I have 2 identical collections of rigid bodies with identical settings. They are on different collision collections yet they are acting different when dropped from identical location. I need for them to be simulated in the exact same way as the previous collection. Why is this happening?

EDIT: The effect I want to achieve is a smoke simulation from shards of a cell fractured object and later hide them and replace with identical shards that liquify: enter image description here

Unfortunately the rigid bodies do not behave the same way when duplicated. I did however found a workaround by creating 2 blend files - one for smoke sim and other for liquid sim. Tho in the future I might look into baking physics as some of you pointed out.

  • $\begingroup$ Maybe edit your question to provide additional details of your simulation - images really help. Also, explain why you need them to behave the same - there could be ways of copying the animation once it’s simulated rather than re-running as a separate simulation - if you need it to behave identically. How are your current simulations differing? Are there just slight differences or are they completely different? Again, images (animated gifs) would be useful to get your explanation across. $\endgroup$ Oct 18, 2020 at 22:34

1 Answer 1


In general, rigid body simulations are not perfectly deterministic. The more complex they are, the more influence the small but unavoidable numerical imprecisions have. You might be able to duplicate simple simulations somewhat reliably, but increasing complexity will mean increasing difference in results between simulation runs.

Therefore, if you want perfectly identical behavior, you will need to bake the dynamics (quite possibly it would be best to bake to keyframes in this circumstances) before duplicating the bodies.


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .