Great suggestions.
I used to use messiah:studio a lot and there all the conversion is done
on the fly inside the nodes. The conversion is very straightforward in
the end and the clutter is reduced enormously. And you can always plugin
a specific converter if you really need to.
There are two other things that make the rendertree in messiah much less
cluttered:
- multiple outputs from the nodes allow you as a developer to give the
user much more choice on what they need. My AoN:studio shaders for
instance (http://www.screendream.de/AoN_studio.htm) have all color,
alpha, raw scalar and a bump output (which can be scaled separately).
This way, you can get a lot out of one node. And other nodes have 10 or
more outputs of different states/aspects of the internal calculation...
- input layer modes. Each input in messiah can have as many connections
as you want, and each one can have it's own "layermode". You hardly ever
need a mixer for basic layering.
What would be extremely cool are groups, like in Digital Fusion for
instance...
Hopefully XSI/Mental Ray will develop that part to be a bit more user
friendly!
Cheers,
Thomas Helzle
Arvid Björn wrote:
I usually stack all nodes that arent branched into little heaps of
nodes, to keep different parts of the tree neatly separated, see
attached. Perhaps a real grouping feature would be nice, like Shake.
peterb wrote:
overall I think there is still room for streamlining XSI in this area.
eg. all the conversion nodes (color2scalar and so) can really clutter up
things.
why have a seperate node to go from one to another?
perhaps this could be added within the connection line,
and if you would right click on the line, you'd be able to see the
conversion properties.
...
------------------------------------------------------------------------
---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi
