mash@mips.COM (John Mashey) (06/10/90)
In article <1029@s6.Morgan.COM> amull@Morgan.COM (Andrew P. Mullhaupt) writes: >Any clock that takes up 300K (or more) had better keep great time >when you consider the fraction of even a large workstation which >will be occupied by it. Having just returned from a trip to read the 100+ new articles in comp.arch.... there's an awful lot being talked about that doesn't seem to have much to do with computer architecture. This seems especially silly, given the number of interesting new machines being announced, current&imminent fights over LIW/superscalar/fast scalar/ superpipelined architectures, etc, etc, i.e., other useful things to talk about. With regard to the clock business.... to be honest, I've never quite figured out the fascination for running such things, given the ubiquitous existence of a servicable substitute that uses 0K and 0 CPU, i.e., a wristwatch... Would anyone who has data care to guess what fraction of CPU cycles and network traffic is consumed by clocks? -- -john mashey DISCLAIMER: <generic disclaimer, I speak for me only, etc> UUCP: mash@mips.com OR {ames,decwrl,prls,pyramid}!mips!mash DDD: 408-524-7015, 524-8253 or (main number) 408-720-1700 USPS: MIPS Computer Systems, 930 E. Arques, Sunnyvale, CA 94086