Jump to content

Atom

Members
  • Content count

    3,396
  • Donations

    0.00 CAD 
  • Joined

  • Last visited

  • Days Won

    116

Everything posted by Atom

  1. I think your premise is that the sphere is solid red all the way through, when in reality it is only red on those thin polygon surfaces. You have to help out the attribute transfer node by filling the sphere with red points.
  2. vdb import into maya looks weird

    That looks right to me. The grey you are seeing is the temperature, which is typically everywhere. Density is often used as a mask.
  3. Clusters along a path might be one approach you could take.
  4. Sticky Particles

    Yes, unpack the Alembic so you can supply the geometry as a collision object to the simulation. Both flip and pop can stick to a collider once a collision has occurred. Take a look at the solvers for each.
  5. #2.) Supplying geometry to the SOP path gives you a one-time injection of particles in that shape. (on the creation frame). With the volume source node, you can "source" inject particles every frame. So, you could emulate the SOP path using the volume source and only activate it on frame #1. The volume source is for continuous activation.
  6. Increase flip vel on static collider

    collisionvelscale is the parameter on the solver. Don't pass that, pass v@v on the surface of the collision object. Without the v@v attribute present, the collisionvelscale seems to do nothing. Once you add the attribute to the collision points, it begins to work. The first GIF I set collisionvelscale to 10.0 This one I set it to 0.1.
  7. scatter points on grid plane

    Try setting your scatter to one count per primitive and set your grid to a 2x2 so it is only made up of a single primitive.
  8. Increase flip vel on static collider

    Try adding point normals to your static geometry, then assign those normals to v@v. v@v = @N; This will provide something other than zero for the solver to multiply against. ap_flip_static_collider_velocity.hiplc
  9. Having lived with this for years (for me since Houdini 12.5), I have recently begun to wonder if it is a viewport bug at all? It is definitely related to the display flag. I wonder if the geo/dop(guide) object gets stuck in memory, or the flag toggle gets inverted? Most of the times, deleting the entire scene view and creating a new one will fix it, but I have seen the bug persist even after trying that solution. Restart always seems to fix it until it appears again. I am less inclined to think it is drivers because I have experienced the same type of viewport error on different machines running different operating systems using different graphics hardware.
  10. It seems to work for me. I can see vorticity on the final surface even without unpacking. Perhaps you have cached something along the way without vorticity? Try removing all Load From Disk check boxes and sim a few frames..?
  11. Rewind and recache? It seems to work when I add it to the attached simulation. ap_flip_viscosity_by_attribute.hiplc Here is a microsolver version here.
  12. Flipbook Export error

    Do you have quicktime pro installed? You may need that. I use quicktime (animation codec) for all my output on windows so I can confirm that it does work.
  13. Karma Noob Issues

    Karma does work with the Indie license, but you may need to recover the render token. Here is an example scene that produces output on my machine. If you can't get it to work, contact support. ap_tyler_bay_shaderbot_v001.hiplc
  14. Pyro sim with moving emitter

    You could try increasing the substeps on the solver, or blurring the initial volume source.
  15. I'm trying to leverage continuous pyro clusters for agent missile launches. The dim pink boxes are my cluster areas. The white boxes I assume are active voxels within the cluster. How do I get voxels with no density to auto-deactivate? I think the cluster is designed to deactivate and release its memory if there is no density. Is that accomplished through DOP masking?
  16. [SOLVED]Remove Continuous Pyro Cluster?

    No, sparse doesn't seem to work in this scenario, which is agents firing in various direction. I pursued that as the first approach, but sparse just makes and increasingly larger area while maintaining divsize. This causes slowdown and memory use which grind the computer to a halt. I think I've figured it out. You have to leverage fluidresize if you are using Continuous clustering. This causes each cluster to collapse after density is no longer present.
  17. OpenGL renderer - line thickness

    It worked for me. That's pretty cool. I didn't realize Wren could render to .eps files.
  18. PolyFrame change curve point normals

    This video has an explanation of polyframe.
  19. possible vellum glitch?

    Yep, send that to support. You have found a bug.
  20. How to emit Particles at Geometry Contact ?

    Maybe these links can help. https://www.sidefx.com/forum/topic/67731/?page=1#post-287800
  21. Few nodes won't work on OpenCL

    Try switching your OpenGL from GPU to CPU under your Houdini preferences. You may be running out of GPU memory.
  22. Flipbook Motion vectors stuttering (ue4)

    That does look like an FPS mis-match. You might want to try kicking out your explosion at 30fps. I faced something similar when I tried to supply a 24fps point cache as a Niagara source. I was told Niagara runs at 30fps. Try searching under Unreal project settings. You might be be able to adjust the engine playback speed and avoid reexporting your explosion.
  23. Python: eval string param as "raw"

    Yeah, just check the docs. parm.rawValue() https://www.sidefx.com/docs/houdini/hom/hou/Parm.html
  24. POP Solver substeps issue...

    Don't substep the DOPNetwork, substep the PopSolver.
  25. Redshift...euh why are you...

    One thing you could try is to export all (or most complex) of your geometry as a .rs file, if you have not already. Then Redshift will read the geometry from the disk and not have to request the geometry through the Houdini API, which might be a single threaded process IDK. This API request might be part of the slowdown.
×