My first computer was an Atari 400 (although I wish it had been an 800, but I couldn't afford it at the time). This was probably around 1982-1983, but my brain has gotten a bit fuzzy so I could be wrong on this. Anyway the first thing I noticed when I brought it home, was that you couldn't do anything except use it as a TV typewriter (my 400 didn't come with BASIC or even a game). So much to my disappointment, I had to wait until my next paycheck to buy a BASIC Cartridge. But when I did, a whole new world opened up for me, and my life as a programmer began. Also not liking the membrane keyboard, I was able to find a used 800 keyboard and hacked my case to accept it (it wasn't exactly pretty).
As time went on I slowly added peripherals to my setup (a cassette deck at first, and eventually a printer, and a disk drive soon followed). My first printer was an old teletype that I picked up for $50 at computer swap meet (it was a tank). This also got me into machine language programming, having picked up a MAC65 cart. I quickly set about writing a new print handler to reside in Page 6 memory (my brother got me started, but I ended up getting it to work). The printer was interfaced to a couple of joystick I/O bits (crude serial interface), and it was the coolest thing when I actually got it to print. I was estatic!
My job at the time, required that I perform performance tests on cryogenic chillers, which was done with very crude equipment and much manual labor. I started thinking that perhaps I could write a program on the Atari that could automate some of this, and soon was born a multi-channel temperature data monitor program, with the ability to cycle the unit from Standby-to Cool-to Defrost, as well as turn the unit off when the test was complete. My boss was quite impressed and footed the bill to buy a couple more Atari's (800XL's), that I soon put to work in the testing area.
Later I really started to dislike the keyboards that came with the 800XL's, and set about to design a method of interfacing an IBM keyboard directly to Pokey's scan lines. The first version was very crude, and was done with a "gates" only method. It sort of worked, but lacked many of the required key combinations. So I bit the bullet and opted for a microprocessor based design incorporating a 6504 CPU, a 6520 PIA, and two other "glue" logic chips to tie it all together. To keep it small and to cut costs, I opted not to include any RAM or the even better suited 6532 RIOT, which would have at least given me some decent variable storage space. Anyway by carefully taking advantage of any unused bits in the 6520, and making no subroutine calls, I was able to write an initial program for the 6504 that actually worked fairly decent, doing so without any RAM. However seeing the need to buffer keystrokes, and liking the idea of having saveable macros, I eventually opted to include some battery-backed ram. This iteration soon became a selling product, and was called TransKey (I later sold the rights to DataQue).
As history has it, I also developed a couple of different Video Genlocking designs (B&W and Color), but then drifted away from all things Atari somewhere around 1995 (yes I became a PC computer user). Anyway after 20 years, I now find myself once again thinking about playing around with these classic 8-bit computers, and that is what has brought me to the AtariAge website and forums.