4

The Motorola 680x0 processor family was used in a variety of systems, including Apple Macintosh, Atari ST, Commodore Amiga, Sharp X68000, and Sinclair QL systems. Had any of these customers wanted a particular feature added to the architecture ("we want ___"), they had enough purchasing power to make Motorola at least consider an idea.

Were any features of any M68k family processors initiated by customer request?

  • Consider only the processors themselves, not peripherals. CPU32, DragonBall, and ColdFire do count.
  • By "customers", I mean those who purchased the processors, not Motorola itself, and not because of programming language features.
13
  • 6
    Most likely the best place to look at such requests would be the early Sun workstations - The Sun-1 went through hoops to implement virtual memory on the 68000 - Early Sun 68k workstations were so tightly bound to the 68k architecture that they will definitely have asked for VM support, restartable instructions and thelikes in talks to their major supplier. It will however, be nearly impossible to find exact evidence of that.
    – tofro
    Sep 13, 2021 at 11:59
  • 4
    See my comment on the question retrocomputing.stackexchange.com/q/21782/11247 . To me it seems that several of the instructions of the M68k family match quite well the needs of the Ada language compilers. Ada was developed at the same time as MC68K. So the DoD may have been such a customer. Sep 13, 2021 at 13:28
  • 1
    Beside never have heard of such requests, it would in fact surprise me. The 68k family is quite a school book, middle of the road CPU. No fancy features, no nifty hacks to squeeze out more performance from less hardware - even less any customer specific functionality.
    – Raffzahn
    Sep 13, 2021 at 23:05
  • 3
    @LarsBrinkhoff Sorry, but that's a myth and simply false. While the Micro/370 uses a partial superset of the 68k bus (32 bit address and data, able to use 8/16 bit motorola devices), is is neither a modification or in any way related to the 68k or made by Motorola. It was a 100% inhouse design by IBM and manufactured there as well. The only relation here is that the CPU architecture was designed by Nick Tredennick, who happened to do the same job for the 68k. The reason to make the bus in part compatible was to allow easy used of 6800(0) peripherals, saving time and money.
    – Raffzahn
    Sep 14, 2021 at 12:34
  • 1
    I was sure I had read back in the day that some instruction or couple of instructions were added to one of the 680x0s for a specific customer. I had a bunch of processor manuals I bought from Motorola directly that I didn't really understand and some were for esoteric derivatives, so my memory is very fuzzy. I tend to think something was introduced in one of the CPUs after the 68000 and ditched again by the 68040. Separately I think I heard that it was added for Ford, but I'd trust that recollection even less. Nov 25, 2022 at 4:38

1 Answer 1

1

The MC68000 was considered by Apple only after the 68k was architected, so no input direct from Apple was used in its architecture. According to Jobs, Apple engineering was sidetracked during that time-frame by a "16-bit" bit-slice architecture proposed by Woz. Also in that time frame, the Mac team initially started with a 6809 prototype designed by Burrell Smith, not a 68k.

Amiga (its Hi Toro predecessor) was founded after the 68000 and 68008 were already well into in production. Atari's 68k design was started even later.

2
  • 4
    Thank you for the answer, but this doesn't answer the the question, which is not just about the 68000, but the entire family that includes the 68020, 68030, etc.
    – DrSheldon
    Sep 17, 2021 at 3:32
  • Unfortunately this just talks about what didn't happen, not what did happen, so doesn't address the question.
    – TonyM
    Sep 19, 2021 at 7:15

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .