GPS Discrepancies

[KTHSullivan]

Out yesterday for my daily dozen, wandered into the Lincolnshire Alps . Cat Eye GPS gave me 28.38 miles and 1840 feet ascending. Mio GPS provided 28.4 miles and 921 feet ascending. Obviously the distance provided by both units were more or less the same. However which one do I believe for height gained. Somewhat slightly spooky that Mio gave me more or less half that of the Cat Eye. Is the Cat Eye measuring the up and down or is the Mio only measuring half of the up? I have looked at the settings on both units and it appears that they are set up correctly.

Any suggestions?

[Jdsk]

Previous discussion of height differences and errors:
viewtopic.php?f=1&t=141339

Jonathan

[Mick F]

Ancient Chinese saying:
Man with one clock always knows the time. Man with two clocks is never sure.

I've compared different GPS devices for ascent and distance, let alone speed. They've never agreed. I had an Edge305 and an Edge705 at one time that didn't agree, and now a Garmin Montana and a Garmin Edge20 ....... and they don't agree either, but I believe the Montana because it's a very sophisticated device.

I drove the support vehicle for some LEJOGers in 2007, and they all had devices of one sort/make or another, and they rode EXACTLY the same route as each other. They compared their figures every evening and at the end too, and none of them agreed at all.

As for ascent as a separate subject, if you have a device with a barometric altimeter to assist the GPS elevation, you will find that the headwind will affect the altitude. Heading into a stiff breeze the hole in the body to sample the air pressure will see a higher pressure and consequently the device will read low. Put your hand in front of the device to shade off the wind, and you will see the altitude increase. The figure when I've experimented, was in tens of feet difference.

A hand in front, will create a lower pressure than ambient, and removing the hand will create a higher pressure than ambient.
You cannot win unless you get out of the wind and remain still to let it settle.

Believe me, I have experimented!

[gom]

My experience, with two of us using a GPS each, is that no one GPS unit will give the same result as another. 100m out over ~1000m climbing not that uncommon. But one giving 50% of the other seems a bit much.
I've always understood that GPS altitude was *much* less accurate than position, so some disagreement unavoidable.
This may be reduced with units that also track the Russian (GLONAS?), European (Galileo), Indian (I think exists) satellites. Don’t know what you have.

Something to check is that the unit settles down to the right altitude when you switch it on. Mine can register -50m at my front door, > 100m to low. Can take some time to get near the correct value.

[gom]

...
Heading into a stiff breeze the hole in the body to sample the air pressure will see a higher pressure and consequently the device will read low. Put your hand in front of the device to shade off the wind, and you will see the altitude increase. The figure when I've experimented, was in tens of feet difference.

A hand in front, will create a lower pressure than ambient, and removing the hand will create a higher pressure than ambient.
You cannot win unless you get out of the wind and remain still to let it settle.

That's an interesting point. Barbag, map over the GPS, etc. could be having a noticeable effect. (thinking of my own observations)

[KTHSullivan]

Man with one clock always knows the time. Man with two clocks is never sure.

Lot to be said for that. In my navigating days I could never remember what chronometer I was supposed to be using for the noon day fix.

[Psamathe]

I wonder if some of the discrepancy is because the different devices are using different ways of measuring altitude (are they?). e.g. some Garmin devices include a barometric altitude sensor whilst others use the GPS and some devices do/can tie your altitude to "ground" based on mapping (which normally means interpolation using NASA SRTM - which can give OK or bad readings depending on the landscape).

Ian

[KTHSullivan]

Just find it a tadge strange that one device plotted more or less half the height gain of the other. I have just plotted the route on CycIOSM an that gives me 805 feet of altitude gain and roughly the same distance provided by both units. So I am inclined to believe that the lower figure is more accurate. Obviously for a definitive answer I would need a Dutch cargo bike so I could lug around a differential GPS set up.

[Psamathe]

Just find it a tadge strange that one device plotted more or less half the height gain of the other. I have just plotted the route on CycIOSM an that gives me 805 feet of altitude gain and roughly the same distance provided by both units. So I am inclined to believe that the lower figure is more accurate. Obviously for a definitive answer I would need a Dutch cargo bike so I could lug around a differential GPS set up.

I have wondered if there is a "right answer". All roads have uneven surfaces so a 1cm edge over a road repair and a few trivial hollows you don't even notice might be considered as being included in ascent/descent. But if a trivial hollow (e.g. 2 cm over half a meter) is not included at what point do you start accumulating something as ascent/descent.

I suspect that most online mapping systems would calculate ascent/descent using NASA SRTM data which is normally a 90 meter grid (sometimes 30 meter grid but the data volumes/storage goes up them). And that means interpolation which can give some horrendously bad estimates e.g. EV6 through Doubs valley where the cycle path is next to the river right beside many steep hills and cliffs so one "point" is sometimes high up a hill, other low down in a valley and you are riding along a pretty level path. Some SRTM data has had vegetation height (trees) removed, others areas haven't (not sure of latest releases) - which means cycle along a level road open countryside into a forest and your calculated ascent will be the height of the trees. My phone app has a setting to use GPS for altitude or to use calculated altitude from mapping (or "best option" available at the time).

Ian

[Richard Fairhurst]

Exactly that. A lot of it is down to the smoothing algorithms that the site applies.

Planning a journey along EV6 through the Doubs valley (from 47.1468,5.7222 to 47.4268,6.6155), cycle.travel reckons 390m of climbing, RideWithGPS reckons 1036m, and Komoot seems to think somewhere between the two, although how the heck you get Komoot to route along EV6 I don't know. Even then I suspect cycle.travel is overstating it.

[simonhill]

Out yesterday for my daily dozen, wandered into the Lincolnshire Alps . Cat Eye GPS gave me 28.38 miles and 1840 feet ascending. Mio GPS provided 28.4 miles and 921 feet ascending.

Any suggestions?

Any suggestions, yes. Do a few more rides to check results. Who knows, this could be a one off funny. Always worth double or triple checking on different rides.

[mattsccm]

I plonked 4 Garmin GPS on the bonnet of my Land Rover once whilst green laning. They all said different things. I knew where I was by the OS map (road junction scarily close to a cross of grid lines) . None were correct with nearly half a mile between the extremes.

[sjs]

I use Locus Map on my phone for recording walks and rides. When I switched the height recording from GPS to data as per the map not only did the height profile of any given trip become much smoother, but also the claimed calorie usage was reduced by a factor close to 2. So I assume it must have been estimating calories based on constant climbing of short but very steep spikes derived from GPS jitters.

[KTHSullivan]

Out yesterday for my daily dozen, wandered into the Lincolnshire Alps . Cat Eye GPS gave me 28.38 miles and 1840 feet ascending. Mio GPS provided 28.4 miles and 921 feet ascending.

Any suggestions?

Any suggestions, yes. Do a few more rides to check results. Who knows, this could be a one off funny. Always worth double or triple checking on different rides.

I intend to now Simon, this was the first time I have had two running at the same time. I am out most days I will see what develops tomorrow.

K

[KTHSullivan]

I plonked 4 Garmin GPS on the bonnet of my Land Rover once whilst green laning. They all said different things. I knew where I was by the OS map (road junction scarily close to a cross of grid lines) . None were correct with nearly half a mile between the extremes.

I had a similar one a few years ago on foot in the Galloway forest. The hand held unit at the time could be set up for OSGB. Always used to carry a map. The GR from the GPS put me 20m from where I was, after I did a resection. Lucky I was not on a mountain ridge somewhere.