Animation Brad

I’m proud to announce that our team, The Fjantastic Fjorgers, won second place in the all-night 32 hour animation competition! The contest involved sixteen three-person teams of CG animators from around the world who forgo sleep and resist several staged distractions for 32 non-stop hours to produce the best character-driven animation.

This was our take on the saddest story ever:
Fjantastic Fjorgers Fjinal Fjilm

You can see all of the films here:
http://www.siggraph.org/s2008/attendees/fjorg/

Anyone attending Siggraph 2008
should definitely come and check out the animation event Fjorg!!

This is a viking themed, iron-man animator competition where 16 teams will be battling out an animation in 32 hours over the next two days. Our plastic swords have been sharpened, the bone horns of our helmets have been shined and we are ready to animate!!!

Last week I was invited to do an article for Pixar Planet. I went to experience Wall-E and the live character show at the El Capitan Theatre in Hollywood. Obviously, I had a great time, I love the theatre, the film and ice cream. You can find my complete article here.

There was a few things I noticed in context
that I personally found amusing:

The El Capitan, like most theaters has moviegoers overload concessions, a large popcorn and a drink were in just about everyones hand. Considering one of the messages in the film, I chose to pass
on the “lunch in a cup”.

What really irked me was what followed. I grabbed my plastic water bottle and went to leave the theater. A worker came down
with a trash bag and glanced at me expectantly.

“I was going to put this in the recycling,” I said,
politely wondering where to place my number 2 plastic.

“Oh we don’t recycle, ” he responded, with a half-smile
like he knew what I was thinking.

This is the theatre that CORPORATE DISNEY owns.

Did they WATCH the film at all ??

Do they realize that THEY, among others,
are a reflection of Buy N Large?

The theatre has a capacity of 1000 seats and five screenings a day. If 75% of moviegoers have a plastic bottle and a plastic popcorn bucket, that is 30,000+ some plastic bottles per week going to a landfill,
coming from the El Capitan Theatre alone.

This weekend 250,000,000 dollars were spent at the movies. Divide by the national average of $6.00 a ticket. That is 42,000,000 some odd people that went to the movies. How many of them had plastic bottles?? How many of those people went and recycled their bottles? Did you?

I’ve been recycling since I can remember, we had recycling education program in my elementary school in the 80’s. This isn’t news. Our school campus where I teach animation finally started a recycling program, as it is the corporate trendy thing to do these days. Better late than never. I guess it’s ok that it’s ‘hip’ as long as it stays and encourages others to do the right thing. It’s definitely time for movie theaters to remove the blinders and realize the amount of trash that they are creating. Watch your own film.

Here is a shot where birdy moves in a quick movement. For this test I’m only looking at frames 688-698. Nothing is polished and clearly there is a lot of work on all of the examples.


Now, this movement is a bit jerky. The head movement doesn’t work for a bunch of reasons. The main point here is that the head makes a squared off path, from 692-693 there is no real curve or arc to the bottom of the move. Some of the reason this doesn’t feel right is that 691 and 692 the bird is frozen. On paper, this would be fine, we call it ‘on twos’ and it cuts down the number of drawings needed. This doesn’t solve the arc problem, but the number of drawings and poses should be fine, and the movement should be ok, and still represent the action. So it should be ok to freeze the model for a frame here and the eye wont notice.

Smoother animation on paper is done on ones, one new pose for every frame, so there is enough drawings to represent the movement. In cgi animation, ‘all’ animation is done on ones. The answer is to move the bird at 692 so that the movement is smoother. Now because of the fast action, the computer didn’t know where to place this inbetween, so I had to do the adjustments to get the arc working better. (body only-don’t look at the feet or wings) The arc feels smoother things are looking better problem solved. when rendered the computer might add a bit of blur to the head on 693 so the movement might feel even smoother. With an inbetween posed on each frame:

The question is can you still have the animation work ‘on twos’, have the bird not move from frame 691-692 and still make the action feel like a nice round arc? Yep.Take a look here:


Look at 691-693. The bird doesn’t move from 691-692. This shot uses extra body parts to define where inbetweens might go to smooth out arcs. This is common in fast zippy actions in traditional animation, and usually incorporates extreme squashes and stretches on the form and shape of the character.
So which is right? or better? or easier?
Well, they both solve the problem ok, as long as the movement ‘feels right’ is what is most important. Easier isn’t ever an option in animation, in both cases I had to manually place something somewhere to get that arc.

All of the books I’ve read on animation mention smears and blurs and adding multiple parts to create fast actions, but none of them really say when or where to use it correctly. It’s rare to find it in computer animation, I’m guessing because the default is to have a new image for each frame. I’ve seen multiples used on traditional animation on ones though, so I’m not sure why the techniques isn’t used more. I’m hoping it doesn’t go the way of the dodo with traditional drawn animation.
I’m also hoping that there will be more answers from the master of the technique himself, Eric Goldberg, in his new book, Character Animation Crash Course.

On to one of the more challenging physical shots. Here is a more ‘refined rough’ of where I’d like the bird to fit in. It looks like I planned the two characters separately. That is not the case at all. This shot was planned with both characters in mind as if I were drawing both characters at the same time for each frame. The dog character of the shot was animated first so that I could work on something while I was getting advice on modifying the bird rig. In a way, it has been easier for me to know where the dog might be before the bird gets placed, but the shot was conceived with both in mind from the beginning. Both are still mailable, meaning the timing of the dog animation might need to change depending on the timing of the bird and bird wings. It’s my intention to pose the bird on the same frames that I posed the dog, so even from character to character everything is all keyed on the same frames.

Recently, I was accepted into the Fjorg! Iron Animator Marathon Competition at Siggraph 2008 in LA.

I’m on a team with Valerie Morrison and Sean Coleman.

FJORG is a competition where 16 teams each work to create an animated sequence of at least 15 seconds in length to a maximum length of 45 seconds based on a theme to be provided at the main event, during 32 straight hours, using only those assets supplied at the event and their own talent and skills. 32 hours of animation goodness.

AnimationMentor has twenty students/alumni participating in seven teams of the 16 selected for the event. It’s going to be great competing head to head with other AM folks and other students from around the country.

One thing that is important to do as an artist/animator, and for people in general is to take inventory. For animation, that means combining shots together and creating a new reel. This helps you keep current, shows how much you’ve grown, and what you’ve accomplished. It also sets in your mind what is yet to be completed or accomplished. The last year for me, got a little sidetracked with other little projects and learning the rigging needed to create the bird for the short animation project. Unfortunately, that stuff isn’t ‘reel worthy’ yet. One or two shots will be cool enough to put on a reel eventually.

The summer is a good time to recompile, reboot, reorganize and replan. With Siggraph and Comic-con coming up, its important to organize and burn some dvds. You can see my slight obsession with design consistency, from this website to business cards, resume, dvd covers and stickers and even neckties. Each of these are hand cut and folded. Soon it seems that websites or digital uploading will be the way to get ones work seen–thank goodness!! Many studios are still looking at dvds. So to market oneself, one has to become a small factory.

After fixing some hitches and glitches,
adding lights and some sound effects,
here is the latest version of the Monkey Flip.

We are calling this ‘final’ unless
there is something major that I’ve missed.
Yes, the monkey sounds are of me.

This is stepped blocking of the flailing scene. This shot is right after the dive backwards shot. This was tricky because everything has it’s own timing and it’s own loops. I think it works ok, but it’s getting late, and my brain is fuzzy. The Idea here is that there will be sky, and some fast moving clouds zipping by. Then the cage will fade into place. (right now it’s visibility is keyed) He needs to move the cage door in this shot at the end so it hooks up correctly with the Cage Closed scene. It might be nice to have a reaction shot of him here, but that isn’t going to happen anytime soon. This shot will be difficult to smooth with alot of arcs to get the leg movements right.

So it’s been a week since I posted. I like to work in big long chunks. Unfortunately life likes to chip away at big chunks of time. I’ve been a little burned about my project, and there is still a ton of things to get finished, including a big wad of animation. Then lots of smoothing has to happen. I did go back and fix sections of my bird model. Most of the chin icicles have been found and removed. I’m still working on some working eyelid blend shapes.

Then as a bigger and better distraction, I dug up this gem.

This Monkey Flip animation I started back in March of 07, and never felt like finishing it. The tail was giving me all sorts of headaches, and there were some big spacing problems. So the last few days, I’ve been polishing and revising. This is the latest test. I’ve been sweating over it so much, it’s hard for me to see any glitches anymore. I have a spot light set up and am looking to put any fishing touches on it and render this out this weekend.

I’ll be back on animation of the bird short in chunks tomorrow most likely. Most of the animation still needs to be blocked before July 15th. I’d also like to do some smoothing by then. At that point I will reassess a new schedule and work on stuff between classes, as the next teaching quarter begins mid July in just a few weeks!!

This is the opening shot. Both characters are in linear, so there is still some smoothing that needs to happen.

I have to go back and pose put the wings. and the eyelids. I’m still having a blendshape issue with the eyelids.

Oddly enough, watching it through now, it feels a bit rushed. For me, I think the opening sequence needs to feel slow and calm, and this feels a bit too fast. I am thinking I can add a pause after he touches the cage before he brings his arm up. (fr 73) I might also add some more alone time for birdy in the beginning. Any thoughts about the timing?? heh –watch the wrist stick at 67 (doh).

More time equals more frames equals more production time. I’m going to sit on this and see how I feel about it later when everything is seen together.

Whoo!! From this view, the cage door has to open more than it will be in later shots. Chalk it up to the first of a few continuity problems. I had more movement on the bird, but it was distracting. The intention is that the lighting and a nice metal screeeeee will help direct the eye to the door opening. I guess I need some more set dressing for the living room; maybe some pillows, some books on the table, a deer head on the wall. *sigh* I have the same problem with my own apartment.

Another amazing cage door sequence ! The back end of the bird and the cage in the back will be out of focus and blurry when rendered. The claw wobbles a bit, but we’ll come back to that. whoo! exciting!!!

As summer gets into full gear, this animation project should be wrapping up. Well, it’s not yet. I still am clinging to the idea that I can get through most of the animation smoothing by July 14. Throughout the upcoming school quarter at AI, I will be doing final animation polishing, lighting, and rendering frames.

However, to speed up the process, I am looking for some help texturing.

The sky needs to be textured in 7 shots. It’s possible to texture
a dome and then import it and adjust it into the shots.
Or I could composite the shots, after rendering.

The ground needs a texture in 2 or 3 shots.
The jet plane needs a texture, maybe the telephone pole
and the bird model should be textured.

The electric effects for the electrocuted scene
need to be created in AfterEffects.

I would like to have some white 3D clouds
when the bird is falling in this shot and the shot after.
This can be done in Maya with a smoke emitter, but I’ve never done it.

Lastly, Foley sound effects need to be created
and I’m still on the hunt for a musician for a short score.

If anyone is interested in helping with any of the above things to help complete and polish this project, drop me an e-mail
or leave me a comment below.

This is linear of the open cage door sequence. This shot took extra long to do because the rigger of the bird didn’t know what he was doing and got the rotation axis messed up. The shot itself, I feel it’s really quite -meh. Kind of boring. I guess every film needs a boring shot or two to get the story from place to place. I can’t decide if the composition is boring, or the animation or what. It fact, I’m bored writing about his shot. No wonder no one is commenting. Animation when broken down into the painful frame by frame process that it is -is boring. Animation is not a spectator sport.

This is stepped stage of the first falling sequence. Actually, it may be in linear. Anyway… it’s close to what I imagined, so I’m moving on. I’ll be back for smoothing and fixes later.

Did you see how the fur on the tiger was waving in the wind as she jumped from rooftop to rooftop in Kung-fu Panda? That’s what I imagined for the feathers here, and in the next shot. Not sure this rig can pull it off though.

Here’s an interesting glitch in my bird rig. When you move the main control too far from the Master control ring, (like maybe when you are flying or falling) some verticies on the bird want to hang out don’t below, and I’m getting these really long stands of geometry. This isn’t blendshape related, so it must be a skinning thing, but I have no idea how to fix it. Cool, huh??

This is stepped stage of the hunting sequence. Its funny because this sequence was originally conceived because I had wanted to use a deer rig from an Art Institute graduate Patrick Dorris. The rig looks awesome and I’ve been wanting to animate with it. As it fits in with my story, this shot needs to be fast and I am finding the deer movement is distracting from the bird. So, it’s looking like the deer has to go.

It is important that all the beats here in this shot are reading:
bird flies, one hunter gun, bird reacts, more guns, bird reacts, guns shoot.

These beats have their own timing and
each needs to read with the audience.
I’m hoping that this is the case.

I still need to block an inbetween or two in place,
before switching to spline and smooth.
Theres still a lot to block first though…

This is stepped blocking for the telephone pole scene. I’m really annoyed with the fold under his jaw, but that’s part of the rig and it seems like well beyond me on how to fix it. The eye lids still don’t close right either. Oh and his upper thighs have some big problems with the skinning. But I should be talking about the animation… well this one will turn into more of a post shot. I’ve created most of a skeleton that I can quickly pose and render. Eventually, I intend on putting lightning bolts through him and he shakes and blinks some. The lighting bolts will be easiest to do in Aftereffects. So this shot needs a few more inbetweens, some simple smoothing (hopefully), and some minor shaking at the end. Then a lot of post and render work will be needed.

Plodding along.

As the animation progresses, I’m noticing my blog is being neglected some. Blocking and planning take time, and aren’t the most interesting things to look at during production. Particularly for someone who’s been staring at them over and over. In therms of production, I believe myself to be way behind where I wanted to be. Obviously rig problems had something to do with that. Also the length. there are 16 different cuts and therefore different camera set-up, so it’s a bit complicated.I’ve always been one to push myself to try more complicated things rather then succeeding in simple tasks. Anyway, one frame at time, and careful of those high wires.

This is stepped blocking for the airplane scene. This is really a ‘throw-away scene’ for a laugh. A lot of the animation was worked out in the animatic. Now all that needs to be done is add overlap on the beak, tail, and feathers for the flying. Then a little bit of smoothing. The camera pan is in linear and feels like it. It is close to where I want it. I think I need to add a few frames of the camera close to lead into this some. Because of the time crunch, I can’t be spending a ton of time on these minor shots, I’ve got to spread out and pick my battles some.

This is still in stepped. This feels tons better to me. Erik, I will look into playing with the timing. It may be because it’s in stepped, and I’m curious if it feels the same on ones. Taber, I totally agree. If this was a test i was doing on it’s own, i would probably add a bunch more acting in the beginning. As much as I’d like too, it is important to remember it is in the center of the fast cutting section of the entire short. It has got to be this many frames for the feel of the overall film to be right. It is important to remember where the shot fits into the larger picture: what is the characters mood from the shot just before? where is the viewer going to be looking from the last shot? etc…

Goes to show the importance of planning, and that sometimes it doesn’t always work the first time. This is in stepped now, so it will feel a little choppy. I think this is working much better however. The timing is snappier and the impact reads better to me also. I need to block in the remaining poses before moving on. It seems now that I might need more time for the feather falling, probably another four frames.

Instead of trying to move the rig around,
I went back to some planning in Flash.
The impact seems to be better.
It seems important to have
that white gap at 755.

Part of the reason that the exit read as too slow
was the spinning, so here, I got rid of it.
(and shortened the exit to five frames)

Yes, I also want to keep the feather.
It lingers in right the right spot where
I want the viewer to look for the next cut.
(the telephone pole shot)

Slowly working my way through the blocking of these quick cut scenes. Here is the first pass at blocking the soccer ball shot. It is another fast shot, but there is a lot going on. The spinning of the bird on the ball is different than the animatic, but I think it’s funnier. The big issue was being mindful of the arcs and spacing. I was using the white of the birds eye as a spacing and arcs guide to get the rotation to feel right. Tracking the spacing of that circle and the pupil circle was helpful for this shot. There is something about the take in the beginning bugs me, but I’m going to have to sleep on it or work out another scene and come back to it to figure out what is wrong. It could be the timing of the body bending inward. At some point, I’m going to need to fix the eye lid bllendshapes so the lids close correctly. I’m also going to need to add controls to the bird to get the eyeballs from poking through the top of his head.

Here the first pass at blocking the broom scene. It is a fast scene, and to get it worked out I had to plunk in a bunch of poses on ones. Obviously the bird needs the most work. I’ll be working out the feathers and wings last when the body and everything is timed out right.

Even with the rig ‘mostly fixed’ you can see some of the ‘uglys’ that are arising with this rig. First, the eye lid shape isn’t going over the eye ball. This is quite fixable, and I will get to it soon. Next is the neck, in extreme action type poses, the  neck gets this bag gap underneath it. The short answer at this point is, “don’t do that.” I don’t think the avoidance or fixing a pose around the rig capabilities will work well for this shot. Other options would be to create a control to puff out the neck. This is possible but would take me hours. Another option would be to paint each frame when the entire film is rendered. Each frame. Other things to notice, the leg has this odd connection that isn’t working and leaving this lovely mess on his belly. Also notice in the drawing how big the wings are to his body. That was really unconscious. nothing I can do about that now. All of these things affect the curve and the pose of his body. It’s a different curve and bend than in the drawing. The attitude of the character is often determined by the bend of a spine, or the simple gesture of a head tilt. These are the important thing, and the final pose of the bird needs to reflect the right attitude for the shot.

Yea! Happy to report here that the blendshapes for the birdy rig’s face controls are now working. Well, mostly working. Some of the face controls are a little bumpy and ugly, some of the lip controls are making the left pointer feather flap. So, there is some polishing to do on the rig. In addition, I need to add on the high-res skin for the tongue (which shouldn’t be a big deal.) The good news is that the controls are far enough along that I can block the bird into the scenes!!

Looking back, I’m trying to figure out exactly what I did wrong. Somehow, I moved the rig root controls and rotated the whole thing around while I was working on the blendshapes. This made them flip and invert and basically implode on themselves. So the first solid rule, is freeze your transforms correctly, reset all of your blendshapes to zero, and keep your character in neutral before starting a single blendshape. I’d also be real careful about naming conventions and taking really solid notes, so that you don’t get lost when connecting more than 100 different heads together.

If you’ve been wondering what is wrong with my bird rig, well so have I. I’ve finally got some useful information e-mailed from a former student, Bryan Money.

…your skin cluster has the mesh being scaled -1 in Z and X… which can only be done by scaling in component mode or the joints were grouped and scaled…. which is bad, basically even if you had your inputs in the right order they still wouldn’t work, which they don’t… even with no blendshapes on, rotating the head causes issues with the eyeballs poking through the skin anyway.

Yes. Rotating the head makes the face implode in unto itself.
I haven’t been wanting to re-make all of the blendshapes (again).
Now Bryan was able to give me some helpful advice on how to re-connect everything. It involves flipping and readjusting all 120 blendshape bodies. Fortunately, I don’t have to re-make everything, just reconnect. For an analogy, imagine a circuit breaker box with 100 different wires, then imagine someone snips all of the wires. Now you have to flip all of the wires around and reattach them correctly so all of the switches still do the correct thing.
It’s taken the work time of the last two days,
but I’m almost through, and happy to report that it just might work.

Still blocking around and without the bird rig.
This dog rig is from AnimationMentor.com
It’s got some great face controls.

This shot wasn’t bad because I had planned to get close and really be about the bird, and just the dog’s face. I barely had to use the body controls, and none of the legs were positioned.
…and yes, I had reference for this shot too…