Raf Anzovin

Oh, yeah, there are all kinds of pass throughs and wierd stuff in the movie at this point--we're just getting to the point where most of the animation is done and we can focus on the details. The vibrating trees, BTW, are caused by rendering a bone dynamics simulation through Netrender. We'll have to bake the tree animation first before we do final renders on those shots. --Raf

Ah, but we never did remove it. While we haven't made it as obvious in the TSM rig (because that isn't how most people like to do it) you can still pull the hand off and use it in exactly the same way. You just need to hit the n key to move so you can move the hand control despite the fact that it's chained to the arm bones. --Raf

Sorry for taking so long to get back to you all about this stuff...I've been very busy lately, and I was in New York for the Robots premier recently. (I'll post my thoughts on that in a week or so when everyone else has had a chance to see it). Thanks, Martin! That was something that was quite often on my mind, actually. I think in this case it just comes down to different eyes. To me, that move does look natural. Just a difference in opinion. A remarkable amount of it is just made up on the spot. I didn't have a very good plan for this shot, only a general idea that they had to end by facing off with each other. The animation method I employed here is pretty similar to the technique that Shamus Culhane talks about in his book "Animation, Script to Screen." I started at the begining and I did what seemed natural. Then, of course, there's the whole process of refining what's been animated and getting comments from other animators, which is essential to giving it a polished feel. But the basic movement was done with a streight-ahead keyframe method, without reference to anything else. Thanks! The styles also rely very much on what kind of weapons the character's use. Black is a rapier fighter, so he does a lot of quick stabs and cuts less often. Red has two curved swords, so he tends more twards spinning moves and big swipes. That's changed a bit from what we originally intended. Originally, Red was supposed to be more like a broadsword fighter, but that didn't really feel natural with his design. Yup. Now that I've got out from under all the stuff I was busy with, I'm starting a new thread in the OT section. --Raf

And the latest addition: a finished render http://www.anzovin.com/finalshot.mov --Raf

Yup, we've been shooting a lot of our meetings and sweatbox sessions. Right now it's looking like, rather then a DVD, this is going to be a book, which will come with a CD including the documentary we're making from the footage we're shooting. --Raf

Dimos; I think we're basically both saying the same thing, just from different angles. You know, it really would. I think I'm going to try doing it this way the next time I do a short and see how it works. I do see it slowing things down a bit--having to constantly pass shots back and forth between animators has got to make everything a little more time consuming. (After all, the animator has to polish one character in the shot till the director approves--then hand it off to another animator for the process to start all over again). But I imagine it would still be more then worth it for consistancy of acting on a feature. Also, I'm guessing that there are little spontanious acting bits that are likely to grow from this kind of colaberation by animators that never would if only one animator works on each shot. Ah, but doesn't that start the whole vicious cycle all over again? Money people like my ideas, money people give me money, money people own my soul? Then they get to suck all the life out of my project like we were just talking about. But.....there may be other ways...... And I DON'T feel like I have to go do it RIGHT NOW, or I'll somehow lose the chance. The next ten years are going to be a very exciting time for animation. Things are going to be possible that were never possible before. I'm only 23. I'm at the begining of my career. Nobody ever did anything great by rushing into it at top speed. I've seen some people get into features who--and this is just my opinion--kind of regret that they did. It took something they really enjoyed doing and turned it into a nightmare that they could barely control. To direct an animated feature--to direct a great animated feature--takes the kind of artistic control of a Brad Bird. It takes time to develop that control, that consistent vision. Thanks! I have a whole lot of ideas about what could be done with animation, but I don't think this is really the right thread for it. Perhaps we should create a thread in the OT forum? --Raf

Hmmmm....interesting. Just by complete coincidence, Terry Rossio happens to have been one of the co-writers on The Road to El Dorado--although what he and his writing partner came up with originally was apparently much better then what found its way to the screen. In any case, I don't think I'm waiting in quite the way he means. He's not suggesting you should strive to do things when they are not yet possible, but that you should sieze the opportunity when it does arrive. A certain amount of pragmatism is neccessary if one is not to wear oneself out with quixotic quests. There's plenty to do in the meantime. --Raf

You got it exactly right. Unfortunately, that's largly what I meant by "animation process." To many, those practices have become inextricably linked with the making of animated films--and the ironic thing is that some of them, committee storytelling included, do have roots in the way the old Disney studio operated. So they can sometimes look like wisdom handed down from on high. I think it worked for Disney because he was both a guiding creative force, and he owned the whole thing, too. He could, personally, keep a handle on the madness. Don't worry, nothing confrontational about it. I do have more connections and resources than most do, it's true--but still nowhere near what would be necessary to get into features. Not yet. Well that, I can do. Whenever feature animation does become feasible for us, I plan to be ready. --Raf

There are two kinds of IK you can use in A:M. There's the automatic IK that just lets you pull around bones quickly. This is usefull but it doesn't react to anything else you do--ei. it won't let you move the body around while keeping the feet planted in place. It only calculates IK while you are actually manipulating the bones. For feet that stay planted in place, you need a Kinematic constraint, which will allow the IK to be calculated constantly. Use the Kinematic to attach the legs to a bone or null, which you will be able to move around to animate the feet indipendently of the body. --Raf

I've always wondered about applying this idea to CG. It seems like it would add another layer to the acting to have each animator focus exclusively on one character throughout the film. Unfortunately, it's not cost effective. By contrast, the entire animation team on Robots was just a little over 30 people. Yeah....I'm continually frustrated by all the cool things that could be done with animation and just aren't. Pixar is great--but they're just one studio. They have a certain style, and they won't depart from it to try something drastically different. And they seem to be the only people with a real grasp on story. In the animation industry, we constantly hear the mantra "story, story, story." Only, it doesn't seem to make the stories any better. People agonize over them for months, and they're still weak. I have this sneaking suspicion that the "animation process" everybody uses just plain doesn't work. That it hasn't really worked since the days of Walt, who could pull it together, probably, just from sheer force of personality. It's a relic from sixty years ago, and today what the feature animation industry needs is really simple. It needs to hire some really talented writers with individual voices. And then it needs to listen to them. Hmmmm....that ended up being a bit more of a rant then I intended it to. Some day, I hope I'll be in a position where I can do something about this. --Raf

Well, if it makes you feel any better, Paul, I wasn't born knowing how to rig. (No born geniuses here, unfortunately). I spent something like six months playing with rigs before I really figured it out. The real difference is that, unlike most people, I LIKE rigging, so I was willing to spend a lot of time on it. The best advice I can give you is to have patience. Rigging up a character from scratch in six hours when you've never done it before would be a feat indeed. Even people very experienced in rigging often take a couple of days, at least. It will take you more time then that to get the hang of it. Everyone's advice here is good: start with something simple--maybe just put the bones in and leave them alone for eveything but the legs, which should have Kinematic constraints for IK. This won't be a particularly elegent rig, but it will be usable. And you can build up from there. --Raf

Pretty soon.....we're close to having some shots completely done. True.....but I'm going for that frenetic fighting look. Similarly, you can't always pick out every move in a Jackie Chan or Jet Li movie. I don't really expect everyone to be able to register every move the first time they watch it--it should still be able to put over an overall sense of the kineticism of the fight. Also, a full soundtrack helps a lot. Yes, well, Black has two secrets. One is that he's not really right-handed. The other is how he does that. Men have died trying to find out..... It's all toon rendering without lines, and with different shading gradients. There actually aren't any shadows. Cool. Although I was never a big fan of their storytelling, I thought Dreamworks Animation had a very interesting style going there for a while before they switched to all-CG. What did you do on Road to El Dorado? Thanks! If this project was bigger, we'd probably do some kind fo simulation for that, but at this size it works to just do it by hand. Sometimes, animating little secondary things like that is a lot of fun. --Raf

And here's another near-complete shot. Like the other, it's a work in progress: still a few pass-throughs, temp background, and incomplete facial animation. But lots of swordplay. http://www.anzovin.com/duelfighting.mov

The original Setup Machine did it in a very similar way. There have been a lot of improvements with v2 (stretchiness, for instance), but the basic principals are the same. --Raf

These kinds of spines are called Schliefer spines by a lot of people, after Jason Schliefer, the guy who rigged Gollum (and I probably just totally mangled the spelling of his name too). They're immensly cool. You haven't lived till you've animated with a stretchy Schliefer spine. I'll never use FK for a spine again. For our version, we had to do it entirely with constraints, because the TSM scripting language doesn't support complex relationships. So, in our case there are really three bones effecting the actual spine geometry bones. The Pelvis is the parent of the whole thing, the Torso is at the other end, and in between is another bone halfway up the spine. This other bone is suspended between the Pelvis and the Torso through Translate Tos and Orient Likes, and the spine is then attached to these three bones with a set of Orient Likes that hands off influence between the three all the way up the spine. Then, for stretchiness, we simply add a Translate To, with Compensate offsets, from every spine bone to the Torso and have the influence grow as it moves up the spine. That's all linked to a slider so that you can turn the stretching on and off. One of the things I like to do with it is to completely ignore the Body bone--the one that moves everything--and just animate the characters position with my two spine bones. This is pretty cool, since you can do offsets between when the pelvis arrives at it's key and when the torso does. I think that's where rigging is headed--total isolation for everything. I really saw that taken to the farthest extreme with the Chicken Little rigs they showed at Siggraph. --Raf