I'm a bit confused about correct and efficient way to work with large crowds with Solaris and Karma. SideFX provided to us sopcrowdimport node as a wrapper for "efficient way of importing crowds", but my problem with it is that i don't see a way to utilize renderfarm for parallel caching. The simulation itself is just clip playing, so it can be cached on multiple nodes in parallel. Everything should happen on 1 renderfarm node as it needs to cache animation, geo and final USD as single files.
Potential solution would be to cache per frame usd's and stitch them with Value clips. Pixar's own documentation about ValueClips says:
The USD Value Clips feature allows users to decompose time-varying data across many layers that can then be sequenced and re-sequenced back together in flexible ways. This feature is purely a value resolution -level feature, not a composition-level feature. Value clips allow users to retime sequences in various ways. This allows users to reuse a set of value clips in different scenarios, with only the sequencing metadata changing. At Pixar, we have found value clips useful for efficiently animating medium to large crowds, and for representing very large, simulated effects. For more detail on these use cases, see the glossary entry for Value Clips.
But when i cache per-frame usds they are super huge. Simulation that was ~10gb altogether with coprowdimport for 200 frames becomes 0.5gb per frame.
The question is, anybody figured out efficient and fast way of caching and rendering stadium-sized crowds in Karma?