Help obi fluid vibrate problem

[busraasevimm]

hi, I have a vibration problem like in the video while using obi fluid. 

https://youtube.com/shorts/4kJfk_2vbr0?feature=share

[josemendez]

hi, I have a vibration problem like in the video while using obi fluid. 

https://youtube.com/shorts/4kJfk_2vbr0?feature=share

Hi,

Could you give some info about your setup?:

- Backend being used (Oni or Burst).
- Solver settings.
- Updater settings (type of updater: FixedUpdate, LateUpdate, etc. Amount of substeps.)
- Kind of collider being used for the container (MeshCollider, Distance field, group of primitive colliders, etc).

[busraasevimm]

    josemendez pid='12629' dateline='1654072597']
Hi,

Could you give some info about your setup?:

- Backend being used (Oni or Burst).
- Solver settings.
- Updater settings (type of updater: FixedUpdate, LateUpdate, etc. Amount of substeps.)
- Kind of collider being used for the container (MeshCollider, Distance field, group of primitive colliders, etc).

solver and updater settings are in pictures. the glass has mesh collider, but vibrate continues when fluid is poured to an object has box collider. I have one solver.

[josemendez]

solver and updater settings are in pictures. the glass has mesh collider, but vibrate continues when fluid is poured to an object has box collider. I have one solver.

Changing collision detection to parallel instead of sequential might help. Sequential constraint evaluation might result in jittering, as explained in the manual (see Evaluation Mode at the very bottom of the page):
http://obi.virtualmethodstudio.com/manua...olver.html

Keep an eye on your fluid's viscosity value too. Large viscosity values might also result in simulation instability (jitter), if this is you case you might want to use a smaller fixed timestep: http://obi.virtualmethodstudio.com/manua...rials.html


As a side note, if you're developing for mobile devices using the Burst backend will result in better performance.

[busraasevimm]

Changing collision detection to parallel instead of sequential might help. Sequential constraint evaluation might result in jittering, as explained in the manual (see Evaluation Mode at the very bottom of the page):
http://obi.virtualmethodstudio.com/manua...olver.html

Keep an eye on your fluid's viscosity value too. Large viscosity values might also result in simulation instability (jitter), if this is you case you might want to use a smaller fixed timestep: http://obi.virtualmethodstudio.com/manua...rials.html


As a side note, if you're developing for mobile devices using the Burst backend will result in better performance.

I tried both but not working. the glass has "rolling friction" collision material, I changed it to low friction, and it is fixed. but when fluid is poured, jitter starts again. then I rotate object has box collider, the jitter is disappeared.

thanks for note   

[josemendez]

I tried both but not working. the glass has "rolling friction" collision material, I changed it to low friction, and it is fixed.

thanks for note   

Why would you enable rolling friction on a fluid simulation? that's usually only useful for granulars. Unless you're using granulars in your simulation too, imho you should disable it.

then I rotate object has box collider, the jitter is disappeared.

Not too sure why would rotating the collider would fix this, though. I'm not sure I understand your setup. If problems persist, could you send the project to support(at)virtualmethodstudio.com so that I can take a look at it?

thanks!

[busraasevimm]

Why would you enable rolling friction on a fluid simulation? that's usually only useful for granulars. Unless you're using granulars in your simulation too, imho you should disable it.


Not too sure why would rotating the collider would fix this, though. I'm not sure I understand your setup. If problems persist, could you send the project to support(at)virtualmethodstudio.com so that I can take a look at it?

thanks!

cause rolling friction looks better than the others. but it is not very important, it can change.

the thickness value of object has box collider was 0.1, I changed it to 0.03 and the jitter problem solved.

[josemendez]

cause rolling friction looks better than the others. but it is not very important, it can change.

Rolling friction has no effect on fluids at all. It affects the particle's angular velocity on contact, but fluid particles have no angular velocity. Only granulars do. This is explained in the manual:
http://obi.virtualmethodstudio.com/manua...rials.html

Using rolling contacts only makes sense for actors that make use of particle orientations: ObiEmitter (when emitting granulars) and ObiRod. They are incompatible with ObiSoftbody, and will have no visible effect when used with ObiCloth and ObiRope.

You can leave rolling friction on, but it will just burn performance for no good reason. Also note that you don't have to use one of the sample collision material assets included, you can create your own and change any of its parameters. I mention this because you say "it looks better than the others", which makes me suspect you choose a material among the included ones only.

the thickness value of object has box collider was 0.1, I changed it to 0.03 and the jitter problem solved.

Yes, that makes sense. A thickness of 0.1 (which is quite large!) means particles will collide with the collider when they're at a distance of 0.1 units or less from its surface. This can "sandwhich" particles in between colliders, leading to jitter.

Glad you found the cause! let me know if I can be of any help. cheers,