I've just uploaded an updated SpeedReader to www.ka72.com/
The new B7 version tightens up a few calcs, and provides more "verbose" output. You can reprocess files that have been previously uploaded and see the more verbose results.
Under the hood, I've been working hard to support multi-session files (like SBP files) and multi-file sessions (where you've switched off the GPS, moved to another location, and resumed logging later in the same day with a new file.)
I'm stuck on a few issues, and if anyone has some thoughts I wouldn't mind hearing them.
1) How to tell when a "run" begins and ends? I have an algorithm which is working pretty well, but I'd still like to tighten that up a bit.
2) How to extract trackpoints from doppler data (specifically so I can work out Alpha 500s for GT-11 users more efficiently.)
Hi Dylan,
You can extract trackpoints from doppler data, but then you're completely missing the point of using trackpoints for GT-11's!! The whole idea is that the doppler data is filtered and therefore not accurate. You can't get that data back by integrating the speed.
I think you're going to have to have two arrays, one for doppler, one for trackpoints, and use the alpha algorithm on the trackpoint array. It's quite OK to differentiate the trackpoints to get speed and then run your algorithm on that array -- but not to go the other way.
In any case you'll have to do this for the trackpoint data that Garmin GPX files have.
Regarding run detection, perhaps if there is a change in direction of 90 degrees or more between two points that are 20 seconds apart then that counts as a new run? It might take a bit of fiddling to get right. I would also count any speed less than 3 knots as a new run...
Hope this helps.
All checks out, cool
Should it say top 5 10sec speeds? confused me initially
2 Second Peak (kts): 32.59
5x10 Average (kts): 31.07
Top 5 speeds: 32.0073
31.2143
30.9859
30.9227
30.2404
1 Hr (kts): 12.94
Alpha 500 (doppler): 22.24
Nautical Mile (kts): 21.06
Total Dist (doppler km): 45.69
Great Job!
I've been browsing through some of the submitted files recently, using the "view" feature on the site, and noticed that some of them had really strange "approximate" timezones (visible when you select the "verbose" option.)
I checked out these files, and sure enough, there has been an influx of files submitted from European locations.
Just wanted to say thankyou to those submitters, whoever you are. It's great to have some really varied testing data.
