Particle edit doesn't work on a Baked, dynamic hair simulation
Open, Needs TriagePublic

Description

System Information
Windows 10 Pro 64bit, GTX 1070 Ti

Blender Version
Broken: 2.79 (5bd8ac9)

The Blender 2.79 manual says you can edit a baked simulation (https://docs.blender.org/manual/en/dev/physics/particles/mode.html#bake-the-simulation).
Does this not apply to a hair simulation?
If it did, while being somewhat of a manual process, it would give massive control over hair and the ability to reshape and correct errors based on the guide hairs during the animation.

Exact steps for others to reproduce the error

  1. Apply a particle system to a basic cube, change type to hair and set emission low, say 50
  2. Turn on Hair Dynamics
  3. Under Cache, Bake the simulation
  4. Switch to Particle Edit mode, all the hairs will freak out and stay static while moving through the frames

Details

Type
Bug

The manual is not clear but the ability theorically exists.

When you enter into particle edit mode, you don't see hair guides. You see all locations that keys of hair paths will take during the whole animation.
It is the reason why changing to next frame have no impact. Next frame is already displayed.

So, when you edit what you see in particle edit mode, it has an impact on successive frames.

There is just a "fade time" slider to reduce number of frames displayed. Set it to one, you see only current frame.
It is supposed to help you to smooth transitions before and after edited frame.

The problem is that restriction does not work for tools. So, if you try to select something, you will select things on previous and following frames.
In practice, particle edit mode is unusable to edit baked simulation.
A workaround can be to use a lattice deformer after particle system.

It is a design issue. Hair simulation basics were not satisfying and stable during an enough long period to have people working on its edition.
The edition of baked caches was thought with softbodies or cloth caches in mind which are corresponding to surfaces. But volumes like hair are too complex for this approach.

Thanks for the info Ronan. I'll try some testing with the lattice deformer, to see if that gives enough control.