So still no GTK3? I think when GTK4 is released in a few months, it would be a good time to move to GTK3 as they can still be outdated.
I know I am being a bit bitter for an open source project with less funding, but GIMP was supposed to be a poster boy GTK application in the first place.
This is just a myth. You make it sound like there was some purposeful intention to gimp and gtk. There was not. Gimp was created by some kids at berkeley who created gtk because at the time qt and motif had untenable cost and or licensing, or sucked. When the gnome project started years later gtk was chosen due to licensing considerations since ironically qt was “too free” ie pure gpl which limited its use in closed source software without licensing. At the time this was considered a major problem in unseating windows (back in 1998 when there was actually real vested interest that could be a thing)—- visual studio cost a little bit of money (but it was cheap) and you didn’t have to license anything to make closed source windows apps. You could write a GUI app in assembly and not pay ms a dime if you wanted. The Unix side of things was not so good. Motif was thousands of dollars.
Anyway, turns out maintaining a professional image editing program is a lot of work. While it is true many fanboys named gimp as some kind of foss and gtk showcase its not like most of them were actively contributing to either gtk or gimp.
There was never any longterm large scale commercial investment in gimp similar to how redhat threw massive resources into gnome and gtk.
> gtk was chosen due to licensing considerations since ironically qt was “too free” ie pure gpl which limited its use in closed source software without licensing
This is wrong. GNOME was already well underway when Qt switched to GPL. Before that it used QPL and before that yet another license.
"Sadly, the KDE effort is based on the Qt user interface toolkit which is a non-free library. The main problem is that modification of the toolkit is not allowed. The ability to modify the interface toolkit is a requirement of free software[6] and of the Open Source Software Guidelines[17].
Although Qt is released in source form, permission to redistribute changes made to it is denied."
The "G" in GTK+ stands for "Gimp". If Gimp is using GTK+ ver 2 who are we to say different? Perhaps ver 2 is the version and the rest are just pointless forks...
This is a serious question in general. There will always be s steady stream of significant upgrades/changes of any popular open source project as later developers add in things they consider important. At some point the decision has to be made to stop following those changes. This decision is usually made by default when it turns out that no one can be bothered to follow the changes of a project from another project.
Glad it works on someone else's computer (i'm the branch author).
Actually https://github.com/GNOME/gimp/tree/gtk3-port may be more up to date as I'm just manually backporting to the meson port.
The challenge for GIMP is that it exposes a lot of GTK in its API, which is used by lots of 3rd-party plugins. All which will break when moving from GTK2 to GTK3.
At the same time there is an ongoing transition to use GEGL as the pixel manipulation API, which eventually plugins also need to be updated for.
2.10 is the release which provides all the new APIs for plugins to use, so they can start updating (but existing code will work). 3.0 will then remove the old pixel manipulation APIs, and also transition to GTK3+.
This hopefully keeps the burden on plugin devs to a minimum, and keep the compatibility clear for users (plugins for GIMP2 dont work on GIMP3, and vica versa).
A new Dashboard dock helps with monitoring GIMP’s resource usage to keep things in check, allowing you to make more educated decisions about various configuration options.
That looks so ridiculous and out-of-place in an image editing program. If it weren't for the left side, you'd think it was Visual Studio!
Did you stop to think that perhaps GIMP's developers are in fact its users? I'm an engineer and I draw as a hobby. I don't mind developer-oriented features in my drawing program. If I wanted Photoshop, I would be using that.
That’s all well and good, but if GIMP is targeting devs first who are also artists, Linux remains an OS without an image editor targeted at artists first. IMO that’s an important hole in the software library that prevents wider adoption.
And if you complain that Krita lacks photo manipulation features, you're told "Of course, because Krita is only targeted at illustrators!" The open source evangelists always have a dodge.
Inkscape, mypaint and gimp is pretty much all there is to use professionally on linux. And they complement each other surprisingly well. Inkscape is the main workhorse, mypaint is for sketches and gimp is to prepare some photos and raster images to import into inkscape (you can't do a lot in gimp).
It's been a while since I've used Inkscape to import .ai files, but does Inkscape still have a tendency to mangle them when you try to import them?
If you're moving from one platform to another, you want legacy files to import perfectly, especially when you don't have time to convert old Illustrator assets into svg files.
On Mac/Windows at least, Affinity Designer was the first app around $50 or under that didn't mangle any of the .ai files I threw at it.
What's with the entitlement here? These are people sacrificing their time and going largely unpaid (AFAIK) for their efforts. I for one am happy and grateful that these "dodgy" open source evangelists exist and offer us all a choice for free. If I have a subjective usability problem with their software, far more constructive to take the effort to try and fix it myself than complain like a diva.
The "entitlement" is from people who are frustrated that the open source software movement lacks such a fundamental piece of the desktop software puzzle.
The confrontation comes from the fact that one side insists the software is there (GIMP), whereas the other side view them as disconnected from reality since its use in industry rounds to 0%.
I don't see why perceived usage share in industry should motivate dissing open source devs. GIMP has a user base and fans who obviously think it's good enough for their needs. If the professionals in the industry and their employers have a need for an all-in-one free software that can do everything - image editing, illustrations, maybe throw in 3D and animations while we are at it - then the industry should contribute towards building it. All the building blocks are available, thanks to those dodgy open source devs.
What can be and what is are two different things. GIMP is not a serious competitor to Photoshop, full stop. So long as there is no serious competitor to Photoshop, that's a huge swath of the professional world (and much of the non-professional) that won't ever give Linux the time of day. There's a reason that Photoshop is teetering dangerously close to being a genercised trademark!
And I think the attitude comes in when GIMP is ignorantly held up as if it were a serious competitor. It isn't. It's something else entirely. It might be good enough for people to make memes for internet shitposting, but between the missing features and the alien UI (to the point where a massive reskin [1] was around for a while), it's never going to be used in industry.
If the GIMP devs want to continue down this path, more power to 'em, I just want the people claiming that it's a Photoshop competitor to educate themselves :)
This thread seems to be full of people making baseless assumptions about GIMP that I don't see the GIMP devs themselves make or any open source enthusiast make.
Or that GIMP is targeting graphics professionals or graphics industry? Don't think so... "We do, however, acknowledge the fact that people will treat GIMP as Photoshop replacement no matter what we tell them, and that’s all right with us. You own this software, it’s up to you to decide how you make use of it."
It is what it is. A tool. That happens to be free and is good enough for some. As somebody above said, it's like complaining why emacs / kate / geany or whatever editor can't do something that visual studio can. If the industry wants a serious competitor to Photoshop, time for them to start building it anew or taking existing free tools nearer to that goal or funding open source devs to do it for them. Complaining and ranting against open source devs who have zero obligations to the industry is not going to achieve anything.
Fair enough. _Most_ things said about GIMP in this thread are complete and utter made-up bullshit that won't stand fact-checking. How about that? Better? :)
The entitlement I was referring to was based on the tone of the criticism. "I wish Krita had image editing features" wouldn't have got any reply from me. But adding "the open source evangelists always have a dodge" sounds to me like a personal attack on those devs. Do they owe something to the artist community? Or have they breached the terms of some work-for-hire contract from the graphics industry to code an all-in-one tool? Nope.
What is the "dodging" in limiting functional scope of a tool? Each and every software project I know of places some limits on what it will and won't implement. Even Adobe has a suite of tools where one tool like Photoshop can do things Illustrator can't and vice versa. It's not Adobe being "dodgy" about their tools, and nor are open source projects.
Nothing stops people from forking GIMP, implementing whatever fixes they want, distributing, and even selling it. Refusing pull requests is not a crime.
"Nothing stops people from forking GIMP, implementing whatever fixes they want, distributing, and even selling it. Refusing pull requests is not a crime."
To put that up as the solution is to be fully divorced from reality.
Krita is targeted at artists and illustrators, just like any other specialized tool. People don't look at Vim and complain that lacks desktop publishing features, despite novelists and programmers both needing to manipulate text.
The original complaint regarded tools for artists, and that's Krita. For photographers, there's DarkTable (https://www.darktable.org/), RawTherapee (http://www.rawtherapee.com/), and many others. For my modest needs, I find DigiKam (and occasionally The GIMP) adequate, though I am envious of proprietary tools.
And when we complain about usability issues and lack of features in those programs (such as GIMP!), we're told that there are lots of open source alternatives; for instance, look how slick Krita is!
And thus the cyclical debate continues for another turn while a competitive raster editor remains unavailable in the open source ecosystem.
If you discard feedback based on tone, you will very obviously bias yourself toward only accepting positive or positive-leaning feedback. I'm sure you can see how that would seriously hinder software quality in the long run. And, indeed, we can see that it has in the case of GIMP.
Very few people can devote the significant time necessary to make free professional-grade stuff just for the sake of a wider audience.
And unless you're an artist who does that yourself, it's a bit much to ask others to meet your needs for free like that in the first place.
If non-dev artists want to see that gap close, they need to contribute, not just wait for mana from heaven.
No, that doesn't have to mean actually writing the software. But it could mean paying someone else to do the coding, even if the software is ultimately released as OSS.
You may not mind developer-oriented features, but the vast, vast majority of artists do. This is why Photoshop wins. Adobe has customers, and Photoshop's devs have to either listen to the customers, or be fired.
Which is exactly why a feature like the one we're discussing has been included in Photoshop for ages, along with stuff like custom swap space management for image buffers? :P
> Last time I checked, Photoshop and most of adobe's products barely function on a 4k device, yet Gimp works nicely.
That's a surprise and nice to hear! I thought the support wasn't too great and it requires GTK+3.x or newer. It still uses GTK+2.x (see https://wiki.gimp.org/wiki/Roadmap).
There are some shortcomings on the UI side (you need to take care of DPI setting for starters, and you also need to configure some details separately to have usable button sizes for instance), but I'm using it on GNU/Linux with 3200x1800 13" screen and works just fine.
> Did you stop to think that perhaps GIMP's developers are in fact its users?
GIMP's users aren't the handful of developers who develop GIMP. GIMP's users would be people that switch to Linux and are looking for an alternative to Photoshop. GIMP is for some reason referred to them as the go-to for image manipulation on Linux so they try it, and when they can't use it they either install Photoshop via Wine or go back to Windows.
I don't know why people recommend GIMP instead of Pinta or more usable software, honestly.
> I don't know why people recommend GIMP instead of Pinta or more usable software, honestly.
Try adding a text line, then editing it. See if you still want to recommend Pinta.
I absolutely love the idea of having a simple image editor for users who don't need most GIMP features. Pinta devs got quite a few things right, but it needs SO much more work...
With that logic, you can argue against any debugging facility in software. Log files? Only developers think those are cool, useless!
FWIW, I have missed better tools for performance tweaking in competing (Adobe) products, so while I don't know if the one in GIMP would be useful, I can see why one would want something like it even as an end user.
So performance tools are fine, as long as they are hard to find (=not user-facing)? And the percentage of users that would want them has to live with worse access to them to not scare the others with an extra menu entry?
Kind of a contorted way of seeing this, but I'll go for it.
So, as a user I have to identify bottlenecks, now..? That would be my first question.
Since you're implying that by not thinking the performance indicator is an important feature for the average user I want GIMP to be slower, yes--actually I would definitely opt for a slower, usable program than a fast unusable one, and I never thought speed was a problem of GIMP.
Right now I'm running Wine to be able to edit this PSD file without going crazy trying to use GIMP's interface, so obviously speed is not what I care about. I'm sure I'm not the only one, since other problems and not speed seem to be stopping people from using GIMP as their image editor.
> So, as a user I have to identify bottlenecks, now..? That would be my first question.
Please do pay attention. The release notes specifically describe who did that and for what purpose, and what's in it for end-users. It's all there for you to read. Just please pay attention.
> and I never thought speed was a problem of GIMP
Which suggests that you never used such a basic feature as Gaussian blur in both Photoshop and GIMP.
GIMP _is_ slower than it should be. It actually _is_ a problem for people with modern DSLRs. We want to fix that problem.
Exactly. As someone who did use Photoshop professionally for years, I got used to Gimp's quirks in like a few days, but hit the performance wall right after that. Pretty much nothing I did in Photoshop is possible in Gimp due to performance issues.
Are _you_ paying attention (your tone is absolutely horrible, by the way)? I'm saying that I doubt the project's priorities are in line with those of its users.
Are you sure that performance is a problem for most users? I have talked to people about GIMP before, and it came out 0 times, while usability was the problem that forced most people to install Photoshop or go back to Windows.
In the past 3 days, I've opened pretty big illustrations (about 20,000px* 8000px) on a i35 thinker (mobile phone processor), and I had absolutely no problem opening them, scrolling around, etc.
Doing basic operations like moving, cropping, and resizing layers was so cumbersome that after 3 days I gave up and installed Photoshop 7.
> Are _you_ paying attention (your tone is absolutely horrible, by the way)? I'm saying that I doubt the project's priorities are in line with those of its users.
You don't like my tone, and I don't like your inability to process basic information written in clear English in the first two paragraphs of an article that you are discussing, followed by making ridiculous offensive statements. Are we even?
> Are you sure that performance is a problem for most users?
The majority of people who work on large images and use filters do have issue with performance of GIMP. Helping them is a priority to us.
> Doing basic operations like moving, cropping, and resizing layers was so cumbersome that after 3 days I gave up and installed Photoshop 7.
I'm still waiting for those "4 times fewer steps" in Pinta.
Yes, as project's contributor I find your claims blatantly inaccurate and offensive. That said, my way of handling them is most likely far from being humble.
> Yes, as project's contributor I find your claims blatantly inaccurate and offensive
That's funny. They can't be inaccurate, because I'm talking about my personal experience using the software.
As for offending, I never even talked about anyone--just the project, and you are the one who's offended me personally by saying that I should get a grip of myself, I don't understand basic concepts in English, etc. etc.--which let me tell you doesn't look that good for GIMP as a project.
This is how many open source projects handle criticism. I've seen this before, and I'm not surprised. Like I said, thanks for reinforcing the conviction I have that many projects stay in their silos and don't listen to users--to say it nicely--because they don't owe them anything (which of course it's true, but still).
Regular users have no idea how to or have the willingness to get in touch with you, they'll just uninstall and move on. When someone in the middle like me does offer criticism, he doesn't understand basic concepts explained clearly to him--like you said--so obviously there's no solution.
> That said, my way of handling them is most likely far from being humble.
> This is how many open source projects handle criticism.
No, this is how some projects handle bullshit.
Criticism is fine and even useful.
Lemme explain the difference, because you are still not taking a grip on yourself.
When you think developers introduced a regression and refuse to fix it, and you can use arguments other than 'because I said so' to prove your point — this is criticism.
When you completely disregard all explanations why a certain feature was added, when you completely dismiss people pointing out that no, in fact other software has the same feature, and people do use it, then continue arguing ad nauseam — this is bullshit.
Got it? Good.
Now, you can be constructive and pay attention, and then we can have a conversation, or you can continue dismissing facts and go on living in an echo chamber. Your choice.
That might not be useful for most users, but could be useful for people in resource-constrained environments. I have a friend with a 2010 MacBook Pro, who is constantly monitoring resource usage as he edits photos etc, and has to close apps, delete files to make space for more Photoshop swap, otherwise the computer just dies.
Since it's a removable dock, I don't see what the problem is. It's a choice every user can make. Only users interested in that information for whatever reason will add it to the dock.
Uhm, even Photoshop has something like that. Once you work with high resolution files with lots of layers, you absolutely need to "make educated decisions about various configuration options" or you'll suffer waiting ages for every simple operation to complete.
The text around the image elaborates on its intended, reasonable use.
> A new Dashboard dock helps with monitoring GIMP’s resource usage to keep things in check, allowing you to make more educated decisions about various configuration options.
Dashboard dock
> On the developer side, it also helps us in debugging and profiling various operations or parts of the interface, which is important in our constant quest to improve GIMP and GEGL, and detect which parts are the biggest bottlenecks.
I use GIMP regularly, and always impressed at the quality of the tool and the amount of features it has. Looking forward to the v2.10 release. Anyone know where I can download a Windows build of the release candidate? Can't seem to find a link on the website.
In theory https://www.gimp.org/downloads/devel/ , but it currently says "Note: the currently available package provides GIMP 2.9.8 and has not yet updated to the lastet version, GIMP 2.10.0-RC1. We're working on that, please check back later."
Downvote all you want--I know open source can't handle criticism, but as much as I want to love GIMP, it's completely unusable.
First, it's still in GTK2 so on my monitor it looks zoomed out at 50%.
Then, the UI is the least intuitive thing I've ever seen. To resize and move a layer takes 4 times the steps compared to Pinta (or of course Photoshop).
Shortcuts are the weirdest thing, completely different compared to any other software I've used.
It's such a pity because undoubtedly there is so much work behind it, and the effects are absolutely perfect.
I really don't understand why they don't try to make it usable instead of showing a performance meter.
Eh, not that I disagree with any specific point, but I've certainly been able to use it many times to do real work. Plus the Python-fu was awesome when I needed it.
> Unfortunately, when one tries to say anything the response is "you go fix it then, asshole!" but I don't know how to develop desktop apps, so...
There are other ways you can help!
As someone with experience using graphics applications, you could look into making UI mockups of the changes / improvements you have in mind. Seeing the mockups might get someone interested enough to start work on user-testing, prototyping, implementing, ..., them.
I think this reputation for intransigence was probably merited by ~noughties GIMP team, but it doesn't seem to me to be deserved at present. I think the UX issues that remain are more due to lack of development resources than intransigence.
Edit: Despite my defence of the current devs, I continue to be annoyed by the GIMP cheerleaders who continue to unhelpfully blame the program's shortcomings on its users.
The thing is Adobe users will tend to want to copy that UI, which is easier because they're familiar with it. But if you're familiar with nothing, or familiar with The GIMP then that's likely to be an impediment.
Squaring that particular circle isn't easy IMO.
There are people who still prefer the separate window system, I've always liked the unified one. Simple differences like this can become divisive and split your developer and user groups.
You should try Blender! Some applications are just complicated and need time to learn.
It's hard to develop the effects, not move buttons around.
As for the "go fix it", that's exactly the reaction I was expecting. Open source projects can't handle criticism because it's volonteers working on it, so "go fix it if you don't like it". Not very productive, uh?
Personally, I use Pinta or Photoshop via Wine. Pinta is very good though.
It's hard to develop the effects, not move buttons around.
Speak for yourself. As someone who's tried to write a simple image editor on a few occasions, writing the image effect was by far the easiest (and most fun) part. Almost all the core algorithms are published and well understood and beyond that it's mostly simple linear algebra. There are also plenty of objective measures to judge how good your implementation is.
Deciding where to "move the buttons" to took up almost all of my development time, was incredibly hard, dreadfully boring and what kept killing my interest in the project.
Given the choice between spending 20 hours trying to make the sharpening filters 5% faster or 20 hours trying to make the sharpening filters more intuitive and easy to use I'd choose the former every time and, unfortunately, it seems far too many developers feel the same.
My guess is the reason they haven’t overhauled the UI is they build GIMP for themselves and their own mental model. Having used Photoshop forever, I can’t make the transition to GIMP because it feels unintuitive. Unless Linux is your desktop OS, there are plenty of other cheaper options (than Photoshop), for example Pixelmator (OS X) or Paint.NET (Windows) that feel way more familiar.
Paint.NET is by far the best the best picture editor I've seen. Not nearly as feature-reach as GIMP or Photoshop so not an option for graphics professionals but for casual usage - it's bloody amazing, extremely intuitive in particular. I use Pinta on Linux and Mac - it's not as great as the original Paint.NET but still very handy while GIMP is like vim - you have to study it before you can use it (and you are going to forget what you have learnt if you don't use it regularly).
I have never seen any picture editor, free or non-free that I (not a graphics professional, just a webmaster) would enjoy using nearly as much as I enjoy using Paint.NET. Photoshop looks notably more stylish (Paint.NET looks great too, GIMP looks mediocre from the aesthetical point of view) and feels a little bit more intuitive than GIMP yet still very hard to use until you spend days-weeks-months studying it and gaining experience. Paint.NET is 100% intuitive, no need to study, you just use it, looks great, feels convenient and does almost (for some things I actually use GIMP occasionally) everything I ever need quick-and-easy. The fact it is free is an amazing bonus to such a great product.
I am not a graphics professional either, and in the past couple of years, I've started to dislike Paint.NET a lot (I was a huge fan in the early days).
These are, off the top of my head, what I consider to better than Paint.NET:
Mac - Acorn, Pixelmator, Affinity Photo
Windows - Affinity Photo
If you're just doing simple edits en masse to images (crops and simple adjustments), then I would bypass an app like Paint.NET altogether and say these are better/easier:
Mac - iPhoto, Photos, Aperture, Any Lightroom Clone (there are many)
Windows - PhotoDirector, ACDSee, Any Lightroom Clone (there are many)
I agree with you, and it's disappointing people are downvoting you. I ended up having beers with several core GIMP developers a few years ago, and they outright stated that improving usability was not a goal for them, in fact they would rather not have the burden of users who expected it to be easier to use. Speaking for myself, I've found other graphics programs were very intuitive and helped me be very creative, yet I've tried to stick with GIMP over the years and have never felt comfortable with it. I think it's a shame it has been promoted as the go-to graphics package for the free software world (it's the one that's always mentioned). There are now better alternatives that should be promoted.
When was this? I remember getting a similar impression like a decade or so ago but I think now GIMP definitely has devs that are interested in usability. And usability has certainly improved in the last few years, although it still has some serious annoyances (for my workflow / taste).
Close to a decade ago. But I haven't noticed any usability improvements. I feel like the time spent with GIMP was time wasted compared to other software where I could easily get what I needed done yet progressively gained aptitude for productivity and creativity. vim for example supports this. This is not to completely slam GIMP, it can be fine for software to demand users learn it deeply before doing simple things (draw a filled shape, draw over layers, consistently remove backgrounds, all typical things a user wants to do that are far from straightforward in GIMP), and to be appreciated by some people for its scripting ability and filters, but it's not a good go-to graphics software and more user friendly software should be promoted. Its name should be the first clue to this status.
Yeah, I think GIMP has this "open source best in class" status that it maybe deserved ~20 years ago but now it's kind of punching above its weight in terms of popularity.
That said, I actually think the current devs are doing good work, usability has totally improved and I'm somewhat optimistic that once this big, important, multi-year development branch is out of the door we'll see continued improvements.
"...we should finally close this report as WONTFIX. This has been discussed to dead and there are plans for a user interface that will solve the problems that the MDI approach is supposed to solve."
and then:
"We are currently working on changes to the user interface that will solve the problems that the MDI approach is supposed to address. Window-in-window is definitely not part of this plan. It's not going to happen."
So it wasn't just closed as WONTFIX, it was closed as WONTFIX, because users requested MDI, and instead tabbed interface was introduced.
Fair enough, although I'll counter that GIMP didn't actually get single-window mode until 2011 (a full 10 years after that bug report), and it seems like the GIMP devs spent a lot of the intervening period trying to claim lack of single-window mode as a feature.
Like I say, ancient history now.
But some of us are good at holding onto grudges :-)
Yes, I remember it very well, and it's been reinforced nicely by other comments. It does seem GIMP developers have their own idea of reality. Congratulations on having the most heralded open source graphics editor, by the way.
I too found The GIMP easier than Photoshop to pick up, but that was well over a decade ago. The problem for devs with established UI is when there are significant changes you risk alienating your current user base who have got used to the UI. Like MS Word getting rid of the ribbon UI.
I tend to use Inkscape for creating, GIMP for modifying (eg photoshopping).
What are the other better FOSS alternatives to The GIMP?
Maybe i'm an outlier, but with zero prior experience in image editing software, i've found GIMP UI much more approachable than Photoshop. With PS i felt that the UI was much more tailored to experts or at least to people that knew what they were doing, whereas with GIMP i found the UI to be more beginner-friendly and discoverable.
So yeah, YMMV. What a shocker! :)
> Downvote all you want--I know open source can't handle criticism, but I really don't understand why they don't try to make it usable.
I'm intrigued by this. Do you genuinely think that the GIMP developers don't want or don't try to make it usable or to improve it as much as they can? What do you imagine? That they decide to spend their time contributing to a FOSS project and think "hey, today i'll try to make this thing that i love a bit more obscure and unusable!"?
I'm genuinely curious. I seems totally bizarre to me to think that way :S
> Do you genuinely think that the GIMP developers don't want or don't try to make it usable or to improve it as much as they can?
Ten years ago they made a call to artists to find out what they could improve, then didn't like the feedback they got and told the artists they were wrong. I guess that's how they got such a user hostile reputation.
I think most of the feedback they got was "copy Photoshop's interface", but I have to agree with some of the other users in this thread that Photoshop's interface can be shockingly non-intuitive in places.
Yeah, but it's 20 years of Adobe, which is not a company that I highly respect. IMHO there is a good chunk of Photoshop's interface that is the way it is mostly for legacy reasons. Some random programmer thought the feature would be good in buried down in this dialog box off of a vaguely related menu option and it has lived their ever since.
"Ten years ago they made a call to artists to find out what they could improve, then didn't like the feedback they got and told the artists they were wrong."
You won't be able to prove this claim, because you made that up :)
Of course I can. We did two rounds of interviews with pro users, all done by UX/UI experts. Who then wrote a number of functional specs (see gui.gimp.org). We implemented all specs that were completed, some specs weren't completed, unfortunately, and there's always more work to do in that regard.
Thanks. I was really disappointed about the replies to my message. Because just claiming that "yes, they do not care, it's a fact", doesn't make it a fact. No links provided. Nothing.
(In case someone wonders what's the evidence for claiming that the GIMP devs do actually care about the project: the evidence is the release itself. People don't volunteer to working on FOSS software and not care about the software or its users ;) )
> I'm intrigued by this. Do you genuinely think that the GIMP developers don't want or don't try to make it usable or to improve it as much as they can
Yes. They focus on features _they_ care about, not what users care about. The features announced in this release are good proof of that, but if you look in a comment below there's a first-hand experience of how GIMP's developers are annoyed by usability and the fact that users would want a more usable software.
I far from think that GIMP's developers owe me or anyone else anything. I just think it's a shame that GIMP is recommended so often to people that just moved to Linux (or actually in general), because there are good programs--like Pinta for editing and Krita for drawing--that are at par with Photoshop and are totally usable without installing Photoshop or going back to Windows.
"Yes. They focus on features _they_ care about, not what users care about. The features announced in this release are good proof of that"
I feel almost sorry about wasting my time replying you, because you can't even read the first two paragraphs in the release notes.
2.10 rc1 is not about new fancy features, it's about fixing a shit ton of bugs and adding stuff that improves reporting bugs and making GIMP more stable and responsive. Try wrapping your head around that.
"but if you look in a comment below there's a first-hand experience of how GIMP's developers are annoyed by usability and the fact that users would want a more usable software."
> I feel almost sorry about wasting my time replying you, because you can't even read the first two paragraphs in the release notes.
That's not very nice, uh? I don't think I offended anyone here.
However, I don't have to wrap my head around anything, because my point is that I don't think GIMP's developer's priority are in line with its users. I have hardly seen anyone complain about GIMP's bugs and stability problems (actually on that front it's been working great since I remember), and lots of people installing Wine because they can't figure out the interface.
> "but if you look in a comment below there's a first-hand experience of how GIMP's developers are annoyed by usability and the fact that users would want a more usable software."
It looks made up :)
Since you've gone around the thread saying everything people post is a lie, can you actually state facts since you love facts so much? Can you link to a conversation the devs had about usability, where they were saying that it's important to them, their users, and a priority for the project?
> However, I don't have to wrap my head around anything, because my point is that I don't think GIMP's developer's priority are in line with its users.
Let's see.
Did users ask for 16-/32-bit per channel editing? Yes. Done in 2.10.
Did users ask for a tool that would combine rotation, scaling etc.? By gods, they did! Done in 2.10.
Did users ask for canvas rotation/flipping? My, didn't they just! Done in 2.10.
Did users complain about unusable iWarp filter? Oh yes! Fixed in 2.10, with a new tool that works directly on the canvas.
I can go on like that forever. At least half of https://wiki.gimp.org/wiki/Release:2.10_changelog was implemented from bug reports, and that list isn't even entirely up-to-date, partly because I'm a lazy ass, and partly because I keep stumbling on people like you and can't fight the urge to reply.
> Since you've gone around the thread saying everything people post is a lie, can you actually state facts since you love facts so much? Can you link to a conversation the devs had about usability, where they were saying that it's important to them, their users, and a priority for the project?
I can point you to a whole website we created to handle UI problems and a whole mailing list for the same purpose.
I can point you to specs there that were implemented.
If I manage to find old mailing lists archives, I can even point you to a post by a former lead dev who literally said that GIMP's usability sucks and then implemented single-window mode (much requested by users whose interests we ignore as you seem to believe).
> I can even point you to a post by a former lead dev who literally said that GIMP's usability sucks and then implemented single-window mode
It's ancient and of academic interest only, but here[1]'s a bug report where people ask for a single-window mode and are told it's impossible and undesirable. It stayed open for seven years before being WONTFIXed in 2008. I guess this might be before your time on the project, and of course things have changed since then, but I think this might the sort of thing that led to the GIMP team's reputation for user-indifference.
You can install Photoshop's keyboard shortcuts into Gimp, takes like five minutes. And Photoshop doesn't have an intuitive UI either, no way for a newcomer to start using it on the spot, takes time to learn it. So, no, UI is not the problem, Gimp is roughly on the same level as Photoshop when it comes to UI. And spreading this silly myth doesn't help anyone.
Gimp's biggest problem is performance. Then features.
My problem isn't that the UI isn't "intuitive", it's that it's bad - eg. the annoying "layer boundary" thing and the tendency for shortcuts to do different things depending on where you focused last.
I don't know if UX people have a name for it, but I get particularly infuriated by the feeling that I'm being punished for learning shortcuts. Youtube is the other thing that continually annoys me like this.
Edit: In the interest of balance, GIMP's UI has got much better in recent years (single window mode being a particularly massive improvement).
Edit2: I note that fixing the layer boundary issue is on their roadmap so, although it continues to be annoying for now, they can't be reasonably be accused of being dogmatic or intransigent about it.
Note: you are likely being downvoted not for what you said, but for saying you will get downvoted. It's a frequent enough thing that it gets mentioned in the HN guidelines, last I checked.
In this case however I think it's more open source fanboys no-matter-what downvoting, because I dared criticize an open source project made by poor volunteers that work on it on their free time, and if you say they are not getting something you're entitled and you should go fork it and do it yourself.
I don't think this is accurate. If you look at other comments here, you will see some of them criticizing free and open source software and being highly upvoted.
Not really about GIMP but for people who look for an alternative to Photoshop, I can really recommend Affinity Photo. It's actually on a sale right now on Windows Store at least. Unfortunately it doesn't support Linux like GIMP does but it's a really viable alternative to Photoshop for Mac/Windows users.
I used to pay for Photoshop, but it's so darn expensive and I even had to pay to get out of my subscription plan. Affinity Photo cost like one or two months of Photoshop and isn't a subscription service.
Affinity is a fantastic alternative to Photoshop, GIMP, or any other raster manipulation program focused on photos. I'd like to back up this recommendation: for the (one-time) price, it's amazing. I don't miss anything from Photoshop, and the UI is incredibly effective and consistent.
I guess the people in this thread are only interested in open source alternatives, because I can't imagine anyone objecting to it on its merits!
Maybe the problem exists between the keyboard and the chair, but I've found Affinity very clunky to use and can't figure out how to do even simple things in it.
They have a really great series of free, official videos on Vimeo which go over individual features and workflows. They got me up, running, and comfortable faster than any other graphical tool so far. It's similar to learning Blender through really good YouTube tutorials, except official and more comprehensive, so you spend less time searching.
Is this the development-hell branch that has exciting stuff like 16 bit colour and GEGL everywhere? If so then a release candidate is quite exciting. Hopefully getting this out of the door will make the roadmap a bit simpler and ease future development.
GIMP was the first advanced image editor I have used. I grew quite fond of it and its tools. Nowadays I have moved away from it because it does not play well with macOS, but still return to it when I am on a Linux box.
For me, GIMP has always excelled in making simple actions simple. Even today I am quite surprised that other software has not taken some of the ideas for interface that gimp has.
Mainly:
- The selection tool is just miles ahead any of the software I have used. It makes the crop tool obsolete and every time I have to switch modes just to resize my previous rectangular selection I long for it.
- Resizing and clipping layers is dead simple. Other programs I have used have no "clip to selection" on layers and I have never understood why.
- Color to transparency is a single button
- Masks are actually simple to use as one can draw on them using all tools, not just some arbitrary subset
It has its dose of quirks and the interface is in dire need of a rework, however I must say that despite what many people say I find GIMP quite intuitive to use. I suppose that it is a force of habit.
My main issue is actually very non-technical. After I have switched to macOS I've started to put more and more importance to the consistency of the interface. GIMP is very different as it is not rendered using native Cocoa widgets. Things like non-native open/save dialog, clipboard support and drag and drop. It also felt quite slow, compared to when I used it on Linux.
Dialog windows will sometimes appear behind the app window or other windows, and won't come to the foreground or accept input focus until another app is focused first. This also affects other OSes on which Gimp runs.
I've always been frustrated by the GIMP interface. I started out using Corel PhotoPaint in the 1990s and used it through version 5 which was about the time it lost parity with Adobe Photoshop.
Adobe Photoshop and GIMP's interfaces have always felt clunky and non-intuitive to me. I use Photoshop occasionally at work and GIMP at home and really don't enjoy or look forward to using either. I actually try to stay in Inkscape as much as possible with everything I do these days.
That being said, GIMP is capable and it's fantastic for the price. I just don't like it.
I'm another native GIMP user. GIMP was all I had (I couldn't afford Photoshop, not that it worked in Wine around 2000) and it let me do some interesting work in high school art and photography classes. GIMP's script-fu (Guile Scheme) was some of the first code I really understood and modified.
I still use GIMP to this day, largely because I've never ran into limitations that a little scheme or python didn't solve. I've never had a reason to use something else. I still install GIMP as one of the first apps on Windows or Mac OS machines I need to use for any length of time.
I'm not a heavy GIMP user, but I've found that the macOS builds from Partha's Place work better than those from GIMP's website. They are compiled from the development source every few months.
> - The selection tool is just miles ahead any of the software I have used. It makes the crop tool obsolete and every time I have to switch modes just to resize my previous rectangular selection I long for it.
Did you ever try mtPaint[0]? It has such selection feature too
> - Resizing and clipping layers is dead simple. Other programs I have used have no "clip to selection" on layers and I have never understood why.
In mtPaint just select rectangle and click "Delete" button ;-)
I haven't used that but according to the description in documentation it is not exactly the same. In gimp the selection has smart corners one can drag around, feathering and rounded corners.
If I want to do something similar in Affinity Photo for example, I have to make the selection, go to quick mask mode, and then use the move tool. In Pixelmator it had similar issues.
> In mtPaint just select rectangle and click "Delete" button ;-)
That looks a bit unintuitive to me. I would expect that this would either fill the selection with white (or the set document background color, or with transparency).
I've been using dev-builds from https://www.partha.com/ awhile and branch 2.9 is amazingly fast on macos. It has some minor issues but everything else is amazing.
Side note: GIMP user here that has recently moved to a new image editor called Affinity Photo and Affinity Designer, that is super cheap and I highly recommend it to anyone who's looking for a good quality Photoshop and Illustrator alternative, with new ideas and concepts; They are doing an amazing job!
It's a bit of a dilemma, isn't it. I would love an alternative to Photoshop or Affinity that actually offers the same features and has a nice UI (let's be real, it's something most free applications really lack... and the look and feel of the software I'm using has high priority for me) but those who actually want these things aren't the ones developing, and the ones developing apparently don't place a lot of importance on it.
We really need more interdisciplinary developers and free software enthusiasts. I know I'm already working on it as an artist-turned-coder :D
I like free software. It doesn't have to be free as in price. Various small free software developers are supported by its users, e.g. Liberapay / Patreon / Tipeee. Various things were contributed by ZeMarmot. This project is supported by its users using above methods.
> We really need more interdisciplinary developers and free software enthusiasts.
It seems just having people contribute by paying a small amount per week/month is also pretty good method. This allows the existing developers to dedicate more of their time to the project.
Is there a demo version? I couldn't find one quickly navigating their site. I'm hesitant to drop money on something like this without at least being able to try it out first.
I bought Affinity Designer when I was still on Mac, and it was the first affordable non-Adobe app that I used that could open up Illustrator files without mangling them.
When I eventually switched to Windows, I basically ended up repurchasing Designer and Photo. Their software is great for the price.
I haven't tried Affinity products, but in recent times, I have started actually buying software.
I realized that I am willing to pay around 60$ for a marvelous editor (Sublime Text) or audio recording program (Reaper), in a non-subscription model, with non-cumbersome licence handling. I wish there was something similar for CAD software - something like Fusion 360, but where I can pay 60$ and be sure to be able to use it for the next 5 years.
I've tried to use GIMP over several years but never got used to it. Things like resizing and transforming layers, applying filters, etc. are too cumbersome. Many tools also don't have a live preview, which means there's many attempts before getting the needed result.
Now I'm back to Photoshop CS3 from 11 years ago, yet many years ahead of today's GIMP. It has smart objects and filters, layer effects, live previews and the very good UI means I can get things done much faster.
I googled "Photoshop alternative" and GIMP was suggested in the first five hits (I stopped checking after that).
There's this annoying FLOSS motte-and-bailey conversation that you see over and over again that starts with "X is a Free replacement for Y!" and ends with "Why don't you fix it yourself then?", via "Nobody ever said X was a replacement for Y".
An alternative option for PS is not same as being feature parity replacement of PS.
PS is a commercial image editor owned by a company with billions of dollars and many years worth of commercial development and improvements - it is used by a very specific group of users with specific needs and specific skills. And they don't mind paying a hefty fee for it.
GIMP (and other similar OS projects) is for everyone else, who wants to do light image editing in an amateur setting without having to pay a large amount for it.
Having said that - GIMP has some very powerful features that are as good as PS, but not a feature parity replacement of it. Nor do I think GIMP has the funding or the development mind-share to achieve that.
GIMP (and other similar OS projects) is for everyone else, who wants to do light image editing in an amateur setting without having to pay a large amount for it
Except that isn't really true either. If you want to do "light image editing in an amateur setting" GIMP is massive overkill, and there are plenty of easier to use tools. GIMP lives in a strange limbo between not 'light' enough for light work and not heavy enough for 'heavy' work.
Yes, there are a lot of lighter image editing options out there, and most of them are even free. But at least my understanding is that GIMP is still predominantly used by amateurs. I didn't say it's the only option, it's one of the more powerful image editing options for amateurs.
It was for me. I'm not a professional photographer and I learned GIMP first (because it was free and worked on Linux). GIMP was never lacking anything I needed for some light amateur work or web dev work. Photoshop might be easier to use from 0, but anything that was moderately complicated or repetitive I simply scripted away. I have almost 18 years of script-fu plugins written in Scheme and Python.
On the other side, I learned GIMP as my first graphics editor quite easily, while I got lost every time I sat with Photoshop to do even the simplest things.
I work in a team of two on a game with an artist that uses Photoshop and as I sometimes look behind their arm, I see lots of UI conventions that GIMP just does so much better and easier, especially when it comes to selecting and placing stuff around. However, I do envy some of the features like live effects on layers that would actually make my life as GIMP user so much easier; and using effects in GIMP generally isn't as pleasant as it should be.
For me it looks like both come with their own strengths and weaknesses and both would benefit from copying some ideas from each other.
173 comments
[ 4.7 ms ] story [ 214 ms ] threadKudos to the dev team.
I know I am being a bit bitter for an open source project with less funding, but GIMP was supposed to be a poster boy GTK application in the first place.
Anyway, turns out maintaining a professional image editing program is a lot of work. While it is true many fanboys named gimp as some kind of foss and gtk showcase its not like most of them were actively contributing to either gtk or gimp.
There was never any longterm large scale commercial investment in gimp similar to how redhat threw massive resources into gnome and gtk.
This is wrong. GNOME was already well underway when Qt switched to GPL. Before that it used QPL and before that yet another license.
"Sadly, the KDE effort is based on the Qt user interface toolkit which is a non-free library. The main problem is that modification of the toolkit is not allowed. The ability to modify the interface toolkit is a requirement of free software[6] and of the Open Source Software Guidelines[17]. Although Qt is released in source form, permission to redistribute changes made to it is denied."
This is a serious question in general. There will always be s steady stream of significant upgrades/changes of any popular open source project as later developers add in things they consider important. At some point the decision has to be made to stop following those changes. This decision is usually made by default when it turns out that no one can be bothered to follow the changes of a project from another project.
At the same time there is an ongoing transition to use GEGL as the pixel manipulation API, which eventually plugins also need to be updated for.
2.10 is the release which provides all the new APIs for plugins to use, so they can start updating (but existing code will work). 3.0 will then remove the old pixel manipulation APIs, and also transition to GTK3+.
This hopefully keeps the burden on plugin devs to a minimum, and keep the compatibility clear for users (plugins for GIMP2 dont work on GIMP3, and vica versa).
That looks so ridiculous and out-of-place in an image editing program. If it weren't for the left side, you'd think it was Visual Studio!
https://msdnshared.blob.core.windows.net/media/MSDNBlogsFS/p...
This is another case where developers decide what's cool, and of course pick features that are cool to them, not their users.
Not once in my life would I ever think I'm missing that feature in GIMP.
If you're moving from one platform to another, you want legacy files to import perfectly, especially when you don't have time to convert old Illustrator assets into svg files.
On Mac/Windows at least, Affinity Designer was the first app around $50 or under that didn't mangle any of the .ai files I threw at it.
The confrontation comes from the fact that one side insists the software is there (GIMP), whereas the other side view them as disconnected from reality since its use in industry rounds to 0%.
And I think the attitude comes in when GIMP is ignorantly held up as if it were a serious competitor. It isn't. It's something else entirely. It might be good enough for people to make memes for internet shitposting, but between the missing features and the alien UI (to the point where a massive reskin [1] was around for a while), it's never going to be used in industry.
If the GIMP devs want to continue down this path, more power to 'em, I just want the people claiming that it's a Photoshop competitor to educate themselves :)
[1]: https://en.wikipedia.org/wiki/GIMPshop
Who here is claiming GIMP is a serious competitor to Photoshop, or even a competitor at all? Even the GIMP team doesn't (https://www.gimp.org/docs/userfaq.html#are-you-trying-to-dev...).
Or that GIMP is targeting graphics professionals or graphics industry? Don't think so... "We do, however, acknowledge the fact that people will treat GIMP as Photoshop replacement no matter what we tell them, and that’s all right with us. You own this software, it’s up to you to decide how you make use of it."
It is what it is. A tool. That happens to be free and is good enough for some. As somebody above said, it's like complaining why emacs / kate / geany or whatever editor can't do something that visual studio can. If the industry wants a serious competitor to Photoshop, time for them to start building it anew or taking existing free tools nearer to that goal or funding open source devs to do it for them. Complaining and ranting against open source devs who have zero obligations to the industry is not going to achieve anything.
Further, people have tried to get usability fixes into GIMP, and been shot down, as "that's not what the project wants to do."
What is the "dodging" in limiting functional scope of a tool? Each and every software project I know of places some limits on what it will and won't implement. Even Adobe has a suite of tools where one tool like Photoshop can do things Illustrator can't and vice versa. It's not Adobe being "dodgy" about their tools, and nor are open source projects.
Nothing stops people from forking GIMP, implementing whatever fixes they want, distributing, and even selling it. Refusing pull requests is not a crime.
To put that up as the solution is to be fully divorced from reality.
Just gimme the facts, please.
Krita is targeted at artists and illustrators, just like any other specialized tool. People don't look at Vim and complain that lacks desktop publishing features, despite novelists and programmers both needing to manipulate text.
The original complaint regarded tools for artists, and that's Krita. For photographers, there's DarkTable (https://www.darktable.org/), RawTherapee (http://www.rawtherapee.com/), and many others. For my modest needs, I find DigiKam (and occasionally The GIMP) adequate, though I am envious of proprietary tools.
And thus the cyclical debate continues for another turn while a competitive raster editor remains unavailable in the open source ecosystem.
But not, and I repeat -- not when complaints are made in demanding and offensive tone, without providing useful feedback.
I think Pinta is a really good software for artists.
No, it's not :)
"At the 31st of May, 1999, the KImageShop project officially kicked off with a mail by Matthias Elter."
From the very website you linked to :)
Still, a truly amazing project.
And unless you're an artist who does that yourself, it's a bit much to ask others to meet your needs for free like that in the first place.
If non-dev artists want to see that gap close, they need to contribute, not just wait for mana from heaven.
No, that doesn't have to mean actually writing the software. But it could mean paying someone else to do the coding, even if the software is ultimately released as OSS.
It's astounding that a company that makes graphics software can't support modern hardware at all.
But I guess they are the company that can't fix the tremendous piece of shit known as flash
That's a surprise and nice to hear! I thought the support wasn't too great and it requires GTK+3.x or newer. It still uses GTK+2.x (see https://wiki.gimp.org/wiki/Roadmap).
GIMP's users aren't the handful of developers who develop GIMP. GIMP's users would be people that switch to Linux and are looking for an alternative to Photoshop. GIMP is for some reason referred to them as the go-to for image manipulation on Linux so they try it, and when they can't use it they either install Photoshop via Wine or go back to Windows.
I don't know why people recommend GIMP instead of Pinta or more usable software, honestly.
Try adding a text line, then editing it. See if you still want to recommend Pinta.
I absolutely love the idea of having a simple image editor for users who don't need most GIMP features. Pinta devs got quite a few things right, but it needs SO much more work...
FWIW, I have missed better tools for performance tweaking in competing (Adobe) products, so while I don't know if the one in GIMP would be useful, I can see why one would want something like it even as an end user.
Log files aren't user-facing. A performance meter in an image editing program is completely useless to 99% of its users.
The more commonly used features are, the more prominent they should be in the UI.
If not, the "Compose" or "Send" button in email clients should be as big as "view connection details" (to give you an example).
That's how you reduce clutter. GIMP does that horribly, so they should probably _remove_ something, not add.
Come on, get a grip on yourself.
So, as a user I have to identify bottlenecks, now..? That would be my first question.
Since you're implying that by not thinking the performance indicator is an important feature for the average user I want GIMP to be slower, yes--actually I would definitely opt for a slower, usable program than a fast unusable one, and I never thought speed was a problem of GIMP.
Right now I'm running Wine to be able to edit this PSD file without going crazy trying to use GIMP's interface, so obviously speed is not what I care about. I'm sure I'm not the only one, since other problems and not speed seem to be stopping people from using GIMP as their image editor.
Please do pay attention. The release notes specifically describe who did that and for what purpose, and what's in it for end-users. It's all there for you to read. Just please pay attention.
> and I never thought speed was a problem of GIMP
Which suggests that you never used such a basic feature as Gaussian blur in both Photoshop and GIMP.
GIMP _is_ slower than it should be. It actually _is_ a problem for people with modern DSLRs. We want to fix that problem.
Are you sure that performance is a problem for most users? I have talked to people about GIMP before, and it came out 0 times, while usability was the problem that forced most people to install Photoshop or go back to Windows.
In the past 3 days, I've opened pretty big illustrations (about 20,000px* 8000px) on a i35 thinker (mobile phone processor), and I had absolutely no problem opening them, scrolling around, etc.
Doing basic operations like moving, cropping, and resizing layers was so cumbersome that after 3 days I gave up and installed Photoshop 7.
You don't like my tone, and I don't like your inability to process basic information written in clear English in the first two paragraphs of an article that you are discussing, followed by making ridiculous offensive statements. Are we even?
> Are you sure that performance is a problem for most users?
The majority of people who work on large images and use filters do have issue with performance of GIMP. Helping them is a priority to us.
> Doing basic operations like moving, cropping, and resizing layers was so cumbersome that after 3 days I gave up and installed Photoshop 7.
I'm still waiting for those "4 times fewer steps" in Pinta.
If you're part of the project, your attitude is exactly what I'm talking about, so thanks for reinforcing my convictions.
If you're not, since I'm unable to process basic information I wouldn't want to waste your time.
Let's call it a day.
That's funny. They can't be inaccurate, because I'm talking about my personal experience using the software.
As for offending, I never even talked about anyone--just the project, and you are the one who's offended me personally by saying that I should get a grip of myself, I don't understand basic concepts in English, etc. etc.--which let me tell you doesn't look that good for GIMP as a project.
This is how many open source projects handle criticism. I've seen this before, and I'm not surprised. Like I said, thanks for reinforcing the conviction I have that many projects stay in their silos and don't listen to users--to say it nicely--because they don't owe them anything (which of course it's true, but still).
Regular users have no idea how to or have the willingness to get in touch with you, they'll just uninstall and move on. When someone in the middle like me does offer criticism, he doesn't understand basic concepts explained clearly to him--like you said--so obviously there's no solution.
> That said, my way of handling them is most likely far from being humble.
I guess we agree on this one (yay!)
No, this is how some projects handle bullshit.
Criticism is fine and even useful.
Lemme explain the difference, because you are still not taking a grip on yourself.
When you think developers introduced a regression and refuse to fix it, and you can use arguments other than 'because I said so' to prove your point — this is criticism.
When you completely disregard all explanations why a certain feature was added, when you completely dismiss people pointing out that no, in fact other software has the same feature, and people do use it, then continue arguing ad nauseam — this is bullshit.
Got it? Good.
Now, you can be constructive and pay attention, and then we can have a conversation, or you can continue dismissing facts and go on living in an echo chamber. Your choice.
You both prove my point further and give me something to link to for whenever the attitude of GIMP devs towards users will come up.
Photoshop has it too (https://helpx.adobe.com/photoshop/using/performance-preferen...)
> A new Dashboard dock helps with monitoring GIMP’s resource usage to keep things in check, allowing you to make more educated decisions about various configuration options. Dashboard dock
> On the developer side, it also helps us in debugging and profiling various operations or parts of the interface, which is important in our constant quest to improve GIMP and GEGL, and detect which parts are the biggest bottlenecks.
First, it's still in GTK2 so on my monitor it looks zoomed out at 50%.
Then, the UI is the least intuitive thing I've ever seen. To resize and move a layer takes 4 times the steps compared to Pinta (or of course Photoshop).
Shortcuts are the weirdest thing, completely different compared to any other software I've used.
It's such a pity because undoubtedly there is so much work behind it, and the effects are absolutely perfect.
I really don't understand why they don't try to make it usable instead of showing a performance meter.
Thanks for release!
I've finally come to the conclusion that it isn't me, but it's the usability that is completely flawed.
Unfortunately, when one tries to say anything the response is "you go fix it then, asshole!" but I don't know how to develop desktop apps, so...
There are other ways you can help!
As someone with experience using graphics applications, you could look into making UI mockups of the changes / improvements you have in mind. Seeing the mockups might get someone interested enough to start work on user-testing, prototyping, implementing, ..., them.
Edit: Despite my defence of the current devs, I continue to be annoyed by the GIMP cheerleaders who continue to unhelpfully blame the program's shortcomings on its users.
They don't care.
Squaring that particular circle isn't easy IMO.
There are people who still prefer the separate window system, I've always liked the unified one. Simple differences like this can become divisive and split your developer and user groups.
You should try Blender! Some applications are just complicated and need time to learn.
My knee jerk reaction to your knee jerk reaction is: If its so easy, go fix it!
As for the "go fix it", that's exactly the reaction I was expecting. Open source projects can't handle criticism because it's volonteers working on it, so "go fix it if you don't like it". Not very productive, uh?
Personally, I use Pinta or Photoshop via Wine. Pinta is very good though.
> I really do want to love GIMP, but it's completely unusable.
And you speak of being productive...
I can sugarcoat it, but we're all grown ups here I thought..? I didn't offend anyone.
Speak for yourself. As someone who's tried to write a simple image editor on a few occasions, writing the image effect was by far the easiest (and most fun) part. Almost all the core algorithms are published and well understood and beyond that it's mostly simple linear algebra. There are also plenty of objective measures to judge how good your implementation is.
Deciding where to "move the buttons" to took up almost all of my development time, was incredibly hard, dreadfully boring and what kept killing my interest in the project.
Given the choice between spending 20 hours trying to make the sharpening filters 5% faster or 20 hours trying to make the sharpening filters more intuitive and easy to use I'd choose the former every time and, unfortunately, it seems far too many developers feel the same.
That's quite a statement. I would only say it with the following qualifiers:
Paint.NET is by far the best free picture editor I've seen for Windows.
I've seen plenty of non-free (some very affordable, some less so) editors for Mac and Windows that are way better than Paint.NET.
These are, off the top of my head, what I consider to better than Paint.NET:
Mac - Acorn, Pixelmator, Affinity Photo
Windows - Affinity Photo
If you're just doing simple edits en masse to images (crops and simple adjustments), then I would bypass an app like Paint.NET altogether and say these are better/easier:
Mac - iPhoto, Photos, Aperture, Any Lightroom Clone (there are many)
Windows - PhotoDirector, ACDSee, Any Lightroom Clone (there are many)
That said, I actually think the current devs are doing good work, usability has totally improved and I'm somewhat optimistic that once this big, important, multi-year development branch is out of the door we'll see continued improvements.
I get your point about having to wait for 7 years, but let's just scroll down to the bottom:
https://bugzilla.gnome.org/show_bug.cgi?id=7379#c118
"...we should finally close this report as WONTFIX. This has been discussed to dead and there are plans for a user interface that will solve the problems that the MDI approach is supposed to solve."
and then:
"We are currently working on changes to the user interface that will solve the problems that the MDI approach is supposed to address. Window-in-window is definitely not part of this plan. It's not going to happen."
So it wasn't just closed as WONTFIX, it was closed as WONTFIX, because users requested MDI, and instead tabbed interface was introduced.
Like I say, ancient history now. But some of us are good at holding onto grudges :-)
We started working with Peter Sikking, professional UX architect, around 2006. He departed three years ago or so.
That puts your claim roughly in the time when Pete was extremely active and wrote several functional specs that were implemented to make UI better.
Now tell me, do you feel like you remember 10 years old stuff correctly after several beers?
Didn't that provide a decent migration path for PhotoShop users to GIMP?
I tend to use Inkscape for creating, GIMP for modifying (eg photoshopping).
What are the other better FOSS alternatives to The GIMP?
So yeah, YMMV. What a shocker! :)
> Downvote all you want--I know open source can't handle criticism, but I really don't understand why they don't try to make it usable.
I'm intrigued by this. Do you genuinely think that the GIMP developers don't want or don't try to make it usable or to improve it as much as they can? What do you imagine? That they decide to spend their time contributing to a FOSS project and think "hey, today i'll try to make this thing that i love a bit more obscure and unusable!"?
I'm genuinely curious. I seems totally bizarre to me to think that way :S
Ten years ago they made a call to artists to find out what they could improve, then didn't like the feedback they got and told the artists they were wrong. I guess that's how they got such a user hostile reputation.
GIMP has nothing but to learn from it. There would be nothing negative about copying it outright.
Regardless, I can actually scale objects easily by dragging in Photoshop, at least.
You won't be able to prove this claim, because you made that up :)
(In case someone wonders what's the evidence for claiming that the GIMP devs do actually care about the project: the evidence is the release itself. People don't volunteer to working on FOSS software and not care about the software or its users ;) )
Yes. They focus on features _they_ care about, not what users care about. The features announced in this release are good proof of that, but if you look in a comment below there's a first-hand experience of how GIMP's developers are annoyed by usability and the fact that users would want a more usable software.
I far from think that GIMP's developers owe me or anyone else anything. I just think it's a shame that GIMP is recommended so often to people that just moved to Linux (or actually in general), because there are good programs--like Pinta for editing and Krita for drawing--that are at par with Photoshop and are totally usable without installing Photoshop or going back to Windows.
I feel almost sorry about wasting my time replying you, because you can't even read the first two paragraphs in the release notes.
2.10 rc1 is not about new fancy features, it's about fixing a shit ton of bugs and adding stuff that improves reporting bugs and making GIMP more stable and responsive. Try wrapping your head around that.
"but if you look in a comment below there's a first-hand experience of how GIMP's developers are annoyed by usability and the fact that users would want a more usable software."
It looks made up :)
Edit: formatting
That's not very nice, uh? I don't think I offended anyone here.
However, I don't have to wrap my head around anything, because my point is that I don't think GIMP's developer's priority are in line with its users. I have hardly seen anyone complain about GIMP's bugs and stability problems (actually on that front it's been working great since I remember), and lots of people installing Wine because they can't figure out the interface.
> "but if you look in a comment below there's a first-hand experience of how GIMP's developers are annoyed by usability and the fact that users would want a more usable software." It looks made up :)
Since you've gone around the thread saying everything people post is a lie, can you actually state facts since you love facts so much? Can you link to a conversation the devs had about usability, where they were saying that it's important to them, their users, and a priority for the project?
Let's see.
Did users ask for 16-/32-bit per channel editing? Yes. Done in 2.10.
Did users ask for a tool that would combine rotation, scaling etc.? By gods, they did! Done in 2.10.
Did users ask for canvas rotation/flipping? My, didn't they just! Done in 2.10.
Did users complain about unusable iWarp filter? Oh yes! Fixed in 2.10, with a new tool that works directly on the canvas.
I can go on like that forever. At least half of https://wiki.gimp.org/wiki/Release:2.10_changelog was implemented from bug reports, and that list isn't even entirely up-to-date, partly because I'm a lazy ass, and partly because I keep stumbling on people like you and can't fight the urge to reply.
> Since you've gone around the thread saying everything people post is a lie, can you actually state facts since you love facts so much? Can you link to a conversation the devs had about usability, where they were saying that it's important to them, their users, and a priority for the project?
I can point you to a whole website we created to handle UI problems and a whole mailing list for the same purpose.
I can point you to specs there that were implemented.
If I manage to find old mailing lists archives, I can even point you to a post by a former lead dev who literally said that GIMP's usability sucks and then implemented single-window mode (much requested by users whose interests we ignore as you seem to believe).
Good enough?
It's ancient and of academic interest only, but here[1]'s a bug report where people ask for a single-window mode and are told it's impossible and undesirable. It stayed open for seven years before being WONTFIXed in 2008. I guess this might be before your time on the project, and of course things have changed since then, but I think this might the sort of thing that led to the GIMP team's reputation for user-indifference.
[1] https://bugzilla.gnome.org/show_bug.cgi?id=7379
Their actions and statements really lead us to conclude as much, yes.
Facts, please.
Gimp's biggest problem is performance. Then features.
I don't know if UX people have a name for it, but I get particularly infuriated by the feeling that I'm being punished for learning shortcuts. Youtube is the other thing that continually annoys me like this.
Edit: In the interest of balance, GIMP's UI has got much better in recent years (single window mode being a particularly massive improvement).
Edit2: I note that fixing the layer boundary issue is on their roadmap so, although it continues to be annoying for now, they can't be reasonably be accused of being dogmatic or intransigent about it.
I haven't heard this one before. Could you please provide examples? Thanks.
In this case however I think it's more open source fanboys no-matter-what downvoting, because I dared criticize an open source project made by poor volunteers that work on it on their free time, and if you say they are not getting something you're entitled and you should go fork it and do it yourself.
That's my impression, anyway.
It does? Let's check.
1. Select a layer in Layers dock. 2. Shift+T for Scale tool 3. Drag to scale, press Enter to confirm 4. M for Move tool 5. Move
Please educate me how to make that 4 times faster with Pinta.
I used to pay for Photoshop, but it's so darn expensive and I even had to pay to get out of my subscription plan. Affinity Photo cost like one or two months of Photoshop and isn't a subscription service.
I guess the people in this thread are only interested in open source alternatives, because I can't imagine anyone objecting to it on its merits!
For me, GIMP has always excelled in making simple actions simple. Even today I am quite surprised that other software has not taken some of the ideas for interface that gimp has.
Mainly:
- The selection tool is just miles ahead any of the software I have used. It makes the crop tool obsolete and every time I have to switch modes just to resize my previous rectangular selection I long for it.
- Resizing and clipping layers is dead simple. Other programs I have used have no "clip to selection" on layers and I have never understood why.
- Color to transparency is a single button
- Masks are actually simple to use as one can draw on them using all tools, not just some arbitrary subset
It has its dose of quirks and the interface is in dire need of a rework, however I must say that despite what many people say I find GIMP quite intuitive to use. I suppose that it is a force of habit.
Edit: formatting
This is huge. I've used Pixelmator because I liked its UI, but even simplest tasks like this one were unnecessary hard. I kept Gimp just for this.
Adobe Photoshop and GIMP's interfaces have always felt clunky and non-intuitive to me. I use Photoshop occasionally at work and GIMP at home and really don't enjoy or look forward to using either. I actually try to stay in Inkscape as much as possible with everything I do these days.
That being said, GIMP is capable and it's fantastic for the price. I just don't like it.
I still use GIMP to this day, largely because I've never ran into limitations that a little scheme or python didn't solve. I've never had a reason to use something else. I still install GIMP as one of the first apps on Windows or Mac OS machines I need to use for any length of time.
https://www.partha.com/
Did you ever try mtPaint[0]? It has such selection feature too
> - Resizing and clipping layers is dead simple. Other programs I have used have no "clip to selection" on layers and I have never understood why.
In mtPaint just select rectangle and click "Delete" button ;-)
[0] https://github.com/matyler/mtPaint
[1] http://mtpaint.sourceforge.net/handbook/en_GB/chap_04.html
If I want to do something similar in Affinity Photo for example, I have to make the selection, go to quick mask mode, and then use the move tool. In Pixelmator it had similar issues.
> In mtPaint just select rectangle and click "Delete" button ;-)
That looks a bit unintuitive to me. I would expect that this would either fill the selection with white (or the set document background color, or with transparency).
We really need more interdisciplinary developers and free software enthusiasts. I know I'm already working on it as an artist-turned-coder :D
I like free software. It doesn't have to be free as in price. Various small free software developers are supported by its users, e.g. Liberapay / Patreon / Tipeee. Various things were contributed by ZeMarmot. This project is supported by its users using above methods.
> We really need more interdisciplinary developers and free software enthusiasts.
It seems just having people contribute by paying a small amount per week/month is also pretty good method. This allows the existing developers to dedicate more of their time to the project.
When I eventually switched to Windows, I basically ended up repurchasing Designer and Photo. Their software is great for the price.
I realized that I am willing to pay around 60$ for a marvelous editor (Sublime Text) or audio recording program (Reaper), in a non-subscription model, with non-cumbersome licence handling. I wish there was something similar for CAD software - something like Fusion 360, but where I can pay 60$ and be sure to be able to use it for the next 5 years.
Now I'm back to Photoshop CS3 from 11 years ago, yet many years ahead of today's GIMP. It has smart objects and filters, layer effects, live previews and the very good UI means I can get things done much faster.
There's this annoying FLOSS motte-and-bailey conversation that you see over and over again that starts with "X is a Free replacement for Y!" and ends with "Why don't you fix it yourself then?", via "Nobody ever said X was a replacement for Y".
PS is a commercial image editor owned by a company with billions of dollars and many years worth of commercial development and improvements - it is used by a very specific group of users with specific needs and specific skills. And they don't mind paying a hefty fee for it.
GIMP (and other similar OS projects) is for everyone else, who wants to do light image editing in an amateur setting without having to pay a large amount for it.
Having said that - GIMP has some very powerful features that are as good as PS, but not a feature parity replacement of it. Nor do I think GIMP has the funding or the development mind-share to achieve that.
Except that isn't really true either. If you want to do "light image editing in an amateur setting" GIMP is massive overkill, and there are plenty of easier to use tools. GIMP lives in a strange limbo between not 'light' enough for light work and not heavy enough for 'heavy' work.
I work in a team of two on a game with an artist that uses Photoshop and as I sometimes look behind their arm, I see lots of UI conventions that GIMP just does so much better and easier, especially when it comes to selecting and placing stuff around. However, I do envy some of the features like live effects on layers that would actually make my life as GIMP user so much easier; and using effects in GIMP generally isn't as pleasant as it should be.
For me it looks like both come with their own strengths and weaknesses and both would benefit from copying some ideas from each other.
[0] http://appimage.github.io/apps
[1] https://github.com/aferrero2707/gimp-appimage/releases/tag/c...
[2] https://github.com/aferrero2707/gimp-appimage/releases/downl...