Originally Posted by
tomato coupe
The most accurate way to calibrate a rotation sensor is to use GPS to determine tire circumference.
This is easy to do but it might not be correct at all (it depends on getting a good signal). One might get a better number riding in a straight line.
Originally Posted by
tomato coupe
As a consequence, distance measurements based on the sensor will not be as accurate as those based on direct GPS measurements, if the distance is greater than that used for calibrating the sensor.
Garmin says otherwise. The GPS distance doesn't include the effect of elevation and the GPS might not get a good signal and the GPS distance likely smooths out some of the movement of the bike.
The rotation sensors are not inaccurate (as long as a good roll out number is obtained).
=============================
Given the variability of GPS tracks, it's not clear that distance from a well-calibrated rotation sensor would necessarily be different in accuracy.
https://media.dcrainmaker.com/images...6/image-49.png