I've been thinking about NTP servers without easy access to a window
or rooftop. They wouldn't be able to use GPS, but they could have
some sort of local higher quality frequency standard to rely on. They
would still need phase from a network source (and thus be Stratum 2+),
but the
Quoting Miroslav Lichvar <mlich...@redhat.com>:
On Tue, Nov 29, 2016 at 09:54:36AM +0100, Miroslav Lichvar wrote:
On Mon, Nov 28, 2016 at 01:38:35PM -0600, Dan Drown wrote:
> Additional information:
>
> https://dan.drown.org/stm32/adev-mix.png
> ^ comparison of various ade
After letting it settle down after starting, the initial results
covering 14 hours: https://dan.drown.org/odroid/run5/tempcomp.png
Shows the tempcomp.log(purple, left axis) vs tracking.log(green, right axis)
The tempcomp factor has moved 200ppb (with some high frequency
movements), while
Quoting Miroslav Lichvar <mlich...@redhat.com>:
On Wed, Dec 28, 2016 at 05:57:15PM -0600, Dan Drown wrote:
That looks like this (green=Linux system's clock, purple=TCXO's measurement
of Linux system's clock).
https://dan.drown.org/odroid/run7/tcxo-acc-1024.png
I've adjusted the starting
Quoting Miroslav Lichvar :
I setup a quick benchmark of heating the CPU and on-board crystal with a
bunch of cpu wasting processes. CPU core went from 51C to 70C (as measured
by the on-CPU temperature sensor):
https://dan.drown.org/odroid/run5.7/tempcomp-cpuheating.png
The
This adds the "disablerxadjust" config option to hwtimestamp. This
uses RX timestamps at the 1588 point (after SFD/before ethernet
header) instead of the NTP point (after ethernet CRC). At 1G, this
changes the timestamp (and RTT) by 752ns. To be symmetrical, you'd
want this enabled on
Quoting Miroslav Lichvar :
I guess most people here don't follow the NTP WG list. There is one
feature proposed for NTPv5 that I think would make a big difference
for chrony:
https://fedorapeople.org/~mlichvar/ntp-freq-transfer/
Very interesting, thank you for gathering this data.
I'm