Uncanny Valleys

ex-machina

It feels, now, so naive to believe that higher resolutions, faster framerates, and more polygons could ever lead to something which appears truly real. Every step we take toward making things look realer becomes a step into the uncanny valley: The more realistic the texture and motion, the faker the seams start to seem, unraveled at the edges. When we film in high resolution, it can’t fail to become more obvious that we’re building sets and canning dialogue instead of recording a reality – when we render at 60 frames a second, the visual gaps between our animations and the motion of muscle and bone become stark.

In the long run, I don’t believe that we can approximate reality by means of picture quality or polygon pushing – or, even, by means of technique and artistry.

This relates to what I wrote about last week. The first stroke of creation still matters deeply, and determines the final form of the piece in an inescapable manner. The more we try to hew to reality through artificial means, the more the gaps between reality and our representation of reality will begin to show.

As we seek reality, if we continue long enough, our methodology will begin to drift away from mimicry and towards emulation – that is to say, it will no longer be sufficient to create a model, texture, and animation that looks like a creature in motion, but will become necessary to create a simulated creature operating under the rules of reality. Follow the thread long enough and realism becomes simulation, inevitably. There’s no bridge across the uncanny valley, just art on one side and reality on the other – one, a representation of something external to itself; the other a system no longer beholden to aesthetics.

If your standards for realism come high enough, the only way to fulfill them is to create reality. So, the question is, what is it we actually want when we say we want things to look better, to look realer? Do we really want Turing machines running virtual flesh bodies, ensuring each motion is motivated by an actor with real wants and needs, each muscle jiggling and snapping as limbs flex? Or is what we want, not reality, but the same old fake worlds with more pores, with higher thread counts, where everything is just a little bit shinier and we can pretend that reality is what we make of it?

Leave a Reply

Your email address will not be published. Required fields are marked *