I’d just settle for the time being correct. Hasn’t been since the last FR meltdown.
The time is always correct. And always accurate.
But the measurement may not be the same as you get from NIST with proper protocols.
And what does it matter? You using FR to do real time milli-second life support functions?
Accuracy has it's place. If you need time that agrees with the ISO standard, don't rely on FR.
Otherwise, folks might think you were pretty up-tight and not tightly wrapped.
/johnny