Register | Login
Attackpoint - performance and training tools for orienteering athletes

Discussion: climb

in: RLShadow; RLShadow > 2016-06-27

Jun 28, 2016 1:19 AM # 
I've noticed similar things when doing laps at Bristol - Aries is similar to your loop, with 5-6m difference between the low and high points. Several years ago (and with the 305), I think climb was calculated OK but the first three laps always had a more squashed elevation profile than the rest. Now (and with the 310XT), the reported total climb is very low even after 20 laps. I assume some kind of smoothing is going on, but I'm not sure if AP's algorithm has changed over the years or whether it has something to do with the 305 vs 310XT.
Jun 28, 2016 11:59 AM # 
There are discussions on climb in AP. In general, it goes off of the USGS data, and not the GPS, but you can change settings. On one of my running routes through powder mills, when I run up Corderoy Rd, AP always blips me up and down a short hill which isn't there (but was there when they made the USGS base contours).
Jun 28, 2016 2:28 PM # 
Correct, AP does go off of USGS data, rather than the GPS altitude which can be very noisy especially when there isn't a whole lot of real climb going on. (When there is significant climb, like, say, climbing or descending Bristol Mt or Pikes Peak, the GPS elevation is quite useful because the actual climb is way more than the noise.) So basically, it makes sense to use USGS data. But what I'm puzzled about (and this is confirmed by Stina's experience at Bristol Airies loop) is why it did show in this case about a 15 foot climb in the one "dip" in the loop I was doing (and that makes sense, based on my impressions of the actual climb -- not much, but it was in a short distance so it was noticeable). And I did 3 laps, but the total climb was essentially what I would have done in a single lap.
Jun 28, 2016 2:40 PM # 
I do recall seeing at some point that there is smoothing done - the reported climb isn't just the sum of the elevation gain between successive track points because then noise in the track could result in the accumulation of lots of tiny little climbs. So there must be something about the pattern of a repeated small hill - and that being the only climb to speak of in the track - that doesn't quite make it past the smoothing threshold.

This discussion thread is closed.