Re: [ntp:questions] NTP absolute accuracy?
Unruh wrote in message news:rl5hm.50089$db2.40...@edtnps83... [] As I tried to emphasise, if the round trip is not symmetric, then neither ntp not chrony can compensate for that lack of symmetry, and the absolute time will be out. If occasionally it has an assymetric round trip, then ntp will probably elimate it, and chrony may or may not, depending on how you set up chrony, but typically chrony will be more sensitive to that asymmetry. Eb, say on exactly every 5th query, an extra .3ms is added to the return trip, and lets say that the minimum round trip is .15ms plus or minus .1 ntp will always throw out those fifth cases. but will have an error around .05 ms. chrony will not, and without them would have an error of .02ms but with them would have an error of about .05ms, just as would ntp. Ie, in this case the chrony advantage would be more or less erased by those asymmetric round trips. There is no definative answer. It depends on the situation. Many thanks - that's clearer. David ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
David Lord wrote: How do you get the time difference between your GPS and system time? Include the GPS in your ntp.conf, but mark it with noselect on the server line. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
David Lord sn...@lordynet.org writes: Unruh wrote: David Lord sn...@lordynet.org writes: David J Taylor wrote: Unruh unruh-s...@physics.ubc.ca wrote in message news:qg_gm.50009$db2.46...@edtnps83... David J Taylor david-tay...@blueyonder.not-this-bit.nor-this.co.uk.invalid writes: David J Taylor david-tay...@blueyonder.not-this-bit.nor-this.co.uk.invalid wrote in message news:%xqgm.126$ym4...@text.news.virginmedia.com... Unruh wrote in message news:34jgm.50898$ph1.36...@edtnps82... [] Note that chrony will give you a factor of 2 or three improvement over ntp in the errors, assuming that the roundtrip is equally split on Linux or BSD. For those without wide-bandwidth academic connections - those folks on cable or ADSL - how good is an equal split round trip assumption? David Thanks, David, David and Jan. A few milliseconds is what I had expected, so if you are on a consumer line, what implications does that have for unruh's comment? Cheers, David Which comment? The one I quoted: Note that chrony will give you a factor of 2 or three improvement over ntp in the errors, assuming that the roundtrip is equally split on Linux or BSD. On consumer-grade circuits the assumption about equal round trip is unlikely to be valid. I hope that David Lord can make some tests, and I await the results with interest. Both fileserver using ntpd and desktop using chrony had only my local ntp servers as sources so I've restarted both with three remote sources, one common source at my isp, and alternate sources from two different sites. What I'm most interested in is the effect of temperature as it's one of two periods in year that gets largest temperature variations and I've already had to restart the servers as ntpd was unable to keep up with the higher minpoll values I had set. I'm now not quite sure how best to compare stats from chrony vs ntpd. Well, the way I did it was to attach a GPS PPS to the computer and use the offset between it and the system time as a measurement of the accuracy. (the gps was NOT used as a source for eitehr ntp or chrony). How do you get the time difference between your GPS and system time? I wrote a parallel port interrupt routine ( well adapted on) which read the PPS interrupt and time stamps it. Ideally that timestamp should be exactly on the second. It is usually off by a few tens of microseconds. That is the time difference. At the moment there only seems to be 100us between offsets and apart from this being insignificant I wouldn't know which was nearer to ntp time. ?? sure you would the kernel time is ntp time. The offsets are the difference between the time ntp gets from sending out and receiving back a packet from the remote site. This is not ntp time That reading goes through the clock filter, and then is fed into the feedback loop to disciple the computer clock. That computer clock time is the closest one can come to something called ntp time. Unlike chrony, ntp does NOT use the time readings from the remote site to do a best estimate of the true time, and then drive the kernel time toward that. It assumes at each reading that the output of the clock filter IS the best estimate of the the true time, and slowly drives the clock to minimize that. Slowly means that in the end, the time ntp approaches is a sort of mean of those readings. Chrony on the other hand uses from 3 to 60 of thepast readings to find the best estimate of what the true time now is, and finds the difference between the system clock and that true time. It then corrects for that difference both in offset and rate. It is a very different philosophy. My adsl connection has also been off already for 30 minutes after I'd set this up. One of my servers that gets time from MSF and also has both test systems as sources. Both test systems also have the common source of my isp's timeserver so I'm expecting to get something meaningful from that. MSF? The problem is that I have no idea what the accuracy of any of those items is. YOur ISP's timesever may be a stratum 7 getting time from a bunch of bozos. Or it may be stratum 1 getting its time from a well implimented GPS clock. You are interested in the difference between your computer time and true UTC. the only way you will get that is by having a local source of true time ( eg a GPS with a 1usec or so accuracy). I used a GPS 18LVM, but of course even there I am not sure that the internals of the GPS receiver do not introduce microsecond sized time errors. I have never calibrated my GPS against some better true time standard. I might be able to get GPS to both systems but not for a while. One GPS receiver should be able to be split to supply time to both machines to usec accuracy. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
[ntp:questions] Frequently asked question about error bounds?
What is the correct error bound between client and server? In the RFC documentation I read one thing and in many theads another. I'm confused so please help me. Which is correct? 1) In the documentation for RFC1305 p. 102 it can be read that the true offset between client and server must lie somewhere in the correctness interval, defined by I=[theta - delta/2 - epsilon, theta + delta/2 + epsilon] 2) And in threads and websites I usuallly see the statement that the error is bounded with half the roundtrip. And it sure look like that when scatter diagram(wedge plots) with offset as a function rtt are viewed. --- To bound the error to the root of the synchronization subnet. Here, upper-case variables are used relative to the primary reference source (s), i.e., via a peer to the root of the synchronization subnet. Since offset, dispersion and delay(rtt) are all additive, you easily sum up all variables from the primary server to server i and achieve OFFSET sub i, ROOT DISPERSION sub i and ROOT DELAY sub i to the root. The synchronization distance, sometimes called the root distance, is calculated with DELTA/2 + EPSILON and represents the biggest statistical error. So the true offset relative to a primary reference server must be contained in the interval [OFFSET- SYNC.DIST, OFFSET +SYNC.DIST.] If the 1) alternative is right, all the other would be much more consistently. Thanks in advance ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Frequently asked question about error bounds?
berra.84 wrote: 1) In the documentation for RFC1305 p. 102 it can be read that the true offset between client and server must lie somewhere in the correctness interval, defined by I=[theta - delta/2 - epsilon, theta + delta/2 + epsilon] This is (more) correct. I'm not sure if it fully accounts for some of the details of the intersection algorithm. One would need to double check whether it was valid during startup transients and in the light of the averaging of multiple sources that is actually used to estimate the time. 2) And in threads and websites I usuallly see the statement that the error is bounded with half the roundtrip. And it sure look like that when scatter diagram(wedge plots) with offset as a function rtt are viewed. This is an oversimplification that is quite often valid in cases where the error is significant, as delay tends to dominate dispersion and most of the delay, and in particular, nearly all its asymmetry, is associated with the retail hop. Another factor, I think is that the RFC attempts to be a mathematical treatise, and the use of lots of Greek letters, rather than spelling out what they mean, turns people off, so very few users can actually read and understand the RFC. Please note that RFC 1305 is de facto obsolete, and current versions of NTP actually correspond with a new, draft, RFC. Also note that this is not a frequently asked question, although maybe it should be. ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
Unruh unruh-s...@physics.ubc.ca wrote in message news:iebhm.50125$db2.41...@edtnps83... [...] The problem is that I have no idea what the accuracy of any of those items is. YOur ISP's timesever may be a stratum 7 getting time from a bunch of bozos. Or itmay be stratum 1 getting its time from a well implimented GPS clock. ntpq -p will tell you that. And ntptrace does exactly that, climbing the chain up from you to stratum 1. Groetjes, Maarten Wiltink ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Frequently asked question about error bounds?
berra.84 berra...@spray.se writes: What is the correct error bound between client and server? In the RFC documentation I read one thing and in many theads another. I'm confused so please help me. Which is correct? 1) In the documentation for RFC1305 p. 102 it can be read that the true offset between client and server must lie somewhere in the correctness interval, defined by I=[theta - delta/2 - epsilon, theta + delta/2 + epsilon] 2) And in threads and websites I usuallly see the statement that the error is bounded with half the roundtrip. And it sure look like that when scatter diagram(wedge plots) with offset as a function rtt are viewed. That is the delta term above. But it is clear that if the clock you are comparing to is wildly off, then the roundtrip time does not encompas the error. Lets say that the server gets its time from a GPS NMEA which has a best accuracy of about 1ms ( and often worse) but the roundtrip time is 200usec. Then the accuracy will be a few ms, not 200usec. That is epsilon. If the 1) alternative is right, all the other would be much more consistently. It is right. Thanks in advance ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
David J Taylor david-tay...@blueyonder.not-this-bit.nor-this.co.uk.invalid writes: Unruh wrote in message news:iebhm.50125$db2.41...@edtnps83... [] MSF? See: http://www.npl.co.uk/science-technology/time-frequency/time/products-and-services/msf-radio-time-signal UK 60KHz radio time signal. Ah, OK. Its accuracy is probably not much better than a few msec I assume. GPS is a few usec. Since the network gives accuracies of better than a few ms, MSF is not a good way of testing the accuracy of network set clocks. David ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
Unruh wrote in message news:vcjhm.50153$db2.9...@edtnps83... [] UK 60KHz radio time signal. Ah, OK. Its accuracy is probably not much better than a few msec I assume. GPS is a few usec. Since the network gives accuracies of better than a few ms, MSF is not a good way of testing the accuracy of network set clocks. At source, it's recently been within about 10 microseconds: http://www.npl.co.uk/science-technology/time-frequency/time/products-and-services/msf-bulletins http://resource.npl.co.uk/time/bulletins/msf/msfbul_08_2009.pdf In the field, the accuracy will depend on the receiver you use, the propagation, and the quality of signal you have. A few milliseconds is probably a fair estimate, although it would be interesting to see some measured values. Cheers, David ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
David J Taylor writes: UK 60KHz radio time signal. Bill Unruh writes: Ah, OK. Its accuracy is probably not much better than a few msec I assume. Should be as good as WWVB, which is good to within 100usec. The propagation delay is highly predictable at 60KHz. These stations can also can provide extremely aaccurate frequency references. -- John Hasler jhas...@newsguy.com Dancing Horse Hill Elmwood, WI USA ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
I wrote: Should be as good as WWVB, which is good to within 100usec. But evidently many commercial receivers are only good to a ms or two. -- John Hasler jhas...@newsguy.com Dancing Horse Hill Elmwood, WI USA ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] Running two ntpd systems in parallel
Please see my message to hackers@ today pointing to a prototype ntpd that disciplines a synthetic clock and does not require root privileges: http://lists.ntp.org/pipermail/hackers/2009-November/004633.html Cheers, Dave Hart ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
I wrote: Should be as good as WWVB, which is good to within 100usec. The David Woolley writes: Most receivers only use the slow code (including the simple hardware solutions for ntpd). I'm assuming a real receiver. With the single-chip designs intended for atomic watches you'll be lucky to get with a second. For high accuracy you would have to sync to the carrier, and possibly use the fast time codes. With WWVB I'd use an analog PLL to recover the carrier and a DPLL to get the PPS. -- John Hasler jhas...@newsguy.com Dancing Horse Hill Elmwood, WI USA ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
David J Taylor david-tay...@blueyonder.not-this-bit.nor-this.co.uk.invalid writes: Unruh wrote in message news:vcjhm.50153$db2.9...@edtnps83... [] UK 60KHz radio time signal. Ah, OK. Its accuracy is probably not much better than a few msec I assume. GPS is a few usec. Since the network gives accuracies of better than a few ms, MSF is not a good way of testing the accuracy of network set clocks. At source, it's recently been within about 10 microseconds: Sorry, at 10usec, the distance away of the transmitter must be less than 3 km. Also your system needs to see the start of the tone to 10usec which means that the tone would have to be about 1MHz which is a bit beyond audio. http://www.npl.co.uk/science-technology/time-frequency/time/products-and-services/msf-bulletins http://resource.npl.co.uk/time/bulletins/msf/msfbul_08_2009.pdf In the field, the accuracy will depend on the receiver you use, the propagation, and the quality of signal you have. A few milliseconds is probably a fair estimate, although it would be interesting to see some measured values. Cheers, David ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
Bill Unruh writes: Sorry, at 10usec, the distance away of the transmitter must be less than 3 km. 10usec at the transmitter. Also your system needs to see the start of the tone to 10usec which means that the tone would have to be about 1MHz which is a bit beyond audio. There isn't any tone. It's the UK equivalent of WWVB. http://resource.npl.co.uk/docs/networks/time/meeting3/martin.pdf -- John Hasler jhas...@newsguy.com Dancing Horse Hill Elmwood, WI USA ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
John Hasler jhas...@newsguy.com writes: Bill Unruh writes: Sorry, at 10usec, the distance away of the transmitter must be less than 3 km. 10usec at the transmitter. Also your system needs to see the start of the tone to 10usec which means that the tone would have to be about 1MHz which is a bit beyond audio. There isn't any tone. It's the UK equivalent of WWVB. How is the beginning of the second ( an hour) marked? The bandwidth of whatever marks it has to be pretty narrow, or the transmitter would interfer with everything around it. Ie, that means a large time uncertainty. Ie how fast is the carrier actually switched off and on. It cannot be instantaneous or the signal would spill out badly into neighbouring bands. http://resource.npl.co.uk/docs/networks/time/meeting3/martin.pdf -- John Hasler jhas...@newsguy.com Dancing Horse Hill Elmwood, WI USA ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions
Re: [ntp:questions] NTP absolute accuracy?
Unruh wrote: John Haslerjhas...@newsguy.com writes: There isn't any tone. It's the UK equivalent of WWVB. How is the beginning of the second ( an hour) marked? The bandwidth of whatever marks it has to be pretty narrow, or the transmitter would interfer with everything around it. Ie, that means a large time uncertainty. Not if they do the same spread spectrum modulation of the carrier as emplyed by the German DCF77 transmitter: By phase-locking a receiver to that signal, you can get down to the 10-30 us or so range. (You still have to be close enough to the transmitter so that the propagation path is perfectly predictable, otherwise you'll see systematic excursions as a function of the time of day/night.) Terje -- - Terje.Mathisen at tmsw.no almost all programming can be viewed as an exercise in caching ___ questions mailing list questions@lists.ntp.org https://lists.ntp.org/mailman/listinfo/questions