I tend to plan routes on the OS app but tend to track rides on Outdoor Active. I have noticed when doing the same route the height gain is significantly different on the two apps , often 15-20%. Anybody any idea why and which would be more accurate ( I would assume OS)?
The bloke who runs cycle.travel has done a lot of work on this. (I seem to recall that currently the app is more accurate than the website?)
I think it's - broadly - to do with the way they handle contour lines. In theory, if you cross a contour line, you've gained 10m, and if you wibble about between two lines, you're not gaining or losing anything. In practice, it's a lot more complicated than that. He did post some updates on his patreon about it, but I let my subscription lapse when I was too ill to ride so I can't view them now!
It's easier to imagine elevation models as being made out of really big Lego bricks. If you go straight up a steep hill, the height difference between the bottom and the top will be pretty close to reality. If you are almost going horizontally along the side of a slope, wandering up and down by less than the height of a brick, then the elevation model won't credit you with any ascent, but it could add up to quite a lot.
Different apps use different elevation data and then your device might be using barometric pressure or satellite based elevation calculations too. For consumer priced devices and apps, they can all be regarded as "accurate enough".
Scotroutes +1
There's probably a hundred different ways you could calculate elevation gained / lost, they'd all be equally wrong.
Also consider that the coastline paradox applies just as well in this instance, the answer for how much climb and descent there is on any circular ride is both 0 and infinite because the difference between tour first and last waypoints is zero, but if you had an infinite number of waypoints then you would also have an infinite amount of height gained and lost.
Given what others have said about crossing the contour lines you could assume the one with the greater elevation figure is the more accurate and then stick with that app by way of comparison with other routes.
That's what I'd do unless I had good reason to think otherwise as any 'smoothing' will only reduce the height gain/loss. As above fpr why.
All these things are an approximation - I recently spent a few hours poking around how 'windy ' predicts the temperature and whilst it's normally very reliable on the coast it gave numbers for temp. different to anything else inland. The answer was in how it used a model and how it smoothed that
I’m fairly sure OS maps assumes a uniform gradient between contours and extrapolates from that. In think strava must do something similar as if i do a ride and then use strava elevation data it agrees quite well the OS data.
If you plan in strava and ride in strava you get the same result. So that’s the standard i use. In the end in think you have pick one App and go with it. For me it’s strava.
There some excellent point about what we mean my elevation data. It’s hard to pin down what you actually mean.
Finally not all ascent is the same. 100m of ascent at a constant 15% takes less energy than 10 lots 10m hills where you start each one with a speed boost from the last down
I do feel sorry for event organisers who must get it in the neck when there published elevation gain has no agreement with some tired entrants head unit
Finally not all ascent is the same. 100m of ascent at a constant 15% takes less energy than 10 lots 10m hills where you start each one with a speed boost from the last down
Depends on the terrain more I think.
On the road a long gradual hill isn't so bad, you just get into a rhythm and keep going, you could ride those hills all day. The same elevation with a 15% gradient is over quicker but hurts your legs far more and is harder to repeat. Same for the descents, 6 miles of soft pedaling on a main road off the brakes gives you a much better average speed than dragging your brakes down a 1in3 hairpined green lane.
And for the same elevation, offroad 15miles at a trail center is easy, 15 miles of mud, rocks or anything else is significantly more energy sapping. Especially when climbing. "Winch and plummet" on fire roads before descending technical trails makes light work of a lot of elevation compared to trying to do the same ride wholly offroad.
All the above is true, but not quite what i was thinking.
If you repeatedly go up and down small hills you get a free roll up the first bit
I tend to plan routes on the OS app but tend to track rides on Outdoor Active. I have noticed when doing the same route the height gain is significantly different on the two apps , often 15-20%. Anybody any idea why and which would be more accurate ( I would assume OS)?
I'm assuming when you the gain is different you're NOT comparing the planned with actual? A gps when riding will record a lot more points than you'll put in when plotting a route on a computer, but gps accuracy for height can be poor. When I upload a ride to strava it always "corrects" the climb - you've now got a more detailed route than anything you've planned but put through their map algo.
If you go on OS Locate you can see how inaccurate the GPS measured height is compared to where you are on the map, to do with the spheroid model used for the earth maybe?
to do with the spheroid model used for the earth maybe?
It shouldn't matter, mapping is projected onto whatever sphere best fits your location.
Morel likley to be just down to the fact GPS satelites are spread out around you so the distances are large and therefore so is the time difference they rely on. Whereas vertically they have far lower resolution, needs at least one more satellite to get a fix and all the usual things about having a clear view of the satellites etc make a bigger difference.