Bake results out of sequence

[robcat2075]

I have a scene with particles that, when baked, has the particles apparently jumping back and forth from where they should be in time.

 

For example "falling" particles might go down, down, up, down, down, up... instead of down, down, down, down...

 

Is there a cause?

[robcat2075]

Here's a sample PRJ. Can you bake the first two seconds and get expected results?

 

partilclebakingtest.prj

[jason1025]

Ill test this later.

 

 

My guess is that the project was not baked. I have had this issue happen when particles are not baked in conjunction with using Net render.

[robcat2075]

I've tried it twice. Unbaked it progresses fine. Baked it moves oddly.

[NancyGormezano]

I've not fooled with fluid material particles much. However I tested for a chor length =24, deleted all your dynamic settings in the instance, and instead, changed the fluid material settings to be the same as your initial value instance settings (except left cull, recursive solving = OFF), and I was able to get it to behave more properly. I usually find that whenever particles seem to be misbehaving (hair and other things) that I have better luck if I change the properties in the original material, rather than the instance.

 

However, even tho I could get the fluid to behave better, it appeared that rendering (1 pass) does not use the baked values, and would recompute, as the render time would go up in "computing fluid systems"

particle24baked0.mov

[jason1025]

Ok 15.J works fine baking this project. 16.A has the problem you describe.

 

I recommend an error report.

[jason1025]

work around. Bake in 15.J than render in 16.a

[NancyGormezano]

work around. Bake in 15.J than render in 16.a

 

I found that 15j would use the baked results when rendering (render time = 1-2 sec/pass for all frames), 16 does not seem to use them (render time increases from 1 sec to 9sec/pass).

[robcat2075]

I found that 15j would use the baked results when rendering (render time = 1-2 sec/pass for all frames), 16 does not seem to use them (render time increases from 1 sec to 9sec/pass).

 

this is another thing I was wondering about. Even when the number of particles isn't increasing the render time seems to keep going up and up.

[robcat2075]

And... the actual baking files seem to be... 60x bigger... in v16 than in v15? Something is amiss...

[robcat2075]

It is now "assigned" so Steffen is on it.

 

How did we not catch this earlier? Are there any other show-stoppers out there?