Article View: sci.virtual-worlds
Article #2895Dimension International
From: John Styles
Date: Tue, 19 Jan 1993 19:10
Date: Tue, 19 Jan 1993 19:10
61 lines
3787 bytes
3787 bytes
I am surprised that Dimension has not responded to these comments about
the software as I know they have an on-line presence (the same system
that I am on).
cdshaw@cs.UAlberta.ca says:
>>All the graphics were designed by Dimension International and
>>basically the whole thing was a walk-through environment rather than
>>immersion HMD-like stuff. There was also no way for the player to
>>change their perspective other than to turn their figure around.
>
>In other words, the usual low-budget VR system - 10K polys/second,
>fixed environment, the user controls only the (strictly limited) rate
>of change of position and orientation.
The software
a) is NOT just static walk-throughs. I realise that you have
a disadvantage in discussing something that you haven't seen. Far from it, the-
y
have all sorts of clever demos including one where lots (20ish) of dodgem-cars-
drive
round all following the same program but with randomised initial conditions
and interacting with each other, the sides of the track and a ball they're
'kicking' round. The user can also interact with things either by mouse clicks-
or by the viewpoint walking into them - the viewpoint can be attached to movin-
g
items (e.g. follow one of these cars round) and it can display multiple viewpo-
ints
simulateously
b) is relatively cheap UKP3500 (double this if you want to program it
in 'C' as well as its built in language [see c below])
c) it runs under DOS on PCs. Dimension were a games company before and
I suspect the software has large amounts written in assembler so I don't
forsee them porting it in the near future (I have no connection with them
so this is just an educated guess).
d) is function rich. Certainly comparted to WorldToolKit (i.e. it has a certai-
n
amount of dynamics / gravity / coefficients of friction built in). It
has a built in programming language ('C' like) in which you can specify
actions attached to objects (e.g. 'on mouse click' do this, 'on colision'
do this).
e) has an object editor (i.e. unlike Sense-8's thing you don't have to
muck around with DXF files, you can design them in the system and can even
fly around models (though in this mode they are just fly-through) whilst
positioning objects etc).
f) supports flat-screen (SVGA with some subset of SVGA cards) only. The input
device is typically Space-Ball. I realise that whether this sort of thing
is 'really' VR is a theological issue in this conference but I have to say
that...
g) on flat screen it looks a hell of a lot better than WorldToolKit in terms
of frame rate and that's with WorldTookKit using a special display with
RISC procesor. Needless to say a 486DX2 66MHz helps (!)
I have no idea what the polygon per second rate is. When I get my hands on
a copy I shall have a go if someone will tell me what a good test is.
This is not to say that WorldToolKit lacks merit, it's just that for my
money Superscape (the name of the product) is Cheaper and Faster and has
far more features - provided of course you want to run on a screen and on
a PC. It is almost at a price whereby I might consider buying one with
my own money... be the first on your block with a VR system!
Technical Details
Message-ID:
<1juvusINN1ou@shelley.u.washington.edu>Path:
arc1-ca.pugleaf.net!sparky!uunet!europa.asd.contel.com!howland.reston.ans.net!spool.mu.edu!uwm.edu!ogicse!news.u.washington.edu!stein.u.washington.edu!hlabShow Full Headers
Complete Article Headers
Path: arc1-ca.pugleaf.net!sparky!uunet!europa.asd.contel.com!howland.reston.ans.net!spool.mu.edu!uwm.edu!ogicse!news.u.washington.edu!stein.u.washington.edu!hlab From: hpengwyn@cix.compulink.co.uk (John Styles) Newsgroups: sci.virtual-worlds Subject: Dimension International Message-ID: <1juvusINN1ou@shelley.u.washington.edu> Date: 19 Jan 93 19:10:00 GMT Article-I.D.: shelley.1juvusINN1ou Organization: University of Washington Lines: 61 Approved: cyberoid@milton.u.washington.edu NNTP-Posting-Host: stein.u.washington.edu Originator: hlab@stein.u.washington.edu