RE: Schematic View

Date : Tue, 4 Mar 2008 22:16:58 -0500
To : <XSI(at)Softimage.COM>
From : "Luc-Eric Rousseau" <lucer(at)Softimage.COM>
Subject : RE: Schematic View
Yep, I get what you mean, thanks for the feedback.

 

________________________________

From: owner-xsi(at)Softimage.COM [mailto:owner-xsi(at)Softimage.COM] On Behalf
Of Bradley Gabe
Posted At: Tuesday, March 04, 2008 8:53 PM
Posted To: xsi
Conversation: Schematic View
Subject: Re: Schematic View
  

I think the theme is summed up from my earlier reply to your email:

On Fri, Feb 29, 2008:

I would hate to see the schematic broken by making it too detailed.
However, based on how I use it now (which is heavily influenced by how I
used it in si3d) for keeping track of hierarchy, scene organization, and
relationships at the x3dobject level (aka rigging), it wouldn't kill the
schematic if we could make relational changes by clicking and dragging
on the hierarchy, expression, and operator lines we already see in the
interface.


The ability to "rewire operators" is not the same as making "full
operator connections." I think we are getting confused again in
semantics and pedantics.

By "full operator connections" you may be thinking something like the
render tree or moondust node graph, where it is required to see many
potential input and output channels on the nodes, and have the ability
to add new nodes, and convert between data types, etc. I think of that
kind of work as node-flow programming, the place where you build new
operators or edit operator architecture.

What we've been asking for in the schematic is rewiring existing
operators, not designing new ones or fundamentally modifying them. Once
an operator has already been assigned (and I'm including constraints,
expressions, SCOPs, default XSI operators, and custom moondust
operators), the schematic currently displays a simple relation line
linking the X3D Node where its operator exists to all the other X3D
Nodes that have some connection to that operator. At this point, all of
the local address ports have been set for the op to function properly,
such as kinematics driving a constraint, or some parameter driving an
expression. It should be possible to disconnect the link, and drag and
drop it over a new X3D Node. If that new node has the same matching
local address to the original, then the operator port gets rewired to
the new X3D Node, otherwise not.

The more I think about this, the more I realize it's going to be a
necessity for Moondust. In a production environment, you will have those
who are comfortable programming in a node-flow environment ( TD's), and
others who need to use the custom operators, but won't be comfortable
outside the explorer or schematic (animators). If you have a custom
Moondust operator that connects to a dozen meshes, curves, and nulls in
your scene, it would be far more workflow efficient if you could select
those operator link lines in the schematic and move them between X3D
nodes, rather than trying to find the right input nodes in the Moondust
tree.

-Brad

On Tue, Mar 4, 2008 at 7:35 PM, Raffaele Fragapane
<raffsxsilist(at)googlemail.com> wrote:

Fair enough Luc, but that was Brad's recap not the backbone of the whole
thread. I would say the largest majority of the opinions on the thread
seem to point toward the need of better layout tools first and foremost
(or at the very least not disagreeing with it), regardless of what view
will one day replace or complement the schematic. Same goes for hdn and
maya's brief mentiones, I don't think they were really the key of the
thread nor the recurrent theme :)

 

On Tue, Mar 4, 2008 at 11:25 PM, Luc-Eric Rousseau <lucer(at)softimage.com>
wrote:

From: Raffaele Fragapane

I don't know how it keeps happening, or if somebody is making a
conscious effort not to read the sense in what's being written, but
nobody seems to want full operators connections in the schematic, don't
know where you read that Luc.


Search the XSI List archives here or use the advanced search form to search across mailing lists. Searching help is available.
This site supposedly brought to you by Benjamin Grosser and the Imaging Technology Group.