I'm running into a peculiar issue with my shader otls. I'll try to explain on an example.

I create a simple shader starting from surface model in the material pallete. I dig in, add a bunch of nodes, export parameters and end up with simplified version of mantra surface. Now I need to make HDA from this so I right click on it > ‘Create digital Asset’ and save it. All the otls we use live on central network location and we use combination of HSITE and otl scan path, to load it all in. That all works fine, however the I'm noticing incredible slowdowns in the viewport when I apply this shader from network located otl. When I stick the otl on local drive everything is fast and I get >120fps, when using identical otl from network it drops to around 12-20fps.

The strangest thing I find though is, that this doesn't happen if I first wrap the shader in subnet and then convert this subnet into HDA. Loading such otl from network doesn't slow down and stays above 120fps.

So the question is. Am I doing something wrong? I assumed that converting shaders into HDA directly instead of wrapping it in subnet would be preferred. Also when I apply ‘subnet shader’ to an object, it seems to ignore any OGL settings and only display plain the basic ogl shading in the viewport.

I've attached simple example scenario file. Just convert both shaders into otl saved on the network drive and try applying.

Any pointers about this would be greatly appreciated. It's simple to work around it by applying shaders from within the subnet, but it is certailny undesirable behaviour.

Cheers

Milan