| From: | Richard Tomasso rtomasso@*******.com |
|---|---|
| Subject: | UV Hosts (was Quantum Matrix) |
| Date: | Wed, 13 Oct 1999 10:53:35 -0400 (EDT) |
> I just thought of a possibly relevant example. I have been using
> a math program (doesn't really matter which one, most do it), which has
> the capability to recognize when something is outside of it's precision
> range. When this happens, the computer expresses it differently.
> Ex. .333333333 is very accurate, but not correct. 1/3 is correct, but
> can't be stored in the same method as the decimal approximation.
Assuming the computing power can calculate rational numbers as fast as
real numbers, and there was a reason we needed near-infinite accuracy, sure
this would be a logical way to go.
> What if the UV calculated an object, but let the brain to the
> computations. As an example, the UV host would define the
> corners of a box, and give that info to the brain. The brain then
> constructs an image of a box. A normal host would define where
> the box is, up to it's level of resolution, then let the decker interact
> with that. This would allow the UV host to say, define a grain of
> sand, then only have to deal with keeping track of it, as opposed to
> a normal host which would have to keep track of it, and what each
> piece looks like.
Except this would require more brain-work and thus make it harder for the
decker to navigate. Plus, if the image is processed and figured out in the
brain, then the ASIST and/or cyberdeck doesn't know where it is or what all
its features are, and interaction wouls suffer. Assuming we figure out how
to wire into the brain's math centers so we can give it inputs properly.
Besides, it's very likely that most basic solids are transmittes as coordinates
plus color data, letting the deck or ASIST fill it in locally and dump it
to the brain.
