By the looks of it the Karma ROP doesn't respect any of the render flags? I'm converting from an redshift based workflow with a complex scene and using extensive automation to render out subsets that use this flag it's a bit of a roadblock. Rebuilding it all in Solaris is a bit to much r&d..

I did notice the karma rop being a wrapper around a lopnet and digging in there confirms that the renderflag is not ending up in the USD data. I guess I could hack something in there to make it use that flag but I'm hopping I've simply missed a checkbox somewhere.. any clues on how to do this comfortably?

For now this works to generate a list of candidate objects in the Karma rop without the need to modify the karma hda itself which is nice.

Hi Jonathan, can you share a simple hip file that demonstrates the issue? Our goal is for Scene Import (which is what sits inside the Karma ROP) to give out-of-the-box results as similar as possible to what Mantra did. If we've missed the translation of a flag/setting, it'd be good to know so we can address this.

robp_sidefx
Hi Jonathan, can you share a simple hip file that demonstrates the issue? Our goal is for Scene Import (which is what sits inside the Karma ROP) to give out-of-the-box results as similar as possible to what Mantra did. If we've missed the translation of a flag/setting, it'd be good to know so we can address this.

Thx for stepping in! And ofcoarse while making a repo scene I found it works just fine. And when I revisited the production scene it was working normally in there as well.

I do notice 'the' viewport bug where objects stop updating in the viewport happens quite a lot recently, could that have been at play here? Not sure if that affects the Karma viewport as well? (It's the issue that the Lab's 'reset viewport' feature fixes)

Jonathan de Blok
I do notice 'the' viewport bug where objects stop updating in the viewport happens quite a lot recently

If you can trigger this during a screen recording, I'd love to see it so I can investigate further.