Live data from Hacker News

25 Years of Krita

krita.org

91–100 of 101 posts

Re: 25 Years of Krita

#91

Krita is a great replacement for photoshop.

Call me paranoid, but I think it's harmful for Krita when someone frames it as a Photoshop alternative - because it really isn't. It is a program targeted specifically at digital painting. Yes, it can do some "basic" image manipulation like Photoshop does, and even you can do some animation stuff too - but it doesn't intend to target everything image manipulation wise as Photoshop does. I for one as a graphic designe…

Well, as Halla, the Krita maintainer, I kind agree. Image manipulation is not a goal. But for animation, we have a very specific goal. What we want to see is someone doing a looney Tunes like hand-drawn animation in Krita. And I've seen some, so mission accomplished!

Re: 25 Years of Krita

#92
post #37

What a self-indulgent writeup. The article came nowhere near answering the central question: what are the devs doing other than constantly changing the name of the project and how is this thing better than gimp.

Oh, noes... A post that has this in the second paragraph:

"I'll hope you, dear reader, will forgive me for making this a really personal post; a very large part of my life has been tied up with Krita, and it's going to show."

ended up in something "jmix" thought was "self-indulgent".

Re: 25 Years of Krita

#93
post #6

Wow, is it just me or does that screenshot of 2005 Krita 1.4 UI[1] look absolutely amazing? All the icons are very distinctive and clear on what they do and the colors make them pop without stabbing your retinas. Perfect. Feels like the mid-2000's was the peak in software UI design and it slowly started going downhill afte that, with companies chasing the soulless corporate mono-chromatic flat design inspired from mo…

Agreed. I suspect the shift in design occurred when the base assumption moved from "most users have never used a computer before" to "most users are familiar with computers". IMO Windows Interface Guidelines from 1995 are still relevant today: https://ics.uci.edu/~kobsa/courses/ICS104/course-notes/Micro...

To be honest, I still reference my paper copy of the Windows 1995 Interface Guidelines from time to time, when people propose new features for Krita. They are not always useful, but at least they don't change all the time.

Re: 25 Years of Krita

#94
post #30

Earlier quoted context omitted.

That's just any big framework. Porting a large codebase between major releases of GTK (or Javascript frameworks) isn't much better.

Depends on the framework. Qt5 was released in 2012. All of my LCL/Lazarus projects from that time open in modern straight-out-of-git Lazarus just fine. I'm pretty sure 99.9% of my LCL/Lazarus projects from 2005 (when Qt4 was released) also work out of the box, with the exception of those that tried to use strings as byte buffers (fixing this is a quick search and replace) since the datatype changed (this was a compil…

Yeah. We need to do lots of low-level stuff, and things related to GPU have changed a lot from Qt to Qt.

Re: 25 Years of Krita

#95
post #83

Earlier quoted context omitted.

I really think it's holding desktop Linux back. I was making windows apps when I was kid using Visual Studio. There is no way I'd be able to with any of Linux frameworks. The closest thing would probably be to use Java.

qtcreator actually supports a workflow quite similar to Visual Basic 6 in terms of making guis/adding functionality to buttons, etc

*right clicks* *context menu covers entire screen*

You can't drag and drop widgets in the hierarchy view, which sucks when a container has a width/height of zero because it has no widgets inside of them yet.

It's not terrible, but it could be better.

I can't be the only one feels this is weird, right? Like, you have a team who can create an entire IDE, even rendering widgets and dragging them around, but somehow things like organizing a context menu in submenus and making the designer intuitive is too hard. And this is a consistent theme in every one of these apps. The things that should be the hardest to program are programmed, the easiest to implement quality-of-life improvements are lacking. GPU-accelerated video transitions in a timeline based video editor? Easy peasy. A status bar that actually displays statuses? Absolutely impossible to implement.

Re: 25 Years of Krita

#96

Earlier quoted context omitted.

There seems to be a phenomena I observe in the FOSS community where we reflexively respond to suggestions despite them being far beneath their FOSS alternatives. Only after digging or running into issues, do you find the good stuff. Brand name -> Low quality, highly mentioned -> High quality, rarely mentioned Photoshop-> GIMP -> Krita Windows -> Debian-family/Ubuntu/Mint -> Fedora I almost feel like a religious Zealo…

David Revoy doesn't recommend Fedora 40 for digital artists any more. The main reasons are regressions in tablet software with Plasma 6, and Krita not being a Wayland app. [1] https://www.davidrevoy.com/article1030/debian-12-kde-plasma-...

He added an edit that the tablet software in Plasma 6 still works as it did before when run in X11. Only if you switch to Wayland, or your distro does it for you by removing X11 packages, will you get the (for the time being) less capable version.

Re: 25 Years of Krita

#97
post #28

Earlier quoted context omitted.

You are right, your theory works but it looks like you never tried OpenSUSE. It’s basically Fedora but with great administration tools, a better package manager (but it also uses rpm packages) and with optional AUR-like community packages. OpenSUSE >>> Fedora >>>> Debian Family

I wonder if there OpenSUSE distros for the RPi or if I’m best sticking to Raspbian due to the hardware quirks.

https://en.opensuse.org/HCL:Raspberry_Pi4

It exists !

Re: 25 Years of Krita

#98

I never knew that the project has been around for so long, let alone that its older than myself. But what I can clearly read from the blog post: Qt likes to break compatibility a lot between their major releases...

Qt hasn't been terrible in terms of upgradeability between major versions for many use cases, but Krita got the short end of the stick.

Krita's scope and architecture requires both cross-platform compatibility and deep integration with the OS. For the latter, they need some degree of going past Qt's cross-platform abstractions, and Qt happens to have made some substantial changes to the way they build on the platform's graphics APIs.

As long as you're sticking with Qt's cross-platform API, you're good; if you ignore it and go full OpenGL or Vulkan for all UI, you're good; it's when you try to mix both that things get hairy.

Re: 25 Years of Krita

#99
post #98

I never knew that the project has been around for so long, let alone that its older than myself. But what I can clearly read from the blog post: Qt likes to break compatibility a lot between their major releases...

Qt hasn't been terrible in terms of upgradeability between major versions for many use cases, but Krita got the short end of the stick. Krita's scope and architecture requires both cross-platform compatibility and deep integration with the OS. For the latter, they need some degree of going past Qt's cross-platform abstractions, and Qt happens to have made some substantial changes to the way they build on the platform…

By going full Vulkan, you mean do nothing at all in Qt, or do you mean a particular way to use Qt but not some parts of it?

Re: 25 Years of Krita

#100
post #48
post #33

Earlier quoted context omitted.

It's not, it's just different. You mostly get newer stuff and more breakage.

So it's like Debian Unstable?

A bit, but it feels a lot more alien. If you're moving through the Debian/Ubuntu family, the base system is quite similar, apt is always there for you etc... . In Fedora, it's different.
Post reply on HN