I'm writing a data field that displays average stride length per interval, for interval sessions where running is separated by rest. I'm backing out number of steps by multiplying elapsed time in minutes by average cadence (and also by 2 to transform from revolutions to steps). From there I calculate stride length by dividing distance by number of steps.
I'm using a .fit file that I collected from a recent track workout. One thing I was surprised about when I looked at Connect online is that for some of the intervals my average stride length was very long (> 2m). Initially I put this down to my great running form :), but when I replayed the fit file in the simulator, I saw that at the start of some of the intervals, it seemed like the number of steps wasn't increasing (consistent with this, the cadence in the info struct was often reported as zero for some time at the start of an interval). This led to the number of steps being too low, and hence stride length being too high.
Can anyone comment on this? Does the watch go into sleep mode and then the step measurement takes 5 to 10s to get going again? I have the vivoactive in case that makes any difference.
thanks,
Tony.