Hi,
I've been reading some really large point caches, particularly
RealFlow BIN and PRT. We are doing this as part of our Fury 2
differed particle cache loader we've implemented for Maya based on
Partio. We are doing differed loading so that only a preview is
loaded into Maya for positioning purposes. But I am finding that
loading ~ 1 GB BIN and PRT caches that are produced by RealFlow are
fairly time consuming. I know that this size of particle cache is too
large to load into Maya, but it is a reasonable particle cache to load
into a particle-based renderer like Fury.
Has anyone yet ran into performance problems with Partio? Has anyone
used it to load similarly large caches?
So there is straightforward optimization solutions, but also I am
wondering whether to add multithreading to the library -- but that
presents cross platform issues -- I guess we could utilize IlmThread
or OpenMP?
--
Best regards,
Ben Houston
CTO, Exocortex Technologies, Inc.
http://www.exocortex.com