Jump to content


  • Content count

  • Donations

    0.00 CAD 
  • Joined

  • Last visited

Community Reputation

7 Neutral

About kfu

  • Rank

Personal Information

  • Name
  • Location
    New York
  • Interests

Recent Profile Visitors

653 profile views
  1. Delete volume attribute

    you can set it to zero using volume wrangle,say v@vel.z=0; for example if you want points stop moving at that dimension
  2. I guess that was from the 'progressive' settings when I render the gif out from ae media encoder. Never heard of this term before, nice to know!
  3. lol, yeah, pop axis force. Why not? Are people tend to more technical or using filed means you are not trying hard enough?
  4. hey, just a quick update that I made the memory usage down to 30GB and now renders fine, pretty quick too for the main pass.
  5. yeah, I will also do a subdivided fit based on size to reduce the inside face poly counts, that would help too.
  6. XD yeah, kind heavy to work with, the debris will assemble into one rock, so they are all shattered geometry from the big rock. I had other orbiting geos that was instanced. I think maybe I subdivided too much for the inside faces. Thanks for the help!
  7. hey, I send you a message with file attached, hopefully you got it
  8. didn't use any displacement, might need double check with the packing, dose packed fragments offer more benefits when shattering one geometry compare to packed geo
  9. thanks for all the tips, haven't tried any other render yet, from what I understand maybe because of the last one you mentioned(crazypants large scene bounding box), do you mind take a look of my file, might need to cache some geo on your side.
  10. I had a scene with 5507 packed fragments(shouldn't be considered as heavy scene,right?), when I render it out using mantra, it took me 92GBs memory. When I include some other geo inside the scene, it won't even render out Render settings are default 3x3 with all limit set to 0. Geo material are pre-defined custom shader packed inside. So I need to turn on save all material when rendering, don't know if this eats up a lot memory? Btw, is there any difference to load the packed fragments as packed disk primitives? From my test the ram usage is the same. Can anyone help me with the issue, I can send u the file if necessary. Much appreciated!
  11. For each - randome point group.

    Here group.hipnc
  12. For each - randome point group.

    you can always add iteration number to the seed value as a rand factor. By the way I find this method still not so random base on the same forloop order, I will see what I can come up with
  13. For each - randome point group.

    Here, I may complicate things using vex, but this works fine. group.hipnc
  14. it should be a vector attribute, basically a initial position attribute stored in points
  15. Glue Constraint Problems

    Rigid body solver node is fine, it contains two types of solver in the node. But usually we use bullet for most cases so I just choose to use the bullet solver node. I never use the RBD FracturedObject, assume it the same with little difference,you can check Houdini document for it.