[Question] Why QT rewrite?

[dani3l0]

Quite a long delay, but I just updated my flatpak apps xD

Might be related to #4499

Not a complaint, i still love the app, but just a question: why dropping gtk4 in favor of qt6?

Couldn't find information anywhere why rewrite happened 😢

I'm asking, because gtk4 worked well and looked great without issues, and will miss it. KDE's QT theming is a bit unpleasant, blue-ish background and light borders on each component makes UI kinda cluttered and confusing.

When it comes to functionality, app still works perfectly fine 😁

---

My environment

Fedora Linux 43
GNOME 49 on Wayland
i5-14500, 64GB RAM
Easy Effects: 8.0.5
KDE Flatpak runtime (Wayland)
KDE Frameworks: 6.19.0
Qt: Using 6.10.0 and built against 6.10.0
Flatpak (flathub)
Build ABI: x86_64-little_endian-lp64
Kernel: linux 6.17.9-300.fc43.x86_64

[wwmm]

Not a complaint, i still love the app, but just a question: why dropping gtk4 in favor of qt6?

I will try to answer that in a way that does not give fuel to the people that live in the imaginary Qt vs gtk war. But I have the feeling that it will give them fuel anyway... Well... Whatever... Long story short I had two main reasons.

When I started this project the only way I found to do what this app needs to do was using gstreamer. And when you use gstreamer choosing gtk as toolkit makes a lot of sense. We can use gstreamer and Qt but using it in gtk is a lot more straightforward. So I used it. And to be honest I liked using gtk a lot while easyeffects was relatively small. But the journey stopped being pleasant when the application became bigger.

Gtk is a very capable toolkit. It really is. If the programmer really wants you can do what is needed both on gtk and Qt. There is not a lot that you can do with one that you can't with the other. But the fact gtk uses C and xml can get quite tiring and time consuming when the application is not small. It was not that I could not do what I wanted. I could. But the bigger easyeffects got the less enjoyable using gtk became. And when you are doing something for free in your free time the journey has to be fun. Otherwise there is no point. It becomes a job you do not want to do.

The second big reason was the libadwaita situation. It started to feel uncomfortable. Not because I do not like the library. I do. But because using it like I use any other library was a HUGE mistake. That is not the vision behind libadwaita development. I may be mistaken but the mindset in gtk land seems to be "gtk as a solid foundation and libadwaita with the nice stuff". They have all the right in the world to do that. It probably works well for them.

Initially I tried to write in pure gtk some of the nice things adwaita does. It is what I saw some people say at the time. If your focus isn't only gnome then do everything in pure gtk. I tried and it was painful and unfun. And in the end the result was nowhere near as good as libadwaita. So at some point I just gave up and decided that I would not reinvent the wheel and used libadwaita. I should not have done that.

As time passed I started to see more and more at random corners of the internet the statement "if its developer is using libadwaita is because he only cares about gnome". That was never my intention and this made me feel really uncomfortable. I just did not want to waste time reinventing the wheel. So I did the adult thing to do in these kind of situations. I moved my project to a different toolkit.

Some people will probably say "but the situation is the same now". Not really. I seriously doubt we will ever see Qt becoming just a solid foundation for KDE/Kirigami. It is not a development model that makes sense there. And if there is some kind of disaster in KDE/Kirigami land that requires us to move away from them the truth is that writing custom widgets in qml with the needed functionality is a lot more enjoyable to me than trying to do the same using gtk and xml. But again if I can avoid reinventing the wheel I will avoid it. I am aware of the fact some people do not like to have KDE dependencies. But the alternative would be rewriting functionality that is already implemented there. I do not intend to do that.

I'm asking, because gtk4 worked well and looked great without issues, and will miss it. KDE's Qt theming is a bit unpleasant, blue-ish background and light borders on each component makes UI kinda cluttered and confusing.

I have nothing against making EE as integrated to different desktops as possible. But there is a technical challenge involved that goes beyond the application code. Some people may not remember it but when we moved to gtk4 easyeffects looked absolutely horrible on KDE. It looks better now not because gtk did something amazing that allowed me to make easyeffects look better on KDE. But because KDE put effort in making gtk4 apps look better when running inside KDE. Gtk and Qt do not really offer any magic button the app developer could use to make this whole theme situation plug and play.

For many years I endured complaints like "Why did you write such an important app in gtk and not Qt?" from people that hate gtk. So I was expecting the same to happen after moving to Qt. The amount of complaints is mostly the same. Just its source has changed. After using Linux for more than 20 years I've got tired of this whole Qt versus Gtk nonsense. I just use whatever I think is the best tool at the time and move on with my life.

[dani3l0]

Thank you for very detailed answer, really appreciate it.

Closing the issue, as you nailed the question 😆 Everything is clear now.

[sandikata]

If you talk about Vulkan+Nvidia problems, they're not related to GTK in any way :)

[sandikata]

• GTK is more direct than QT.

QT add another layers between sound card and easyeffects, which is equal to more latency (which is not desired efect in Audiophile environment). Such silly translation from QT - Qtmultimedia - GSTreamer - End Application and so on :)

In that Regard, GTK is still much better.

Would be better to left and GTK version as well.

Not just to force peoples install a lot of fat and ugly QT/KDE dependencies just for a single app, it's massive waste of time specially on gentoo, to build all this useless stuff (by any other applications).

[wwmm]

QT add another layers between sound card and easyeffects, which is equal to more latency (which is not desired efect in Audiophile environment). Such silly translation from QT - Qtmultimedia - GSTreamer - End Application and so on :)

You are definitely confusing things. I used GStreamer when I started this project. The current implementation uses PipeWire native filter objects. There is no difference in this regard between the Qt implementation and the one that used GTK. No extra latency. We do not even use Qt's multimedia framework...

Not just to force peoples install a lot of fat and ugly QT/KDE dependencies just for a single app, it's massive waste of time specially on gentoo, to build all this useless stuff (by any other applications).

And people that hate Gtk say something similar about the version based on Gtk... I can see why both sides would feel this way. But the implicit idea I should force myself to work with a tool I am not willing to use because of that is honestly very unreasonable.... That will never happen. And for that reason I didn't delete the old gtk code https://github.com/wwmm/easyeffects/tree/gtk4-legacy. People that consider the Gtk version better for them can just build a package from the legacy repository. Just do not expect me to work on it.

[Mlocik97]

I'm working on gtk fork https://github.com/Mlocik97/easyeffects-GTK tho I didn't had much time yet (mainly last month I had a lot of stuff to do). I plan to just make GTK front-end and re-use back-end as is in this codebase, backporting it.

[wwmm]

I plan to just make GTK front-end and re-use back-end as is in this codebase, backporting it.

Feel free to ask questions about the new code in the master branch if you need. Just using a gtk window on top of the current backend is going to be a little tough because just like the backend in the legacy branch uses gtk signals and events the same thing is done now with the Qt code. And there are subtle differences in behavior between gtk and Qt when it comes to handle signals and slots.

My first thought if I were to do something like that would be to backport the changes done to plugins backends and new plugins like the crusher to the legacy branch.

[sandikata]

And people that hate Gtk say something similar about the version based on Gtk... I can see why both sides would feel this way. But the implicit idea I should force myself to work with a tool I am not willing to use because of that is honestly very unreasonable.... That will never happen. And for that reason I didn't delete the old gtk code https://github.com/wwmm/easyeffects/tree/gtk4-legacy. People that consider the Gtk version better for them can just build a package from the legacy repository. Just do not expect me to work on it.

Anyway you brake something which was working, for some weird and nonsense reason. Or just because KDE fanboys crying all day long to change it to their liking?

It could be just because of that KDE is not capable to make GTK apps look nice on QT engine :> Or who knows. But for me specially, the new version is just for introducing bugs... And i'm sure that i'm not the only suffering from random crashes with QT version. So the logical move is to use GTK version, or not to use easyeffects at all (if gtk version disappears) :>

It's not a "hate" or whatever, it's deserved criticism :)

I know that you're doing it in your free time, i do respect that. But don't respect the list of reasons behind this change.

[wwmm]

I know that you're doing it in your free time, i do respect that. But don't respect the list of reasons behind this change.

Dou you? It doesn't seem so
. Considering the kind of comments you have been spamming and the statement that my reasons are "nonsense" the only message you are actually sending is that you actually think I should force myself to do something I do not want.

You know that is not going to happen. I am not going back to the gtk version just because you don't like qt. So what exactly do you intend to do here? Waste people's time spamming your comments about how much you hate "kde fanboys" through all our discussions?

If you somehow expect me to abandon my project because of your unsatisfactory you are wasting your time and mine.

[sandikata]

Nope, just your total ignorance on any report for crashes with QT version. Many peoples have suffered them. You said in one of the reports that the KDE fanboys asked you for that, because they hate gtk, which is invalid argument, if some app is working :>

And yes, i do respect! But maybe not in way you'd like.

[wwmm]

Nope, just your total ignorance on any report for crashes with QT version. Many peoples have suffered them. You said in one of the reports that the KDE fanboys asked you for that, because they hate gtk, which is invalid argument, if some app is working :>

And yes, i do respect! But maybe not in way you'd like.

As I said in another discussion open an issue providing information about the problems you are having. Within my possibilities I will try to help like I have always tried to help other users. But if you just want to say in several posts you don't like qt we are not going to get anywhere. Nothing I say will make you like qt. And nothing you say will make me go back to gtk. It will be just a huge waste of both of our time.