Re: GATORing Point and Edge Clusters

[Raffaele Fragapane <jaco(at)thejaco.com>]

clusters as of now ARE coherent, but they are indexing based (you know 
this, but I'm posting in more generic terms really)
and the trick with materials is nice, but it won't do points and edges 
like in the original request. You can do points transferring WMaps or 
envelope clusters eventually (I seem to remember), but it starts to get 
a bit tricky.
points are easy to abstract, and for polys (that are a lot more painful) 
you kind of have a workaround, but when it comes to edges and samples it 
gets quite painful, and samples are still a nightmare to deal with (and 
sadly they are also the best way to define univocally a set of edges, as 
the owning points and polys system has exceptional cases)

spatial clusters can be very convenient sometimes, a good example of 
that is HDN, where in place of cluster you have groups that can be 
dynamic, which means you can have a cluster of points with an expression 
that will always manage a certain isoparm, or use a lattice that's 
animating a mesh, to also drive  a group in the deformed mesh by volume, 
and apply a smoothing to it.

you can do a lot of that already in XSI with minimal writing and using 
weightmaps, which I know a fair few of us have written propietary tools 
for, but it's just not the same, especially when then you have to jump 
through hops in the SDK because you need to constantly translate and 
cast data.

spacial clusters though fall apart in a number of other cases, as they 
aren't as nice to define arbitrary selections, unless you tolerate a 
considerable data overhead.

HDN's system is the one to take as an example here anyway.
the way groups can be defined procedurally or explicitly  or both is, 
simply put, perfect in every way.

XSI has an infinitely superior algorithmical solution then anybody else 
with GATOR, and there must be an internal rayCaster that's not mental 
ray for things like shrinkwrap etc, which seems to be pretty fast.

working with 4.2 here is painful compared to the few things we can use 
5.1 for.
innovation wise v5 was a huge step ahead in my opinion, all it takes now 
is wrapping and exposes these things in a better way, which I really, 
really hope is the next focus. It would be an excellent situation to be 
in seen what the competition is (not) doing.

congrats if you managed to read through all that, that was my friday 
evening mailing trip, now I'm off to the pub : p

******************************
|     Raffaele Fragapane       |
|     Rising Sun Pictures      |
| "Remember, TD is for TopDog" |
******************************



kim aldis wrote:

> its quite neat in that it transfers in an animated way. Try this:
>  
>     get a cube, put a cluster material on the top face (sorry Bernard ;) )
>  
>     get a sphere, Gator it to the cube and transfer materials.
>  
>     now animate the sphere's rotation through 360 degrees, render and 
> see the cluster move over the sphere to remain matched with the cube. 
> You will need to render it, though. It doesn't updated in playback.
>  
>  
> I'm thinking this would be a good way to get coherent clusters, weight 
> maps and UVs onto metasphere meshes and realflow imports. Sweet, huh?
>
>     ------------------------------------------------------------------------
>     *From:* owner-xsi(at)Softimage.COM [mailto:owner-xsi(at)Softimage.COM]
>     *On Behalf Of *Alan Jones
>     *Sent:* 19-May-2006 10:45
>     *To:* XSI(at)Softimage.COM
>     *Subject:* Re: GATORing Point and Edge Clusters
>
>     Gator does transfer clusters. If you want to get it to transfer
>     them then apply a material to the cluster you want transferred and
>     then they'll be brought across when you bring materials.
>
>     Cheers,
>
>     Alan.
>
>     On 5/19/06, *Matthias Worch* <langsuyar(at)ocrana.com
>     <mailto:langsuyar(at)ocrana.com>> wrote:
>
>         Yes. I can assume that if a point existed on the source, I
>         want to match it to a point on the target - even if the target
>         has considerably less vertex density and that target point is
>         far off the mark. A tolerance would be nice, of course, but I
>         don't think GATOR allows for that.
>
>         --
>         Matthias Worch
>         http://www.worch.com
>
>         b> You trying to match point per point?
>
>
>         >>  -------Original Message-------
>         >>  Does anybody know if this is possible? The cluster doesn't
>         >> show up as a transferrable attribute in the GATOR dialog,
>         and the
>         >> help also doesn't mention it.
>         >>
>         >>  --
>         >>  Matthias Worch
>         >>   http://www.worch.com
>         >>
>         >>  ---
>         >>  Unsubscribe? Mail Majordomo(at)Softimage.COM
>         <mailto:Majordomo(at)Softimage.COM> with the following text in body:
>         >>  unsubscribe xsi
>         >>
>         b> ---
>         b> Unsubscribe? Mail Majordomo(at)Softimage.COM
>         <mailto:Majordomo(at)Softimage.COM> with the following text in body:
>         b> unsubscribe xsi
>
>         ---
>         Unsubscribe? Mail Majordomo(at)Softimage.COM
>         <mailto:Majordomo(at)Softimage.COM> with the following text in body:
>         unsubscribe xsi
>
>
> ------------------------------------------------------------------------
>
> Internal Virus Database is out-of-date.
> Checked by AVG Free Edition.
> Version: 7.1.392 / Virus Database: 268.5.6/338 - Release Date: 5/12/2006
>  
---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi