I know there are other threads on this topic, but it warrants bringing up again. As a newer skydiver (<200 jumps) who has also been graced with a coach rating, I’m obsessed with fall rate. How fast do I fall in suit 1 vs suit 2. How much weight do I need to put on to fall with that beef cake over there? How much fabric do I need to add to fly with the skinny dude?
With that in mind I offer action 271387, which I submitted for review. Here’s a visual analysis of the data being presented.
Issue one: vertical speed appears to be an average. It doesn’t take twenty seconds to reach peak velocity, and it doesn’t take this long for my canopy to open. The green and black lines show a clear opening event, and the blue line should match… instead the vertical speed graph uses a different algorithm and is completely inaccuragte.
Issue two: the altitude graph shows me landing at the point I turn from downwind to base. I’m 600 feet up, but the graph shows me on the ground.
You can use an alternative path. Download the action as a flysight file and then import it into flysight viewer. It basically just downloads the gps data of your flight. You load that into flysight viewer and it gives you the data for that point in time. There is also baseline.ws if you want a web view of it. Unfortunately, there is no one app or website that is the best currently for skydiving data. They all have their strengths and weaknesses. The cool thing is dekunu does this automatically for you and gives you a good summary view.
Okay - super interesting, @kat00. I did as you suggested and I’m definitely getting better visualization of elevation and where it I was on my flight path using the FlySight tool. Unfortunately vertical speed is still crap… which makes me wonder if FlySight is expecting one measurement and getting another.
The data IS there… I took the CSV and did the math myself and am able to get much more accurate results. The raw data shows about ten seconds from pilot chute out to fully inflated parachute. The top graph is from Dekunu Cloud, and the bottom is the raw data presented at about a one second average.
Awesome work! I’m curious as to why the graphs on the Dekunu site don’t just show us the raw data in a graph. Clearly the raw data points are all there and being correctly recorded. One would think it would be quite simple to take that raw data and dump it into a graphical display as you have done. Your homemade graph is much more useful than the one on the website. Hopefully Dekunu can fix their site to display the correct data.
Thanks, @jphart03. What’s interesting to me is that the horizontal speed doesn’t have the radical smoothing out. I wonder why one does but not the other?
