Krita has turned out to be the easiest and most intuitive program to use that I have tried. I use krita focused on practicing animation, since it is very comfortable to use the Timeline and the Onion-Skin. I also use Vector Layers a lot for illustration. Because they are really useful because of the versatility with which they work and the ease of changing shapes and colors
But one day I realized a little problem. Paint layers are especially good for brush effects and animation sketches. But, to do the Line-Art, it is very impractical. I tried the Brush Stabilizers, but was not happy with the result. I got the idea to do it with vector layers by editing shapes. But I found out that you canât
So I thought of another way to do it with Vector layers. I thought to make a Vector layer for each frame, then convert them to Paint layers and then add it to the Timeline. But I realized that that was a lot of slow and unproductive work.
That is why I make this request. If you could animate in the Vector layers, and also with all the normal timeline options (such as duplicating that frame, the vector shapes are copied) it would be very useful in the workflow and would save a lot of time and effort. Making each frame easier and faster to edit. combining the way vectors and timeline work would be a hugely useful tool.
PS: I realized that today October 28 is the International Animation Day⌠interesting day to make an Animation feature requestâŚ
We designed the animation support in Krita explicitly for hand-drawn animation. While at one point we might add frames to vector layers, we wonât be adding rigging, bones or automatic interpolation between frames.
Youâre missing two interesting alternatives, btw: Pencil2D and OpenToonz.
Thatâs because weâre not just developing whatever people ask for: weâre working with a vision of what we want Krita to be, and our vision for animation is that Krita isnât a replacement for flash, weâre making Krita for people who want to create classic hand-drawn animations.
But itâs a big part of animation. Thatâs like refusing to have vector functionality in a program because itâs supposed to be a raster drawing app.
I understand not just adding anything people ask for, because it would end up being a mess and it takes time to make things actually good, but that explanation doesnât make sense. Having basic tweening doesnât mean that you have to âreplace Flashâ (and I didnât mention rigging or bones).
Just say itâs because you donât have the means and manpower, so you have to make choices, Iâm sure everybody would be understanding.
But by the way I forgot, thanks for the two suggestions.
I think you guys are misunderstanding a bit. with vector animation I meant simply Krita default vector functions + timeline. NOT rigging, bones or automatic interpolation between frames. those are specific functions that only a program that is dedicated exclusively to vectors needs. not like Krita. I never said it as a replacement for flash, since I have never tried it. and I also donât think that Krita needs rigging, bones or automatic interpolation between frames because the workflow style Iâve had it doesnât need those functions. and as the same request guide says, you have to think about the worflow. not in random functions just 'cause yes
I did propose building simple vector animation functionality and more keyframing options myself. Even with the focus being primarily on raster animation, there are still many great use cases for vector adjustments being tweenable (grabbing a node and moving it elsewhere to dynamically change the shape of an object to a different shape). Even having that alone would open up a lot of possibilities. Still, I can imagine anything that would involve interpolation and other high level vector/mesh editing would probably be a major numeric release (like Krita 6.0) since I can imagine it wouldnât be as simple as âadjust what already existsâ but also making it work with how Krita renders images. and as aluded to, you technically can do all this stuff in Blender by just exporting a drawing or animation as a PNG sequence and importing it to blender and going to town. (I did that myself, it works alright).
Plus amusingly I recall 3 years ago, the developers said they wouldnât add enhanced animation support (e.g. cycle frames) and now they have one of the few animation software that can support multilayering and cycles. Anything is possible, but given the nature of the project, it may take a solid proof of concept without completely destroying the vision. As I can agree that Krita is strongest as a Raster application with some vector features, but comparing it to Clip Paint Studio, I can also say that at least Krita is made âBy artists for artistsâ as CSP does have some questionable design philosophies and wonât adapt to modern trends, while Krita at least can being open source. Though it could just ultimately be a matter of a volunteer making a proof of concept that works, than expecting the base team to handle it.
Now you can mark some frames on the timeline as a cycle.
Though Iâm confused reading that @Kao_Ninjaratzu got a negative response when asking about cycle frames 3 years ago, because while it was very much on hold, and there were other animation features before that - clone frames and animated transform mask - and for sure no one was working on that, I believe cycle frames were also kind of planned, though since no one was working on that, maybe it was put into the âinfinite future TODOâ so thatâs why they got a negative response? Like, what the developer said was more like âno one is working on that and weâre not planning to start working on anything currentlyâ (which was true because before Emmet and Eoin came, no one was working specifically on animation)?
As a developer, what I can say is, no one will tell you âyeah sure we plan to do itâ unless itâs one of the biggest pain point - like text tool or shortcuts or, in the past, resource system etc. - or unless they are working on it at the very moment or plan to do it in next week or at least month. There are no actual roadmaps or anything. Especially not three years in the future There are of course both advantages and disadvantages of that, but note that quite a lot of new features come from either volunteers or GSOC students, and we canât control what they want to work on (only suggest), so we canât plan for those features. We canât plan even for core team features, because the funds are not enough to be so sure of the future to make precise plans. And there are always stuff like âoh no, sip 5 stopped working with our Pythonâ, âQt 6 removes ANGLE, what do we do?â, âgcc 11 is an awesome speed improvement for our smudge brushes, great, oh no, on Windows it actually slows everything downâ, and also stuff like âok so this week weâve got ten people making MRs, whoâs gonna review all of them, and are we sure our GSOC students are doing fine, did anyone check with them recently?â etc.
In any case, Iâm interested in this part:
I wonder, what questionable design philosophies are in CSP? (Iâve never used it, unfortunately).
I donât know what the international version of csp is like. But my domestic csp has a serious problem: if you download resources, you will need to manually move from the warehouse to a suitable location. If you download 100 brushes, you need to move 100 times⌠And, resources such as automatic actions also need to be moved by yourself, donât know why?
Ah, I didnât get the negative response myself, but I saw someone asked about it on the forum years ago. I think the issue was the way it was worded made it sound as if they were not going to add such a feature at all, and itâs kind of the reason I jumped into the fray a year ago as I genuinely still believed that Krita was never going to add such a feature unless I or a volunteer did it, causing me to push for how the software update system works and take a look at the code. I was really happy to see that the OâNeal brothers were working on it though and working on enhancing the animation system to make it more user friendly, being Animators themselves and since they had more experience with working on Krita, and were redoing the foundation, I decided to step back and let them update the foundation before proposing any updates myself (that and I got a little zealous, but I totally understand, I didnât want to seem like a fanboy dictating what should be done.). Though it is easy for a developer to say âNo we are not planning to work on this.â Given Kritaâs open source nature, I still wouldâve liked for an indication that I or someone else could try to build it myself. I can agree that itâs a pain when people expect developers to do a ton of extra work they werenât planning and scope creep being a thing (Iâve been working as a professional software engineer for 6 years and currently work at Microsoft. I know the struggle. ) But having gotten more used to how this forum works, it does seem like any discussions about new features can at least be talked about in concept and theory with developers providing feedback on whatâs already there and how difficult it would be, without promising to do it themselves, but at least open the door for a volunteer to take the mantle. Or just even stating that âWe can do that later, but right now we need to set up the foundation or it would be a huge mess in the code base.â I can totally respect that.
As for the CSP, so I did buy the EX version (on sale, but still spent $100 on itâŚ) and compared it 1:1 with Krita. Despite costing more and being more commercial, there are some features that I do like that Krita doesnât have such as the ability to convert every line drawn into a vector, butâŚEDITING the vector seems to be a design philosophy that goes completely against the whole point of vectors, âprecisionâ. you canât seem to edit the anchors and their angles and you canât use âVector fillâ for some reason. Also when I looked up their animation system, there seemed to be no intention of adding Cycle frames (a system I litearlly only saw in one other software, RoughAnimator, and that was made by some dude as a personal project yet is one of the best animation softwares Iâve ever usedâŚif you donât intend on shading the animation.) The fact that CSP is close source and developed privately means that whether a feature would be added or not is purely up to the developers with no way for artists to step in and add the features and tools they need for thier workflow. This is something Krita does allow and why to this day despite spending hundreds on other software, Krita is still my number 1 go to for art. Adding clone frames has just bumped Krita up to being slightly better than RoughAnimator in terms of 2D animation work flow due to the features it allows on top of being an excellent drawing app. A few UI tweaks and a few more controllers will sky rocket it to being the best on the internet. (I know Iâm using a lot of subjective terms, but I want to be clear I spent the last year testing all of these softwares meticulously, Krita and Blender have the most potential. CSP is a good second place, Most others dontâ even allow you to set up layered frames or they can animate, but cost a fortune to do rigging and meshing, something blender can do for free if you know how).
But above all else, since Krita is open source and made âBy artists for artistsâ, that means Krita is more likely to have tools and workflows that let us artists choose how we want to make our art, and not try to dictate to us how we are supposed to create our art. Unlike Photoshop or CSP which we pay for them to tell us how to paint (itâs pretty unorthodox, though the UI is clean but a lot of functions are hidden behind sliders). So thatâs ultimately what I mean. But reasonably, as a developer, I can agree that itâs up to us artists to figure out how to add these features and tools to the software without disrupting other methods of creation. Iâm totally up for just discussing what is feasible.
Well one way is you can copy and paste frames as âclone framesâ. Those clone frames are exact copies of the previous frames down to the pointer. So if you just repaste those frames, they will repeat. Itâs not exactly âCyclesâ like where you increase a number until you want the cycle to end, but the fact that it copies frames from earlier and when you edit those frames, it edits all other frames that are clones means you can do cycles manually and more flexibly. Thatâs mostly what was needed.
Ahhhh, âClone Keyframesâ on the frame context menu in the Timeline followed by âPaste Keyframesâ. I should have paid more detailed attention to changes in 5.0.0. Thank you
Theyâre mutual clones too. You can edit any one of them and all other clones plus the âoriginalâ will be changed.
Please do consider this a joke, but you know, I could be cheeky and say: you can spend hundreds on Krita too, if you want Krita Development Fund, or even the one-time donation, is very much accessible and waiting for all those hundreds (as I said, joke. There is a reason Krita is free as free beer).
Is that a praise or a critique? Itâs a bit unclear for me if you say that here it shows well how the communication should happen, or that we still have a way to go. I must admit I do catch myself on being a bit too negative and shooting features down just because I have no time to do them and canât even think of adding them to the TODO list
Clone frames were already in development (though on hold) those three years ago⌠I thought Emmet and Eoin actually finished cycles, but it seems they didnât. Itâs a feature that will probably come in the future, then It wonât require cloning and it will show on the timeline how itâs cycled. I remember some video or screenshot⌠you can see it here: WIP: Clone frames and cycles (!157) ¡ Merge requests ¡ Graphics / Krita ¡ GitLab
That is why I made a smaller proposal that would eventually lead to this feature and others.
I think the basis should be made first before doing stuff with animation and it.
Because as it is now animating vector is not very painterly and you might be best with blender to do it that way.
There are of course both advantages and disadvantages of that, but note that quite a lot of new features come from either volunteers or GSOC students, and we canât control what they want to work on (only suggest), so we canât plan for those features. We canât plan even for core team features, because the funds are not enough to be so sure of the future to make precise plans. And there are always stuff like âoh no, sip 5 stopped working with our Pythonâ, âQt 6 removes ANGLE, what do we do?â, âgcc 11 is an awesome speed improvement for our smudge brushes, great, oh no, on Windows it actually slows everything downâ, and also stuff like âok so this week weâve got ten people making MRs, whoâs gonna review all of them, and are we sure our GSOC students are doing fine, did anyone check with them recently?â etc.
) But having gotten more used to how this forum works, it does seem like any discussions about new features can at least be talked about in concept and theory with developers providing feedback on whatâs already there and how difficult it would be, without promising to do it themselves, but at least open the door for a volunteer to take the mantle. Or just even stating that âWe can do that later, but right now we need to set up the foundation or it would be a huge mess in the code base.â I can totally respect that.
Krita Development Fund, or even the one-time donation, is very much accessible and waiting for all those hundreds 
It wonât require cloning and it will show on the timeline how itâs cycled. I remember some video or screenshot⌠you can see it here: