Comment by mrandish

As I teenager I got my first computer in 1982 - a 6809-based 4K model from Radio Shack named just "Color Computer" (because all their other models were B&W)! I taught myself BASIC and then assembler on that machine. Because I had nothing to compare it to, I didn't know the 6809 was the most advanced of the 8-bit CPUs, with 16 bit registers, layered interrupts, separate user and system stack pointers and a highly orthogonal instruction set enabling re-entrant, position-independent code, relative addressing, indirect pointers and complex addressing modes. In 1985 I naturally graduated to the 68000 based Amiga and was immediately at home with its powerful addressing modes and deep architecture. It was only decades later when I bought 6502 and Z80 systems as a retro collector and did some assembler on them that I learned just how spoiled I was with the 09!

There was even a multi-tasking, multi-user operating system called OS-9 created for the 6809 that was quite UNIX-like. Businesses actually connected serial terminals and supported four or more simultaneous users doing work all day on these little 8-bit, 64K micros. It was extremely capable and even quite elegant in it's architecture.

Unfortunately, in the 70s Motorola misjudged how large the market for personal microcomputers would grow and over-priced the original 6800 (1975: basic 8-bit) and 6809 (1978: advanced 8/16 bit). Even though the 6809 was more than double the clock for clock performance of a 6502 or Z80, at four times the price, it was a tough sell to consumer computer makers. By the time Motorola lowered the price, it was too late as platform choices had been made and the 68000, the 6809's 16/32-bit big brother, was just around the corner. A key reason Jobs may have been able to cut a killer deal to put the 68000 in the Mac was simply that Motorola had been losing almost every big CPU design win based on their earlier mis-estimates of the market.

But if you made a bar bet today to do something challenging on a 1970s 8-bit CPU, you'd win by picking a 6809 or, even better, its lesser-known CMOS version the 6309, created as a second-source part by Hitachi. Being CMOS the 6309 was operationally identical to the 6809 but could run at 3.5 Mhz vs the 6809's 2 Mhz. The 6309 also has a 'secret' alternate mode that saves cycles over the 6809 on many common operations as well as adding several new instructions including a hardware divide. The paper being in Japanese, I was surprised they didn't use a 6309 since it can easily be swapped into any 6809 socket for a quick boost .

Little known history: Apple's original prototypes for the Macintosh were actually based on the 6809 before Jobs negotiated his legendary discount deal for 68000 CPUs. Most pinball machines of the 80s, 90s and even early 00s were based on the 6809. I have a Simpson's Pinball Party machine released in 2003 based on a 6809. Quite a long life for not only an 8-bit ISA, but in the same 1970s package, voltage and clock speed. Interesting to see such an ancient CPU as the brain of a $3,000 modern design driving a 144-pin surface mount FPGA next to it (which probably has orders of magnitude more gates)!