Re: Getting SITransform objects from component clusters

[squid <squid(at)3d-palace.com>]

ah, i'm on the same page now :)

digging through your cluster and using the Geometry_V1 subset (wich all 
seem to have localrefframe methods)
is going to be slower than using the 1st edge normal of the poly 
projected onto the normal plane (and crossing to get the z basis).
especially as Geometry_V1 mothods can't be access directly through 
another other language than vb.

i tend to use the latter of the two almost exclusively. not that my code 
should be the benchmark for any aspirations :)

if i had to stick my neck out, i'd say that si uses the 'majic forumula' 
plus a bit of best fit plane.

i'm always happy to be wrong though.
:)
Bradley Gabe wrote:
> Clusters do not have a kinematic node or attached center, but they do 
> have a local reference frame. When you apply a constrain to cluster, 
> that local reference frame must be a full transform matrix which is 
> calculated very quickly, and drives the position, tangency, and up 
> vector of the constraint.
>  
> I could write my own method that converts a component cluster into a 
> point collection, and then there are a number of ways to churn out a 
> transformation from that data, but the performance is probably going 
> to be sluggish.
>  
>  
> There is a command that takes objects as input and spits out a 
> bounding box. I was hoping for an analog that does that for local 
> reference frames (whatever code the SRT box is using internally to do 
> that).
>  
>  
> On 10/6/07, *squid* <squid(at)3d-palace.com <mailto:squid(at)3d-palace.com>> 
> wrote:
>
>     if the cluster has a center then :  (
>     myCluster.CenterReference).Kinematics.Local.Transform
>
>     or ( myCluster.Parent3DObject).Kinematics.Local.Transform
>
>     both seem as nippy as each other to me.
>
>     i have a sneaking suspicion i've not understood the question though
>
>     Bradley Gabe wrote:
>     > If I have a pointer to, say, a polygon cluster, what is the
>     quickest
>     > way performance-wise to get an SITransformation object for that
>     > cluster's local reference frame?
>     >
>     > I'm having no luck hunting this down in the SDK docs.
>     >
>     > Thanks!
>     >
>     ------------------------------------------------------------------------
>
>     >
>     > No virus found in this incoming message.
>     > Checked by AVG Free Edition.
>     > Version: 7.5.488 / Virus Database: 269.14.2/1052 - Release Date:
>     05/10/2007 18:53
>     >
>
>     --
>     Jon Swindells
>     squid(at)3d-palace.com <mailto:squid(at)3d-palace.com>
>
>     ---
>     Unsubscribe? Mail Majordomo(at)Softimage.COM
>     <mailto:Majordomo(at)Softimage.COM> with the following text in body:
>     unsubscribe xsi
>
>
> ------------------------------------------------------------------------
>
> No virus found in this incoming message.
> Checked by AVG Free Edition. 
> Version: 7.5.488 / Virus Database: 269.14.2/1052 - Release Date: 05/10/2007 18:53
>   

--
Jon Swindells
squid(at)3d-palace.com

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