[sci.virtual-worlds] Polhemous tracking devices

jorice@maths.tcd.ie (Jonathan Rice) (09/21/90)

In <7467@milton.u.washington.edu> zik@bruce.cs.monash.OZ.AU (Michael Saleeba) wr
ites:

>Does anyone know how polhemous (sp?) tracking devices work?
                      ^^^^^^^^^
                      Polhemus
Sorry, not really. I was told it was some sort of system using an AC
magnetic field.

>It seems to me that these devices are even
>more impressive than datagloves; apparently being able to sense location, 
>direction, and rotation in three dimensions.

Sure they can, but they can't measure finger positions (unless you put one on
each joint, which is overkill). A position and orientation sensor such as the
Polhemus is part of most gloves - VPL's Dataglove actually has a Polhemus
sensor mounted to the back of it, while Mattel's PowerGlove uses an ultrasound
ranging system.

>Also, does anyone know who produces these devices commercially and how much
>they cost?

I don't have any price for the Polhemus, but I don't think you're talking
cheap here. I do have some info on a full plug-in-and-go RS-232 position and
orientation device from Ascension Technology Corp. of Vermont that uses a
*DC* field. This apparently gives better immunity to eddy currents in metallic
objects in the sensor's vicinity. Anyway, that device costs $5000. Not cheap.

The beef:
        The Bird (TM) 6D graphics input device
        Ascension Technology Corp.
        P.O. Box 527, Burlington VT 05402
        Tel: 802-655-7879

There's also this similar device called the Flying Mouse (TM):
        Flying Mouse (TM)
        SimGraphics Engineering Corp.
        1137 Huntington Dr., Suite A-1,
        South Pasadena, CA 91030
        Tel: 213-255-0900
        Fax: 213-255-0987
I don't have a price for that. It uses a Polhemus sensor.

-- Jonathan

Jonathan Rice, jorice@cs.tcd.ie

Disclaimer: I've no connection with Ascension or SimGraphics. I've a slender
enough connection with anything in the real world, come to think of it.

murray@vs2.scri.fsu.edu (John Murray) (09/24/90)

In article <1990Sep21.143120.21614@maths.tcd.ie> jorice@maths.tcd.ie (Jonathan R
ice) writes:
>
>In <7467@milton.u.washington.edu> zik@bruce.cs.monash.OZ.AU (Michael Saleeba) w
r
>ites:
>
>I don't have any price for the Polhemus, but I don't think you're talking
>cheap here. I do have some info on a full plug-in-and-go RS-232 position and
>orientation device from Ascension Technology Corp. of Vermont that uses a
>*DC* field. This apparently gives better immunity to eddy currents in metallic
>objects in the sensor's vicinity. Anyway, that device costs $5000. Not cheap.
>
>The beef:
>        The Bird (TM) 6D graphics input device
>        [..address + phone deleted..]

I was told at SIGGRAPH that The Bird also uses a Polhemus sensor.  On the
other hand, the blurb I've got in front of me touts it as "the only 6D
input system that works...near metal and other obstructions," so you may
be right about the DC. (maybe Polhemus makes both, now?) Also, the $5000
price tag is the single-unit price. 2-4 units saves 20%, 75-99 saves
55%, >99 "consult Ascension."


>There's also this similar device called the Flying Mouse (TM):
>        Flying Mouse (TM)
>        SimGraphics Engineering Corp.
>        [..address + phone deleted..]
>I don't have a price for that. It uses a Polhemus sensor.

The price I was given at SIGGRAPH was $4K, no specific mention of volume
discounts. The Flying Mouse was being demoed in the Polhemus booth.

In my (humble) opinion, one big difference between these two devices is
ergonomic. Both are standard mouses on the ground, and 6D mouses in the
air. (I'm pretty sure) both have RS-232 interfaces. The Flying Mouse is a
largish thing, about like a DECStation mouse with wings that fit nicely under
the thumb and pinky, while The Bird is a tiny thing that made my hand cramp
just looking at it.

Notice how everything using the Polhemus sensor has a pretty hefty price
tag on it? My guess is that nobody gets the sensors for less than $1K or so.
I'd like to know the single-unit price on it if anyone happens to know.

SimGraphics had an engineering proto of a Flying Mouse with a hacked-up
tactile sensing. Next year, maybe. Interesting, interesting...

>-- Jonathan
>
>Jonathan Rice, jorice@cs.tcd.ie
>
>Disclaimer: I've no connection with Ascension or SimGraphics. I've a slender
>enough connection with anything in the real world, come to think of it.

John R. Murray          | "They call me Mr. Know-it-all, I am so eloquent.
murray@vs2.scri.fsu.edu |  Perfection is my middle name!
                        |   ...and whatever rhymes with 'eloquent'." - Primus

pepke@SCRI1.SCRI.FSU.EDU (Eric Pepke) (09/24/90)

The Polhemus tracker is made, surprisingly enough, by Polhemus Navigation 
Sciences (which used to be Polhemus Data Services, I think.)

Byte magazine had a little summary of the Polhemus tracker in the same 
issue that described the DataGlove, July 1990.  The transmitter has three 
orthogonal coils of wire.  So does the receiver.  Each of the 
transmitter's three coils is pulsed, and you get amplitude measurements 
for each of the three reciever's coils, for nine measurements.  That's 
enough.

I think the tracker runs a couple of $K.  Beacuse everybody uses their 
tracker, I guess they've got a pretty tight patent on it.

Eric Pepke                                    INTERNET: pepke@gw.scri.fsu.edu
Supercomputer Computations Research Institute MFENET:   pepke@fsu
Florida State University                      SPAN:     scri::pepke
Tallahassee, FL 32306-4052                    BITNET:   pepke@fsu

Disclaimer: My employers seldom even LISTEN to my opinions.
Meta-disclaimer: Any society that needs disclaimers has too many lawyers.

cdshaw@cs.UAlberta.CA (Chris Shaw) (09/27/90)

In article <8073@milton.u.washington.edu> you write:
>In article jorice@maths.tcd.ie (Jonathan Rice) writes:
>>*DC* field. 
>>        The Bird (TM) 6D graphics input device

The Bird uses a continuous version of electromagnetic sensing (what someone
refers to as DC field), whereas the Polhemus uses some sort of pulsed sensing.
I don't know the electronic details. 

>I was told at SIGGRAPH that The Bird also uses a Polhemus sensor.

No. The Bird uses Ascencion's "own" sensor technology.

>other hand, the blurb I've got in front of me touts it as "the only 6D
>input system that works...near metal and other obstructions," 

True. However, when I used it, it seemed to me that the cable was artificially
too short, and that the effective range of the Bird was about 2.5 feet,
less than Polhemus' 4-5 feet. Also, either the demo program introduced lots of
lag or the Bird was heavily lagged. Either way, I think that the Bird's 
metal immunity is bought at the expense of limited range and slow response
time.

>(maybe Polhemus makes both, now?)

No. Although there's a legal battle underway. The president of Ascension
is an ex Polhemus employee. You can guess the stolen technology arguments.

>Also, the $5000 price tag is the single-unit price. 2-4 units saves 20%,
> 75-99 saves >55%, >99 "consult Ascension."

The Ascension guy's dreaming. He's got to get his software up to snuff first.
Anyway, do you plan on buying more than 4?

>>There's also this similar device called the Flying Mouse (TM):
>
>In my (humble) opinion, one big difference between these two devices is
>ergonomic. Both are standard mouses on the ground, and 6D mouses in the
>air. (I'm pretty sure) both have RS-232 interfaces. The Flying Mouse is a
>largish thing, about like a DECStation mouse with wings that fit nicely under
>the thumb and pinky, while The Bird is a tiny thing that made my hand cramp
>just looking at it.

I thought that the mouse sizes were about equal, but the SimGraphics thing was
better because it fit the right hand better (or worse, if you're left handed).

>Notice how everything using the Polhemus sensor has a pretty hefty price
>tag on it? My guess is that nobody gets the sensors for less than $1K or so.
>I'd like to know the single-unit price on it if anyone happens to know.

Basic Polhemus Isotrak is $3000. That's one source & one sensor. If you're
going to use more than one sensor, get one of the multi-sensor single source
models. It's a lot more convenient, and you have only one reference frame
instead of N for N sensors.

>SimGraphics had an engineering proto of a Flying Mouse with a hacked-up
>tactile sensing. Next year, maybe. Interesting, interesting...

Actually, it was another company who was doing the Flying Mouse hack with
Sim Graphics' help. Anyway, I think that SimGraphics' software is a whole lot
better than what Ascension is offering. Ascension is offering some doofus
serial line driver software for PC's I think (like Polhemus with Isotraks),
while SimGraphics has integrated its Flying Mouse into the Iris GL
Graphics Library, and other machines also, I'm told. You can get Flying Mouse
events in the GL event queue if you want. That alone (i.e. painless system
integration with no kernel hacks) is almost worth the price of admission.

>John R. Murray          | "They call me Mr. Know-it-all, I am so eloquent.

-- 
Chris Shaw     University of Alberta
cdshaw@cs.UAlberta.ca           Now with new, minty Internet flavour!
CatchPhrase: Bogus as HELL !

cdshaw@cs.ualberta.ca (Chris Shaw) (09/27/90)

>In article jorice@maths.tcd.ie (Jonathan Rice) writes:
 
>>*DC* field.
>>        The Bird (TM) 6D graphics input device
 
The Bird uses a continuous version of electromagnetic sensing (what someone
refers to as DC field), whereas the Polhemus uses some sort of pulsed sensing.
I don't know the electronic details.
 
>I was told at SIGGRAPH that The Bird also uses a Polhemus sensor.
 
No. The Bird uses Ascencion's "own" sensor technology.
 
>other hand, the blurb I've got in front of me touts it as "the only 6D
>input system that works...near metal and other obstructions,"
 
True. However, when I used it, it seemed to me that the cable was artificially
too short, and that the effective range of the Bird was about 2.5 feet,
less than Polhemus' 4-5 feet. Also, either the demo program introduced lots of
lag or the Bird was heavily lagged. Either way, I think that the Bird's
metal immunity is bought at the expense of limited range and slow response
time.
 
>(maybe Polhemus makes both, now?)
 
 
No. Although there's a legal battle underway. The president of Ascension
is an ex Polhemus employee. You can guess the stolen technology arguments.
 
>Also, the $5000 price tag is the single-unit price. 2-4 units saves 20%,
> 75-99 saves >55%, >99 "consult Ascension."
 
The Ascension guy's dreaming. He's got to get his software up to snuff first.
Anyway, do you plan on buying more than 4?
 
>>There's also this similar device called the Flying Mouse (TM):
>
>In my (humble) opinion, one big difference between these two devices is
>ergonomic. Both are standard mouses on the ground, and 6D mouses in the
>air. (I'm pretty sure) both have RS-232 interfaces. The Flying Mouse is a
>largish thing, about like a DECStation mouse with wings that fit nicely under
>the thumb and pinky, while The Bird is a tiny thing that made my hand cramp
>just looking at it.
 
I thought that the mouse sizes were about equal, but the SimGraphics thing was
better because it fit the right hand better (or worse, if you're left handed).
 
 
>Notice how everything using the Polhemus sensor has a pretty hefty price
>tag on it? My guess is that nobody gets the sensors for less than $1K or so.
>I'd like to know the single-unit price on it if anyone happens to know.
 
Basic Polhemus Isotrak is $3000. That's one source & one sensor. If you're
going to use more than one sensor, get one of the multi-sensor single source
models. It's a lot more convenient, and you have only one reference frame
instead of N for N sensors.
 
>SimGraphics had an engineering proto of a Flying Mouse with a hacked-up
>tactile sensing. Next year, maybe. Interesting, interesting...
 
Actually, it was another company who was doing the Flying Mouse hack with
Sim Graphics' help. Anyway, I think that SimGraphics' software is a whole lot
better than what Ascension is offering. Ascension is offering some doofus
serial line driver software for PC's I think (like Polhemus with Isotraks),
while SimGraphics has integrated its Flying Mouse into the Iris GL
Graphics Library, and other machines also, I'm told. You can get Flying Mouse
events in the GL event queue if you want. That alone (i.e. painless system
integration with no kernel hacks) is almost worth the price of admission.
 
>John R. Murray          | "They call me Mr. Know-it-all, I am so eloquent.
 
 
--
Chris Shaw     University of Alberta
cdshaw@cs.UAlberta.ca           Now with new, minty Internet flavour!
CatchPhrase: Bogus as HELL !