Path: utzoo!utgpu!jarvis.csri.toronto.edu!mailrus!tut.cis.ohio-state.edu!bloom-beacon!mit-eddie!uw-beaver!cornell!rochester!pt.cs.cmu.edu!andrew.cmu.edu!po0o+ From: po0o+@andrew.cmu.edu (Paul Andrew Olbrich) Newsgroups: comp.graphics Subject: Re: 3-D perceptual abilities Message-ID: Date: 27 Jan 89 20:53:19 GMT References: <6382@thorin.cs.unc.edu> Organization: Carnegie Mellon, Pittsburgh, PA Lines: 46 In-Reply-To: <6382@thorin.cs.unc.edu> (in response to using stereo views to check for asteroids...) > The stereo effect was supposed to make asteroids stand out from > the flat starfield. As it happened, they did not seem to stand out > very much for me, and I reverted to using it as a blink microscope > instead (good exercise for the eye muscles :-) Do many people have > this type of problem? > -- > Jon Leech (leech@cs.unc.edu) __@/ > SUSHIDO: the Way of the Tuna I noticed that I seem to be somewhat better than average at understanding 3D relationships. In a high school geometry class, for example, I remember being rather surprised when the teacher had a hard time drawing a cube on the blackboard as a 2D projection, (even without perpective) ... I even figured out how to do a similar projection of a 4D hypercube in 2D, and once made a 3D projection of a 4D hypercube out of toothpicks. (This had no forth-dimensional "perspective" ... It was essentially two cubes, one offset diagonally with the corresponding verticies connected by toothpicks. If you make one of these and do it well, it creates some neat effects that I noticed by accident. I had it placed on a table, and once I was walking by and glanced over at it, and noticed that it had collapsed. The cubes seemed to be both flat on the table, lying ajacent to each other. In reality, it was just my viewing angle making them look that way! Everything lined up and fooled my vision. After more experimentation I realized that from the correct angle it looks like two cubes stacked vertically. Also, viewed "edge on" it looks like a single cube. Later I noticed that if I placed it on a polished surface, and viewed it from the two stacked cubes point of view, the reflection was of a single cube only. From the "one cube" point of view, the reflection was two cubes. This is fun at parties!) I also wrote a hypercube simulation program (ideas taken from Sci Am) that did rotation in real time (after calculating the screen coordinates for about up to 30 seconds before each rotation). It did 4D perspective so it was a more "realistic" (yeah, right) model than the toothpick one. Rotating through a 4D axis is hard to visualize at first, but I found that I really had a good mental idea of what was happening. I attempted to do a red/blue 3D glasses version, so that I could essentially have a 3D projection (with 4D perspective) model, but I was doing it on a vintage IBM PC and the colors didn't match up. Whatever. - Drew --- Drew Olbrich po0o+@andrew.cmu.edu "A brush that does not work is not a brush." -- me