From: Forrest Bishop (forrestb@ix.netcom.com)
Date: Wed Jul 31 2002 - 20:44:02 MDT
----- Original Message -----
From: Michael Wiik <mwiik@messagenet.com>
To: extropians <extropians@tick.javien.com>
Sent: Wednesday, July 31, 2002 10:16 AM
Subject: Re: FUTURE SHOCK/STASIS SHOCK
> "Avatar Polymorph" <avatarpolymorph@hotmail.com> wrote
> It's like Id's FPS games, from Wolfenstein 3D to Doom to Quake II. W3D
> was blocky, somewhat nightmarish. Doom was better, less 'scary' in a
> visual sense than W3D (and I don't mean the game universe was less
> frightening, I meant the dreams that resulted: dreaming of answering the
> knock on the door and finding a nazi guard who's face is like a dozen
> pixels). When I got Quake II I also got my first 3D card, and thought,
> wow! this is much better than software rendering, but found it
> disturbing that as I approached objects, they got *blurry*. Things
> aren't supposed to get blurry when you get close to them. It's not the
> way the world works. Better visually perhaps, and ignorable, but at
> least the predecessor games still had definite 'limits', in that you
> knew where one pixel started and the next began. I'm wondering what the
> idea that things should get blurry when you get close to them is doing
> to my brain.
>
> As we get more and more saturated with digital movies (especially as
> most theaters are not equipped to show them and show film prints
> instead), and VR, we're losing more and more visual information. Until
> digital imaging gets as information-rich as real film (like 'Patton'
> above), it seems to me we'll be in an 'Uncanny Valley'.
The low resolution of a digital image seen close up can be worked around in several ways. The VR engine can use an adaptive
polygon-count on the 3D model the camera is looking at. As the camera (the VR viewer's 'eye') approachs the simulated object, the
object's geometry is finessed by increasing the number of polygons (or faces) that describe it. 3D StudioMax (and probably most
mid-to-high level animation packages) has this feature. It is computationally expensive. The bit map or algorithmic mapping applied
to the object can also be re-processed in a variety of way- interpolating pixels in the original bitmap image, fine-graining the
generating algorithm, blending in higher resolution bitmaps or additional algorithms, etc. I used a few of these tricks on the
Freitas' "Microbivore" animation, as the camera operates continuously over several orders of magnitude scale. Again, computationally
expensive.
Forrest
-- Forrest Bishop Chairman, Institute of Atomic-Scale Engineering www.iase.cc
This archive was generated by hypermail 2.1.5 : Sat Nov 02 2002 - 09:15:49 MST