I think we have to remember it's not a tape measure, ultrasonic measure or laser measure.
The technology is very new and still in the process of refinement. As I mentioned on page 2 of this thread, it can theoretically be thrown off by something as simple as lumping in jump time with sea level riding when determining sea level height. The designers had to be very aware of the differences between the different versions and I can't say I agree with their publicity not being clear on the measuring differences, switchkites did a similar thing releasing an Element V6 by a different designer and choosing to keep it quiet. But the kite handles better and the woo3 shows higher jumps.
If we had a better standard than the woo to measure against then we'd be able to make more informed decisions.
My guess is it took woo sports between the woo1 to the woo3 to make better adjustments for removing in-jump time from water level height averaging... I could be wrong but it fits, I can spend 3% of my time in jumps according to the woo2 which would raise the average height over that time and reduce my detected height if not corrected. Or if I look at a local rider Garrick's in air time on the woo3 vs the woo2 the woo3 shows about twice the air time!!
So that could be a woo3 calculated actual in air time of 5% to 8% of total session time which has to heavily affect the woo2 average height if not compensated for. It's potentially even worse, Any averaging calculation will have to ditch old values after a while, say you deploy your kite 5M above sea level and power up your woo before taking your board down to sea level, say the woo is averaging for a minute before your first jump, by the time you jump, your woo might be subtracting 2 or 3 meters from your actual height. Woo sports need to use an algorithm that ONLY averages height to determine true water level during non jumping, non shore time.
But here is the open question to Woo sports, can you release a firmware update to correct the earlier product jump height readings?
And if not, what's the explanation, are you stuck with accelerometers with chip manufacturer height averaging that you can't figure out how to correct?
Or are there other factors?