the nodes grouping thing you talk about is also present, and better
implemented then in USAnimation, in AVID|DS (although it was born with a
different purpose)
in DS you can group nodes and actually even render them as a cache, so
that if you modify the tree before that branch it will read the images
cache instead of re-rendering the whole branch.
it's also possible in Houdini's halo I believe (without the caching),
where you can group nodes into one and create a CHOP to wrap them it
with an interface that only exposes the input and output channels you
want to use
it's not particularly new, and it's such an intuitive and logical way to
work that I'm still surprised this kind of feature isn't a standard in
every nodal comp app by now.
******************************
| Raffaele Fragapane |
| Rising Sun Pictures |
| "Remember, TD is for TopDog" |
******************************
Greg Smith wrote:
I agree on basic nodes, if you can replicate the functionality of a
big node with that of a simple nodes, you have that flexible level of
abstraction. I have to say though, after getting more exposure to
compositing in general, one of the most interesting concepts in tree
organization came from the most unlikely of places. When I worked for
Disney, we used USAnimation for compositing, a sub-par compositor for
sure, but the way it orginized its networks was cool. It worked in
multiple levels of networks where in your root tree, a subtree would
be represented as a single node, as you clicked in that node, it would
open up that subtree. Reminds me of how windows explorer works.
Granted it doesn't give you the ghestalt of having every component on
one layer, but if a simple and clean flow is really important. That
feature seems pretty nifty!
Just thought I add my nonsensical ADD banter :-p
Greg
---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi
