Jump to content

All Activity

This stream auto-updates     

  1. Past hour
  2. Stiffness, vellum vs wire solver

    Thanks, had my hopes up for vellum aswell, since this was pre 17. But never made it work out as i wanted. Also tried pop-grains without any luck.
  3. Today
  4. Clouds

    Thank you, glad you like it. I will try to post the animations as are visually more effective than these captured frames.
  5. dynamic goal position for flips - how?

    Another quick tweak - after a secondary delay, you can set the particles to 'ballistic' (after turning on Force Override on the particle motion tab of the solver), so that they'll no longer collide with each other. That seems to get the final few stragglers back into position relatively smoothly. flipreset_ballistictoggle.hiplc
  6. dynamic goal position for flips - how?

    Here's one approach. I store initial position with one pop wrangle (if I were setting up from scratch, I'd do this outside DOPs), then switch on another pop wrangle at an arbitrary frame. In the second wrangle, I just add a force to each particle in the direction towards its initial position. To keep them from moving forever, I just test against distance and if it's 'close enough,' move it to the initial position and set 'stopped' to 1 so it stops simulating. Note, it's not perfect because some particles get blocked from their original position by the already settled particles. You could potentially sort this by directly pushing position, maybe after an additional delay. But hopefully this gets you started. An additional, critical detail is to turn off reseeding! Particles generated mid-sim to fill in density gaps aren't going to have well defined initial positions, not to mention even initially-existing particles may be culled. flipreset.hiplc
  7. Hey magicians, I'm trying to figure out the best approach to make something like this, idea is having some lines that came from back and try to catch the airplane flowing around it with a nice flowy motion: So far I tried having some pop emitter at the back, and using attractor / collider to get the particles flow in that direction, with no luck. I also did some manual curves to use with curveforce but i'm trying to make it more procedural, avoiding manual curves. I also played a bit with find shortest path, wich is the closest, but wanted to make something in pops to get more flowy movement. Here are some frames of what I get with shortest path: Any thoughts or ideas will be super helpful, feeling stuck here. Thanks!
  8. Getting Creative with Vex and Python

    I think there is no magic behind this, you just need tons of practice, repetition usually works when you learn new stuff. If you are not able to solve any simple task by yourself, it means that you are not ready yet, repeat more and more tutorials, the are plenty of them now. ABC: always be coding!
  9. Yesterday
  10. VEX infection

    oh....is that all...never knew that. Thanks.
  11. VEX infection

    ah, it was about the mountain node, in that case just add Normal SOP in Point mode before Scatter SOP so that your points have N attribute that Mountain can use
  12. Hej guys - I am looking for a way, to add a goal position for flips after a specific time. What I basically want: the fluid is "falling down" ......after 50 frames.....flow back to the initial state ( but dynamically - each point exactly to its original/initial start position) I know this is well possible with grains via these attributes: targetP , targetweight ,targetstiffness With "popsteer" and "popattract" i could not got is work, because the particles never really stopped moving smoothly. I attached the file:scene.hiplcscene.hiplc Some an idea? Best jack.
  13. VEX infection

    it's because mountain doesn't work on a set of points. try this simple test, Grid, mountain, works perfectly fine. Now scatter points on the grid and feed that into same mountain....mountain refuses to work. here's one way to get around it, 'surface' it first with tri2D..then add to convert them all back to points....I'm just guessing at what you really want...dunno. vu_infection_2_mod.hiplc EDIT: I think in my groupexpression @Cd.r>0, use @Cd.r>0.1...it's more accurate I think. And I'm not too savvy on the actual infection algo...if your question is more about this rather than the mountain, then I really don't know.
  14. It isn't really necessary to use fancy tricks for varying copy shapes in Houdini 18... the new Copy SOP has a pretty straightforward workflow for this. Pack each of the objects you want to instance and merge them, then on your template points create an integer attribute that will refer to the primitive number from the merge that you want to copy to each point. Then on the Copy to Points SOP just activate the "piece attribute" parameter and give it the name of the integer attribute you created on the template points.
  15. VEX infection

    since the file seems to work and your "infection" spreads from start point within the painted shape I'd consider it working if it doesn't work as you want it to maybe it will help if you describe how you actually want it to work
  16. RBD Transform Pieces

    Looks like there's a bit of trouble with your setup (unless I'm confused by what you're trying to do). In particular, when you copy the pighead to your pop-points, you're copying the packed geo - so there aren't any primitives; just points being copied onto other points. When you get to the loop, you promote name from points to prims - but there are no primitives. Your loop, consequently, only runs over a single element. The main result is that every instance you spawn into the dynamics network has the same name (you can see that if you look at the fetch_transforms point data). This means transform doesn't know how to match the pieces...
  17. RBD Transform Pieces

    Hi guys, Playing with some RBD stuff for the first time and having a little trouble using the transform pieces SOP with my setup. I'm creating geometry from scattered particles, creating a sphere proxy for my collisions and the setup all seems to work fine (although this doesn't mean i've set it up correctly). I'm bringing point transforms back in which also looks good but the transforms don't match when using the transform pieces SOP. Probably something i'm overlooking (maybe to do with orient attribute on incoming geo) but i'm at a loss as to what at the moment. Name attribute seems to match on both so i'm not sure it's that unless setup incorrectly. Also, the way i'm copying geometry to points means is by scattering each frame, so my transform pieces only show on their corresponding frame - a workaround for this would be great. I'm thinking a Solver SOP and merging last frame with input geo off the top of my head. HIP file attached Thanks! RBD_Test.hiplc
  18. Best Linux Distribution for Houdini ?

    But Linux does have considerably better memory management. Which in turn allows you to get more out of your machine.
  19. the instance node in your file is not loading in houdini H18
  20. VEX infection

    oh, please... could somebody help me. start to become grey hairs... The solver wont work as I want to. Where is my mistake? infection_2.hip
  21. Best Linux Distribution for Houdini ?

    That's not why people use Linux. As an individual user there are few benefits like more privacy, more control, and less bloat-ware. For a studio there are many benefits related to scalability and flexibility that Windows either can't do or does terribly at like... Hugely scalable and high performance parallel file systems like Lustre and Gluster Powerful configuration management tools like Salt, Puppet, Chef, and Ansible Integrated support for enterprise hardware like high speed network adapters and disk controllers No silly artificial hardware limitations on the number of processor sockets or quantity of memory Full control over any aspect of the operating system like automatic updates and desktop interfaces Ability to automate just about anything that happens on the machine I'm not saying people shouldn't run Linux as an individual. There are lots of good reasons to run Linux as an individual. Just saying that it's not faster.
  22. Hi, I was wondering if it's possible to create a button on a node that will allow you to select geometry in the view port from a specific group select attribute
  23. Hello; I'm looking for some good tips or tutorials about how to use "attributes" tab in the "Soft Transform" SOP? Thanks for helping.
  24. Hello everybody, basically in h18 there is a sop based approach for fracturing , so my question is , is there an effecient fast way of making secondary fracture on impact from a "cached fractured bgeo sequence", so there is the cached fractured geometry some pieces are moving and hitting the ground and some are static stay at the top (regular rbd sim), to get a secondary fracture from the ones which hit the ground in an efficient fast approach , do you have any ideas with h18, the imapct data give points which have been collided , so how do i get that data to fracture the cached bgeo sequence, would be great to get any ideas with the sop bullet solver, or even using the usual dop aproach , the voronoi fracture configure object seems to be a bit unstable/slow (may be i am wrong) , one more question , with the vellum style approach 3 inputs , in the sop bullet solver , if i jump in the node and if i wanted to add a voronoi fracture configure object , would i connect it to the first input ? Thanks in advance
  25. Stiffness, vellum vs wire solver

    So I'm not crazy then Thanks Thomas! That looks cool btw!
  26. Getting Creative with Vex and Python

    Hi, So i've started the journey to get into coding for short while now. And everytime I follow along a tutorial I am able to play around with new functions or parts of code I just learned. I'm able to alter the code but what I'm failing at is to write something else that is based on what i've learned. I can also read some of snipets others write but getting there myself is the issue. Which annoys me and im struggling with this a lot. I think im having trouble visualizing along the coding. Not sure if thats the point. So im looking for a way to get a better understanding on how to creatively use VEX and python. So code in general. Hopefully Im making my self clear and hopefully someone knows how to point me in the right direction. Many thanks in advance, M
  27. Stiffness, vellum vs wire solver

    Seems about right. Did some rnd on this, and went with rbds. Easier to interact with other objects and get stiff branches. Downside is the amount off setup.
  1. Load more activity