[ntp:questions] Re: difference between M12 & M12T GPS receiver
Wolfgang S. Rupprecht
wolfgang+gnus20040202T143637 at dailyplanet.dontspam.wsrcc.com
Mon Feb 2 22:59:35 UTC 2004
"Maarten Wiltink" <maarten at kittensandcats.net> writes:
> "Wolfgang S. Rupprecht"
> <wolfgang+gnus20040201T190459 at dailyplanet.dontspam.wsrcc.com> wrote in
>> I'm on a very asymetrical line to the internet 256kbps clocking in the
>> upstream direction and 1.5Mbit/sec clocking in the downstream
> It may be choked differently, but when not saturated, is the upstream
> actually slower?
I've been able to confirm the total bandwidth of my incoming and
outgoing direction, but I can't think of a way to test for delay
without buying a used Redback (or similar ISP-side DSL termination)
and setting up a short test network here.
My understanding from reading various ADSL whitepapers is that ADSL
has independent clock rates for the outgoing and incoming
serializer. * The standard clock rates for TX are 128kbps, 256kbps,
512kbps, 768kbps. The standard clock rates for RX (that I know about)
are 1.5Mbps, 3Mbps, 6Mbps.
For my 90-byte (on the wire) ntp packets with my observed 256kbps
clocking, I should be seeing a shift-out delay on the TX of:
(/ (* 90 8) 256e3) 2.8125 milliseconds
For the incoming RX side with my 1.5Mbps clocking I should be seeing a
shift-in delay of:
(/ (* 90 8) 1.5e6) 0.48 milliseconds
For a difference of:
(- (/ (* 90 8) 256e3) (/ (* 90 8) 1.5e6)) 2.3325 milliseconds
This is roughly consistent with the offsets I'm seeing when polling
external stratum 1's and 2's.
If anyone can think of a way to test the RX vs. TX delays, I'm all
* Sort of like the bad old days where tty/serial ports had independent
clocks for TX and RX, and some modems were indeed clocked at different
rates for rx and tx.
Wolfgang S. Rupprecht http://www.wsrcc.com/wolfgang/
The above "From:" address is valid. Don't mess with it.
Gripe to your senators about spam: http://www.wsrcc.com/spam/senators.html
More information about the questions