Incorrect route elevation profile

AFAIK, the recorded elevations are determined from the barometric pressures read along the way.  If there are changes in ambient pressures due to weather fluctuations they will result in elevation changes although one may remain at constant vertical position.

To determine the effects of ambient pressure changes on recorded elevations, I use Digital Dutch:

https://www.digitaldutch.com/atmoscalc/index.htm

I think you're right but off point CowboySlim.  Changes in barometric pressure would account for some error if the device were set to record elevation by barometric pressure, but I think the errors Jesus is reporting are way too big to be explained by changes in barometric pressure.   I almost always get crazy elevation errors in my 66i tracks.  This has been true since I got the device and remains true in version 5.40.   Tracks can show elevation changes of several hundred feet in track points that are only a few minutes apart. And that's with your feet on the ground.  Not in an airplane.  This is a problem that can and hopefully will be solved in future software updates.  In the meantime, users should be warned not to use tracked elevation for trip planning.  It's usually drastically wrong.  What Jesus reported is typical of my experience.  And I don't think changes is barometric pressure explain the extent of these errors.

Hi, Jesus.  I assume you're still getting the same gross elevation errors.  This is typical of the 66i in my experience.  Recorded tracks show crazy sudden changes in elevation. I think from your message that you assume the 66i is getting its elevation information from the map data.  I don't think that's right.  I'm not sure exactly what the 66i is doing.  But I usually get errors as great or greater than those you reported in my recorded tracks.  Theoretically, there are three sources of elevation information.  The stored map elevation can be used from a horizontal GPS location.  Another is from the measured barometric pressure.  The third is to calculate the position in three dimensions from the received GPS signals and subtracted mean sea level.  I think you assumed incorrect that the 66i was getting elevation information from stored maps.  I don't think that's right.  If it were, we wouldn't be seeing the kind of insane errors that are typical.  On foot walking, my tracks often show changes of many hundreds of feet in elevation in a segment of track points that span only several minutes and cover a horizontal difference much less than the recorded change in vertical distance.  The 66i still has firmware problems.  Garmin has corrected some of the most glaring.  Hopefully the programmers there will get to the elevation problem sometime soon.

Don’t know if there’s a f/w issue there or not. The vertical error in a GPS fix is much larger than the horizontal error. IR devices try to conserve power as much as possible. Speculating that that might mean settling for a lower quality fix, which would result in even larger vertical error.

I don’t even know for sure that the GPS is the source of elevation data here. Nor do I know how the auto-calibrate settings affect barometer-derived altitude on this device. You might get a more definitive answer from tech support. If you believe there’s a f/w problem, you might want to report it anyway.

Edited to add: To be clear, I suspect that the actual elevation profile for the route came from DEM data on the map you used to PLAN the route. However, the remaining ascent/descent is going to involve the device’s idea of your current altitude. As jlg2 mentions, the error would not be anywhere near that large if the device was using horizontal position plus DEM data. This is why I suspect either GPS derived vertical position with high error or badly/uncalibrated barometric altitude.

Hi, Jesus.  I posted to your thread earlier but forgot to mention that a lot of the problems with what the 66i calls "elevation" can be avoided by using what I think it calls "gps elevation," which does not rely on the barometer.  The gps elevation uses a straight mathematical calculation of the device's position in 3d space compared to the mean sea level mathematical surface.

All,

I think people are misunderstanding this post. There isn't a problem with the accuracy of GPS or barometer elevation. The problem is with the elevation profile of the active route, which is based on map data.

Hi, Jesus.  Can you clarify where the discrepancy comes from.  Elevation information can come from a route profile, from a track profile, from reading the handheld device GPS elevation, from reading the handheld device barometric elevation or from reading the elevation from the contours on one of the map displays.  I read your original post a couple of times, and I'm still not sure what is conflicting with what. 

I'm using routes, not tracks. Therefore, the profile is built using the elevation data on the map. I was using OpenTopoMap for Italy. The contours always matched barometer/gps elevation, therefore the contours are correct. However, the profile plot drawn on the elevation page was incorrect.

I see.  As you say, that is strange.  I assume it has nothing to do with barometric or GPS calculated errors. Is it a possible unit discrepancy.  Do the errors match the feet-to-meter ratio?  3.28 or 0.305?

Hi Jesus, After my earlier reply today, I noticed Garmin put out a new firmware version to the 66i.  And the update notes say that route displays were corrected.  It's possible that fix applies to the the conflict you noticed.