mark@hubcap.clemson.edu (Mark Smotherman) (11/15/90)
by chuck@trantor.harris-atd.com (Chuck Musciano): >> And to keep this >> thread valid for comp.arch, what were David Letterman's Top Ten computer >> architectures? > > Ed Borasky's Top Ten Computer Architectures: > > 1. Babbage's Analytical Engine (first computer) > ... You folks must not watch David Letterman. I doubt he would give a serious list for "top ten", rather it would be something like ... 1. A Honeywell mainframe (6000?) with extra bits in the floating-point accumulator for more precision in intermediate calculations - but with no way to save these in memory. Thus they will disappear at random, depending upon interrupts and context switches [1, p. 137]. (Ever thought of floating-point expression evaluation as being a critical section in need of interrupt disabling???) 2. ... Followups to folklore group? -- [1] R. Dewar and M. Smosna, Microprocessors: A Programmer's View, McGraw-Hill, 1990. -- Mark Smotherman, Comp. Sci. Dept., Clemson University, Clemson, SC 29634 INTERNET: mark@hubcap.clemson.edu UUCP: gatech!hubcap!mark
lindsay@gandalf.cs.cmu.edu (Donald Lindsay) (11/16/90)
In article <11635@hubcap.clemson.edu> mark@hubcap.clemson.edu (Mark Smotherman) writes: > (Ever thought of floating-point expression evaluation as being a > critical section in need of interrupt disabling???) Sure. The LSI 11/23 treated slow ops (like floating point) as restartable. Since some ops took a lot of microseconds (50?), a few thousand interrrupts/second could stop the user program in its tracks. -- Don D.C.Lindsay
meissner@osf.org (Michael Meissner) (11/16/90)
In article <11635@hubcap.clemson.edu> mark@hubcap.clemson.edu (Mark Smotherman) writes: | 1. A Honeywell mainframe (6000?) with extra bits in the floating-point | accumulator for more precision in intermediate calculations - but | with no way to save these in memory. Thus they will disappear | at random, depending upon interrupts and context switches [1, p. 137]. | (Ever thought of floating-point expression evaluation as being a | critical section in need of interrupt disabling???) One of the early Data General Eclipse's had this 'feature'. My memory says it was the s200, but it may have been an s150 or some such instead. -- Michael Meissner email: meissner@osf.org phone: 617-621-8861 Open Software Foundation, 11 Cambridge Center, Cambridge, MA, 02142 Considering the flames and intolerance, shouldn't USENET be spelled ABUSENET?