I want to get the second and the third output of the hda in PDG workflow, but it seems the PDG just process the first output.

Hi Eric,

PDG is a templated processing graph - as such, when nodes connect, it's really talking about the flow of the workitems. With that semantic in mind, the nodes like split can can multiple outputs, because there are two ways the workitems can flow with a split node. This concept is quite distinct from the multiple results within a workitem, which is probably what you're thinking of when you mention multiple outputs of the node. To address the the multiple outputs (bake results) of a workitem, please use @pdg_input.0, @pdg_input.1, @pdg_input.2 etc from a downstream node to refer to the upstream workitems's various bake results.

That said, there is an outstanding RFE (RFE 94236) to actually capture multiple bake results of HDA Processor, so it can be addressed downstream as @pdg_input.0, @pdg_input.1 etc. Right now, for HDA Processors, it is indeed only grabbing the first output of the HDA itself.

Hi, Ken Xu:
Thank you for your quick reply, I'll try to merge the outputs and split them later.

I couldn't figure out kenxu's solution, I did make it work with a partion expand and a split though. Thought I'd post for future reference.