Lars Kjetil's student blog

So, yeah. I was supposed to take today off from animation and all that, but I got incredibly bored and decided to challenge myself with setting up a ballerina pose as quick as possible, as they are sort of complex, but easy to read and remember. Sure enough, I got the pose roughed up fairly quickly for one that has not done this much before. (Well, the rig is easy to use; what can I say?)

But then I noticed something: I could never get it to quite work as I wanted, and the posing slowed down and ceased to be speed-posing. It always looked rather static, and never quite like it was supposed to be. I realized what I think would make everything so much more dynamic, and possibly save the pose: Bends. (Or whatever they’re called: I’m sure the pros have a more suiting name for it)

While this is nothing but a quick exploratory suggestion, I’d love to hear from you suggestions on how you would solve the same problem (or just this one pose), with or without post-kinematic bending, or even why you think it’s a good or bad idea.

I can imagine this working really well in movements with lots of follow-throughs/overlapping/inertia/whatever. It’s probably something we’ll touch in the basics (I mean, I’m pulling this from good ol’ cartoons after all!), but I’d like to explore and understand this concept deeper, what it does, why it does it, and how it does it. Perhaps take a look at Mrs. Incredible again? =P

(Now that I think about it, I think I recall reading that it could be used for even minor movements to make it easier on the eye?)

Just my two cents.

Our task for today – as far as I understood it – was a small precursor to the first project. We’re to make a quick heavy-lift animation, while documenting everything we do and how we do it. A rather refreshing start, in my opinion. I wanted to see if I could use my old rig I used last year –

But, alas, I couldn’t get my hands on the last version, and it had severe problems with skinning and missing digits as well. (Not to mention, his pants are missing)

Nevertheless, I tried posing it quickly.

It worked, and I’m somewhat used to this rig, but I didn’t want to risk it breaking down on me completely, so I got that other max rig thinger and set up my UI like I prefer animating with it.

That is, I like having access to the full curve editor, at least when fixing the animation, and have a flat shaded small view-port to keep an eye on the silhouettes (Wish I could make it black or something, so I couldn’t see the eyes). I haven’t begun animating yet, but as I’m going along and posing, I’m trying to find an attitude, or a personality. I think up traits and small gestures and secondary motions that could be performed to enhance it, such as per haps edging his fingers towards a more comfortable holding position, or just trying to nudge the ball to one side to get a better grip on the other, or things about the way he goes about bending down, et cetera. Perhaps I could even give him a feminine attitude. (Naw.)

And to be honest, I haven’t really bothered looking into planning the whole ordeal very thoroughly – especially not for this task – where the goals will be obvious with only a few key poses. I’d make him act it more out if the task was bigger and all that; make him study the ball a bit, perhaps, relax his shoulders before going down, and sway and stutter and try to keep his balance while not crashing into anything. In other words, I’m not using many external tools such as sheets and drawings to figure this one out: I believe the three view-port panels I have at the moment to be enough. No reason to over-complicate it.

Since this rig is using custom attributes, and those are a pain to convert tangents on en masse, some of us guys here in Stavanger, figured that for us who enjoy using step mode, simply staying in normal tangent mode for now and cramp it together so that there’s one key pose per frame, would be close enough. (Seriously, is there no good way to convert the tangents on custom attributes and the likes en masse, like you can with controllers?)

As said, time is compacted down into only the necessary keys, and to keep things simple and “final”, I always select everything and press “k” on my keyboard to put keys in everything so I don’t have to worry about that when I start timing (hoping that the custom attributes won’t cause much trouble).

I’ll often feel tempted to add in one extra key somewhere to remind myself of subtle movements I want, but I’m starting to learn that this doesn’t help at all, and only makes things harder. Better jot it down somewhere else! I’m hoping that working this way – focusing on key poses and possibly some break downs – will let my animations be more focused on beats and rhythm and readability, rather than make them mechanical and stiff and horrible to read; but I honestly don’t know until I’ve tried some more. (It’s not like I could read that out of a dope sheet or something like that [yet, if ever].)

Anyways, I’m currently not very innovative with the use of tools and such, but I figured sticking an object on the spheroid for when I moved and rotating it would help me figure out the placement of the hand in accordance.

I hesitate to use link constraints on the hands for two reasons: One, since I’m not used to this rig, I’m afraid it would break down. Now, that’s easy to figure out with a little bit of experimentation (Hold & Fetch, anyone?), but then there’s: Two; I want the interaction between the ball and the hands to be organic and non-constrained. I’m sure someone good would be able to do this without making it look like the ball suddenly got ultra-light or something like that, but I think I’ll keep it simple here.

One thing that I noticed while using this rig, is that how I animate a character is very biased by how it looks: Even if this character looks very plain, it looks – to me – as if it already has some color to it. Or predefined personality, if you’d like. I don’t know if this is a good or a bad thing, or if this is one of only a few that does this, as I’ve already seen this used a bit elsewhere on the net.

Anyways, curves are being made.

And then some breakdowns. I use “k” to put keys on everything there, as well. I like having the keys “finalized”, so I don’t have to worry about unwanted blending. Only thing is, I should be pretty sure about the surrounding poses before I lock the breakdowns/inbetweens, so I don’t have to fiddle too much with the curves.

This is when I realize that I probably should have timed it before adding more stuff. And so I remove all the keys I just added that I haven’t changed again, and start spacing them out roughly. Very roughly, as it’s getting late, and I’m slow, and the deadline is soon. Silly me.

While I animate and notice how I forgot about inbetweening the fingers and such, it occurs on me that perhaps I should animate in passes? But that clashes slightly, perhaps, with my preferred method of animating, much like Chris Kirshbaum did in his Character Animation: Mechanics and Emotion over at ‘The Gnomon Workshops’.

Anyways, here is the story so far:

Meet Guy: He sees a spherical object. He’s incapable of expressing his wants with his face, because this project strictly prohibits the use of facial animation. Nevertheless, his urges are as strong as any one’s – with or without a face – and he really, really want to pick it up and hold it close to his chest.

So he decides to pick it up!

Hang on …

Just got to get a good grip, first …

There we go!

Hnggh!

Nnnnh~

Watch your steps, Guy!

Good catch, Guy!

The ball is yours now, Guy! Oh, joy!

Well, anyways I think that’s all I have time for today. I’ll bring it with me home, but we’ll see if I’ll get anything more done. It’s just playing with the curves from now on, anyways. I suppose I need practice with that, too. Anyways, got a train to catch, see you all la

The red circles are the keys we found, while the crosses are those that are copies of other frames or doesn’t show up in the loop at all.

These are the key frames we could agree on. We found three poses that we considered to be “key” to the animation: the first extreme, prior to the anticipation drawing. (Besides, it’s of slightly better quality than the rest ;O) The second one was when the impact had been firmly established and he once again changes direction. We figured the last one was necessary as an extreme as well, to complete the animation and all that jazz.

Then we watched The Jungle Book from 0:30~0:46. And here are what we found.

It was hard getting the exact frames we wanted, but you get the idea.

P.S. Teacher, give feedback. :V

In this post I’ll talk a little about what I expect out of myself throughout this whole year at 3DFP.

Ideally, I’d like to experience every aspect of a production. Anything from brainstorming and research to visual development and storyboarding/beating, cinematics, sound work, scene layout, composition, modelling, texturing, skinning, animating, simulations, integration, and so on. In other words, I’d love to be able to make short movies on my own, or even be able to be a one-man studio or a somewhat (but not too much) all-round guy. Realistically, I know that I will have to focus on a set of skills rather than all of them at once, and so some aspects may suffer for the advancements of others.

Assuming we get a really good workshop going on, and have some sort of system, I believe that working together with others – either as a team, or if that’s not viable, through constant polling and criticism – will let us get even closer to that complete pipeline product, not to mention the obvious increase in efficiency (depending on who you work with) and man-hours. Simultaneous asynchronous development and happy accidents could really help spice things up a bit and give it a more organic feel. I’m all for sandboxing; But I’m also all for keeping to that initial red line – that thought, idea, message, feeling or what ever you want to call it that possessed you to make what you’re making, that you want to so much to share with the audience. Or even – as someone else said – “To dream aloud.”

What I’m hoping the most, is that there’s enough time to go into the more technical stuff, so that we are in fact able to accomplish all this. Of course, I will also need to practice on the aesthetics, possibly even more so than the technical parts. Yes, I realize that riggers would hate to receive a horrible mesh to work with, but you have to start improving somewhere, and I think it’s what ends up on the screen that is the most important, with group morale following in very, very close on second place, even surpassing it at some points.

One thing that have bothered me is that I seem to have used 3Ds Max less and less lately, instead using stuff like Zbrush. These are entirely different packages, and Max is essential to what we’re doing. I really should squeeze it back in to my daily use-tools for more and varied use. I feel like I have a lot to catch up on if I’m ever going to get to compare myself to professionals who have many years of experience in their field. When I was little, I used to be a higher-than-average generalist, as long as I liked it, with relatively little effort (either that, or most others didn’t try hard enough); yet these days I can’t grow fast enough. I’m by no means a genius at what I do. But things, as they say, take time. And patience. I’m nowhere near considering giving up. Anything I learn may somehow help me in the long run. (… and thus I become a generalist again?)

Anyhow, on the more technical side, I spent some time looking around at alternatives, and found that 3Ds Max is still my preferred (although not perfect, of course) platform for 3D and animation. So other stuff like Maya and whatnot will have to wait until I got free time. Last year, I also didn’t explore plug-ins and such tools nearly as much as I could. I’m hoping to start making my own textures and sounds and HDRi maps, and OpenEXR for composition, and …

Anyways, for this year, there’s a whole lot of inspiration to be found out there, and I intend to find all of it. /END CHEESY

Actually, it’s not as much 3D as animation and CG in general that has pulled me into this line of study. Disney has always fascinated, to mention the obvious one. I also thoroughly enjoy researching lots of stuff like, say, planetology, which teaches anything from geology to chemistry and biology and whatnot, which could serve well in almost any fantastical setting. The fantastical as well as the realistic – often interchangeable – is probably my main drive. You could say I enjoy science fiction the most, but everything inspire.
I enjoy anything from set extension to flash animations and rotoscoping, writing, concept, * design, compositing, lighting, etc. I do enjoy all this, but I don’t think I’m good enough just yet to start with the neon arrows pointing to me, saying “3d artist” – I have yet to find an area to specialize within, as well, in case I end up in some pipeline work – after just one year of beginning studies. (3DDA, Noroff Stavanger)

I don’t have anything to show off at hand at the moment, other than perhaps five hours of me staring (and sculpting slowly away) at a spider-like scarab-ish ant-esque while listening to weird, random music that nobody complained about at http://www.livestream.com/nukyface. I barely knew/know Zbrush. I’ve worked some more on the creature since those days, and just got ZBrush 4, but don’t know how far to take it – if I should go through the whole process up and including perhaps camera matching and animating it.

I’m hoping this year as a web student in a classroom together with some of last years’  classmates will be most of all a creative thing. As long as I can be creative and properly convey a story or a meaning, I wouldn’t care whether I ended up in big production pipeline or grass root projects, living on bread and water.

Running the risk of spreading things a little thin, I’ll close of this post with an apology for the lack of eyecandy. Empty hard disks and all that.