The watches/clocks they are talking about listened to WWV, a set of radio stations transmitting from Fort Collins, Colorado. The system long predates the Network Time Protocol you're referring to. Radio controlled clocks/watches had no means for accounting for latency.
Yeah, but you need to factor in the distance to the transmitter. Going to add at least a few microseconds to your time accuracy!
Latency is accounted for in the sync process
Sync process? The other comment was talking about the old receivers for the atomic clocks on SW/MW frequencies. It was a one way thing.
Now in theory if a receiver also had GPS they could account for the distance. But, then they'd get far more accurate time from the GPS receiver so..
The watches/clocks they are talking about listened to WWV, a set of radio stations transmitting from Fort Collins, Colorado. The system long predates the Network Time Protocol you're referring to. Radio controlled clocks/watches had no means for accounting for latency.
Ahh OK my bad. I've only worked with NTP for a long time and wasn't aware of the earlier stuff.
If you're interested in prehistory, listen to this for a couple minutes.