Hey all,

What would be the best workflow for having an animated vehicle coming from Maya into Houdini and then passing through to LOPs "/stage" IF they have lights? For instance car headlights.


We are ingesting caches from Maya as alembic or ideally USD soon (not yet tested in Maya) into Houdini, assigning materials based on 'shop_materialpath' attributes and then reading that in LOPs.

We just need lights now on the vehicle.

Any ideas of the best route?

Your best bet may be to create the light rig with constraint to the car directly in Solaris. The light rig should update with any animation updates to the car, and you could reuse it in shots. Essentially the light rig would just re-constrain to the car automatically. The specific kind of constraint may vary depending on what's being animated (if it's just a transform cache vs actual deformation of the points). Could be a parent constraint, or perhaps a surface constraint.

Tim Crowson
Your best bet may be to create the light rig with constraint to the car directly in Solaris. The light rig should update with any animation updates to the car, and you could reuse it in shots. Essentially the light rig would just re-constrain to the car automatically. The specific kind of constraint may vary depending on what's being animated (if it's just a transform cache vs actual deformation of the points). Could be a parent constraint, or perhaps a surface constraint.

Thanks for this Tim, we are heading down that route this week so will possibly throw some questions your way if its not obvious, thanks for the quick reply!

I'm tasked with the same challenge.

I do use the reparent constraint LOP for this. Using the transforms/locators from the Alembic file. It does the job, however I'm seeing viewport performance suffer severely. Going from 25fps to 9fps for one simple constraint.

We could insert the light under the transform in the scene graph. However, this is messy since we need lights to live under /lights

Any insights as to why performance degrades?

Ha, I had the constraint on the light layer before sublayering into the main scene.

If I add the constraint after layering in the lights, the performance does not degrade.

Andy_23
Any insights as to why performance degrades?

Making the nodes time dependent before the sublayer means it has to recompose the sublayer for every update. otherwise the sublayer and composition can happen only once.