|The basic TV Typewriter II was based on a design that I did for Datapoint Corporation during the 1971 time frame. The features were very similar
to those included in the Datapoint 3360 Autotab terminal for large data
retrieval systems. During 1973, I was the engineer for a sign company
and we needed a method of sending data to large message board signs and
since the Datapoint terminal was very expensive I designed a very cheap
version that could store the sign data on the screen and send it to the
signs to display the messages. The sign company version stored 2000
characters on the screen and with embedded control functions could send
the data to the signs as either flashing or scrolling messages. After
building 20 of the 300 kilowatt message board signs the company went
into bankruptcy, due to legal battles over distribution rights.
At the bankruptcy sale Dan Meyers saw the terminal and offered me a deal on designing an educational terminal for royalties of $5 per terminal. I had to simplify the terminal by reducing the memory to 1024 and reducing the display from 80 characters and 25 lines scrolling to 32 characters by 16 lines and two pages. He wanted very loose spacing on the parts and very heavy lands to make it easy for inexperienced people to build. He also wanted to make the design modular so that he could sell separate features to make more money. The design was finished by July of 1974 and submitted to Radio-Electronics. It was finally published in January of 1975.
About this time Motorola approached Dan Meyers and offered him the design for the computer system with the MikBug ROM. He took the Motorola design but messed up by using tin plated Molex instead of gold.. Gimix did follow this path and while not producing as sophisticated a product did produce a much higher quality product. Dan Meyers always wanted his products to look inexpensive and unsophisticated and he refused to come off this approach even after going into business computers. In 1976 I sold him the design for the terminal used by the sign company, but he never produced it. He later told me that he bought it to prevent anyone else from getting it as he had the CT-1024 redesigned to do 64 characters and scrolling (not what I would have called scrolling) and my royalties ended.
I do not know exactly know much I got from SWTPC in royalties as most of my income was from repairing terminals that customers could not complete. I charged $25 dollars to repair the CT 1024 Boards and did possibly 10 per week. Most of the problems were with the soldering and transistors inserted wrong or in the wrong place. I never did get an accounting of the sales from SWTPC and just received a check at the end of the month. In one of the conversations with Dan Meyers he told me that he was making more money on this product than any other that he had sold previously
In 1977 I showed him a hard disk system that I had set up for a programmer that was sharing my office with me. It was a 20 megabyte removable cartridge system. He was interested but did not want it because I was using an Intel DMA chip in the design. During 1978 I got some of the Motorola DMA chips (6844) and by this time I had enough experience to design a really good hard disk system. The system used a 6800 microprocessor for internal functions and the DMA chip for data transfer. The design was for daisy chaining multiple 15 inch Marksman platters for mass storage. About 33% of the chips were for expansion and he never produced the multiple platter design, but sold the system as the CDS-1 with 20-60 megabyte platters. The availability of the hard disk allowed him to sell the UniFLEX system in the 1980 time frame. I collected royalties on this system from 1979 through 1983.
The CDS-1 was much more profitable for me than the CT-1024. The royalties were for each hard drive board and an additional fee to check out and format the disks. I did at least 5 hard drive systems per week but spent most of my time writing assembly level programs to format, check and refine the operation of the hard disk. The first ROM program for the hard drive was written by Norm Rietzel who was the head programmer for SWTPC. This version was only slightly faster than a floppy as it only read one sector per revolution. I rewrote it to completely fill the buffers on each revolution with multiple sector look ahead.. The operating system could not take advantage of the disk read speed. The UniFLEX operating system used the hard disk mostly for swapping to support the multiple user function. The CDS-1 was very profitable for SWTPC as he sold them for $4000 or more, depending on size.
Wayne Green visited with me (and SWTPC) on several occasions. He originally started Byte magazine and later Kilobaud but is most famous for 73 magazine (Radio Amateur). He lived and worked out of Peterborough N.H. which is where I grew up until 13 years old. SWTPC would have been much popular and successful if Dan Meyers and Wayne Green could have gotten along.. They disagreed on cost of software. Wayne wanted to start a system for users to get free software and Dan insisted that the user had to pay for any thing that worked on a SWTPC computer.
TV Typewriters Page
Michael Holley's SWTPC Collection Home Page
This page was last edited July 20, 2013