The top-secret prototype that went missing

11/3/2010 8:14 AM EDT

I was working for Tandy at the time this machine was being developed. The company was paranoid beyond belief about it because information on the earlier machine had been leaked to the press. The company became divided into the knows and don’t knows. Even in Tech Support, only a few people knew about it. Customer service was totally in the dark, except for the director--a guy named Bill.

One day much to my consternation a guy named Kam from customer service came by and asked, “What’s a Model 16?” I had to keep a straight face, play dumb and say "What’s that?" Kam told me that a customer had called in on the 800 number complaining that he his Model 16 kept getting disk errors.

I should explain that the Model II and the Model 16 were the same as far as the Z80 side of the computer went. But the disk drive in the Model II was an AC-powered unit with 8-inch, single-sided media that ran continuously and on access loaded the head. The Model 16 used a DC-powered drive that was half the thickness of the Model II and on drive access would start rotating. The drive was much faster in stepping, but it had that startup delay to spin up on first access. As a result, if you tried to run the Model II operating system, you would get disk errors all the time and had to retry to keep the drive spinning.

Since I wasn't allowed to talk to anyone else about it, I went over to speak with Bill in customer service. Mystified, we decided to call the customer to confirm he actually had a Model 16. The stock number and serial number checked out. Sure enough, he had one--in Lancaster PA of all places! How the heck?

The customer had purchased what he thought were four Model II computers for his business. He set up three of them at work and took the fourth computer home, which happened to be the Model 16.--in a Model II box! We couldn't figure out how he had it and he didn't know what the heck he had. And while he said he’d pay any difference in price, he would not give it back.

We made some quick decisions, and sent the customer an NDA. We brought Kam back in and swore him to secrecy. The customer was given Kam's name and direct phone number and told not to talk to anyone  at Tandy but Kam, Bill, and me about it. As soon as the customer sent us back the signed NDA, I gave him patches that would allow for a ‘special’ version of TRSDOS that would work on both models. It would step the drives at the slower Model II rate and also put an access timeout in place to prevent the disk access errors.

This became known as TRSDOS 2.0d for ‘dual.’ I did it for the repair centers so they could have a single operating system that would work on any of the machines instead of having multiple disks on hand. I later had to add reverse copy protection, but that’s yet another story!

The customer then became one of our "field testers," and as far as I know, he never leaked a word and was never charged for the difference.

Later on I pieced together what had happened: The origin of this mystery dated back to 1979, when the Model II was announced. As a test, Tandy delivered ten units to selected cities. I was sent down to Miami to oversee the delivery of the machines there.

Originally I was supposed to just look and observe, but the first two machines that were opened up at the computer centers were damaged beyond belief. The problem was the mechanical design.

The card cage was fastened to a metal plate on the computer's plastic case. On the left side was a fairly large power supply and on the right the 8-inch drive, all  in a vertical orientations. In shipping, the metal plate and case would flexed, and the top of the disk drive would swing back and forth, breaking pieces off the case. I had to make repairs to the first two machines just to get them functional and also ordered new case parts. At that point I elected to drive around Florida to check out each machine upon delivery. Good decision, as every one of them needed repair. I also got to talk to all the customers, apologize, and make sure everybody was happy.

Then I got to Key West, where the last Model II was scheduled for delivery. But. it wasn’t there. It was Friday night, so I had to wait until Saturday morning to even be able to call anybody to inquire about it. Not bad, being stuck in Key West for a weekend. I tried to go out snorkeling to Sand Key, but Hurricane Fredrick bore in on us. The weather started getting rough, the tiny ship was tossed. We made it back to Key West with half the people in the boat having lost their lunch en route.

Monday morning I found out the Key West machine was never shipped because it was a 32K version instead of a 64K version. I got back to Miami safely and the trip was deemed a success. Braces were quickly added to the machine design and the problem was solved.

Naturally the factory didn’t want another disaster like this with the Model 16 so when it was close to release they quietly boxed up ten units--in Model II boxes--and shipped them to Philadelphia with routing tags so they could be shipped right back. Good idea, actually. But when the machines came back, there were nine  Model 16s and one Model II. The crew at the Philadelphia warehouse figured they were all the same and just sent a count of boxes.

The factory went crazy about it, but no one told anyone in management that they had lost a machine. Since it was in a standard Model II box it eventually made its way to a computer center, where it was sold to the customer. Mystery solved.

Mike Yetsko attended Lehigh University in the early 1970s and worked for Tandy for 9 years through the early days of the TRS-80. He currently works at Wavemark designing RFID products and has been a HAM radio operator for over 25 years. He enjoys skiing and gets out on the slopes as much as he can.