Re: Weird Realflow problem in XSI 6

Date : Tue, 9 Jan 2007 21:36:36 +0000
To : XSI(at)Softimage.COM
From : "Alan Jones" <skyphyr(at)gmail.com>
Subject : Re: Weird Realflow problem in XSI 6
Hi Matt,

This is half the problem and half the solution.

The actual motion vector data in itself isn't an issue at all - it's
right there in the realflow mesh data. XSI however doesn't have any
way to directly say these are the motion vectors for each vertex - so
in order for XSI to discover these motion vectors you need to deform
your mesh by them based on the time. Not a big issue in itself either.

Where it gets complex is your mesh topology is changing, but when?
does it change at frame n or frame n+0.5 or some other point. To solve
this you need to know what time you want the mesh at, as well as a
base reference time (as Halfy suggested). This way the base reference
time is the frame for which you pull the topology, and the time you
want the mesh at determines where your motion vector shifts each of
points so that XSI will discover the motion vectors itself.

I haven't looked into it any further than that - so I've not found
what the solutions are for the remaining issues (nor what all the
issues involved are).

Cheers,

Alan.

On 1/9/07, Matt Lind <speye_21(at)hotmail.com> wrote:
If I have it straight, it's up to the translator to provide motion vectors
for each vertex of a mesh.  mental ray computes an interpolated motion
vector at each ray intersection at render time.  In the case of realflow,
the motion vectors are included (or can be) when you export the mesh from
Realflow and send it to XSI.  This is how we were able to do motion blur for
the milk and other liquids on 'Barnyard' using our own motion vector shader.


Matt

------------------------------
Matt Lind
Animator / Technical Director
SOFTIMAGE certified instructor:
   SOFTIMAGE|3D
   SOFTIMAGE|XSI
Matt.Lind(at)Mantom.net



Date : Tue, 9 Jan 2007
To : XSI(at)Softimage.COM
From : Kim Aldis
Subject : RE: Weird Realflow problem in XSI 6

Possibly Alan knows more about the  way motion blur works between mental ray
and Realflow. It should be OpenSource though really. That way you'd get the
best of everyone's knowledge, in theory at least. If they had any sense at
all they'd release their code to OS. I'm sure they could do it without
letting anything important out.

_________________________________________________________________
Get FREE Web site and company branded e-mail from Microsoft Office Live
http://clk.atdmt.com/MRT/go/mcrssaub0050001411mrt/direct/01/

---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi

---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi


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.