home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!spool.mu.edu!darwin.sura.net!haven.umd.edu!ni.umd.edu!sayshell.umd.edu!louie
- From: louie@sayshell.umd.edu (Louis A. Mamakos)
- Newsgroups: comp.protocols.time.ntp
- Subject: Re: uVAXII clock
- Date: 19 Nov 1992 13:56:17 GMT
- Organization: University of Maryland, College Park
- Lines: 23
- Distribution: inet
- Message-ID: <1eg6e1INNnr4@ni.umd.edu>
- References: <1992Nov18.140049.199016@rrz.uni-koeln.de>
- NNTP-Posting-Host: sayshell.umd.edu
-
- In article <1992Nov18.140049.199016@rrz.uni-koeln.de> schittel@geo.Uni-Koeln.DE (Christoph Schittel) writes:
-
- >Under VMS a uVAX keeps time very well. What about
- >Ultrix on uVAX? Is it a good idea to use an old
- >uVAXII (in constant temperature environment) as stratum
- >1 server?
-
- Given that a MicroVAX-II CPU's clock only has a resolution of 20ms,
- this is some new interpretation of "keeps time very well."
-
- The problem with the MicroVAX-II interval timer (which is used to keep
- time on the running system by periodically generating interrupts) is
- that you cannot determine the time remaining until the next interval
- timer interrupt, so you stuck with the resolution that corresponds to
- how frequently the interval timer is programmed to generate
- interrupts. This is (on 4.3 BSD systems) 20ms.
-
- While it may be that the clock doesn't drift very badly, the fact that
- the resolution is only 20ms somewhat precludes it as a good source of
- time for others.
-
- louie
-
-