[-empyre-] CG and all things fuzzy

[futurenatural]

Yello,
having your Computer Science Department teach computer animation does  
seem unusual. I wonder if this is specific to Cornell?

This question reminds me of something you (Renate) posted last week  
when you talked about how you found your students were concentrating  
too much on the "programatic aspects" and ignoring whether the final  
animation was really working? This split is certainly a widespread  
issue and I find it begins as a consequence of the way that animation  
software is designed and used. As the discussion is moving more  
decisively towards "problems with the digital" I thought I would put  
down a few thoughts now.

When using software there is a strong tendency to "animate with the  
numbers" instead of looking at the screen at what is actually  
happening. So if you want to animate an apple falling onto the ground  
you know when it has hit the ground because the height value of the  
bottom of the apple equals the height value of the ground plane.  
Except that it often doesn't look right. You might have to add  
shadows or textures or make the ground reverberate in a way that it  
doesn't actually do in reality. But the student will sometimes say it  
must be right because the numbers say the apple is on the ground!   
Another example is anatomy. In computer animation you tend to keep  
the basic anatomical body the same all the way through because there  
is something in the rational world of software design that insists  
that you cannot suddenly loose or gain an arm. But of course  
traditional animators do this all the time, if characters get into a  
fight for instance. There are lots of examples like this.

My personal issue is with keyframing and the timeline. Keyframing is  
the most popular way of digitally animating and appears to be a  
straightforward reflection of traditional practice of keyframes and  
inbetweening. Except that it isn't because the computer's automation  
of inbetweening means that these intervals between extreme frames are  
no longer subject to conscious design as they were when there were  
people actually drawing them. The result is often a lot of  syrupy  
animation as the software is obliged to smoothly interpolate between  
as few keyframes as possible. But this isn't a problem with the  
technology as such. It is perfectly possible to animate "manually" in  
most software programs. You could simply just create a keyframe on  
every frame. But in practice most people don't even think of doing  
this because "that's the computer's job".

This isn't really a problem in interface design. It's more to do with  
the concepts that we are interfacing with (although not exactly that  
either). The problem is not to resist software as to find a different  
angle on it. Building your own Open Source software or DIY media is a  
much talked about option but not open to most people. And in any  
case, how do I know what I want software to do?

Lev has pointed out how some animation qualities like continuous  
variability of the image arise out of software design. And this  
software design also supports and extends older practices like the  
layering of cells that Tom used to describe the qualities of "limited  
animation" and the "time-image". But all of this is complicated by  
the fact that the fundamental assumptions of software design are not  
necessarily constant nor is its conventional usage (hopefully,  
otherwise we're stuck). It is not a matter of the different technical  
abilities of software or of how much it costs, but of how easily a  
technical potential can be perceived by the user in a way that  
motivates engagement. This is the point at which the "techno- 
aesthetic" is worked out - different motifs that permeate  
technological, social and cultural levels.

That's more than I intended to write so that's enough for now.  And  
it's not even "my week" yet...


Richard


On 13 Feb 2010, at 04:37, Renate Ferro wrote:

> Dear Paul and Suzanne,
>
> Can you both talk about how CG fits into your animation programs?  At
> Cornell, Computer Graphics and 3D animation is taught by Computing
> faculty.  It is in the art department where students, particularly
> recently, have been creating stop action, frame by frame, roto- 
> scoping,
> drawing based and a medley of other fuzzies. Whether working  from
> photography based or original drawing. their novel, quirky rendering
> styles, interdisciplinary interests and criticality make their work  
> fresh
> and innovative.
>
> How does it work in the UK?
>
> Renate
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: https://mail.cofa.unsw.edu.au/pipermail/empyre/attachments/20100214/f6c13525/attachment.html