On Mon, 17 Feb 1997 12:25:58 -0500, "Perry E. Metzger" <perry@piermont.com> wrote:
> Martin Husemann writes:
> > The (classical) question is: should I hand out the timestamps as a seperate
> > device, run the standard nntp deamon to use them, which in turn calls the
> > appropriate kernel functions to adjust the time?
> >
> > Or should I shortcut this and adjust the time right inside the
> > kernel, without any device and userland daemon?
>
> I wouldn't bother. Line latency distribution is probably high enough
> to negate any advantage you would gain. What *will* help is the kernel
> PLL stuff, since it will keep the clock better disciplined between
> adjustments.

The timestamp is coming over the 16Kbit/s D channel, which is designed to be  
a very low latency out-of-band signaling channel.

Why would latency be high on the D channel?  Assuming you're not abusing it  
for something like a low volume X.25 channel, by definition, there should  
never be any significant latency.

--
Christian Kuhtz <ckuhtz@paranet.com> (work),   <chk@gnu.ai.mit.edu> (personal)
UNIX/Network Specialist,   "A German in the U.S., speaking for himself *gasp*"
Paranet, Inc. Rocky Mountain Branch, http://www.paranet.com/  MIME/NeXTmail Ok