Don't know if anyone wants to dive into this one.

Problems with my digital asset.

First how does it work ? There is a Python SOP on the obj level ('shardgen1')that has several parameters. First the align<->build parameter is toggled off. The screen shows two tubes that are used to position the spot for a hole in eg. a wall (here inside wall object on obj level). Then you toggle on align<->build and the wall appears with a hole in it. So far so good. Problem is, the script code inside the ‘shardgen’ SOP also creates a bunch of shards that sit inside the hole. They don't appear though (not anymore that is, yesterday all was going fine). What happens is this: when you go into the dopnet1 (inside shardgen1) and you mliddleclick on ‘shards’ Houdini tells you "no objects were processed on the last timestep'. This should be the about 2 hundred shards. (the visualisation of the shards is done after the dop stage through a dopimport SOP) When you click on the display flag of ‘shards’ or when you change any parameter everything recooks and the shards appear magically where they should be. So, why does it not (and it did before) cook the shards properly straight away ?

Hope this makes sense.

It seems the culprit is a destruction routine in the shardgen Python code: whenever align<-> build is switched off, all shard geometry is destroyed and then remade when it is switched back on. It seems the dopnet registers the destruction of the shards, but doesn't immediately register the shard geo been made again. This is then why it doesn't display nor simulate.

This is really weird behaviour. I need a workaround since shards geo with Python code slows down aligning the hole so must be destroyed when switching.