5
$\begingroup$

In the following figures we show the results of our attempts in order to mix a key-framed animation with a fluid simulation. As you can see, in the top one the obstacle is static (fix) but in the bottom one it turns counterclockwise slowly.

The Static one took only 5 minutes to be baked, however, the dynamic one took 65 minutes on the same computer system!?

We need to know:
1. First, whether this massive time difference is usual even though there is no significant difference between the two simulation?
2. Is there any tip / trick to speed-up the baking of mixed animation/simulations?

Our final simulation (for education purpose, btw) will be much more complex in terms of dynamic and the number of objects in it.

static object

dynamic object

$\endgroup$
4
  • $\begingroup$ Did you enable Export Animated Mesh? $\endgroup$
    – gandalf3
    Commented Dec 22, 2013 at 22:24
  • $\begingroup$ @gandalf3 No we didn't. Should we? For the two simulation everything is exactly the same except in the dynamic one we have used key-frames to rotate the obstacle. $\endgroup$
    – Developer
    Commented Dec 23, 2013 at 2:51
  • $\begingroup$ @Developer Not need do it. Thats is for calculate animations with parents, objects with armatures and so on. $\endgroup$
    – user2266
    Commented Jan 28, 2014 at 1:37
  • $\begingroup$ @Developer I was wondering because enabling it will drastically slow down the simulation, so I thought it was a possible cause. $\endgroup$
    – gandalf3
    Commented Jan 28, 2014 at 3:14

1 Answer 1

1
$\begingroup$
  1. Yes. The liquid system must translate one moving object to the domain matrix, making it process for every frame, and calculate the new differences and forces produced by the push of the object over the liquid. Of course, this requires more time. I guess if the obstacle is really complex (more faces), this also requires more time.

  2. If you can use lower resolution, less to time to bake. Use lower resolution for tests and increase the resolution for the final product. Take a look at my answer on your other question.

$\endgroup$

You must log in to answer this question.

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