ron@vicorp.com (Ron Peterson) (09/15/90)
There seems to be some interest so here's a rehash of what I've seen
so far on hooking up a PowerGlove as a low cost VR input device:
So...for all you VR hardware hackers..
I just bought a Mattel Power Glove for Nintendo systems and
spent last evening taking it apart. It turns out that it is NOT a
toy version of a DataGlove (surprise, surprise). It does not use
fiber optic based sensors to detect finger motion; instead it uses
some sort of electrical sensor. I can't tell exactly what type since
it is imbedded in the plastic of the glove but there are wires leading
out to the fingers so it's probably something that changes resistance
or produces voltage when flexed. It doesn't detect flexing at each
joint; it just responds to thumb curled/uncurled, index finger
curled/uncurled and last three fingers together curled/uncurled.
These motions are interpreted as on/off for A, B, and C buttons.
It uses two ultrasonic transmitters (in a module mounted on the back
of the glove) and three ultrasonic receivers (arranged at the corners of
your monitor) to detect up/down left/right ,in/out, and bank left/right
motion. There is a custom IC under the forearm mounted keypad that is
hooked up to a 10Mhz crystal. I suspect that this is a microcontroller
that performs all the motion sensing operations. The glove plugs into
a junction box that also connects to the ultrasonic receivers and the
output to the Nintendo is a 7 pin joystick type connector. This causes
me to believe (and it is also implied by the instructions) that the Power
Glove mimics a joystick (or other standard Nintendo controller.) It
has 15 or so programs that can be called up to configure the way in
which it simulates a joystick. (A joystick has four switches
that represent four directions. The Power Glove can detect motion
in six directions as well as hand rotation. There are several ways
the various Power Glove motions can be mapped to the four joystick
directions.)
A very recent issue (September?) of Byte magazine has an article
describing several devices for interfacing the human hand (as an
input device) to a computer, and includes mention of the Mattel
PowerGlove.
A sidebar to the article gives some simple specs and software
for interfacing a PowerGlove to an IBM PC. Given this you
could probably connect it to other computers also.
There seems to be a "hidden" high-resolution mode to the glove
which can provide distance, orientation and finger flexion data
rather than just a simple emulation of a joystick but no one
has published info on how to access this mode yet. (See article below.)
>Article 259 of sci.virtual-worlds:
>Path: vicorp!uunet!samsung!uakari.primate.wisc.edu!dali.cs.montana.edu!milton!j
munkki@hila.hut.fi
>From: jmunkki@hila.hut.fi (Juri Munkki)
>Newsgroups: sci.virtual-worlds
>Subject: Re: Mattel Power Glove Specifications
>Message-ID: <9008111818.AA01693@santra.hut.fi>
>Date: 11 Aug 90 18:18:17 GMT
>References: <5488@milton.u.washington.edu> <618@entec.Wichita.NCR.COM>
>Sender: hlab@milton.u.washington.edu
>Organization: Helsinki University of Technology, FINLAND
>Lines: 40
>Approved: hitl@hardy.u.washington.edu
>
>
>In <618@entec.Wichita.NCR.COM> jsurine@entec.Wichita.NCR.COM (James Surine):
>>Howard Eglowstein from Byte magazine who wrote the article reverse engineered
>>the power glove to get the information to write the article. He is going
>>to reverse engineer the "enhanced mode" when he gets a hold of that new game.
>>(That is if the article doesn't get canceled by the editor)
>
>If the article gets canceled, maybe we could get it published here. I'll
>have a PowerGlove to work with in a few weeks and it would be so much easier
>to use it, if I just had some kind of documentation. I wouldn't mind
>contributing a few $$ to someone for revealing this information, although
>I feel that it should have been freely available from the start.
>
>Someone wrote that the increased resolution near the screen wouldn't be of
>any use because the user's hand would be too close to the screen. As far as
>I know this can be solved by placing the detector somewhere else than around
>the screen. Shouldn't be any harder than moving a mouse to a new place.
>
>The address and FAX number of the company that designed the PowerGlove
>was posted in comp.graphics and I've been thinking of sending them a
>FAX. Has anyone tried this? Here's the information, just in case
>someone wants to try to contact them and missed it in comp.graphics:
>
> Abrams/Gentile Entertainment, Inc. (AGE)
> 244 West 54th Street, 9th Floor
> New York, NY 10019
> (212) 757-0700
> (212) 765-1987 FAX
>