I've been given some FIT files to check. These didn't come from Canmore GPS units.

I've observed that the doppler distance, particularly when gybing, is consistently larger than the trackpoint distance, by a significant margin (often over 5m/s). As a result, they may record spurious results for Alphas and longer distance divisions.

This underlines the importance of using reliable, tested devices.

Cheers,

Dylan

Thanks for that Dylan.

Can you tell us what device the files are from please. Also, are the trackpoints and Doppler speed points at fixed 1 second intervals?

Device is a Garmin watch (D2 I think.) Yes it was 1s fixed intervals.

There was a gybe I was inspecting where in doppler distance, the length of the gybe was over 100m, but in trackpoints the length was less than 80m.

My Garmin fortrex had a track smoothing feature. Basically it would fill in data if it changed satellites and got a different tracklog point to previously. ie make stuff up.... I take it the Canmore records raw data. Just speculating but this may be more an issue with Garmin rather than the Fit file format

Select to expand quote

Dylan72 said..
Device is a Garmin watch (D2 I think.) Yes it was 1s fixed intervals.

There was a gybe I was inspecting where in doppler distance, the length of the gybe was over 100m, but in trackpoints the length was less than 80m.

One of the issues we found with using doppler distance for Gybes (Alphas) is that any error in the doppler direction is accumulated. That means you can end up with a track that puts you somewhere very different from what the positional data tells you! This is why we use the doppler speed for Alphas but the positional data for the proximity.