freeman@spar.SPAR.SLB.COM (Jay Freeman) (05/03/88)
A couple of months ago the person in charge of the NASA/Ames effort to
develop a helmet-mounted visual interface to a world of computer-generated
objects gave a talk at Xerox Parc. I went (I'm not at Parc, but it's just
up the hill and this talk was public). I reviewed the talk internally, in
connection with a project that we have here at Schlumberger. Portions of my
review may be of interest to this group ...
================
At the Xerox PARC Forum on 4 February 1988, Dr. Michael W. McGreevy of
NASA Ames Research Center described the "virtual workstation" project ...
They started in early 1984 ... The technical approach is remarkably like
our own, right down to using the Polhemus sensor, the VPL DataGlove, and the
disassembled guts of pocket LCD televisions. ... Notable technical
highlights included:
(1) Specially built video preprocessor/mixer/what-have-you
electronics, which does interpupil adjustment, image predistortion and
stuff like that in hardware; also, they did not need to use our
"high-tech optical coupler" hack. This preprocessor was a well-built
modular system about the size and configuration of a small
oscilloscope, with printed circuit boards (not wirewraps) and no loose
ends. McGreevy stated that a project goal was to put one of these on
the desk of every principal investigator at Ames.
(2) Use of a dedicated graphics-processing computer for
image-generation. They had switched processors several times -- the
one most recently used was something called an "Iris" (?);
(3) Location of a commercial source of very wide-angle lenses for
their helmet-mounted display. I asked for details -- seems somebody
wanted to build a high-quality wide-angle stereo camera system, to
photograph and view scenes like the "ViewMaster"s some of us remember
from childhood, but with very wide angle. The lenses Ames uses are
part of the system, BUT the vendor seems to have folded, and it's not
clear that even Ames is ever going to get any more;
(4) Procurement of specially-built LCD displays from the
(Japanese) vendor, via personal contact and a close working
relationship. Having the right displays makes considerable difference
in perceived image quality.
(5) A new viewing metaphor: in addition to the "helmet-mounted
display" metaphor, and instead of our "fishtank" metaphor, Ames used
what I might term a "lorgnette" metaphor. They took the optics, video
display and Polhemus sensor from their helmet-mounted display and put
it on a spring-loaded arrangement like one of the Luxor lights that
Andy has had jumping around, and put it on a desk. You move the
"head" of the unit around, and peer through it as you like, but you
don't have to wear anything or carry any weight. I kind of liked that
idea.
...
They also had some whizzy demos. But I did not see any signs of
software that seemed particularly impressive or more advanced than our
own 3D interface software ...
...
-- Jay Freeman
-------