Unwanted ground collision
#26
Still not working very well. I started this project almost a year ago, but my simulation is still looking like this. Client is not happy with me. I'm under pressure to get something to put on my bill.


Attached Files Thumbnail(s)
   
  Reply
#27
It's difficult to tell what you're trying to achieve from that screenshot...
  Reply
#28
Is there any ideas what to do to get some interaction with the particles and rest of the scene? Fluid source system works fine, but the PhysX does not in the very same scene. On the previous screenshot the PhysX version was used.


Attached Files Thumbnail(s)
       
  Reply
#29
Try to setup a scenario that acts like your customers NDA machine in a simple way. Make a simple PHX simulation (short and lowres). Make a simple tyFlow sim (also short and lowres). Try what you want to achieve with that test sims. So you can work more easily without having to deal with massive data. So you can easier find out what is affecting each other and tweak parameters.

Another good method to affect particles by a fluid sim is when you export your PHX sim as PRT and then load that PRT with tyFlow and use the "particle force" operator to drive your current tyFlow sim with the particles from PHX.
  Reply
#30
Thanks

I used a setup scene.
https://vimeo.com/683703290

It works there fine, but it does not where I want it to.
  Reply
#31
Is there any development on this issue? I just can't get the simulation through in a way that's acceptable. I have already spent a lot of my savings to finish the project.


Attached Files Thumbnail(s)
   
  Reply
#32
Hmm well you haven’t really explained what the problem is in a way that I can understand.
  Reply
#33
I think the simulation is constantly crashing due to RAM limitations. At the moment I have to work with laptop having 32 GB of RAM. It's not easy to buy hardware in this periphery. I have tried since January, but so far I haven't got any quotations from computer sales people.

Anyway I tried to create several caches as a workaround. I thought, changing the feed value would do the trick, but it seems not to. Both systems are identical as seen on attachment. What is the feed value supposed to do anyway?


Attached Files Thumbnail(s)
       
  Reply
#34
The seed value affects the variation parameters that are greyed-out because none of the grid thresholds are enabled.

Have you tried simulating with RAM cache disabled? I don't see enough particles in your screenshot to eat up 32 gigs of RAM, but if you're simming a long frame range with RAM caching enabled, RAM can get eaten up over time. You can disable RAM caching from the grid right click menu.
  Reply
#35
Thanks,

Enabling those channels didn't change anything. I would like to see the seed value affecting in which voxel each particle is born. When that's possible I would get more dense particle system if I create more cache objects.

Switching cache off, will make the simulation much slower.
  Reply
#36
There's attached two screenshots of an identical tyFlow systems. The other is working perfectly, while the other is not working at all. Where I should start looking to fix the other system. As we see. it totally ignores PhysX elements.


Attached Files Thumbnail(s)
       
  Reply


Forum Jump: