just to clarify for people who didn't watch the video, he was answering a question from the audience at the end of a 25 min presentation, this was not the presentation. How Bifrost is similar or different to ice is something a lot of us want to know.
I don't think you understood the difference, btw. The compilation is an implementation difference, but how the graph works is also different. The scene data flows through Bifrost nodes and modified immediately; in ICE the computation only happens at the set data and the connections are just logical links, so you can't really view the data of a graph in the middle without rewiring something to a set data. It's more difficult to understand and debug a graph because of this. On Aug 15, 2014 2:09 PM, "Eric Turman" <[email protected]> wrote: > That is a reasonable statement, Luc-Eric. However, given the proper > resources such libraries could have been generated for Softimage. Moving > forward though, a large amount of presets for Bifrost would be a good thing > for Maya users. > > I have to agree with Simon too in that the way the Bifrost data flow was > presented felt more like it was and underlying "Bifrost compiles rather > than traverses to reduce overhead" as opposed to "this is easier and more > intuitive to use." When you deal with nodes, you still have to know your > math and logic or you won't be able to do anything worthwhile with it > regardless if it is ICE, Utility nodes, Bifrost, or Houdini. I do believe > it would have been a better, more professional presentation if he focused > on what Bifrost can do as opposed to taking the time to compare it to ICE. > > -=Eric > > > > On Fri, Aug 15, 2014 at 12:40 PM, Luc-Eric Rousseau <[email protected]> > wrote: > >> I think the "hard" comment relates to wanting to have more built-in >> functionality in Bifrost, and workflows in the viewport and outside >> the node editor. This is a counter point to the artists who are not >> really interested in just getting a library of hundreds of nodes to >> connect and figure out. They want to be able to open the box and fix >> whatever is there, which bifrost will allow, but it should already do >> a lot, and do it well, outside of the box. You should not have to >> program to do common/simple things. >> >> On Fri, Aug 15, 2014 at 2:30 AM, Sebastian Kowalski <[email protected]> >> wrote: >> > https://www.youtube.com/watch?v=DucKeXM_gHM&feature=youtu.be&t=26m >> > >> > >> > >> > > > > -- > > > > > -=T=- >
