[ntp:questions] Isolated Network Drift Problem

Richard B. Gilbert rgilbert88 at comcast.net
Wed Nov 26 01:11:54 UTC 2008


Cal Webster wrote:
> On Tue, 2008-11-25 at 17:52 -0500, Richard B. Gilbert wrote:
>> Steve Kostecke wrote:
>>> On 2008-11-25, Cal Webster <cwebster at ec.rr.com> wrote:
>>>
>>>> ... We'd have to run a long cable [to the GPS18LVC] though,
>>>> the computer lab has no windows and the roof is the most logical place
>>>> to put an antenna. I don't think USB will handle that much line loss.
>>> It is RS-232, not USB.
>>>
>> RS-232 specifies a 50 foot maximum!
>>
>> IF you use top quality cable, take care that your cable runs avoid the 
>> fluorescent lighting fixtures, etc, you can use a great deal more than 
>> that.  I've used more than 100 feet on occasion.  If it doesn't work, 
>> you will get no sympathy.
>>
>> I'd suggest that you put the computer on the top floor somewhere and use 
>> Cat-5 cable for your RS232.
> 
> I was thinking I'd try the rather crude but elegant solution here: 
> Using a Garmin GPS 18 LVC as NTP stratum-0 on Linux 2.6
> http://time.qnan.org/

What's crude about that?  GPS gives you time accurate to within about 50 
nanoseconds.  The process of getting the signal into your computer 
degrades the accuracy somewhat but time to within 500 microseconds or so
is more than adequate for most purposes.




More information about the questions mailing list