After Copy to Points, I duplicate the whole setup using Copy and Transform because this workflow is more convenient for my setup.

I currently have 3 spaceships, but the problem is that they all fire at exactly the same time, perfectly synchronized.

I need to add a different time offset to the shooting animation for each copied spaceship, so each ship fires at a different time.

I can’t figure out the best way to do this, especially since the ships are duplicated with Copy and Transform.

Could someone please suggest how to set this up?

Houdini 20.5
 

For a project, I'm using a Copy to Points SOP with a piece attribute to scatter two different animated mesh types onto points. Each copied mesh has a unique id attribute.

I would like to scatter 100 points on every individual mesh instance after they have been copied to the points. Currently, if I use a Scatter SOP with a Force Total Count of 100, the points are distributed across all meshes combined, rather than 100 points per mesh.

Because there can be hundreds of animated mesh instances in the scene, I'd prefer to avoid a For-Each loop if possible for performance reasons.

What would be the best way to generate a fixed number of scattered points (e.g. 100) on each copied mesh instance?

Cheers!

Here's the steps :

• I simulate first the ship as an RBD on low res flip just to get the general motion, but here the waves (from ocean wave node) are wedged into 3 variants (to test I have input one value that is very high)
• then I take those 3 variants points from the RBD-FLIP sim, create a VDB (using collision soucre) for each of them so 3 VDB variants.
• Finally I input the VDB into the main flip sim as deforming object with its geometry of course . Here I wedge the velscale of the collider (3 variants) as well.

So if I understand correctly I will have 9 wedges ? (3x3) this is what tops shows me as well, I'm testing just 10 frames ,as I'm starting in this context yet when I verify the OPENGL renders they are sometimes mixed up the falling boat(from the big ocean wave,shows a very calm wave)

here's my setup ( up till the fetch VDB BOAT it's correct , I have put ROPGL nodes to verify this) , I think I got it all wrong below that ? I'm not sure how to fix , I have watch the sidefx intro to pdg in their portal but I still don't know well the nodes :/

let me know if there's something more I need to provide to debug this.

I built a procedural city and created a random attribute to control the height of the buildings, which works correctly.

However, when I merge/assemble all the buildings together at the end, the randomization no longer works, and all the buildings end up with the same height.

Does anyone know why this happens and how I can preserve the random height of each building

PDG.zip

I don't want to use UV Rasterize Attribute because it’s too expensive for me when I need to increase the divisions.

I haven't found any node in COPs to do this.

Any help would be appreciated.

Thanks!

sample attrib on uv - No rasterize.hip

I am having this model of a full armor I want to rig / animate.

At first I used the "bone capture biharmonic" method to skin the armor which gave me overall deformation of the different pieces.

Next I used "capture packed geo" so the pieces stay rigid which worked really well and gave me exactly what I was looking for.

The issue is now that there are pieces like the hand of the character which I want to deform in a more organic way like a normal hand would.

Because the hand is not made of different joints like a robotic hand, what would work with the capture packed method. I would have to split the fingers up in segments to deform them, what would look weird / break the whole hand.

Is it possible to use different capture methods for different parts and combine the bone capture attributes later for deforming?

So for example use biharmonic capturing for the organic parts and packed capturing for the rigid parts?

Or would there be a different approach / workflow?

Looking forward to your ideas. Thanks!

while the volume from attribute (after pointvelocity 0,-3,0) shows:

Any idea how to get the volume from attribute to align with the 0,-3,0

it seems to be inventing some other velocity that I'm not aware of, I gave it just 0,-3,0

velocity_mismatch.hiplc

I'm using an RBD Packed Object with a SOP Solver. I want to apply an upward velocity only once when a piece becomes active.

DesTruction.hip

Has anyone else encountered something like this, or is it just me? I'm using the latest build of Houdini 20.  

https://youtu.be/FKZ_0Y17yE4?si=ccFXDxuKr3cnt1mU

Here is the images of what I mean. Transform a sphere to this shape, modified it without point count or point number changes  Ignore the center core.  I'm having a hard time getting that nice round bevel around middle section

geo_fix_stretch.hip

Houdini 20.0.547
Fedora 44 KDE wayland
Mesa 26.0.8

 

If I do check the Re-evaluate will the simulation stupidly recalculate the interpolation that has already been calculated by the Collision Source, or will it gladly use the already available interpolation in the file cache from the collision source?

If I do not check the Re-evaluate ... will it yes or no use the existing interpolation from Collision Source

Why am I asking the question? because Sidefx documentation clearly states for static object under re-evaluate SOPs to interpolate Geometry: "Normally when a solver asks for geometry data in a sub-step, the simulation will simply linearly interpolate position data from integral frames. However, this is not exact. Turning this option on re-evaluates the geometry network for each substep. This is more accurate, but can be very expensive."

So does the simulation or does it not stupidly recalculate an already calculated interpolation, if this Re-evaluate thing is checked?

I have an asset that I imported into Houdini that has pieces with reverse Vertex Normals inside a for each loop per connected piece node, I have written a vex code that finds their bbox center of a piece and tries to determine which pieces get flipped according to the dot product of the point positions and the the bbox center point and have gotten no result out of it

And this is what I could come up with for islands with volume. I have absolutely no idea as to how to handle flat plane like structures. Maybe compare their vertex normal orientations to the centroid of the entire enclosure? But then again my code doesn't do the job

Any help would be appreciated.

AJ

flip_reverse_prims.hiplc

I'm looking for vellum net football. like this picture. 

I’m experiencing an unexpected behavior when rendering with Karma (both CPU and XPU) using the MaterialX Standard Surface shader, specifically related to the subsurface scattering parameters.

Here’s my test setup:

• Base Color: Red (RGB = [1, 0, 0])
• Subsurface Color: Green (RGB = [0, 1, 0])
• Subsurface Intensity: 1
• Radius: {1, 1, 1}
• Scale: 1

Based on my understanding, with these settings I expected the material to display Green (because Subsurface Intensity is 1) as a result of the subsurface scattering. However, the rendered result comes out completely white—almost as if the output is dominated by the scattering (i.e., the Radius parameter), with no visible contribution from the Base or Subsurface Color.

I’m wondering:

1. Is this behavior a known limitation or bug in Karma’s implementation of the MaterialX Standard Surface shader (particularly the subsurface scattering part)?
2. Could there be another explanation or parameter interaction that might cause the subsurface effect to override the expected color ?

Any insights or advice would be greatly appreciated.

Thank you in advance for your help!