Hair System Overhaul Roadmap

Lukas Tonne has posted to the Code blog about the plans for the hair system project for Gooseberry. I’ve already put my two bits in the comments (PLEASE just copy as much workflow paradigm as possible from Yeti!). It’s an interesting read at least, for a feature that has never really been usable in production in Blender.

This is incredibly exciting! I just hope Lukas can manage to implement it. A strong, production capable hair solver along with artist-controlable tools doesn’t sound like something one can knock out in a couple of weeks. :slight_smile: But I remain hopeful!!

The commit logs show that Lukas has already poured a ton of work into this new system, so he knows exactly what he is doing and I wouldn’t doubt his experience with complex systems such as this.

He’s the type of developer that can and will get things done, it’s no wonder that he also has depsgraph refactoring and the particle system overhaul on his plate as well.

Yeah, knowing that it’s Lukas on the case makes me very hopeful about the project. Can’t wait!

Well, I was pretty happy with the hair as it was. But I only ever do still renders…

Very cool news.

Yeah, for stills Blender’s grooming tools and cycles rendering are awesome. As soon as you want that hair to animate though… UGH! After Gooseberry, we just might have ourselves a top-level hair and fur pipeline for production!

So what’s Yeti’s workflow? Have heard a lot of people clamouring for it in words but no actual examples of what makes it so good to use.

after watching youtube videos of yeti workflow , I definitely prefer Blender. I ’ m sorry but I don’t get what is so special about yeti. Hair Farm has a different approach , but am I missing something about yeti ?

Have the possibility to use curves as guide and/or a map for the clump will be really good !

Think of yeti like Cycles nodes for hair. Any parameter can be plugged in to control any other parameter. Child groups can drive other child groups, mesh curvature can be used to automatically drive the direction of hairs. It’s limitless control from a handful of simple nodes. Blenders hair tools are a joke in comparison.

Yup, I doubt that we will ever reach Yeti’s level (unless Lukas has 10 secret twins that are all as good as him and want to work on Blender) but this is already a great improvement :slight_smile:

This sounds great!

Really pleased to see that the proposed system allows animator driven control and tweaking as well as simulation… I can hardly wait!

Lukas does a lot of work with the nodal system of Blender so I wonder if this is the approach he will take considering he was been experimenting with particle nodes for some time now. This is just the pilot they are working on so the could a chance to push things further if the community gets behind the feature length movie.

I asked this very question on the blog and he said yes nodal approach is a goal eventually, secondary goal, first he needs a robust system as its core which is higher priority

We would like to “nodify” the hair system eventually. There are quite a few effects that can be combined, and nodes have the advantage of clarity because the user explicitly adds individual effects, instead of using a large one-for-all system and disabling parts of it. However, getting the solver stabilized, completing physical effects and improving the editing tools has higher priority.

In general I think nodal animation tools would be pretty cool. So, this will be interesting to see.

Awesome,but this will be merged to master or will remain only a branch?

Possibly related question: on the buildbot there is now a test build version. Is that the gooseberry branch? Or something else?

@@PyC0de it shall be merged to master, there is no point in keeping it in a dev branch forever :wink:

@@fahr), etc… So it’s not just Gooseberry related.