Re: [ntp:questions] how does jitter and round trip time affect the accuracy of the local clock?
Miroslav Lichvar wrote: In a simulation with 500us exponentionally distributed jitter and clock wander insignificant to the PLL time constant, the RMS time error is about 40 us for the standard PLL and 80 us for the Linux PLL. The 99th percentiles are about 100 us and 200 us respectively. Significant was a fuzzy term, because I didn't want to do the actual maths or simulation, but the key to the real point was your use of the word percentile: with a statistical system, there will not be a hard limit on the error. ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] how does jitter and round trip time affect the accuracy of the local clock?
On Sun, Jun 26, 2011 at 09:05:59PM +0100, David Woolley wrote: If the jitter is of the order of 500 microseconds, and your delays are perfectly symmetric, and there is no clock wander (in particular, the temperature is tightly controlled), the error will exceed 500 microseconds, a small but significant amount of the time. That wouldn't be a very good clock discipline if it wasn't able to keep the clock error significantly below the jitter in these ideal conditions. In a simulation with 500us exponentionally distributed jitter and clock wander insignificant to the PLL time constant, the RMS time error is about 40 us for the standard PLL and 80 us for the Linux PLL. The 99th percentiles are about 100 us and 200 us respectively. -- Miroslav Lichvar ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
[ntp:questions] how does jitter and round trip time affect the accuracy of the local clock?
How doe Jitter and round trip time affect local time clock time acccuracy? When looking at the jitter, what values would tell me that the phase noise would make the local clock in accurate? Are there any Rules of thumb to go by when looking at jitter and delay to determine local clock accuracy Kindly Chip ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] how does jitter and round trip time affect the accuracy of the local clock?
On 6/26/2011 11:36 AM, chipper wrote: How doe Jitter and round trip time affect local time clock time acccuracy? They don't *affect* the accuracy, they measure it! When looking at the jitter, what values would tell me that the phase noise would make the local clock in accurate? Are there any Rules of thumb to go by when looking at jitter and delay to determine local clock accuracy Kindly Chip How close do you need the time to be? Many people live most of their lives with clocks that may well be off by a minute or two. Other people need numerous clocks to agree with each other +/- a few seconds, milliseconds, or microseconds and don't care that the time is not correct as long as all the clocks involved agree within the specified tolerance. Still others need to know the time +/- a few microseconds or +/- a few nanoseconds. Specify your needs. If you can't figure out how to meet your needs, post a description of your requirements and somebody might be able to give you a clue or two or three. ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] how does jitter and round trip time affect the accuracy of the local clock?
chipper wrote: If the ntp client needed to be withing .5 milliseconds of the server,how Is the server keeping perfect time? could I tell if the jitter and delay were to much to support this kind of accuracy? What percentage of the time can you tolerate the time being out by more than 500 microseconds? Do you have perfectly symmetric network delays? If the jitter is of the order of 500 microseconds, and your delays are perfectly symmetric, and there is no clock wander (in particular, the temperature is tightly controlled), the error will exceed 500 microseconds, a small but significant amount of the time. The smaller the jitter, the less the time, but it will probably never go to zero. If the delay is less than 1ms, and the server clock is perfect, it is unlikely that the error will exceed 500 microseconds, although there may be some overshoots. Most people operate with delay desired accuracy, in which case you are at the mercy of the network delay asymmetry. ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] how does jitter and round trip time affect the accuracy of the local clock?
Richard B. Gilbert wrote: very high quality. In particular, GPS sends a pulse per second and one edge of that pulse is accurate to about +/- 50ns. The rest of the GPS signal tells you which second is being marked by the specified edge, leading or trailing (I've forgotten which). GPS doesn't do this. In fact GPS relies on the fact that there are significant time errors between the times received from different satellites. It uses pseudo random ranging codes, not single pulses. Some GPS receivers, synthesize a pulse on the one second mark from their solution for the time. ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] how does jitter and round trip time affect the accuracy of the local clock?
On 6/26/2011 3:52 PM, chipper wrote: If the ntp client needed to be withing .5 milliseconds of the server,how could I tell if the jitter and delay were to much to support this kind of accuracy? Thanks Chip snip Try it! If you are using an NTP server on the internet, it's going to be difficult. You may find that time quality is excellent between 2300 - 0700 local time and very poor for the remaining sixteen hours! If you have a GPS receiver, LORAN receiver, etc, you can get time of very high quality. In particular, GPS sends a pulse per second and one edge of that pulse is accurate to about +/- 50ns. The rest of the GPS signal tells you which second is being marked by the specified edge, leading or trailing (I've forgotten which). ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions
Re: [ntp:questions] how does jitter and round trip time affect the accuracy of the local clock?
If the ntp client needed to be withing .5 milliseconds of the server,how could I tell if the jitter and delay were to much to support this kind of accuracy? Thanks Chip Richard B. Gilbert wrote: On 6/26/2011 11:36 AM, chipper wrote: How doe Jitter and round trip time affect local time clock time acccuracy? They don't *affect* the accuracy, they measure it! When looking at the jitter, what values would tell me that the phase noise would make the local clock in accurate? Are there any Rules of thumb to go by when looking at jitter and delay to determine local clock accuracy Kindly Chip How close do you need the time to be? Many people live most of their lives with clocks that may well be off by a minute or two. Other people need numerous clocks to agree with each other +/- a few seconds, milliseconds, or microseconds and don't care that the time is not correct as long as all the clocks involved agree within the specified tolerance. Still others need to know the time +/- a few microseconds or +/- a few nanoseconds. Specify your needs. If you can't figure out how to meet your needs, post a description of your requirements and somebody might be able to give you a clue or two or three. ___ questions mailing list questions@lists.ntp.org http://lists.ntp.org/listinfo/questions