Re: spacer rigs
| Date : Sun, 4 May 2008 18:57:47 -0500 |
| To : XSI(at)Softimage.COM |
| From : "Bradley Gabe" <withanar(at)gmail.com> |
| Subject : Re: spacer rigs |
On Sun, May 4, 2008 at 6:00 PM, Eric Thivierge <eric(at)xsidatabase.com> wrote:
Hey Brad,Quick question for you, if you rig to Nulls and not to bones, do you have a specific setup to parent them or constrain them to the bones of a rig? Or is it just straight constraints?Thanks,Eric T.
Depends on how you set up your rig. I tend to split my rigs into 3 branches using mostly pose constraints to wire the connections:
- Animate
- Deform
- Geometry
My entire Deform branch includes the envelope nodes, most of which are constrained to a control system in the Animate branch of the rig. But then, you saw this setup at the Production Challenge. :-)
The advantages of this kind of abstraction include:
- I can plot fcurves on my deform rig, then delete the Animate branch of the rig
- I can mesh cache performances to the geometry, and delete the Animate AND Deform branch
- I can swap out animation controls in the middle of a shot, and my deformation should still all work
- I can swap out or enhance deformation controls in the middle of a shot without affecting the animation
- Many 3-node IK-chains are replaced with single nulls, thus reducing the total number of nodes in the rig
There are more, but these are the main advantages.
Eric ThiviergeQuoting Bradley Gabe <withanar(at)gmail.com>:A side not to Dave's question:The static kine state parent technique is something I demo'd at the TD Lovetour a few years ago, and also at the Production Challenge last year, andprobably will demo agian at the next Production Challenge.I've found it useful for a number of effects, but none so much as allowingyou to interactively find a joint placement sweet spot.Let's say you've been tweaking the weights of your characters thigh to hipsarea, and no matter what you do the geometry simply crunches too much whenyou lift your character leg. If you do all your envelope weighting on nullsrather than bones (which I do for many reasons including this one) you canvery quickly give your thigh deformer a parent with linked static kinestate. Then lift your character's leg again and translate the thigh deformerparent. In real time you will see how changing your joint position affectsthe resulting deform, and you can very quickly hone in on the best spot.I've found that this technique alone has allowed me to get much betterdeforms out of thighs, biceps, elbow, knees, so much so that I've been ableto simplify my rigs and drop a few secondary systems I used to use for shapecorrection.-BradOn Sun, May 4, 2008 at 1:01 PM, Steven Caron <carons(at)gmail.com> wrote:i guess he didn't provide a link actually...stevenOn Sun, May 4, 2008 at 10:54 AM, Steven Caron <carons(at)gmail.com> wrote:> Also Raffaele did you get a chance to read the xsi-blog eric linked you> to? Have you downloaded the project files to see what it does?>>>>> On May 4, 2008, at 10:06 AM, Raffaele Scaduto-Mendola <> raffaele(at)turbolinea.com> wrote:>> Its a bit funny to hear I have maya-centric lingo since I have been a> > softImage user since 1994.> > Just have a talk with Helge, since I asked him to such a tool when I> > was character sup on Barnyard, an XSI production.> >> > RSM - character setup "masochist".> >> > Steven Caron wrote:> >> > > I am having a hard time understanding what you are describing> > > raffaele. It might be all the maya centric lingo.> > >> > > I think if david uses a higher level description or a better example> > > with context we can help get him what he needs.> > >> > > s> > >> > >> > > On May 2, 2008, at 11:26 AM, Raffaele Scaduto-Mendola <> > > raffaele(at)turbolinea.com> wrote:> > >> > > This doesn't solve the problem, but I did want to mention it should> > > > any coding guru/plugin expert might be following this thread.> > > > I have also been thinking about this problem for a bit. In my> > > > previous jobs I had access to custom deformer that offer a base.> > > > You could then rig/deform the deformer without moving the base,> > > > and you would get a normal deformation, you could deform the base only so> > > > the mesh would transform through the deformer, or you could move both and> > > > the mesh would not deform.> > > >> > > > If you ever take a look at the SharkTale pirhana or sea horse> > > > crash sequence you can see the result. I used this setup mostly to> > > > blendshape deform the character to its basic shapes but then have secondary> > > > "influence like" deformers to control the lip, and have the control move> > > > (but have no effect until you moved them ) with the blendshapes.> > > >> > > > Influence deformer from my understanding are global based, so the> > > > mesh heiarchy doesn't affect them.> > > > Now what would be nice, ( and maybe theirs a way to do this using> > > > moondust), is to have a optional "influence base" you could optionally> > > > connect to any given influence.> > > > Then you could get that control of when and when not to deform.> > > >> > > > Just some tips, that might help figure this out.> > > > RSM> > > >> > > >> > > > Eric Thivierge wrote:> > > >> > > > > Hey David,> > > > >> > > > > If I'm understanding you correctly, have you looked at Steven> > > > > Caron's post on XSI Blog about dorritos? Is this what you're looking for?> > > > >> > > > > Thanks,> > > > >> > > > > Eric Thivierge, XSI Database Admin> > > > > www.xsidatabase.com> > > > > eric(at)xsidatabase.com> > > > > Forum Username: EricTRocks> > > > >> > > > >> > > > > David Gallagher wrote:> > > > >> > > > > >> > > > > > Oh, I guess one method would be to parent the second mesh and> > > > > > all the joints as a whole to the traveling mesh, but I'm looking for a way> > > > > > to more discretely parent controls around the body.> > > > > >> > > > > >> > > > > > David Gallagher wrote:> > > > > >> > > > > >> > > > > > > Does anyone have any ideas of workarounds for "spacer rigs".> > > > > > > It's common in the Maya world at least to rig a second mesh and pipe in the> > > > > > > result as a live blendshape. So, the second mesh is skinned to joints, for> > > > > > > instance. Some of the joints, you might want direct control over, but they> > > > > > > can't leave their location. In order to get them to ride with the main rig> > > > > > > but only engage when actually directly used, you make "spacer" controls that> > > > > > > mirror the actual joints, parent them to some part of the body, then pass> > > > > > > their numeric transform values only back to the real joints sitting at> > > > > > > global center.> > > > > > >> > > > > > > I think this is pretty standard, but I've heard of> > > > > > > alternatives, such as inverting the shape from certain deformations,> > > > > > > cancelling out the unwanted deformations.> > > > > > >> > > > > > > Anyway, I'm interested in any alternatives to spacer rigs> > > > > > > that are used.> > > > > > >> > > > > > >> > > > > > > What would be optimal, is if you could transform the same> > > > > > > joints in two different ways. One which just changes their location in> > > > > > > space, and another which makes the deformation.> > > > > > >> > > > > > >> > > > > >> > > > > > ---> > > > > 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> > > >> > > ---> > > 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> >>Administrator, XSIDatabase.com
---
Unsubscribe? Mail Majordomo(at)Softimage.COM with the following text in body:
unsubscribe xsi
- Follow-Ups:
- Re: spacer rigs
- From: Eric Thivierge <eric(at)xsidatabase.com>
- Re: spacer rigs
- References:
- spacer rigs
- From: David Gallagher <daveg(at)blueskystudios.com>
- Re: spacer rigs
- From: David Gallagher <daveg(at)blueskystudios.com>
- Re: spacer rigs
- From: Eric Thivierge <eric(at)xsidatabase.com>
- Re: spacer rigs
- From: Raffaele Scaduto-Mendola <raffaele(at)turbolinea.com>
- Re: spacer rigs
- From: Steven Caron <carons(at)gmail.com>
- Re: spacer rigs
- From: Raffaele Scaduto-Mendola <raffaele(at)turbolinea.com>
- Re: spacer rigs
- From: Steven Caron <carons(at)gmail.com>
- Re: spacer rigs
- From: "Steven Caron" <carons(at)gmail.com>
- Re: spacer rigs
- From: "Bradley Gabe" <withanar(at)gmail.com>
- Re: spacer rigs
- From: Eric Thivierge <eric(at)xsidatabase.com>
- spacer rigs
| DATE: | << | >> | THREAD: | << | >> | INDEX: | Main | Thread |
|---|
- Previous by Date: Re: spacer rigs
- Next by Date: Re: spacer rigs
- Previous by Thread: Re: spacer rigs
- Next by Thread: Re: spacer rigs
- Index(es):
| Search the XSI List archives here or use the advanced search form to search across mailing lists. Searching help is available. |