Re: Escher perspective

Andy Norris (andyn@texas.net)
Thu, 04 May 1995 13:58:21 -0400

At 01:30 PM 5/4/95, Andrew C. Esh wrote:
>On Thu, 4 May 1995, Andy Norris wrote:
>
>> If you assume a, um, monographic perspective, then it might be
>> possible. What it would mean is that instead of generating a 3D world
>> up front and just navigating it in real time, you would have to have
>> some kind of transform that generates the world on the fly as you
>> navigate it. This isn't how we normally think about a virtual world,
>> but there's no real reason you couldn't build one this way (technical
>> limitations not withstanding).
>
>I'm not denying that this is possible. Of course it is. It just isn't 3D.
>How many times have you walked through a room full of inanimate objects,
>and one of them was constantly reorienting itself so as to always present
>the same face to you? My cat does this, but he doesn't count. Am I going
>to have to build Escher effects by defining a cat, upon whose face is an
>impossible Escher object, and have him stare at you as you walk around him?

I didn't necessarily mean the same view constantly turned toward you.
It may not be possible, but what I was thinking of was some kind of
effect that looked different from different locations, but sustained
similar illusions. I can't construct an Escher example offhand, so
how about the following:

AcidSpace, in which some of the objects warp toward you as you move, and
others kind of warp away. they don't curve uniformly like a fisheye lens,
but they do adjust themselves based on your location. This can't be
represented by a single 3D model, because the object location is dependent
on the viewpoint. (Also, at unusual moments everything suddenly becomes
covered in ants, but that's another story.)

>Oh man. Now I've drifted into the Surrealism discussion, haven't I?

I certainly have. :)

--Andy