This is absolutely fascinating. I've now been looking at Alex Colville's paintings and trying to work out what it is about them that makes them look like CGI and how/why he did that in a world where CGI didn't exist yet. Here's what I've got so far:
- Total lack of atmospheric perspective (things don't fade into the distance)
- Very realistic shading but no or only very faint shadows cast by ambient light.
- Limited interaction between objects and environment (shadows, ripples etc)
- Flat textures and consistent lighting used for backgrounds that would usually show a lot of variation in lighting, colour and texture
- Bodies apparently modelled piece by piece rather than drawn from life, and in a very stiff way so that the bodies show the pose but don't communicate the body language that would usually go with it. They look like dolls.
- Odd composition that cuts off parts that would usually be considered important (like the person's head in the snowy driving scene)
- Very precise drawing of structures and perspective combined with all the simplistic elements I've already listed. In other words, details in the "wrong" places.
What's fascinating about this is that in early or bad CGI, these things come from the fact that the machine is modelling very precisely the shapes and perspectives and colours, but missing out on some parts that are difficult to render (shadows, atmospheric perspective) and being completely unable to pose bodies in such a way as to convey emotion or body language.
But Colville wasn't a computer, so he did these same things *on purpose*. For some reason he was *aiming* for that precise-but-all-wrong look. I mean, mission accomplished! The question in my mind is, did he do this because he was trying to make the pictures unsettling and alienating, or because in some way, this was how he actually saw the world?