Re: Time off by 6 seconds

On Mon, Sep 25, 2023, 1:33 PM Jack Woehr via MIDRANGE-L <
midrange-l@xxxxxxxxxxxxxxxxxx> wrote:

On Mon, Sep 25, 2023 at 11:25 AM (WalzCraft) Jerry Forss <
JForss@xxxxxxxxxxxxx> wrote:

Every week the IBMi is behind by 6 seconds.
Any idea why there would be such a drift?

Because the crystals sold to generate time pulses are off by a few shades
of a Hz which has been the case for 50 years.

But 6 seconds per week is quite a lot. It's much more than I would expect.
Even the chintziest wristwatches do not lose time that fast.

I would be interested to see whether that rate is steady. Like, if you can
afford to not sync it one week, will it be 12 seconds off the following
week?

I would not be surprised if actually what happens is that it stays 6
seconds off, or at least loses time much, much slower.

And it wouldn't matter if they were exactly right, because NIST and others

change the time from time to time.

But, to my knowledge, they do not change the time in big chunks. I would
think they make adjustments that are fractions of a second. Maybe a second
at a time, at most.

John Y.