Welcome to od|forum

Register now to gain access to all of our features. Once registered and logged in, you will be able to contribute to this site by submitting your own content or replying to existing content. You'll be able to customize your profile, receive reputation points as a reward for submitting content, while also communicating with other members via your own private inbox, plus much more! This message will be removed once you have signed in.


  • Content count

  • Joined

  • Last visited

  • Days Won


Atom last won the day on August 17

Atom had the most liked content!

Community Reputation

494 Excellent

About Atom

  • Rank
    Grand Master

Personal Information

  • Name
  • Location

Recent Profile Visitors

6,391 profile views
  1. I was working on a scene where I wanted lines to wrap around a geo shape but lie upon the surface. FindShortestPath is used. Here is a basic setup with UV mapping setup as well. Currently it is setup in lightning mode where the lines change per frame. This is accomplished by adding $F to the random seed of the scatter. To make the lines stand still simply remove that expression. ap_connect_points_around_geo_surface.hiplc
  2. I wish I could save an image with my HIP file. Why can't we lock a COP based File node yet?
  3. I wondered about that as well. Personally I don't use .sim files. I think they are just left over from an older time. I just write out the .bgeo. sc sequence from the import node. Did you alter any single setting in the simulation? Perhaps Houdini has detected that the current settings and .sim file settings are different and it simply trying to be accurate?
  4. If you are referring to surfacing, try setting Influence Scale closer to Droplet Scale on the particlefluidsurface1 node in the fluid import node. The default of 3.0 makes fluid a bit thick IMHO. Make sure to read the tool tip, however. Influence Scale must always be larger than Droplet Scale. But you can get them closer than the default.
  5. The pos input must be driven by P. Try connecting N to amp and you might get a varying result on your normal, but why would you want that? There are a lot of tutorials available. Have you tried exploring the Learn menu on the SideFX.com website?
  6. Are any of your files larger than 2GB? A company where I worked at we had the same exact problem. Old school networks may still have that 2GB limit. You may want to simply invest in a 128GB USB 3.0 local drive, or larger, and try writing your sim file to that. Report the network problem to your school's IT department. They may have a solution.
  7. Are you saving to a network drive? Some networks can't save a single disk file larger than 2GB. Try setting your particle separation really high to see if you can make it through the frame range you require. Then, using the voxel estimator, dial in a lower particle separation until you find the breaking point. You may have noticed in the GIF image I posted that the difference between 0.025 and 0.1 it 100,000,000. So the particle separation value is EXTREMELY sensitive.
  8. I recently started adding a single float channel to my FLIPObject. Then I place this expression in the field. It takes into account the size of the fluid box and the particle separation to produce a number that is the approximate voxel count for the fluid area you have defined. ch("../flipsolver1/limit_sizex")/ch("particlesep")*ch("../flipsolver1/limit_sizey")/ch("particlesep")*ch("../flipsolver1/limit_sizez")/ch("particlesep") This will generate a number in scientific notation. Basically +6 at the end mean a million voxels, +7 means 10 million voxels and +8 means 100 million voxels. If you are in the +8 range or higher your computer may not have enough memory for your simulation, so try a higher value for particle separation to lower voxel count. Can you place your system specs in your signature? Also, you must leave OpenCL checkbox off to do any hi-resolution work in pyro or flip. flip_voxel_estimator.hiplc
  9. When I am working with FLIP I never save .sim files. The best work flow suggest to me was to save out the compressed cache first as .bgeo.sc. Then save out the surface cache second. So it is a two step process. The fluidimport node is where you will find these two surfacing support nodes. The FileCache are put in place by the shelf tools so if you used those for your initial setup then they should already be in place. The reason you want to save out the compressed cache first is so you can have a bit more interactivity when dialing in the surface characteristic on fluid surface node. If you don't have a compressed cache you will find that every time you move to a new frame, you will have to sim again. Also whitewater references the compressed cache node for it's surface so once you have that cache written to disk bothe fluid surface and white water will respond much quicker. Also it allows you to sim at a little higher resolution because you don't have to hold both caches in memory at once.
  10. I came across this HIP file posted on Discord and thought I'd re-post it here. It is a way to deform one object based upon the proximity of another without a solver or a simulation. If you are the author of this HIP file feel free to chime in and take credit! CheapAssCollisionDeform.hipnc
  11. You are out of disk space?
  12. That looks a lot like Matt's example file. You can review it here. http://www.tokeru.com/cgwiki/index.php?title=HoudiniDops#Pop_advect_by_volume
  13. The simplest way is just to model the "spout" and set it as a collision object. Then the fluid will be forced into the shape of the pouring vessel.
  14. The old AMD chips work fine. The new Ryzen works fine. I don't encounter any problem with AMD CPU processors.
  15. Ok, here is my rework of your scene, not sure if it is right. Here is what I changed. Inside the popnet you want to keep an eye on how the merges work. Some merges work based upon left input affecting right input and others work based upon a mutual relationship. In shelf tool configurations, static is always on the left of what it is colliding with. Then RBD is mutual. I re-wired the nodes to reflect this. I also added gravity to the scene. Once you have a popnet output with colliders, you have to make sure you don't import the collision geometry as the popnet result. So alter your ObjectMerge on the popnet to remove the ground plane and the bench like so. This will give you only the particles as the result. * ^rbdobject1 ^groundplane1 ap_Hand2.hiplc