Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > comp.arch.embedded > #28040

Re: clock synchronization over IP

Show all headers | View raw

On Mon, 17 Sep 2018 01:45:53 -0700 (PDT), gp.kiwi@gmail.com wrote:

>On Monday, September 17, 2018 at 12:49:25 AM UTC+12, George Neuner wrote:
>> On Sat, 15 Sep 2018 15:08:11 -0700 (PDT), gp...@gmail.com wrote:
>> 
>> 
>> >Thanks, I did eventually discover PTP.  The hardware timestamping
>> >appeals to me but PTP is apparently harder to set up and you need
>> >routers and switches that support it. 
>> 
>> PTP doesn't need any special network hardware - it implemented over
>> UDP.
>> 
>> George
>
>After googling for information it seems that some hardware does
>support hardware timestamping and there are implementations of both
>NTP and PTP that use it.
>https://en.wikipedia.org/wiki/List_of_PTP_implementations
>
>The main wikipedia article on PTP actually gives (almost) no
>indication that hardware timestamping gives better accuracy and
>claims that on a LAN, it can achieve sub-microsecond accuracy. 
>Without hardware support I suspect that kind of accuracy requires a
>low traffic LAN, although it's possible that it keeps trying until it
>achieves a really short round-trip time.

You need enough protocol packets to get through (to every node) in a
timely manner to correct for the node's clock drift.  But what you
need depends on how good (or bad) is the clock.  You could need
several packets/second, or you might need just a few packets/minute,
depending ...

Obviously you need to start with high precision time source ... an
easy way to do it is to buy a network switch that has PTP built in.
But in a simple setup your time source could be just some node
designated to be the timekeeper.

You don't need routers/bridges to implement the protocol unless there
is significant congestion and lots of variance in packet time across
the network.  A few milliseconds here and there will be sorted out by
the protocol.

I think you said you needed +-20ms ... you didn't say how large or
complex the network, but if the cross section is under 10 milliseconds
(a pretty large wired net), you should be easily able to achieve your
desired resolution just with software PTP.

YMMV,
George

Back to comp.arch.embedded | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

clock  synchronization over IP gp.kiwi@gmail.com - 2018-08-24 17:54 -0700
  Re: clock  synchronization over IP Paul Rubin <no.email@nospam.invalid> - 2018-08-24 18:28 -0700
  Re: clock  synchronization over IP wzab01@gmail.com - 2018-09-15 09:00 -0700
    Re: clock  synchronization over IP gp.kiwi@gmail.com - 2018-09-15 15:08 -0700
      Re: clock  synchronization over IP George Neuner <gneuner2@comcast.net> - 2018-09-16 08:49 -0400
        Re: clock  synchronization over IP gp.kiwi@gmail.com - 2018-09-17 01:45 -0700
          Re: clock  synchronization over IP George Neuner <gneuner2@comcast.net> - 2018-09-17 11:30 -0400
            Re: clock  synchronization over IP gp.kiwi@gmail.com - 2018-09-17 15:04 -0700
              Re: clock  synchronization over IP George Neuner <gneuner2@comcast.net> - 2018-09-17 22:10 -0400
                Re: clock  synchronization over IP graeme.prentice@gmail.com - 2018-09-18 04:23 -0700
                Re: clock  synchronization over IP Andrew Smallshaw <andrews@sdf.org> - 2018-09-18 13:19 +0000
                Re: clock  synchronization over IP Mikko OH2HVJ <mikko.syrjalahti@nospam.fi> - 2018-09-19 18:56 +0300
                Re: clock  synchronization over IP upsidedown@downunder.com - 2018-09-22 11:51 +0300
          Re: clock synchronization over IP Clifford Heath <no.spam@please.net> - 2018-09-18 21:10 +1000
  Re: clock  synchronization over IP Chris <xxx.syseng.yyy@gfsys.co.uk> - 2018-09-21 15:00 +0100
    Re: clock  synchronization over IP Dimiter_Popoff <dp@tgi-sci.com> - 2018-09-21 17:12 +0300
      Re: clock  synchronization over IP Robert Wessel <robertwessel2@yahoo.com> - 2018-09-21 09:58 -0500
        Re: clock  synchronization over IP Chris <xxx.syseng.yyy@gfsys.co.uk> - 2018-09-21 16:26 +0100
      Re: clock  synchronization over IP Chris <xxx.syseng.yyy@gfsys.co.uk> - 2018-09-21 16:02 +0100
    Re: clock  synchronization over IP George Neuner <gneuner2@comcast.net> - 2018-09-21 17:15 -0400
      Re: clock  synchronization over IP Chris <xxx.syseng.yyy@gfsys.co.uk> - 2018-09-21 22:35 +0100
        Re: clock  synchronization over IP "gp.kiwi@gmail.com" <graeme.prentice@gmail.com> - 2018-09-21 15:16 -0700
          Re: clock  synchronization over IP Chris <xxx.syseng.yyy@gfsys.co.uk> - 2018-09-22 00:11 +0100
            Re: clock  synchronization over IP "gp.kiwi" <graeme.prentice@gmail.com> - 2018-09-21 16:35 -0700
              Re: clock  synchronization over IP Chris <xxx.syseng.yyy@gfsys.co.uk> - 2018-09-22 14:19 +0100
              Re: clock  synchronization over IP gnuarm.deletethisbit@gmail.com - 2018-09-22 07:22 -0700
            Re: clock  synchronization over IP graeme.prentice@gmail.com - 2018-09-21 19:40 -0700
        Re: clock  synchronization over IP George Neuner <gneuner2@comcast.net> - 2018-09-21 22:17 -0400
          Re: clock  synchronization over IP Chris <xxx.syseng.yyy@gfsys.co.uk> - 2018-09-22 13:58 +0100
          Re: clock synchronization over IP Brett <ggtgp@yahoo.com> - 2018-09-22 18:50 +0000
    Re: clock  synchronization over IP upsidedown@downunder.com - 2018-09-22 11:56 +0300
      Re: clock  synchronization over IP gnuarm.deletethisbit@gmail.com - 2018-09-22 07:31 -0700
        Re: clock  synchronization over IP upsidedown@downunder.com - 2018-09-22 20:02 +0300
          Re: clock  synchronization over IP gnuarm.deletethisbit@gmail.com - 2018-09-22 14:40 -0700
    Re: clock  synchronization over IP gnuarm.deletethisbit@gmail.com - 2018-09-22 07:06 -0700
    Re: clock  synchronization over IP Mikko OH2HVJ <mikko.syrjalahti@nospam.fi> - 2018-09-22 20:16 +0300
      Re: clock  synchronization over IP Chris <xxx.syseng.yyy@gfsys.co.uk> - 2018-09-22 22:22 +0100

csiph-web