I’m always looking forward to SIGGRAPH every year, both for reasons of catching up on latest product news and research papers, but sadly enough, this year feels quite empty. It’s as if it never happened. Several interesting products (Mari, RealFlow, various render engines) had long been announced and shown before the show, taking quite a bit the wind out of the whole event.

Additionally, what news we got on Lightwave, ZBrush, Houdini and Autodesk‘s products was not particularly spectacular or interesting to me as a user nor on a geek level. It feels like everybody is mostly trying to defend his market position with only minor enhancements that might still come in handy in your day to day work, but do not necessarily expand feature sets. One such thing would obviously be the whole Lightwave 10 situation with the VPR renderer – as I wrote, useful, but not necessarily the ultimate incentive for an upgrade so any user who already owns FPrime will have to think hard about shelling out the money. In itself it’s not a bad thing of companies focusing on more practical workflow things, but still, somehow I’m craving for those times when literally SIGGRAPH reshuffled all the players involved and you generally were a bit more excited. Well, there’s always next year and, which is also quite noticeable, there have been no news on some programs (modo, Cinema 4D), which leaves room for something to look forward perhaps later this year.

Research papers, as far as I have looked them up and as my puny brain understands them, were also a bit odd this year. Several of them focused on topics that are not necessarily relevant to an end user and are perhaps a bit too esoteric to ever become relevant for mainstream programs even. Of course there was once again a ton of fluid simulation stuff and the notorious foam bubble noise simulator (which seems to become a running gag), but lots of it focused on alternate implementations or resolving quite specific detail problems in the underlying algorithms. The same always strikes me when seeing those behind the scenes presentations about how specific effects in movies were created – sure, interesting to see the science at work, but 99% irrelevant for your average Joe who doesn’t have the resources to even run the simplest fluid simulation and constantly struggles with deadlines, because the render farm is always too small…

Mmh, this second day doesn’t really offer any exciting news, but I’ve got to write something, so let’s see what we have. First, SideFX now officially presented Houdini 11. Considering that they had an announcement already almost 2 months ago, the Wow-factor isn’t that great. I also wasn’t particularly convinced by their sample videos, but then again those may have been rushed and not show the full potential.

Another, perhaps more relevant news (for the future at least) is the release of some specs for Alembic. Could be interesting, if it really can replace the various deformation, particle and point caching methods and formats used in different 3D apps, but that may be a while. At least they seem to have some good ideas on how to avoid point/ vertex list bottlenecks (the inefficiency of relevant sorting algorithms being a major reason, why many 3D apps eventually will still be slow as molasses when those lists need to be reorganized).