Let’s attract the attention of developers and let them know that we need this tool inside the editor!
Are you all really comfortable doing animation with code?
1 Like
Hi,
For gui animations, there is the great Panthera editor/extension.
You can also animate gameobject, but I didnt use it much for that, but it more to animate game object together, not to do sprite animation.
For sprite animation, there is the great SpriteLoop editor that @Sam is working on it (search forum)…
You didnt precily tell us what kind of “animation” you were thinking of?
But if you want new feature in editor, you should open an issue in the github defold project, it will be more efficient than mysterious sentences in the forum
4 Likes
I’ve been summoned! 
I would say “super profesional” games use tools like Spine as it’s not common to ask your artist to animate in your game engine. There are workflows to animate both 2D and 3D stuff in Blender. Here we have extensions like Panthera , I’m building my own. There is also Rive extension for UI and general animations (Duolingo uses Rive). You can always animate your characters with Sprites/Images in any drawing software and import them as Tilesource or Atlas animations. Not every project requires animation timeline. 
2 Likes
I mean Animation Timeline as in Unity, so that it is not a third-party application because it is not convenient.
1 Like
What if I just want to quickly move, rotate and scale a sprite? Do I need to go to another program for this? This is very inconvenient! It’s just that you’re all programmers and don’t understand how convenient it is to animate simple things using Timeline. Artists will understand.
1 Like
Yes. For this kind of stuff, there is Panthera. It is an external program, but it works perfect with Defold. It would be cool, to have an integrated timeline animator.. but I think there a too much external tools, who do the job perfect. Panthera 2.0 Editor by Insality
3 Likes
There is a Github issue about it. You can ping the team there as they removed the Feature Requests category from the forum on purpose.
I can only speculate that there are more important features and tasks at hand and not so much demand for this.
2 Likes
The tone of this thread is a little harsh, let’s keep it friendly, people!
I agree a built-in timeline editor would be a timesaver, and would allow quicker iterations when it comes to GUI or go animations. Let’s hope it gets added. Would be good to see if the “team” has considered this.
The mentioned tools are all great to use in the meantime.
7 Likes
Not sure if you guys have seen a recent addition we did on GitHub where we attach a priority (1-5) and size (1-5) to all issues. The “Timeline animation editor” issue has been assigned a priority of 5 (highest). We haven’t estimated the size yet, but it’s a fairly large task. So yes, it’s on our radar, but it is early days and we need to discuss and design it before we can even start with the implementation. Also, the question is what the minimal product would be?
10 Likes
Moving, rotating, scaling and transparency of a game objects and groups. This is enough to get started.
2 Likes
Sorry, I haven’t noticed it. The priority field is not visible in the Github mobile app. I can see it on the web now. 
I guess there will need to be a new resource type for animations, similar to an atlas, but dedicated to animated properties, keyframes, and animation names. Something like an animation collection or animation atlas.
For the MVP, I imagine a timeline editor with multiple timelines, one for each animatable property.
Core functionality should probably include:
- Create/delete animations
- Set animation length (max frames) and FPS
- Add/remove keyframes
- Drag keyframes on the timeline
- Basic interpolation/tweening between keys
For the initial version, I would skip more advanced timeline interactions such as:
- Multi-select and bulk editing across keys/timelines
- Timeline zoom in/out
- Curve editor / custom easing graphs (or plug-in the exiting curve editor if posible)
2 Likes
That’s very possible! Because guys are working on such an extensibility to allow to create custom data components as the final vision of this direction. Light components pave the way here.
5 Likes
Huh, that’s annoying… their mobile app is not that great tbh.
Thanks!
3 Likes
That’s a very good point, because I think that what is really needed is a way to animate any data in space and time. then having an API to access this data component and play it, or wire it to our code.
It is good to have a cutout animation editor, but you quickly need use cases beyond “a simple animation of sprites”.
In a full game, you quickly need
- sprite animations with data anchors in animations (to parent game objects, send events etc..), that’s the basic
- object complexe animations: like path following animations for enemies, walk wandering behaviours etc.
- multi-objects orchestration: think full cutscenes with camera movements or GUI animations
So you need data visually but also any data through time, but you need an API to use it in code.
Because once a game starts to be big or when you have complex objects, you will do most thing through code, so the “sprite animation” use case is the least of your problem.
So I think the strategy should go first as defining a generic “data animation component” with an API to use it.
Then add a visual editor to display this data in the editor, eventually automatically wire it to sprites
2 Likes
And just to go beyond animation, but about Editor improvements, I think that if many external tools are popping everywhere in the community like Panthera, SpriteLoops, the physics polygon editor example etc.
It is because we cant extend the editor ourself through Lua. Editor script have their limits.
If we had an editor api to create panels, rendering views, UI components or tools (like Blender can be extended), and if we could do it efficiently in Lua, all theses tools wont even exists and we already had all of them as integrated Defold editor extensions.
As a longrun Defold strategy, and as Defold team has limited ressources, I think the first priority should be to give its community the power to build Editor extensions in Lua.
You do that, and you’ll have an army of developers right arround the corner
5 Likes