tvb@LeapSecond.com said: > Just in case we have some newcomers to the thread I'd like to point out > that this recent series of measurements of RS232 / NMEA have no bearing at > all on the quality of the timing output. Timing NMEA is more of a > curiosity; something to measure at the hundreds or tens of millisecond > level. As anyone knows, the real timing output of these receiver is the > 1PPS pulse itself, which is good to the tens of nanoseconds level. So a > factor of a million different. I think that depends on how far down the nutty scale you are operating. I think all the investigations of timing over serial ports are useful. You can get sub millisecond timing over RS-232. That is interesting for things like NTP if you don't have a PPS connection, so it helps to know which chips are good and which ones to avoid, Serial ports are getting rare. Most USB to RS-232 chips are slow with a 1 ms polling rate. I think AdaFruit has one that polls 8x faster. That should get sub-ms timing if coupled with a good GPS chip. -- These are my opinions. I hate spam.