Reinterpreting Location on an ICE topology "clone"

18 views
Skip to first unread message

Tyler Fox

unread,
Feb 20, 2014, 3:50:58 PM2/20/14
to soft...@listproc.autodesk.com
Hi guys.

I have an object.
For each vertex on that object, I look for the closest location on a second object and store that data.

Now I want to reinterpret those locations on a clone of the second object (using the "Reinterpret Location to New Geometry" node)
If it's just a straight clone, the data is reinterpreted just fine.  But I can't use a clone for this application, so I'm using an ICE tree to copy the topology.  However, the reinterpretation fails on that ICE clone.

Does anybody know of a way to make this work?

~T.Fox

Alan Fregtman

unread,
Feb 20, 2014, 4:17:04 PM2/20/14
to XSI Mailing List
Do you really need to copy the topology? Maybe you can clone, freeze modeling and use ICE to match pointpositions? That has worked for me in the past.

Tyler Fox

unread,
Feb 20, 2014, 4:30:51 PM2/20/14
to soft...@listproc.autodesk.com
I'm trying to keep the relationship between the original and the clone, so freezing is bad in this instance.

I want to do it this way so I can replace the original with anything I want at any time, and I won't have to rebuild all the crap I made on my cloned object.

I mean, unless somebody knows how to point the input port of a CopyOp to a different object?

~T.Fox

Eric Thivierge

unread,
Feb 20, 2014, 4:34:16 PM2/20/14
to soft...@listproc.autodesk.com
CopyOp can be made in ICE using a the input mesh > get
data['pointposition'] > switch context > set data['pointposition']. You
can then make a compound and swap the input as needed.

Eric T.

Eric Thivierge

unread,
Feb 20, 2014, 4:34:38 PM2/20/14
to soft...@listproc.autodesk.com
This is, if the geometry is a duplicate.

Tyler Fox

unread,
Feb 20, 2014, 5:13:22 PM2/20/14
to soft...@listproc.autodesk.com
Ooh, I just got an idea.  And my test works!

Instead of taking meshes and getting the relationship data between them, I add one more layer.  So I have an object that just sits there being the "master",  an ICE topology clone being the "chooser" and then an actual clone of the chooser being the deforming object. Then I calculate the locations data on the chooser objects (instead of the master objects like I was before).

This allows me to use the location data at any point later in the tree because I'm still using the regular CopyOp where it counts.

Thanks for letting me talk at you guys!  :-D
~T.Fox


Reply all
Reply to author
Forward
0 new messages