Re: [time-nuts] Network Time Puzzle

2019-05-28 Thread K5ROE Mike
I would broaden your experiment to try many different remote servers.    Maybe using different chunks of the global NTP pool (https://www.ntppool.org/en/use.html) It could be traffic shaping on your ISP , an ISP in the middle, or one on the remote end. It could be traffic prioritization

Re: [time-nuts] Network Time Puzzle

2019-05-26 Thread Peter Martinez via time-nuts
Steve: Many thanks for that link to your paper on NTP http://leapsecond.com/ntp/NTP_Paper_Sommars_PTTI2017.pdf I read the entire paper, not just figs 7 and 8, and it explains everything. I was beginning to doubt my own sanity because several authorities that I had consulted told me quite

Re: [time-nuts] Network Time Puzzle

2019-05-26 Thread Steven Sommars
See figures 7 & 8 in http://leapsecond.com/ntp/NTP_Paper_Sommars_PTTI2017.pdf When a router forwards an NTP packet multiple potential egress links may have equal cost. In order to distribute the traffic across the egress links the router can use the IP addresses and UDP source/destination port

[time-nuts] Network Time Puzzle

2019-05-26 Thread Peter Martinez via time-nuts
Greetings, Time Nuts, from a new member. I have two old Windows XP laptops on which I can lock the timing to GPS, which means I can read the time at which things happen to a few microseconds. I thought I would modify some of my old NTP software, both client and server, to make use of this