Finding the moves

Down below are you can see a rough setup for the animation. Sofar I`m trying to figure out only the moves. Camera will be adjusted once I finish the choreography. Also, I made a union-skin like picture of the fall. Because it would be too small to see if I had put it in the vid like that, I actually do prefer the camera position from the side.

Falling

I felt it was time for another update, so here`s a small one. Some time ago, I finished the final rig for animating and started to block in the final thing. Along the way I came to realise this project isn`t exactly as straight-forward as I thought when starting it. Main focus is the movements of the characters. In their movements their motives should become clear. Though they are pretty straight forward, the turning point has proven quite difficult so far. It is a full reversal of motives between the two characters. The other day I had someone check out my project who had absolutely no previous knowledge of it. She suggested to let the dancer just take a long deep dive and having the robot still catch her. As I thought about it, this event would obviously show the dancer and the audience that she`ll be safe, no matter what. So here`s a picture of the deep dive, from my rough (cut up) model. I`ll stick with this solution for now, though at first I had in mind it would be more interesting to have the dancer devieate more and more from what the robot intends untils she has fully taken the lead.

Blocking...

Since the previous method of animating obviously didn`t work for me, I`ve resorted to a more conventional method where I place the keyframes first (without letting maya interpolate the motion). This gives a much clearer idea of where the accents are and makes it way easier to time...tho I`ve done a pretty rotten job at it sofar. I`d rather not continue on it now, since it works so well I am affraid I`m gonna have to redo the whole thing when my final rig is done...
So, planning for now is first finish the final model and rig, then block in the whole animation. Time. Refine, refine refine...
The beauty of computer animation is, that I can refine the parts I`m happy with, leaving room for drastically altering the rest. I can even refine the model when I`m halfway done with animating. So yeah, tho I was a bit reluctant toward a `non-material` and over technical way of working, I am beginning to see the obvious benefit of being able to experiment throughout the entire production. I have yet to see if I`m going to be able to take full advantage of it, but I am being optimistic about it now.

Another animation test

This is the second animation test I did. Used simple block with a heel-piece for the robot for now. Quite frustrating, the robot was really hard and I`m still far from satisfied with the result. Why is this so frustrating? The robot as a character moved very confident/straight. With as few keyframes as neccesary and without follow-through. Yet, because I have only some blocks to work with for now it is really hard to imagine them as being part of a machine...so, some arms to come.
As for the dancer...I changed the arm controlls to FK wich works better, except that I need to correct the shoulder so that it doesn`t move the entire arm anymore (this isn`t something you can see in the animation, but something I noticed while animating). The animation now, is a lot of swaying back and forth. And it seems more like she is pulling her foot out of a jam than that she is pushing off from the platform. For a lot of movements that originated from the hips I corrected the spine so she would be in balance. Now, that is pretty much `on key` with the hip motion. It`ll probably be better when that is off some frames to create follow-through.
But, by far the most frustrating thing of all is that I haven`t been able to put any character in it. Yes, it moves and perhaps convincing too...but no character; I`ll need to work on that. But I guess that I haven`t exactly made it easy for me...after all, how do you portray a clear character that is all alone in a very empty space?

Chocobobobobo!

Got new paper. Not fancy paper or anything, just plain white for now...to practise. It is way stronger than the coarse paper I had, so that`s an improvement. The downside is I can only get sizes up to 30cm square...
Tried and obviously didn`t succeed completely. I could blame the paper, but I wont. It wasn`t precisely square (off by a mm or so) wich has HUGE consequences for the final result. The toes didn`t turn out as pointy as they needed to be (mainly because there are some really small folds in there) and also the back of the head needs to be pointier. But the worst thing and I dont really know how that happened, one leg is longer than the other... Ah well, this was my third attempt at a Kamiya design, so I guess I could have done worse. It`s very fun to do tho, trying to figure out how to read the instructions. Going from form to form where only at the end you see how everything comes together.

Cloth test

A short cloth test. For the dress of the dancer. This is just a single render-pass, the background is composited. I`m still not entirely satisfied with the final look. The dynamics is fine, but additional post-processing will probably improve the look. At least now I know the nCloth module of maya works at a acceptable speed. And setting up the cloth is reasonably intuitive.

Designs

I meant to do some animating today, but got distracted in the designs again. What started as a quick model for a plateau (I needed for an animation test) ended in a short research for a style to fit with my machine. This is what resulted from that; a plateau with chainlink top, a cogwheel and a piston:

Although I like it, it somehow seems to Burtonesque and perhaps to fragile to fit with the character. Also, the warped shapes will most likely keep the cogs from interlinking properly. The chainlinks will find their way to the final design, I`m sure of that. The previous design evolved as well last week, but I forgot to post it. A shot of the base and of the first two joints: