[ntp:questions] Garmin 18 LVC: whether to fudge

Hal Murray hal-usenet at ip-64-139-1-69.sjc.megapath.net
Mon Feb 16 01:26:16 UTC 2009


>Since a regular ntp packet is just 48 bytes without extensions, the idea
>that ADSL would make a difference is rather unlikely. The asymmetry is
>related to the size of the packets but it affects TCP rather than UDP.

I'm not sure what sort of asymmetry you are thinking of for TCP.

I'm thinking of ADSL where one direction has a different bandwidth
than the other.  The delay due to transmission time is bits/bandwidth.

That's not the speed of light delay, it's the time it takes
for the bits to go past any point on the wire.  It's how long
the packet would be if you looked with a scope.  It's the same
reason that 100 megabit ethernet is faster than 10 megabit ethernet.

If we round up a 48 byte NTP packet to 64 bytes to make the math
easier, that's 512 bits.  So our packet will take 1 millisecond
at 512K bits/second, 2 ms at 256K and 4 ms at 128k.

If the downstream link is 512K and the upstream is 128K, then
it's 1 ms down and 4 ms up.

The OP said ADSL, but didn't give the numbers.  I picked those
to be reasonable (at least in the right ballpark) and easy
to work with.


-- 
These are my opinions, not necessarily my employer's.  I hate spam.




More information about the questions mailing list