Core 4D

Just this moment everyone is going batshit crazy over the latest announcement on Maxon‘s corporate blog touting a massive rewrite that has been going on for years. I say: So what? This is just a confirmation of what anyone has suspected/ guessed anyway. It’s amazing how naive people tend to be and then rave over such non-news. The signs have all been there, or did anyone think those new interactive tools in the last few releases fell from the sky? As I like to say: My mom could have told you!

Now Maxon may not be the most forthcoming and popular 3D company on the planet, but they aren’t stupid, either. In light of what e.g. Autodesk are doing with MAX and Maya (parallelized geometry processing, GPU accelerated deformations, hybrid massively parallel simulation frameworks), it was only a matter of time. It’s critical to their business, both in terms of maintaining their current user base and potential future growth.

From a technical point of view, I think it’s a good idea to manage expectations. Not only will it take time to convert all code as already stated in their blog, but people shouldn’t expect this to be anything glitzy. Such “core” rewrites are mostly about invisible stuff like drawing calls, traversing large data trees, scene graph optimization, unified deformations, overall data structures, program-internal communication and all those other good things that you take for granted when you click around in your favorite tool/ toy.

Interestingly, and I do think that’s one of those things that people never “got”, this is basically the same what NewTek tried when they were working on CORE (yes, somehow any mention of “core” automatically brings up these discussions about history repeating now ever since this), they just completely were in over their heads and underestimated how much work it would take to turn a proof-of-concept into a functional app.

So what’s the skinny? At this point it’s impossible to say, but I would predict that in R18 Maxon are going to make up a bit and give us something bigger than the disappointing R17. My guess would be that they’ve rewritten the MoGraph module using this new architecture for better performance, more clones and some new tools that e.g. could benefit from last year’s spline tool renovation. It also is not as outdated as e.g. Hair or Thinking Particles, so any conversion work might be easier. Then again, though, they could be totally out of their minds and have written their own version of a 3D dildo-generator. Either way, these are intriguing news and most importantly it will be interesting to see how competitors respond to this…

%d bloggers like this: