I've seen tutorials, that do this with the Copy Stamp node, but is it possible with the Copy to Points node as well, to define a group of objects for houdini to pick from to get some variation in my scatter? For example to scatter 5 different tree models onto the terrain to get variety to a forest.

The reason being, I think to have heard, that Copy Stamp is a legacy thing, working less effectively than Copy to Points and might be deprecated in the future.

watch this:
https://vimeo.com/213127548 [vimeo.com] around min 40
and
https://sidefx.com/tutorials/houdini-16-masterclass-compiled-sops/ [sidefx.com]

you don't necessarily have to utilize compiling, however the workflow you are looking for is explained there in a great detail
compiling is the next step to make it multithreaded and much more efficient though

cool! I'm gonna go watch these right now. Thank you.

Hi, I'm back again with a similar problem. I started using Redshift, and it doesn't seem to be taking Packed Primitives into account when rendering and treats the Copy to Points node as standard geometry.
It does support the Instance node on Object level though, so back to my question about instance variation. What's the workflow, if I need to substitute the Copy to Points for the Instance node?

The best thing I could come up with was to split the point cloud into a number of groups, that corresponds to the number of prototype objects and create an instance object for every single unique prototype. But this sound like a lot of work and isn't very flexible, should the number of unique prototypes change.

I'll be really thankful for any better idea.
Thanks in advance.

ivanmalek
Hi, I'm back again with a similar problem. I started using Redshift, and it doesn't seem to be taking Packed Primitives into account when rendering and treats the Copy to Points node as standard geometry.
It does support the Instance node on Object level though, so back to my question about instance variation. What's the workflow, if I need to substitute the Copy to Points for the Instance node?

The best thing I could come up with was to split the point cloud into a number of groups, that corresponds to the number of prototype objects and create an instance object for every single unique prototype. But this sound like a lot of work and isn't very flexible, should the number of unique prototypes change.

I'll be really thankful for any better idea.
Thanks in advance.

That is indeed the required workflow for redshift to properly use instancing. An instance object per unique instance. I'm not sure if redshift supports instance files, but maybe look into using a .rs files instead of objects. Someone with more redshift experience can probably answer that. Since this is a redshift workflow specific question, you could try posting it over on that forum too.

I was looking at this video earlier. So I have things kinda working.. i have random scale, rotation.. and even object selection.

What I don't have is any control like I had before. The Stamp function sends an attribute up stream. I am scratching my head about how to put the $ID value into the switch node.. basically copy mesh #5 when there is a point attrib of ID#5.

Like I do with stamp(“../copy4”, idval, 0) on my “old” copy stamp setup.