Re: Article on Bifrost

[Stefan Kubicek]

But it can't be used to build custom modeling tools, deformation tools, rigging tools, etc, and all the cool stuff that ice is capable of doing outside of simulation.

Why can't it? I don't remember anything in that article that rules that out. From what I read, Bifröst is an architecture which already implements fluid simulation nodes as a proof of concept. I can't think of any reason as to why it could not also implement nodes for parametric modeling or deformation effects. It's just a high-performance dependency graph, pretty much like ICE, Fabric Engine, Maya, Houdini,...

So i guess Softimage will still be used for years to come, because i don't see bifrost anywhere near a replacement for ice. At least not in its present form.

I think it's only as good as it's frontend. If it's cryptic and hard to use it won't find many adopters. ICE has pretty good usability, but I bet there are things that Bifrost might have learned from it.

And why in the world do they keep pushing for that annoying "node editor" gui in maya. Am i the only person that finds it incredibly clunky compared to the ice interface?

I find it a huge improvement over the other node-based editors in Maya (Hypershade, Hypergraph, etc). It has the potential to merge all these editors into one homogenous node graph editor. The only thing I see that can be distracting is that Maya nodes do have a lot more inputs and outputs than the average ICE node usually has, which can be irritating and harder to understand (what goes where, which connections make sense and which do not, etc...).

Cheers,

Octavian

--

-------------------------------------------
Stefan Kubicek
-------------------------------------------
keyvis digital imagery
Alfred Feierfeilstraße 3
A-2380 Perchtoldsdorf bei Wien
Phone: +43/699/12614231
www.keyvis.at ste...@keyvis.at
-- This email and its attachments are --
--confidential and for the recipient only--