From: brain@msen.com (Jim Brain) Subject: INFO: The PowerGlove saga Date: 28 Dec 1994 14:37:05 -0500 Organization: Brain Innovations, Inc. Finally, for Christmas, I managed to get my hands on a copy of both Linda Jacobson's "Garage VR" book and Joe Gradecki's 2nd book. I even managed to check out my .00001 second of fame on page 137 of Linda's book. Oh well. The following can be printed out if you care, and stuffed in your Powerglove archives, or hit "n" now. Any, as I am reading the "Cockpit" section in Linda's book, I thought I might add a bit, as I remember it, from my memory: Linda notes that Jay Eric Townsend claims that the original PowerGlove Serial Adaptor (PGSA) posesses "an almost mythical status." Well, here is the scoop: The PGSA was a small interface packaged in a 3" by 5" by 1.5" black hobbyist box. The unit was built on a 4.5" or so by 2" printed circuit board, and contained a power supply connector, (Small mono earphone jack, as I recall, but not sure), a 68705 microcontroller, a rs232 db-25 connector and a NUBY (Then called some other name) extendo cable. The unit had on board 5 volt regulator. Also, it had the RS-232 interface, but I am not sure if it was a MAXIM chip or the old 1488/1489 chips. Anyway, I know this because I saw the unit that Greg Newby at the UIUC has (He has two, but I think one is just a printed circuit board. Alas, when I looked at the PGSA unit in 1992, Greg noted that it was not working. I believe it had a voltage regulator out. (If I remember correctly it had taken his SGI serial port with it, but can't be sure). When the newly formed Special Interest Group for COmputer Architecture was having early meetings, it was noted that all of the projects we wantd to work on requires some major money input, so we started looking for fund- raiser projects. At the time, Greg Newby was looking for someone to start re-manufacturing the PGSA boxes. Through ACm member Ben Gross, we were approached to try and do the task. At the time, Mike Schaffstein, Mike Stangel, and I were in the group. I believe Stangel was the chair. We all started having meetings with Greg and Ben to determine how we should proceed. After initial contacts, I received one of the boxes, plus run down on the box. Just a few of the boxes were ever made ~100-300. This explains the amateur look of the box. Greg was still in contact with Chris Gentile at the time, as they had worked together at Greg previous university job. Anyway. Greg started me off on this newsgroup, as well as pointing me to the ftp site karazm.math.edu (something like that, I can't ever get it right). There, I found the PG timing diagram, and Ron Menelli's (got his name right) HC11 code. Greg also gave me a copy of the PGSA code as best he had it. I looked over all of it. After a while the PGSI core consisted of Myself and Mike Schaffstein. Mike suggected that I do pricing on the units and feasibility. After looking both the PGSA and the Menelli box over, I decided to merge the two units. Why? Well, even though Greg had supplied the AGE PGSA code to me, it was incomplete, and I did not have the compiler I needed. However, at the time, I had easy access to HC11 development tools and a dev board, and had written substantial HC11 code for other concerns. However, the PGSA output was more complete than the Menelli box. Also, the HC11 runs twice as fast as the HC05 in the PGSA, and the HC11 has a built in serial port, which the HC05 doesn't have (The AGE code replicates the serial port in SW). So, I took Ron's code, and started hacking it to combine the best of both worlds. At that point, I took over the dev of the unit. I worked on cosign for most of the summer of 1992 on squeezing as much as I could into the 2K of EEPROM on the HC11E2 I had gotten fom the UIUC labs. My goal was to allow people who had AGE code to be able to use the box, but allow people who used Menelli code to be able to continue using the Menelli commmands. (20/20 hindsight shows that I should have scrapped the Menelli commands, since the AGE commands were easier to code up in SW) Also, I wanted to get the frame speed of the glove up. A while down the road, I noticed a lot of people wanting to use Greg Alt's Shutter glasses circuit, so I incorporated it into the PGSA II, as it was called. Work continued. The unit was completed and sent to production in March 1993, after enough prepaid orders came in. As it was my last semester, I moved Heaven and Earth to get as much done before I graduated. Needless to say, the untis arrived during finals wek, so we finished the assemblies right before I had to leave. Some who have been here a while remember the bi-weekly progress reports. (nuff said) The name was changed, since Greg said he never did like the PGSA II name, initial two-glove support was planned for in the hardware, but was never finished due t some problems with another graduate student who was working on a similar unit that would provide 2 glove support (It's been a year and some, and I never saw the other unit come out). The hardware supports the second glove, but the SW needs to be upgraded. I noticed that the Menelli code was made from Atari ST code from someone in Europe. I did receive the AGE PGSA programmer's book (yes, there really is one), but it didn't have the byte stream coming out of the glove, just the buffered byte stream. The raw stream sae from the Menelli code and some Dave Stamp readme files on the PC parallel port interface. Since I had looked at the AGE code, and some of the proprietary deglitching routines in the PGSA were used in the PGSI, we paid AGE some royalties. (Yes, even the original PGSA glitched, at least it had provisions to get rid of glitching) Along the road, I met up with Michael Starks, of 3DTV. Now, I seem to remember him talking about how he was involved in some of the early PG dev, but he is the guy to ask. One thing I always wanted to know but never found out was what the bytes were in the init string. In the finished PGSI, I used the AGE init string, although it diffes a little from the Menelli one. Niehter made much difference, though. I have been told that the CPU in the PG runs at .5 MHz, so that is another piece of trivia. One problem I always had with two glove support, was the interlacing of the gloves. You couldn't read both at once, since they interfered with each other. I am intrigued by this Yellow Jack IT thing. So, that is how AGE, Greg Newby, PGSI, PGSA, Jim Brain, Michael Starks, Ron Menelli, his Box, the ACM at UIUC, Greg Alt, Dave Stampe, Chris Gentile, and this group all fit together from my perspective. When I didn't have time to respond to my mail (I used to get 100 of msgs a day), I was contacted by people who wanted to stock the units, decveloper's, some people with a low cost head trakker in mind, and numerous others. A while back, I was going to try and follow up, but even though I saved every piece of mail on VR I ever got, the mail from the beginning was out of date even by the time the product went to the PCB shop. Ahh, fond memories. Its a shame that gloves are so hard to find. Its also a shame I stepped out of the VR world (no pun) for even a little bit, since things move at lightning's pace in here. For those who do use the PGSI or have one, I am working on drivers, since people have noted that being a bad point of the unit, and it is mentioned in Linda's book. Anyway, I ahven't a clue when they will be done, but I though I would at least let you know. I recommend the book(s). I finally get to see Joe GRadecki's picture, since he published my article, and se some of the other faces. And Eric Townsend, if you are listening, you can have this post for the glove FAQ if you wish. -- Jim Brain, Embedded Systems Designer, Brain Innovations. brain@msen.com Dabbling in VR, Old Commodore Computers, and Good Times! "The above views DO reflect my employer, since I am my employer" - Jim Brain