Below is a screenshot of of my HDA properties with 2 input nodes set. My intent is to better understand how to control the different ways in which nodes may communicate. I hope to be able to plug in 2 separate pieces of geometry and set parameters which will determine the behavior of the asset. I am unable to remember the process for creating nodes from within my subnetwork capable of accessing the data sent in via the input nodes (and detailed in the Input/Output tab). My subnetwork is currently empty as I'm stuck not knowing how to link to the two pieces of geometry input.




And a screenshot of my Input/Output tab


Any and all help/direction is greatly appreciated. Thank you in advance.

Do you really want to use Object nodes? Typically, assets that process geometry from their inputs are implemented as geometry nodes, not object nodes. Data flow is only vaguely 'implied' by object nodes, and it's entirely up to the asset author to implement it.

This is usually done with expressions on the node to track what nodes are connected to which input, and then expressions on object merges inside the asset to fetch the connected objects' geometry. See the Houdini groom tools for an example of how this works.

Hmmm...I'll look into expressions then.

Thanks for setting me on the right track. It's funny. I knew the first thing someone was going to mention was the fact that I'm doing this all on the object level. *I actually chuckled. You're absolutely correct in questioning me, though (many thanks for the concern too). Every tutorial I have ever seen has followed a workflow of manipulating all geometry from within one geometry node. In fact, that's usually how I work. I only tried this way to "see if I could" just once. Really just testing the waters.

My next steps:

-Transfer everything back into a single geometry node
-Create a subnetwork which I will save as an HDA
-Attempt your suggestion with expressions from there
-finally report back!

Thanks again!

hwilliams
Attempt your suggestion with expressions from there

I believe the suggested expressions were for Object level trickery

In SOPs the geo naturally flows in
And of course you can have additional logic inside and in ui using expressions, but for simple setup you should be fine with either direct geo connections or Switch If logic to handle some cases like empty or not connected geo

Yeah, so I read what you said about sops and decided to adhere to the path of least resistance. Houdini is challenging enough without intentionally making it harder on myself lol. Subnetworks it is.