Particle size changes in simulation
#1
Hi,

I am having this issue with my simulations (see attachment vid) where the size of my particles is changing for no apparent reason? At first i thought it was because my mesh operator wasnt at the bottom and set to tri mesh so i have redone the sim. I have just complete this test but half was through the size changes again. I have attached my flow which is pretty simple. What am i missing here?

Would appreciate any help as i need to get this rendering tonight. Sad


Attached Files Thumbnail(s)
   

.mp4   sand.mp4 (Size: 4.46 MB / Downloads: 44)
  Reply
#2
That looks weird indeed.

What happens if you add a scale operator bellow shape?
Does it fix it?
Do you need Particle Bind for this sim? If you turn it of, does sim looks different, and does it fix it?

Ultimately, since you need to render this tonight, you might try this:
- Disable mesh operator (you won't render this flow)
- right click on empty space in the editor, choose preset: simple reference flow
- This will create another TyFlow, with just 2 operators, that are basically just play your first flow.
- Disable Scale in both operators.
- Add scale operator and set it to the size you like.
- add mesh operator and render that second flow.
  Reply
#3
I think it's a moblur thing...

Your particle count is changing and you're rendering as meshes, which means moblur will not be properly computable due to the changing topology (always render as instances when possible).

If this size change is happening around frame 400, it's because that's when the topology stops changing (no more particles are born) and thus the moblur vectors aren't getting messed up.
  Reply
#4
Thanks for reaching out. I wanted to come back for a bit of resolve and to document my solution.

 It turns out it was the Corona material that was causing the issue. It had some pretty weird/complex displacement and bump maps going on. I have no idea it was changing at specific points on the timeline but replacing the material seemed to fix my issue.

Thanks again.
  Reply
#5
Oh weird...well, glad it's solved.
  Reply


Forum Jump: