Ok. Clubspeed data is more clear now. The PureLaunch will measure
clubspeed but has a deviation for each measurement of around +/-
10mph. When using a consistant moving object (the calibration tool) it
is possible to use all points (between 35-80 points) measured to
calculate an average. That rather simple calculation will produce a
clubspeed which will be within 1% of the real value.
However, where the calibration tool produces a radar image of a moving
object with a consistant speed in real life the club has almost never
a consistant speed. In fact it can accelerate AND deccelerate before
impact (like when fitting the mat before the ball). Due to this unknow
"pattern" a simple average calculation is not possible.
An other issue is that Zelocity decided to not include the full moment
of the impact in the data. So the calcualtion has to "guess" when the
impact really happend. If before impact the club is really
accelerating but we stop calculation to early we will "measure" a
speed which it to slow, when we continue the calculation for to long
we will measure a speed which will be to high. IMHO the firmware for
the PureLaunch should be re-written to include the impact data thus
creating more chance to improve calculations
In the next beta I will include clubspeed data before impact in a
graph. It will explain the issue more clearly then I can do using
words :-)