On 10/06/2017 03:42 PM, ben via cctalk wrote: > On 10/5/2017 4:19 PM, allison via cctalk wrote: >> On 10/05/2017 03:46 PM, Chuck Guzis via cctalk wrote: >>> On 10/05/2017 04:22 AM, allison via cctalk wrote: >>> >>>> Funny the market knew of the 386 in the fall of '85 but it would be >>>> three years before I'd see >>>> one in the field. Disks and CPUs lagged the introductions by >>>> years due >>>> to cost. >>> It was hard to rationalize the extra cost of a 16MHz 80386 when there >>> was little software or performance gain over a fast 80286 box when >>> running MS-DOS--the dominant OS of the day. >>> >>> I recall an Intel engineer opining on the subject. "We give you a >>> 32-bit advanced architecture CPU and you p*ss it away running DOS." >>> >>> Compatibility is a tough mistress. >>> >>> --Chuck >> >> Moore's law only worked for hardware, software lagged typically two >> years behind. >> >> Of course when we did get something else Venix and winders winders was >> the winner >> and a poor one at that. >> >> Allison >> > > Did anything ever use *ADVANCED ACHITECTURE(?. > Games don't count here. > Ben.
Fred hit the nail on the head. How did computer with less than 64K of memory deal with millions of subscribers? Good code that used the disk as addressable space. By the late 80s I was solving problems on the trailing edge using less speed and power. That and a being a bit crafty. The trick was to not allow creeping featurism one plus the poor thing to death. Allison > >
