All Activity

not

A kind sir helped me with a tip to use attribute transfer node instead of vex. Been a minute since I picked up Houdini, apparently. Updated is attached. The forces are being transferred so the issue now will be figuring out how to break constraints / have gravity affect the pieces only once they reach a force threshold, or something like that. customForces_v09.hiplc

In the dopnet > sop solver > wrangle, it seems the nearpoints function only actually finds neighbors of point 0 for some reason... That might actually be my issue rather than not reading each frame.

Fixed! Fantastic, thanks so much for the advice, and the file. All makes sense.

Hey thanks! I'll check this out and report back :-)

uv_transfer_with_seams.hip try a for each loop

Wondering if there's a technique for this? I've got a megascan model with a bunch of UV islands. Convert to vdb, do some stuff then convert back to polys with the UV data along for the ride, then extract the UV data and re-apply it to the mesh. Works great, except at the edges of the UV islands - the UV points are all connected so there's glitches. I've made an edge group for all the stretched edges, selecting by length, so I've got them isolated and the UV islands look intact. But after that I'm running out of ideas - hopefully this has been solved before! Thanks :-)

Thanks! ill check out your notes, normally i would work with particle Cd attribute but i was worried that i wasnt the best approach if i wanted to see that mixing inside the fluid, not only over the surface so i thougt that maybe doing the process with volumes would allowed me later to obtaint a volume field that i could render but after some research, i think i might need a differetn sim witch pyro to "emulate" that effect...

Hello, I have a primitive string attribute: Bridge_Small_Dark_01 I need to procedurally extract the part of the string before "_Dark". Using split it seems it only accepts singular symbols as a separator (eg "/"), I need to find a way to use the specific string as the separator. Using s@name = split(s@name, "_Dark")[0]; does not work. Is there another function I should be looking into? EDIT: using "_" as separator and using [0] + [1] etc in split would not work since not all assets have the same number of words before "_Dark" EDIT2: Big thanks to Tomas Slancik for answering this on sidefx forums. Answer below: use re_split() [www.sidefx.com] string s = "Bridge_Small_Dark_01"; s[]@split = re_split( "_Dark", s );

Hello, I'm trying to create an outward glass shatter effect similar to the helicopter crash scene in The Matrix. So far I have rbdmaterialfracture > rbdconfigure > rbdbulletsolver and on a separate branch is a point scatter with velocity attributes I want to use to affect the glass sim, but I'm having a few issues with that (I'm a sim noob). The first problem is that I can't get the bullet solver to read the velocity each frame. It seems to read the first frame in which there's a non-zero velocity value but stops after that. If I crank the velocity on the scatter points, it seems like mayyyybe the other frames are being brought over, but not as clearly as the first frame and the sim sometimes just stops after a few frames... Secondly, I'd like the glass to have glue constraints and I just learned glue can't be broken with just forces, or they'd have to be manually broken. There's this tutorial that seems to be exactly what I'm looking for, but I get lost when trying to follow it in 20.5. Does anyone have any resources that could help me bridge the knowledge gap there? He breaks glue constraints in a sop solver within the bullet solver at 16:22. https://www.youtube.com/watch?v=fzRJcGFuPXU&t=982s Attached simplified file with my issues. Inside the bullet solver are only my best attempts... Right now I'm just trying to get the velocity transferred each frame so I turned off constraints in the rbdmaterialfracture node and gravity on the solver. It might not come through in the project file but a visualization marker for @v would show the effect I'm going for on the point scatter and also the transfer problem on the bullet solver. Any help would be greatly appreciated - Thanks for taking a look. Robert customForces_v06.hiplc

timeshift and then attrib copy to anim. . maybe send an scene example then i can fix it

Hi Everyone I'm studying the FLIP solver a bit in depth in my free time, going node by node and reading a bit of theory. However, there is a part that I can't understand and I hope you can help me understand the logic. I often read in the literature that in transferring velocities from Lagrangian particles (v@v attribute) to Eulerian grids (v@vel field) a sort of weighted average is made taking into account the velocities of the nearby particles. Example: https://nccastaff.bournemouth.ac.uk/jmacey/MastersProject/MSc12/Ioannidis/Thesis.pdf (page 43), here an abstract: In other more detailed cases the kernel functions are described numerically, for example: https://nccastaff.bournemouth.ac.uk/jmacey/MastersProject/MSc16/01/thesis.pdf (page 12) Or http://alexey.stomakhin.com/research/siggraph2013_snow.pdf Given this, let's say I have a point cloud with certain attributes (v@v, @mass, @pscale), my main questions are: - Does the Gas Particle to Field, when rasterizing the velocities of a point cloud, already perform a weighted average taking into account the velocities of nearby particles like described in above abstracts? - If so, how and what are the mathematical operations behind each option? For example, reading the houdini documentation, does the Accumulated option take into account these kernel functions: - What could be a code/pseudo-code to compute the weighted average of the contribution of nearby particles based on their mass, their distance from the current evaluated cell using a kernel function of a specified radius? - If I wanted to "build" a Gas Particle to Field from scratch, for example to implement the functions and methods proposed in the literature (like the ones proposed above) in which language would it be possible to do it? VEX, Python, OpenCL, C++? Any information will be really useful, thank you very much everyone.

Thank you, I watched a tutorial on how to bake the camera in Houdini, and then I used 'Scene Import (Camera)' in LOP to import it.

Houdini has sooo many cool nodes, thanks for sharing the file. I learn new things every day.

I usually do it like this. gradient.hiplc

Amazing, thanks @Librarian!! I'll start working my way through it.

check this VEX. inertiatensor.hipnc

post scene to check But as I remember you need to bake camera motion to transfer to LOP. Animation from nulls not read

I have this idea in my Head, just not shure if its possible. This is the Situation: I have a cloth sim. I have made a rest state to make the cloth be generally in the place it needs to be in and saved it with a vellum i/o to presesrve all the forces and constraint attribs. From there i do another simulation pass for the simulation that i want to use in the shot. That way i get rid of a lot of simulation time. Now we realised that we could need a higher resolution sim and i would like to avoid having to redo the Rest state. So my wiosh would be to create a new version of the constrainbts, matching a subdivided geo, and transferring all the attributes (scaled appropriately) to the new constraints. I already tried some stuff, copying a bunch of the prim attribs from the constraints, but it mostly pulled itself together. I know this is probably not a simple thing to accomplish, and i wonder if its even worth it. But this could be really useful. Maybe there is someone with some deep knowledge of all the vellum attributes? Anyways I would be super happy about any insight from yall

Or without a double intersect call: vector A = point(0, "P", 66); vector B = normalize(A-v@P); vector C = v@P + B * .1;// 0.1 to avoid overshooting int prim; vector primuv; xyzdist(0, C, prim, primuv); vector surfaceP; prim_attribute(0, surfaceP, "P", prim, primuv); v@field1 = normalize(surfaceP-v@P);

How do I solve the issue where a camera animated in SOP is imported into LOP through Scene Import (Camera) but the look is completely wrong? The camera is animated with two nulls. Looking for assistance.

Hi you can project the direction onto the surface using the Y-direction for example. TangetntField_01_mod.hipnc

Hi guys, I need to create a vector attribute, where all the vectors are aligned along the tangent of the surface, while also pointing toward a specified target point. I believe that combining the Tangent-field SOP with the Normal vector attribute should provide the solution, but I'm not sure how to properly blend them together! vector target_pos = point(0, "P", 66); vector normal = normalize(@N); vector tangent = v@field1; vector dir_to_target = normalize(target_pos - @P); v@FinalVect = ? Thanks for helping. TangetntField_01.hip

Hi Vikas, thank you so much. That is really interesting. What an amazing technique. And well documented. Thank you for sharing, much appreciated! Have a nice day, Hannes.

Here is the file showing my initial vex setup (adding @w to rotation), a bullet sim with Rotational stiffness = 0, which matches my vex setup. And a bullet sim with the Rotational Stiffness = 1 (default) that shows the difference in rotation. I would like to match the Rotational Stiffness = 1 with my vex setup. I hope that helps rotationalStiffness_vex.hip