Garmin elevations in crisis.

[fossyant]

I appreciate for route planning, but this is looking at the device, and I don't pay too much attention as it's either for routing or just recording rides. When planning a ride, I will look at ascent if I don't know the area but that's with the mapping software.

 

[EckyH]

Atmospheric pressure derived is considered more accurate than GPS elevations. Only really a problem if you have a massive low or high pressure system moving rapidly across where you are.

My commute (roughly 25,5km one way) has 141 metres total ascent according to Openstreetmap.
The Garmin 840 I use to record those rides shows different values for nearly every commute.
Maximum were 218metres...

E.

 

[Ming the Merciless]

My commute (roughly 25,5km one way) has 141 metres total ascent according to Openstreetmap.
The Garmin 840 I use to record those rides shows different values for nearly every commute.
Maximum were 218metres...

E.

Right, and do all those routes take an elevation measurement at exactly the same locations every time? Trying to assess elevation gain is like fractals. A different number of trackpoints or measured at different locations and you will get different results regardless of gps or pressure derived.

 

[geocycle]

You need to get to know your route planner! I find Cycle.travel underestimates elevation compared to my wahoo by about 10-15%. It is worse in very undulating terrain and much closer in flatter or more consistent climbing. This is a classic interpolation issue and limited by the size of the servers doing the number crunching. It’s only a ‘problem’ in planning routes for the lunacy climbing challenge!

 

[EckyH]

Right, and do all those routes take an elevation measurement at exactly the same locations every time? Trying to assess elevation gain is like fractals. A different number of trackpoints or measured at different locations and you will get different results regardless of gps or pressure derived.

It's the same real route (not routing in the meaning of navigation via the Garmin device, so no predefined track points) every time. The only deviations are 2-3 metres beside the road to purchase some fruits or vegetables or to go brambling.
On the other hand I know that it isn't flat as flat can be and the biggest mountain here is the "dutch mountain". It's invisible for humans and has no descent, only an ascent as long as the whole distance. In both directions. Probably the bike computer data include that mountain...

E.

 

[Ming the Merciless]

It's the same real route (not routing in the meaning of navigation via the Garmin device, so no predefined track points) every time. The only deviations are 2-3 metres beside the road to purchase some fruits or vegetables or to go brambling.
On the other hand I know that it isn't flat as flat can be and the biggest mountain here is the "dutch mountain". It's invisible for humans and has no descent, only an ascent as long as the whole distance. In both directions. Probably the bike computer data include that mountain...

E.

The point is that it will vary every time and is to be expected. It is not measuring height at the same location each time. A metre here or there every few seconds soon adds up.

 

[Milzy]

Garmin 1040 and other Garmin devices guess the elevation by atmospheric pressure: barometric height "measurement".
Therefore I don't trust them not as high as I could throw them.

E.

Yep, I’m going back to wahoo, should have never left. Bolt 3 probably.