From: CAREY SCHUG via Talk <talk@lists.gtalug.org>
Talk about Hayes refers to 1200 baud (or faster), right? Ah, you young people.
The first terminal I owned was a clone of an IBM 2741, with a built-in modem. Think of an IBM Selectric typewriter turned into a terminal. It ran at 110 bits/second. That was a limitation of the typewriter; the modem itself could run at 300 bits/second but there was no connector with which to access it (we did access it with jumper wires).
Anybody remember 300 baud days? And especially the Potomic Micro Magic PMMI-103 modem that would squeeze 600 baud out of the 300 baud spec, running on S-100? A few CBBSes would negotiate that rate. That modem was the first thing I bought, signing up for a pre-sale discount at a Chicago Area Computer Hobbyist Exchange (CACHE) meeting.
I don't remember this 600 baud rate. From Wikipedia: The (US) Bell 101 data set (introduced 1958) was 110 bits/second. It was used by the SAGE for North American air defense. https://en.wikipedia.org/wiki/Bell_101 https://en.wikipedia.org/wiki/Semi-Automatic_Ground_Environment The (US) Bell 103 (1963) modem was what made the 300 bits/second standard. https://en.wikipedia.org/wiki/Bell_103 The Bell 202 (1976) modem is what made 1200 bits/second standard. https://en.wikipedia.org/wiki/Bell_202 CACHE was pretty historic. The first BBS was from members Ward Christensen and Randy Suess. https://en.wikipedia.org/wiki/Bulletin_board_system
After I received it, I waited till there was an S-100 adapter offered for the TRS-80 Model 1, but that was a fraud; I had to wait till another came out. Talk about fun. Modifying S-100 printer and disk controllers to look like TRS-80, and programming modem drivers in hex to connect it to BASIC. Also putting 64K memory into the base unit and programming the extra 16K as a printer buffer.
Wow. How did you access 64k of RAM when some of the address space was for the ROM? I seem to remember the ROM was at low addresses, complicating CP/M. Often i8080 and Z80 interrupt hardware injected RST instructions. These were 1 byte instructions that could call instructions in one of 8 locations, 8 bytes apart, starting at 0. Fixing the content of those locations as ROM might be a problem. Surprisingly, interrupts weren't very important in PC operating systems in that era. Certainly the main program I wrote for that architecture wanted to control that memory -- a 1-byte subroutine call is quite a bit more compact than a 3-byte call.