Hello everyone.
I'm quite new to usd but love the ability to simply switch renderers in solaris, so I thought I'd do a test setup with some assets from megascans. However, I'm finding the conversion process confusing. So, presumably when using the component builder:
does anyone have an idea as to what an ideal setup, or can point to one, for dealing with all the LODs that come with megascans? And also, what when the asset might contain several "sub-assets" (like 5 rocks, each with x lod's but a single set of textures for all)? Finally there are assets that consist of several parts, making up the one asset...
Since I am new to this I'm not sure about what should be "variants", what ought to be put in "sub-layers" and so on, and how best to do a setup that catches all the above cases. I've found videos showing off the component builder, but they mostly deal with simple cases.
So, again, can anyone show/point to/explain these things to me?

Thanks,
Doug

I'd like to bump this topic, since it's something I deal with too

Variants would be the way to go for "sub-assets" and lods. Variants can be nested, so as you switch a subasset the lods would update accordingly. I'm not sure from the OP what the difference is between a part and a subasset but if you needed three levels of variants part->subasset->lod that would be doable too.

I may be wrong but I think out the box the component builder might only support geometry and material variants. So you might want to get familiar with it to get an idea of good practices and other features (proxy geo etc) but then create your own thing that has a more specific variant structure.

Unlike variants, the sublayer structure is less user-facing and can be done many different ways, depending on how much flexibility you want and how many files you do/don't want to manage.

I`m working on a file that does all the diferend kinds of things I am however stuck in some places I`m mostly making this post to come back here when I`m done.