The good Dr. makes the point that computationally, we have left only
the argument where *best* do we obtain the data to derive power.
PowerTap is direct power with derived location mapped to contour.
Hac4 is direct contour with power derived from direct time/distance.
Hac4's are barometric sensitive, affecting degree of accuracy during
measurement. That same sensitivity provides resolution of gradient
changes w/in a reasonable window. I've watched 12% grades change to
16% w/in 20LF on-the-bike. It's near-realtime altimeter readings.
GPS I've raced with have varied, in one instance, we were unable to
locate the turning mark in one race. GPS can't match the resolution of
the Hac4 within 20 LF coordinates. The best GPS get your location
resolved to within 30-50'.
I share the good Dr.'s apprehension about the Hac4's power data. Hac4
computional algorythm is proprietary. I've not tested its accuracy
against a direct measure like the PowerTap.
In the end, I would choose Hac4 over GPS because Hac4 has better
resolution on-the-ground and leaves only the derivation of power
calculation in question.