From: rwzobel@unity.ncsu.edu (Rick Zobel)
Subject: Re: DESIGN: Architectural Design
Date: Wed, 17 Mar 1993 17:46:23 GMT
Organization: North Carolina State University Computing Center


In article <1o60fkINNpgv@shelley.u.washington.edu> neuron@carson.u.washington.edu (Fernando Diaz) writes:
>One aspect of my research is in creative/conceptual design using VR.
>Right now I am looking into architectural/interior design environments using
>VR technology.

I am interested to hear what you come up with. The preliminary
design phase in architecture remains largely un-aided by the computer
at this point. This is mostly because of a lack of tools which allow
for free flow of ideas from the mind to the electronic media. There
are many tools which work adequately for visualization or construction
drawings, and even some which begin to talk about the needs for
designing (VIRTUS WalkThrough, Alias Sketch, Form-Z...), but these
are still beat for the most part by an HB pencil and sketch paper.
Until there is a reason to switch, the architectural community will 
remain largely unconvinced that computers really have any place
in design (or architecture in general for that matter).

>Does anyone out there in cyberia know about any commercial systems or
>companies involved in this kind of work?

The ones mentioned above are screen based media which do not yet allow
for immersion. I am highly doubtful that any program could make the
transition from screen to immersion without completely re-writing
it, but there are still many lessons to be found in how each of these
accomplished what they did. VR walkthrough programs abound (UNC,
WorldToolKit, AutoDesk, SGI) but these are not design at all really,
just visualization. UNC's 3DM project is close, but it lacks the
tools that would need to be used specifically for design. There
have been on again off again rumors that 3DM would start back up as
a project, but as of yet it is still parked. There are architectural
based research projects (researched by architectural graduate students)
going on in the last year here at NCSU with UNC, there at the HIT Lab, 
Univ. of Huston, CMU, Univ. of Sheffield (UK), Venesuela, Netherlands,
and many others that I am sure I am unaware of. To the best of
my knowledge, CMU and UNC are the closest to working on what it
means to design in immersive environments.

>My emphasis is in communicating architectural spaces as they relate to:
>        user interface issues
>        human factors of vr workstations
>        software tools for conceptual work and design communication
>        effective presentation of visual spaces

The problems that I have discovered that need to be examined in depth
are perceptual, hardware and tool oriented: (Nothing much :)

-Depth perception is off by up to 20% low in immersive environments
when compared to an equivalent space in the real world. (ie. In the
real world someone points to an object and estimates that it is 20
feet away. Put on an HMD and from EXACTLY the same position, the
observer estimates that the object is 16 feet away.) There are
currently three theories, lack of full texture (perceptual),
small field of view (perceptual/mechanical), and lack of freedom
of ambulatory movement (perceptual/physical). There are undoubtedly
others. (The P/M study was done last semester at the HIT Lab)

-Tools are lacking for intuitive spatial creation and editing. This
is no small task unto itself. Immersive media create an entirely
new paradigm for design. What type of tools can take advantage of
the inherent power of the medium?

-Jumping scale is vital, and must be done in an intuitive manner.
Scale essentially becomes another dimension through which movement
is possible. It is very easy to become lost in scale.

-Constrained movement is needed. This can vary from flying down to 
single point of view, but the many different steps inbetween are 
very important to how the space is perceived.

-The user needs to be put into the immersive space. Watch an architect
talk about space without a pencil and paper. They put Italians to 
shame with the amount of hand waving they do. This combined with
the perceptual fact that humans scale the world with ourselves and other
humans. We touch things and see our size relative to its which
enables us to understand how each relates to the world.

These are a few of the basic problems that (IMHO) need to be understood
before a serious effort can be made toward a marketable immersive
design product. I am very interested to hear what other people
have to say about the issues of immersive design.
 
Rick Zobel
NCSU School of Design
Architecture Dept
rick@oscar.sod.ncsu.edu