Post

I totally agree with you. Another issue I have with the existing plugins is that they often break between updates, are difficult to upgrade because of their complexity, or aren’t actively maintained. At the moment, Multimeshes are very limited to code, so the current node is very hard to work with.

Also I forgot to say this, but I’d totally use the tools you showed in the video if this ever becomes a plugin or something else in the future. Loved the simplicity of it.

The repo exists, I made it last year as a quick test, but the implementation is really naive and to optimized at all github.com/gtibo/multim...

I personally just want the current scatter tools gone. I use MultiMesh all over my project and not a single use case involves surface-scatter. I'd rather not have a default assumption at all. It feels oddly specific and biased.

You'd have to collect data about what use cases are most used, and how best to serve those. Instead of just asking "would anyone like this?", which is always guaranteed to get a lot of "yes". My stance though is that it's better to just make different plugins or make the tooling customizable

Just to provide more specifics, I'm currently using MultiMesh for, among other things: - digital glyph rain - fences - seagulls - star dome - paths - leaf clusters They're all manually placed, or follow a special algorithm for placement.

That's why I'm talking about another node (that would use multimesh with the renderingserver for example) but not affect the current multimesh node. The current "issue" with multimesh is that you have to heavily interact with it through code to use it.

The main function of multimesh is to declare transforms with optional color data to display meshes. This functionality should be (imo) accessible in some way other than through code. I 100% agree with you that I should do a survey of some sort to collect different opinions :)

The idea would be to create a node to: - Place items - Modify scattered items (scale, color, etc.) with a falloff option. - Define scattering rules (no overlap between groups) - Split transformations into actual multimeshes (e.g., via an octree to optimize the size of multimesh instances)

Being able to set a random seed has been a consistent feature I've had to add to all my MultiMesh tools. Maybe that would be good to include in the MultiMeshEditor node as a stored property?

At that point I don't have a clear idea in mind But if you're interested I can make an online document that can be edited to write down idea. I was thinking following this process: - Create a survey to gather real use cases - Write a proposal / design document - Get feedbacks

This node would ONLY modify existing properties in multimesh (easily modify transformations and color data, that's all). I know there are scatter plugins, but they offer a lot of features compared to a simple node.

I don't know if many people are in the same situation as me, but would a simple tool like this be useful to you? I'm posting here because I would need some help if someone is interested (mainly on plugin structure and for the octree). I'm more of a graphic design guy that know how to code a bit 🫠