rwessel (robertwessel.delete@this.yahoo.com) on January 11, 2020 7:20 pm wrote:
> Brett (ggtgp.delete@this.yahoo.com) on January 11, 2020 2:16 pm wrote:
>
> > Intel invested in the 432 failure, but also developed the over designed and expensive 286 which never had a
> > successful OS. Let’s say the 286 would have faced the
> > cheaper eZ80 or Z380 without the expensive MMU taking
> > half the die. Intel may still have won, but the cheaper Z80 line would dominate POS system terminals, etc.
>
> Windows 3.x would certainly count as a success, at least FSVO "operating system". Remember that
> Win3x was straight 16-bit protected mode, and ran just fine in both 286 and 386Enh mode. In
> fact it ran the *same* code in both modes. The big advantage of the later was DOS multitasking,
> mostly handled in the 386 version of the DOS extender underlying things in that mode.

The 286 came out in 1982, Windows 3.1 came out in 1992, the 386 came out in 1985.
OS/2 came out in late 1987 but did not get a GUI until the next year.

The MMU on the 286 was always a joke doomed to failure.
Half the die wasted on this MMU while the CPU was used as a faster 8086.

This was a huge opportunity for a improved address extended Z80 to jump in with a cheaper solution that is just as fast with backwards CP/M compatibility while DOS was still running CP/M ports.

I would hope that management today would continue iterating the existing CPU while working on the next bet the company breakthrough as most breakthroughs fail.
Zilog management failed hard.

The 65816 in the AppleIIGS rescued Apple while the company waited for Macintosh volumes to grow.