trek330

I took a 40 mile hilly ride yesterday where elevation gained was listed at about 2000 feet and when corrected with the elevation correction option it soared to over 4000 feet.Is this a trustworthy reading?I did do some serious climbing.

cafzali

When you correct elevation, I assume you're doing it via Garmin Connect, Strava or a similar site? In cases like that, the correction comes not from the device, but from the site itself using map-based data. Map-based data is always more accurate than the barometric altimeter readings from a Garmin.

speedwobbles

Is there any sort of source for that? I find that my "corrected" readings on Strava are much higher than I would expect, while the original readings were close. This is mainly on rides where I start at a know altitude, do a climb to a known altitude and then descend home, so I have a pretty good idea of what the actual elevation change is.

jimc101

I ride with a 200, but if I do a ride with someone with a 500, their elevation gain will always be greater with the 500 than the 200 for the same ride, similar with iphone, riders always seem to have slightly faster times, even when riding at the same pace.

Have found normally a 200 & 800 will give reading far closer than a 500 vs them.

cafzali

The U.S. Geological Survey is the source for all map-based data, meaning the elevation gains have actually been charted by an individual using time-tested equipment. There are issues with barometric altimeters that just don't exist with maps. Ever ridden with someone who has the exact same device as you on the exact same route? I guarantee you there will be 2 different readings, sometimes they'll be off by a decent amount. This mainly stems from the fact that a barometric altimeter has to be calibrated using known data for a particular area before it will really be reliable. Garmin itself points out that the elevation data is "designed as recreational GPS devices, as an aid to navigation. They should not be used for any activity requiring precise measurements such as surveying or weather prediction."

See here:
https://support.garmin.com/support/searchSupport/case.faces?caseId={8b380bc0-1901-11dc-70f0-000000000000**

lsberrios1 

As a reference. I ride with 2 other friends and we all have G 500. We also use strava. When we ride together there is one garmin that registers almost always 200 feet less than the other 2 for the same ride. Given that the same devices were used, and we use the same data tracking website I assume it is calibration and garmins are fairly accurate within 5-10%. That doesnt sounds great but it's not horrible either.

Bacciagalupe

My guess is that Garmin Connect's "correction" is not accurate.

I did a ride this weekend that reported on the device at 1900ft, in RideWithGPS as 2200 feet, and with elevation corrections 5000 feet.

I don't expect those types of measure to be super-accurate anyway, so I just go with the device and/or RWGPS. If it's off by a few hundred feet, I don't really care.

trek330

My correction source was Garmin Connect.That raised me from about 1900 to more than 4000 feet.Correction from Strava was a more modest 2300 feet!What gives?I'm assuming they both use the geological survey for corrections.Strava gave me a top speed of 42 compared to Garmin Connects 38.5.I don't have a speed sensor.I'm not happy with all these innacuracies.

TrojanHorse

It's not particularly accurate but then neither is the barometric estimate, which is obvious if you ride 40 miles down to the beach, turn your GPS off and turn it back on and ride home.

One problem with the map correction is that your GPS position may not be perfectly accurate - so that time tested equipment is measuring the elevation of the road but your garmin thinks you're just off the road. My neighborhood has a measured spike in it when I use the altitude corrections.

RWGPS notoriously overstates elevation gain so if you end up with more... it's probably wrong.

I agree with Bacciagalupe though - if it's off by a few hundred, I don't care.

itrsteve

Ditto on every last account.

nhluhr

Correction is based on DEM which first of all only has a resolution of approximately 30 meters (therefore unable to account for smaller changes) and secondly don't always account for road cuts or bridges, etc. You could ride a bridge over a ravine and the DEM might show that as a very steep descent down into and climb up out of the ravine whereas your barometric altimeter (and even just GPS-elevation data) would not make that same error.

Ridelots24

Garmin elevation correction = Good and accurate

Strava elevation correction = bad

znomit

Ive seen some large spikes with elevation correction on (generally climbing hills), and some large humps and holes when a front blows through when using barometric pressure. Neither is perfect.

Garfield Cat

All this means Google will come up with a new system to be even more accurate than these USGS methods.

SpeshulEd 

Without correction turned on, my 810 thinks my driveway (about 50 feet) is a 2000' climb.

I don't even shift out of the big ring. lol.

lsberrios1 

Well, i've seen people break KOM on a 25% .4 miler at 22mph when using their Iphones as GPS so that means the garmin is pretty accurate.