From: edlantz@aol.com (Edlantz)
Subject: Re: HUMAN-FACTORS: Immersion Studies
Date: 1 Oct 1996 22:11:29 -0400


A few words regarding the debate on the utility of VR in its present form:
 

Mainstream VR businesses have always been very sucessfull at, well, demos.
 It is a fascinating application of technology which I first beheld at
Siggraph 91 and the first CyberArts.  The lines were very long in those
days.  Those entreprenuers who charged potential customers to demo their
equipment (Fakespace?) seem to have done better than those who did not
(Worldesign?).  In engineering terms, VR is the classic "solution in
search of a problem."  Most of us intuitively feel that VR is a very cool
thing, and will become an important communication medium.  But anyone
looking to make a living in the field must first ask: in its present form,
what is VR good for?  

Linda J. listed a number of websites reporting sucessful VR
applications.  I have not surfed all these sites, but would have to
ask - how many of these applications are demos?  How many are
successful demos of a system which works fine, but are performing a
task which can be accomplished cheaper and easier by conventional (or
non-VR) means?  How many of these applications are selling like
hotcakes?  These are questions that a pragmatist must ask.

A few years ago I sought out conventional applications of VR-ish
systems.  I found the best examples in the visual simulation (VisSim)
field.  Flight simulators have been using immersive displays for
years.  The HMD was Ivan's attempt to improve these systems.  So if
HMD-based VR could displace dome-based or collimated displays, then we
would have proven the technology.

Well, HMD's are just now taking over some dome-based training systems.
I was surprised to find that these dome-based displays were far more
capable of simulating reality than an HMD, even without stereoscopy.
But they're not cheap.  Some even have eye-tracked area-of-interest
projections.  Also domes take up alot of room and often have only a
single user, so HMD's are a natural in these applications.

Look at the visual requirements of VisSim systems.  Ultra-wide FOV.
Medium to high resolution.  What other applications require these
things?  Large screen entertainment systems, perhaps.  But what else?
Who is crying out for a wider monitor?  Where is the demand for
immersive environments?  We are stuck with a solution looking for a
problem!

Sorry if I seem skeptical, it's just the engineer and visionary
clashing within me.  I agree that VR will be a most powerful meduim,
especially in the arts.  But as for today's real-world problems, a
classic VR system is often not a good solution.  Yes, engineering CAD
can use wide FOV displays for immersive visualization (they just don't
know it yet).  Is an HMD the best solution?  Probably not.  A wide FOV
projection display such as the Reality Centre will suffice.

I like the spatially immersive display (SID) for such applications
because it is approachable without any special training.  Limited FOV
displays (Virtual Model Displays or VMD's) are great when our
attention is fixed on a model such as a virtual patient.  Why render
an entire room and risk cybersickness when our interest is only on the
model?  Our fixation on HMD technology as "true VR" is in some ways
holding back the advance of VE technologies, in my opinion.  That is
why I like the concept of "Experiential Computing" promoted by Linda
J. et. al.

Sorry, I don't mean to bash HMD's.  They have their place, especially
when full immersion must be simulated with 360 degree freedom of
motion and direct hand-eye interaction.  But if this is not an
important requirement, other displays should be considered.

The problem with immersive environments is that they engage sensory
modalities which are not required to complete most tasks.  The
simplest representation of a thing is usually the best.  If your
problem is simplified by visual immersion, then great.  Otherwise, I
would avoid it.

Ed Lantz
edlantz@aol.com