Path: utzoo!attcan!utgpu!jarvis.csri.toronto.edu!mailrus!cs.utexas.edu!tut.cis.ohio-state.edu!unmvax!ncar!ames!uhccux!munnari.oz.au!comp.vuw.ac.nz!windy!gpwd!gpwrdcs
From: don@gp.govt.nz (Don Stokes, GPO)
Newsgroups: comp.arch
Subject: Re: The Burst Computer
Message-ID: <591@gp.govt.nz>
Date: 23 Nov 89 20:00:02 GMT
References: <23567@cup.portal.com> <1989Nov1.213049.1697@ico.isc.com>
Organization: Government Printing Office, Wellington, New Zealand
Lines: 24

In article <1989Nov1.213049.1697@ico.isc.com>, rcd@ico.isc.com (Dick Dunn) writes:
> mmm@cup.portal.com (Mark Robert Thorson) writes:
> ) ...interesting idea for increasing computer performance:  the burst computer.
> ) This would be a computer which delivers 10X or 100X performance over some 
> ) short span of time.
> Sorry...the original question was kind of interesting; it just hit me in a
> strange way.  I've encountered a few pieces of hardware which were built in
> such a way that an (ab)errant piece of OS code could do physical damage
> (like a disk drive without an inner crash stop!).  These somewhat unhappy
> incidents lead me to suggest that the "bursts" would have to be limited by
> hardware, which would probably get in the way of speed.  Also, seriously,
> there are some doubts about reliability for hardware which is temperature-
> cycled to an extreme.

Just an aside - I seem to recall someone putting a scope on the ZX-81's 
clock line way back when, and noting that the clock speed (a nominal 
3.75 MHz or thereabouts) depended heavily on the temperature of the 
machine.... (also noted that the 9 volt power pack actually put out 18 
volts, which tended to heat the machine more than it should have....).

Don Stokes  ZL2TNM  /  /                                 vuwcomp!windy!gpwd!don
Systems Programmer /GP/ Government Printing Office       PSI%0530147000028::DON
__________________/  /__Wellington, New Zealand__________don@gp.govt.nz________
Do not believe in miracles; depend on them.