Subject: Re: Clockticks lost, why ?
To: Henry B. Hotz <hotz@jpl.nasa.gov>
From: Scott Reynolds <scottr@og.org>
List: port-mac68k
Date: 01/30/1997 00:10:55
On Wed, 29 Jan 1997, Henry B. Hotz wrote:
> Just my $.02: What you guys are talking about is basically NTP with a 1PPS
> time reference derived from the VIA1. It sounds pretty easy in theory.
Yes, that's the direction my thoughts were headed in, to some extent or
another.
> Since NTP tries to compute a drift rate for the system clock and the drift
> isn't constant it won't work *well* without some nonstandard tuning IMHO.
You're correct; the tick interval would be adjusted almost continuously as
a function of the SCSI and serial activity.
--scott