I wonder if the difference in results is partially due to a difference in measurement strategy: a naive way is to wait one minute while counting the pulses whic

I wonder if the difference in results is partially due to a difference in measurement strategy: a naive way is to wait one minute while counting the pulses which will necessarily be slower in response (it could still update the screen at a faster rate of once per second by having 60 different integers but each integer gives the count displayed on the screen when it's last second has passed, but that would still have a slow response in the sense that given a hypothetical heart rate change from a fixed constant initial rate to a new fixed higher rate, like when you got scared by say unexpected fireworks, it would still take a minute for the final rate to show up). A faster response could be got by measuring the time between heartbeats (probably a low but even number of beats).