Animated particle emitter refuses transforms within simulation range
Open, ConfirmedPublic

Description

System Information
Win8.1x64, 3xgtx580

Blender Version
2.79

Short description of error
An animated object with a particle system, for frames within simulation range, jumps back to it's animated location on confirmation on hotkey or widget transforms, as if the frame was changed, when it was not.
Temporarily setting particle system Subframes parameter to 0 seems to be a workaround, eliminating the behavior.

Exact steps for others to reproduce the error

  • (G)grab move the emitter
  • The emitter jumps back on confirmation

Details

Type
Bug
ronan ducluzeau (zeauro) triaged this task as Confirmed priority.EditedSat, Nov 4, 12:09 PM

There are lots of refreshing issues with particles.
Many modifications of emitter can conduct to an invalid cache. Color of cached frames in timeline becomes brighter when cache is invalid.
Calculate to Frame button is here to correct cache when it happens.

So, particles creator anticipated this kind of problems. But I am not sure how people who could have modified subframes will appreciate this.

Particle cache, at least from how it looks, isn't exactly relevant here as the main problem is with inability to move the animated objects themselves.
However if the issue is tedious and as some claim the particle system will be rewritten in observable future, the workaround of temporarily disabling subframes could be acceptable in most cases.

Sorry for my lack of attention. I don't want to confuse people. It is definitely unexpected behaviour.

Concern about subframes is relatively recent.
At creation of particles system for 2.5, default cache step was set to 10 in order to have a fast feedback to animate objects.

And basically, subframes are described in manual only as a factor of quality improvement.
So, proobably most of blender users don't use them or modify emitter animation with subframes ON.

But they do not just allow to avoïd some unwanted explosions, they are correcting trajectories of particles which can result in totally new path of particles flow.
It is a legitimate request to be able to move emitters with subframes ON.

It has been taken into account in 2.8 proposals about physics as a nodal system. But maybe problem could be solved for a future 2.79a.