ll be ok with that. As dt code I’m not perfect, just a simple human trying to get something nice out of the cool app that dt is.
The rants raise valid points but the program is still perfectly usable (albeit with frustration if you used the recent collections module which is now removed).
The change to the core module filmic with V6 and it’s WB sensitivity creates many issues in highlights if you don’t understand what’s going on or how to address them…changing defaults can work or several other proposed steps… I think this has been one of the main things users have noted moving to v4.0
Staying with 3.8 means you miss the other improvements from version 4.0. You can still use the V5 filmic, and even the old WB method, the new stuff is an addition to what is available in 3.8, not a replacement.
And of course dt slows down when you decide to use the laplacian highlight recovery or “diffuse or sharpen”: those modules are known to do a lot of calculations (compared to other modules). That means the CPU or GPU will have to work a lot harder, which means they’ll need more time…
why focussing on somebodys rants?
Either darktable is working fine for you then it’s completely meaninless if somebody complains about it or it doesn’t match your expectations but then an issue report or feature request is far more helpful.
A self-help group of disappointed ones doesn’t improve the tool …
In fact I am not even aware if I am missing something. My PC is pretty old and the rumours about dt 4.0 being slow were one of the arguments against deploying it. My main problem in dt 3.8 are related to recovery of highlights and if this is not improved in 4.0 (quite the contrary it seems), so why to take the risk?
Anyway, I am really grateful for all the work the developers have done and are still doing on dt. Do not get discourage by any rant.
)