The sampling time between each tracklogg is 10 sek, and when checking the data in Oziexplorer it was rather clean. I had the gps mounted on the lifewest, not on the arm, to keep it steady. Sheeting with an arm mounted gps is a good way to get speed glitches.
The 250 meter and 500 meter runs feels reliable but the 100 meters are too short based on the tracklogg sampling rate. The gps said that the highest recorded speed was 22.4, but I think it's safer to refer to the best 250 meter run at 21.3 knots.

The highest speed I recorded with the A-class is 20.5 knots using the spi in high winds and flat water trapezing with chickenline as far back as the hull stretches, and that's the gps instantanious speed!
If I analyze that tracklogg from the A-class run using the software I get the following data:
Best 250 meters run, 19.9 knots
Best 500 meters run, 19.2 knots

The Hobie 16 rocks!!!!! It has higher top speed than my A-class, which was something else than an A-class since I was using a spi on it.

/hakan