View Full Version : Rigging the Spine in max 5
12 December 2002, 08:38 AM
What metod do you like more for animation, spine like max5 tutorial, or maxtd metod or the one there you rotate evryting.
12 December 2002, 12:44 AM
The MaxTD spine rig isn't bad, but I found that I ended up doing too much counter-adjusting with it. SplineIK is decent, but it tends to "snap" around sometimes and I can't seem to make it stop doing that. If you can get past the drawbacks, either of these are still excellent options. Also, take a look at the Spine R&D thread by Eek (http://www.cgtalk.com/showthread.php?threadid=30082) for another possible way to rig the spine.
12 December 2002, 02:08 AM
Thanx, that was good spine rig you have in Spine V2.max...:thumbsup:
12 December 2002, 06:15 AM
Thanks. If you can't figure out how it's put together, I might be able to help. One of these days I'm going to make a tutorial on it.
12 December 2002, 02:43 PM
I hope to se that tutorial soon :thumbsup: , but right now can you point me in right direction I can see all points now...
12 December 2002, 01:23 AM
If you haven't already, take a look at the R4 version of the rig. It's got all the point helpers and everything already visible, which should help in figuring everything out. I've discovered that it's a little hard to explain the rig briefly, so until I've got some sort of tutorial up and running I would recommend just looking at the structure of the heirarchy (figure out what's parented to what) and looking at the properties of all the position and rotation controllers for everything.
I'll try to write up a tutorial over the next week or so.
12 December 2002, 10:11 AM
What's wrong with IKSpline for spine ??? Seems to works just nice, isn't ?
12 December 2002, 04:52 PM
I dont have max 5, so had to find a way in max 4 to do the same thing. Still a few problems though like crunching! and twist issues.
Take a look (http://www.cgtalk.com/showthread.php?threadid=30082)
12 December 2002, 01:31 AM
Like I said before, spline IK is an excellent option if you can get past the little "hurdles" of using it in Max. It just doesn't work completely the way I'd like it to.
The main issues I have with spline IK are twisting and popping. You get two manipulator handles for twist - one for upper, one for lower. The upper works properly, twisting only the upper half, but the lower one just spins the entire chain. That leads to counter-animation, which I don't particularly care to do. And in extreme poses, Max's spline IK tends to "pop" around rather than producing smooth motion. This seems to have nothing to do with the solver settings.
My rig isn't perfect, but at least I can avoid those two problems while still getting a pretty good range of motion in my rig. There certainly are trade-off's in quality between the methods:)
12 December 2002, 10:22 AM
Oh sorry eek :/
LF, yeah that's right :thumbsdow
12 December 2002, 10:33 AM
KiboOst you have great tutorials on you site maybe if you have time you can write one about LFShade spine rigg?
01 January 2003, 08:39 AM
While working on the tutorial, I stumbled upon a way to make the rig work with splineIK. It doesn't solve the "crunching" problem described by eek in his R&D thread (at least not in the way he intended it), but it does fix the twist problem(s). Since it's actually a bit simpler to set up than the Spine V2 rig, I'm going to re-focus my tutorial on this newer version. It'll be Max5 only, though. I'll post the results as soon as the tutorial's done.
01 January 2003, 08:58 AM
01 January 2003, 05:59 PM
I could never get a spine ik to work properly, so I´m using fk.
There´s a version of the spline ik system of max 5 made for max4 by Michael B. Comet (the rigging guru) at www.comet-cartoons.com
There´s also a tutorial on how to make the bones stretchy if needed.
01 January 2003, 08:55 AM
I was checking those spine rigs too. In LFShade's model IK didn't work for the top controller in my case. I think he locked it. Was it on purpose?
Personally I don't think IK spines work well (I'm very new to rigging, sorry). I would like a setup with a ribcage control, though, but I was thinking along the lines of 2 FK controllers - one for the entire spine and one for the top ribcage part. Plus individual manual controls for each bone for tweaking.
I like the hip movements of LFShade's rig, but not the ribcage movement.
01 January 2003, 12:48 AM
Thanks for at least giving the rig a test, dvornik! I don't know what you meant about the top controller being locked, but it's been a little while and my memory of that version of the rig is fuzzy already:p
Perhaps you'd be interested to know that my new splineIK-based version of the rig behaves much better than the old version, and can be easily modified so that you'd have a lower spine that works just like my old rig, and an FK ribcage control with tweakable sub-controls. The only limitation would be that twist could not be controlled per bone (it's a limitation of the splineIK system).
01 January 2003, 08:21 AM
"spine flex - upper" had locks on Move in Hierarchy. That's all. Unlike the original rig.
And of course I'm very interested to see the splineIK spine setup.
01 January 2003, 09:05 AM
LFShade can you post new spine-rigg?
01 January 2003, 07:06 AM
No problem, Namroth, here ya go:
Spine_V2.max - splineIK, works pretty much like the V2 version (http://www.zeetec.net/host/lfshade/spine_v3.max)
Spine_V2_hybrid.max - splineIK with pseudo-FK upper torso (http://www.zeetec.net/host/lfshade/spine_v3_hybrid.max)
Still working on the tutorial here and there. I never realized how difficult it can be to author a clear, concise, easy-to-follow tutorial. I apologize for the delay on that :shrug:
01 January 2003, 08:13 AM
Could you give some basic rundown on important points in your setup? Looks very promising, although I would try to achieve somewhat different functionality.
I think the controller at the hips works in a more natural way with a pivot at the hips. The ribcage would work better if it affected all the spine when it moves same way it does now when it rotates on Z.
It would be nice to have a setup where you could establish a position of the hips and a general direction of a spine first and then adjust the position of the ribcage. Looks like with your method it can be done, I don't know enough about rigging to understand your setup right away.
I worked with a spine that was controlled by a spine controller at the hips wired to control objects on each bone with expressions. The problem with it is there's no IK functionality, so to position the chest where you want it you really have to adjust bones individually.
01 January 2003, 12:42 AM
Download the hybrid rig again; the one I put up last night wasn't set up completely (oops!). The one that's up now has some action in the lower spine when you rotate the chest control, which I think brings it closer to what you might want. Just play with it some, and I think you'll find that the chest is now quite natural and easy to pose.
The pivot for the hip control needs to be where it's at in order to get the swinging action in there. Keep in mind that there's also a control for adjusting the lower spine without the swinging - the small blue circle at the bottom. From what it sounds like, you seem to want a control at the hips that rotates the entire spine, so that you can adjust from there with FK. The problem with that, as I see it, is that you'd end up doing way too much counter-animating. That's not what I wanted, so I didn't include the capability.
Also keep in mind that you're not just limited to rotations. Try moving the yellow/blue circles and the chest control for some different ways to adjust the spine :)
01 January 2003, 02:18 AM
Great stuff. Personally I don't quite see the benefit of swinging motion, but still, overall it has a good potential. The spine bends properly now.
I've made a couple of small changes to illustrate the setup I like. I couldn't figure out how to rotate the spline together with the hip controller, but other than that it's close to the way I would like to use it. I would also like to get an FK-type gradual rotation so the spine bends a bit when you rotate the hips but that may be to much to ask. Maybe not, ribcage controller can be wired to the hip controller rotation.
Great job, really looking forward to your tutorial.
Updated setup (http://home.earthlink.net/~dvornik/cgt/max/spine_v3_hybrid101.zip) based on LFShade's rig
 Oh, one more thing. The fact that the ribcage controller controlls 2 points often results in unnatural spline bends and bone jumps. What's the rationale for 2 points?
01 January 2003, 05:01 AM
The benefit of swinging motion is that that's how the spine moves when you swing your hips and thrust your pelvis. Doing this with FK is just as effective, but it's a bit more tedious to animate.
Your update to the rig illustrates a concept that I was trying to avoid. By having to readjust the ribcage after every rotation of the spine base, you lose the benefit of IK. At that point you might as well just FK animate the whole thing. That's what I meant by having to counter-animate; it's something that makes a rig more difficult to animate with, IMO.
I think I know what you mean about the unnatural bends and bone jumps. I think that's primarily a problem with splineIK in Max, and I've done my best to rig this thing so that it minimizes the effect. Hopefully it's not a situation that will come up very often in animation :)
I appreciate you taking the time to give my rig an honest evaluation, though! I'd also like to give my thanks to Eek, without whose clever idea I would never have devised this thing.
01 January 2003, 05:43 AM
Try to animate the dude stepping forward and bending. Maybe I have to get used to it. I definitely don't think hips should rotate around mid-spine. I know what you mean by avoiding counter-animation, but it also should be not too far from anatomy.
I would like to work in terms of trad. animation more or less (to my limited understanding of it). Define line of action, define orientation of hips and ribcage, position major joints, worry about the small stuff later.
01 January 2006, 02:00 AM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.