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.

vtrvtr

Members
  • Content count

    59
  • Joined

  • Last visited

  • Days Won

    1

Community Reputation

23 Excellent

About vtrvtr

  • Rank
    Peon

Personal Information

  • Name Vitor
  • Location Brazil
  1. Take a look. Changed nodes are either inside the box or colored red Constraint Obj Test 01.hipnc
  2. Sorry if I'm misunderstanding your problem, but you can control the distance per primitive like this: The promote is just there because you mentioned you promoted from points to prim attributes
  3. Check out the file. Red nodes are important, although you should check all What I did 1. Used a camera frustum from the qLib to only use the particles being seeing 2. Changed how the velocity was calculated inside the popnet 3. Used chops to scale the droplets so you get some shape difference 4. Added some mountain displacement to each droplet Maybe that can help you FrozenRain.hipnc
  4. The c. a. pieces sop expects geometries to have a name since it's usually used to connect pieces for rbd simulations. If you right click it and go with "allow edit of contents" (or something of the sort) you will see that some nodes have errors because of the absence of the name att. You can delete those nodes since you're not using them if you want to get rid of the error About the other question see here
  5. Just as an addendum, the very useful bbox functions work in wrangles. E.g ysize = `bbox(0, D_YSIZE)`; Will give you the size of the bound box in the y axis
  6. Going just by eye... Maybe? It's hard to tell just from these small clips You certainly should test with only the tips and if it doesn't go well you can emit from the whole curve
  7. Which part of the video are you referring to exactly? Around 1:40 to 2:00? FLIPs have a built-in viscosity att. You just need to turn it up to make it viscous. You can drive it in many ways. For example multiply it by the particle life, so after X seconds/frames alive the particle would have full viscosity On a side note, this is a very cool breakdown. Usually they go like "and then we added the effect [cue inclusion of a major and complex work]"
  8. I don't think you need to use chops you could use chops With chops I particularly find it a bit roundabout, but it goes like this: Initial setup with a transform sop that will control the scale: A simple chops noise makes it go like this https://i.imgur.com/SRrT3RX.gifv But now you want to modulate by the pointnumber, afaik you need a single channel for every ball which calls for a copy node. Here I'm using a expression to create a channel for each point from 0 to the number of points in the pointsfrmovolume1 node. Then I'm using a simple stamp function to vary the seed of the noise (it's hard to see but each curve is different) Now, using a chop expression you can bring it into sops. Note the $PT After this a simple stamp function in the transform node and that's it. Each sphere has its own curve, which means you can do whatever you want with it. https://i.imgur.com/V4lZkd2.gifv This might be a old work flow, I'm not sure
  9. Sorry, not sure I understand the question. Usually interpenetrating geometry breaks sims. What is wrong is what you just showed. You don't know how to fix it? The pieces needs to be further from each other. Instead of using a copy node, use duplicate with a bbox expression, that way you'll get a perfect fit instead of eye balling it. https://i.imgur.com/wE7yRWM.gifv Besides that you can go the packet geo inside dops and reduce the collision padding so it's more stable
  10. Increasing the subteps, either in whole dopnet or in the rbd solver fixes the sliding This 6 subteps in the dopnet https://i.imgur.com/BtG7rL6.gifv If you play with the substeps in the rbd solver you might be able to get the same without so many subteps
  11. Personally I think the most robust way to fit values in chop is using the icmin/max function. It preserves the curves and you can work with absolute values and it's only one node. It's so useful I even recommend you making a preset for it. It might have some pitfalls, but I'm yet to find any. Like this:
  12. You can also use the pop attract dop. The sop setup is pretty much the same as above https://i.imgur.com/2uM2AhZ.gifv
  13. Can't you just stamp first on the ground and then do the leaves?
  14. Not sure I understand. If you want the env. light you can just use the env. light without setting up any masks. The default "sky light" from the shelf is the sky + env. light
  15. You can check the box to render as background and then except the geo from the env. light using the "light mask" option. Like this It's in the render tab of the geo node at the obj level