There is also this: Game tools for terrains [www.sidefx.com]

Make sure your size parameter field matches the size of the longest side of your terrain (in meters).

The animation editor pane allows you to see a graph of how each channel behaves between frames. The default is to ease in and out of each keyframe and this is the change in velocity that you are seeing. On the right side of the pane, you can change this to a constant slope for selected segments.

See:

The Animation Pane Graph View [www.sidefx.com]

Were you able to get the Scale Source Volume to work? Activation (on/off) works for me but I don't see any effect on changing the Scale Source Volume.

(I am using a grid as an emitter source.)

I was able to make my hda produce deformable clones by saving the original after the generation step. I'm sure the problem is my lack if understanding of the hda process/embedded vs file/python panel save vs Houdini save/presets/galleries/otl managers/properties/etc…

I'm good for now and will look for the masterclass on tge autorig.

Thanks.

At the risk of giving you some bad info because I am just learning Houdini…

I found that you have to put the “.abc” with the file name when you specify the output path. At least the default settings produced an alembic file that was readable in Blender.

Try something simple first.
Set up the animation of an object (translating or rotating).
In the timeline, set the start and end times.
In the Output Network, add an Alembic node.
In the parameters of this node, set the output to render the complete animation (not the current frame).
Set the path for the the output with a file name that includes “.abc” (test.abc)
Click render to disk
Now try importing that file into your other program.

If that works try it with the animation of a simulation.

BTW, alembic files can be large. For my project, I am leaning toward using a workflow in which the Houdini stuff is rendered in Mantra and composited with the output from Blender.

I create a rigged character using the AutoRig panel with the Embed in Hip file option. It deforms and can be posed as expected. If I duplicate the hda or bring it in from TAB > Digital Assets in the network editor. The geometry does not deform with the rig. If I dive into the node, the deform node inside geo_skin appears to be selected correctly and the display geometry switch in the parameter panel works as expected. The geometry just does not deform.

I'm a novice at this hda business so it is likely that I didn't save something correctly. Any ideas of what I should check?

Here is my process for building the rig:
1. Import geometry through FBX
2. Launch the autorig panel
3. Select Operator and point to the geometry node inside the FBX subnetwork
4. Select the biped preset
5. Select Embed in hip file
6. Turn on display deform SOP
7. Create the rig
8. In the Scene View adjust control points
9. Back to autrig panel, File -> Save
10. Generate

The character can be posed.

Duplicate the node. This version does not deform with changes to the rig.


Edit after more experiments… Looks like I need a final step to update and save the digital asset for clones to work properly. As far as I can tell, this involves unlocking the DA and updating the Type Properties?

Space-F is like Space-G but keeps the view orientation.

One suggestion is to position Nulls at points in your scene that you visit frequently and use them as view transport targets. Then pan from there.

Another is to just put cameras at different points of interest.


(Edited to correct Shift -> Space)

Thanks. Good to know.

I imported a rig that originated in MakeHuman into Houdini using FBX. The Find Node function in the network editor does not seem to find all of the nulls or bones when using the filter.

In searching for nulls, it finds all on the right side and only a few on the left (about 150 nodes in total).

Is there a limit on the number of items that may be returned by the Find Node function or could there be something about my file that is causing this behavior?

Houdini Indie 16.0.504.20
Windows 10

Edit for clarity: it can find the null nodes, just not all of them at once.

I have a similar interest (but with Blender instead of Maya). My plan is to render the crowds in Houdini and composite with the rest of the scene rendered in Blender. I'll probably use the compositor in Houdini.