Re: VRE, the Sandbox paradigm, and transportation issues

Torbjoern Caspersen (Torbjoern.Caspersen@ark.unit.no)
Mon, 4 Sep 1995 17:48:01 +0100

>As a person who writes 3d graphics algorithms, I really can't agree
>with your statement that "far more info may be coded" in 3d. I think
>you overestimate how much resolution a computer screen really has.
>Seeing things from far away and projecting things at angles loses a
>lot of information. You gain more info about a given object's 3d
>shape, but you lose info about what's on its 2d surfaces, and you
>still can't see objects that are occluded by other objects. Believe
>it or not, the best informational representation of what's on a piece
>of paper, is a 2d projection. :-)

'As a person who work with 3d form' I'm getting the feeling you're pulling
my leg :)

Of course 2d is best suited for representation on 2d faces. What I saying
is that you can have more information in a 3d space than on a 2d surface.
3d has a timepenalty as it forces you to wander around to get the whole
picture. But as it's one space it's probably easier to maintain a grip on
the structure than if it was spread out on tens of sheets of paper.

I'm not saying that 3d is better than 2d, period. But in certain cases, it is.

>Have you ever seen a 3d visualization of an OS's file system? You
>really can't see everything all at once - all the text on the
>"background" objects is illegible. You may learn something about the
>overall structure of the system, but you don't get any "more" info on
>any of the objects in particular. And for a lot of objects in a given
>scene, your 3d viewpoint doesn't make them any easier to find.

It might be that the overall structure is the most important matter at a
given time. Also, do you have a 2d OS which let you see everything at once?

>VR makes up for these informational deficiencies by allowing you to
>move around in the environment. That takes _time_. Time is what
>consumers don't have a lot of. Since VR can't provide the fidelity of
>real-life objects, what's the justification for spending a lot of time
>wandering around in a VR world to shop? You might as well look at a
>2d photograph, or several 2d photographs taken from different views.
>The fidelity is way better than what VR has to offer, and is more
>likely to affect your buying decision.

Today, several photograps would give far more resolution, but it's just a
question of time before afforable computers can do the same in 3d. The main
justification for VR is that it's not necessarily organised after _one_
subject all the time, like the alphabetic ordering of your average
bookshop.

> Take for instance shopping for books or CD's. At least I got into a shop
> without any clear thought of what to buy, instead I'm looking for something
> to catch my interest, be it through a familiar title, interesting cover, or
> whatever. Think of what a VR system could do for that kind of information
> immersion, walking or flying through the acid-jazz recordings
>
>But is this information access, or marketing flash?

Could be either...

>That would take a lot of _time_ to go through. Why wouldn't someone
>just consult an index of movie titles, and then get a videotape? It's
>much more direct. To get to the history, you have to watch the
>videotapes anyways.

If you know the title, consult the list, if not browse the shelves.
(what history?)

> I know this borders to commersialism, but then again I think we have a lot
> to add to more utilitybased VR systems as well as the artistic ones.

>Thus the sale of goods in Cyberspace is going to have nothing to do
>with the technology of VR, and everything to do with the artistry of
>VR.

And we are [would-be] VR-artists.

>It is important to recognize what tasks an interface is good at, and
>what it is not.

My point exactly.

-----------------------------------------
Torbjoern Caspersen casper@due.unit.no
http://www.stud.unit.no/~casper/
Student of Architecture
at the Norwegian faculty of technology, NTH, Trondheim.