Path: utzoo!utgpu!news-server.csri.toronto.edu!cs.utexas.edu!wuarchive!uunet!mcsun!ukc!uos-ee!cam-cl!news
From: maj@cl.cam.ac.uk (Martyn Johnson)
Newsgroups: comp.protocols.time.ntp
Subject: Re:  clock resolutions etc..
Keywords: adjtime
Message-ID: <1991Jan14.100556.10738@cl.cam.ac.uk>
Date: 14 Jan 91 10:05:56 GMT
References: <9101112100.AA22695@snook.ecs.umass.edu>
Sender: news@cl.cam.ac.uk (The news facility)
Distribution: inet
Organization: U of Cambridge Computer Lab, UK
Lines: 20

Venu S Rayaprolu requests information about the modified clock ticks used 
by adjtime on Ultrix.

On the VAX, the manpage is wrong when it says that ticks of 9ms and 11ms 
are used.  The actual values are 9.960ms and 10.040ms if the amount to be 
adjusted is less than a second, and 9.600ms and 10.400ms if the amount is 
greater than a second.

On the MIPS, the values used are 3.891ms and 3.921ms for small 
adjustments and 3.756ms and 4.056 for large adjustments.

I am not convinced that adjustments less than a clock tick work at all - 
does anyone have evidence otherwise?

(These values were observed by sitting in a tight loop watching the clock 
tick on a MicroVAX II running Ultrix 3.0 and a DECstation 3100 running 
Ultrix 3.1C.  In principle the values are CPU dependent.)

Martyn Johnson
University of Cambridge Computer Lab