[ntp:questions] Windows time question.

David J Taylor david-taylor at blueyonder.co.uk.invalid
Wed May 4 15:01:01 UTC 2011

"unruh" <unruh at wormhole.physics.ubc.ca> wrote in message 
news:slrnis2psa.6vn.unruh at wormhole.physics.ubc.ca...
> Jitter and accuracy are not the same thing. It would be interesting to
> attach the PPS to a real interrupt (ie one which can be serviced on the
> 1usec level) and see what the offset of the usb pps actually is.
> From what I read, the usb interrupts have a minimum latency of 125us but
> more likely much larger ( but less than 100ms).

Bill, I would not be able to see offsets on 125 microseconds on the 
arrangement I have, only hundreds of microseconds.  I don't have the time 
for the tests you propose, either, sorry.  But from the graphs, it's 
obvious that the offset is well under 500 microseconds in the test 
configuration, at least.


More information about the questions mailing list