Fenix 5X - 9.72 Beta Release

Played around with Sporttracks and not really seeing any difference. I've imported all my activities for the last month, and it shows exactly what Garmin displays. Even when I enable the elevation correction in settings, nothing changes. Am I not working it correctly?

Does anyone know of a FW version where everything just worked properly? I'm seriously considering rolling back to an old version, even at the price of all the extra settings.

I'm using the windows desktop version of Sporttracks with the elevation correction plugin. Within the program you can choose between using data from the original source or calculate the data with Sporttracks. In general the altitude correction within Sportracks using the plugin works very well.

What about Strava? There you can re-calculate the distance and elevation...

Strava and the Garmin elevation correction add's way too much vert in my case. I use Training Peaks. But they should address this issue immediately without a smoothly working alti/barometer it's hard to analyze a workout.

I’m not having an issue with my altimeter personally. I compared the same run way back with 6.1 to one this weekend at 9.71 and the profile is almost exactly the same (different calibration so actual altitude numbers are different). Elevation correction is done differently by each application and results depend on the database being used, the location (which will determine the resolution of the data - which can be really bad in some areas like 50plus meters), smoothing algorithms, and the accuracy of your horizontal gps data (consider your gos is 5m off which takes you down a cliff or steep hill by 20 or 30m). There is a reason many of the sites do not apply correction by default to data from watches with barometric elevation. By all means use correction if you have issues with barometric elevation but this is considered the most accurate and responsive than GPS (when working properly).

About accumulated ascent.
1) The only reliable numbers comes from a device equipped with a barometer.
2) For that purpose the best results come from an Altimeter with a single calibration at the activity START. no constant calibration during the activity.
3) The problem with ALL software corrections reside the fact they will all use an AVERAGE altitude in a given cartographic position! It would be impossible to have a database with a CORRECTED altitude for every single inch on the planet.
4) The same kind of problem happens with GPS devices without a barometer because they use a geodesics ellipsoid in his formula ( the globe is not perfected rounded) which results in a AVERAGE altitude for a given position, again...
5) Not to mention that every single software correction algorithm smooth (or not) that out in a different way giving different numbers...
6) in terms of accumulated ascent correction, in MHO, the best results are from Google Earth minus 10%. It´s a hell of an approximation.

Regards,

I’m not having an issue with my altimeter personally. I compared the same run way back with 6.1 to one this weekend at 9.71 and the profile is almost exactly the same (different calibration so actual altitude numbers are different). Elevation correction is done differently by each application and results depend on the database being used, the location (which will determine the resolution of the data - which can be really bad in some areas like 50plus meters), smoothing algorithms, and the accuracy of your horizontal gps data (consider your gos is 5m off which takes you down a cliff or steep hill by 20 or 30m). There is a reason many of the sites do not apply correction by default to data from watches with barometric elevation. By all means use correction if you have issues with barometric elevation but this is considered the most accurate and responsive than GPS (when working properly).

I think it is they way it is sampling, frequency, and their attempt to improve battery. Seeing I have baseline locations and routes established, and two separate devices that both functioned correctly before updates I have controlled testing eliminating both hardware and locations. I also have a few years of data built up, and multiple samples from other devices that ran the same route.

Looking at old activities I see:

No issues on running versions 4.X
Minor elevation issues running 8.1
Major elevation issues running 9.2
Major elevation issues running 9.72

Also, any app is still pulling elevation from the device and doesn't control the sampling frequency or accuracy. It may also be possible you are doing a slower activity like hiking/walking in which the slow sampling would not impact elevation. Just wanted to throw a few ideas out there.

I don't know what ot make of your data either, AgentDBCooper. I am a cyclist and not a runner but I have has mixed altitude performance over a bunch of different brands of cycling computers- Garmin, Wahoo, and Pioneer. In general, on a ride with 4000ft of elevation gain the difference in reported altitude on the computers (and even between friends with the same computer) was between 5 % and 10%. And if I corrected with various sources- Strava, RideWithGPS, etc. the numbers were randomly different by about the same amount. The other thing is that I tended to get the same altitude gain when I repeated the same ride to within maybe 100-200 feet. So, I don't know if the things you are seeing are that unusual at all.

The other thing reported- errors when rapidly changing altitude- seems to be a bigger deal. Especially for someone like me that goes up slowly and down fast.

I do think it is the sampling to save battery at this point. Standard gravel long rides with lower elevation change are off but not terrible. Add a faster road elevation change to gravel and it compounds. Mountain bike and it just falls off. I even paid attention to elevation refresh during the last ride and you can see it doesn't update fast.

The good news is I think if they actually read the forums this can be addressed. I'd rather burn more battery and have accurate workouts. Even on the worst days and 6h+ mtb bike rides my battery still ends at 40-50% starting with a full charge. I'm fine with that.

Here's a few good examples:

115m gravel ride: I read 968ft, two other riders had 1198 and 1277. 200 off.

100m gravel: 1325, others had 1549, 1575, 1503, 1561, 1541, 1519. Only 200 off.

75m mixed road/gravel. Again, I had 1594, others 2056, 2078, 2256, and 2077, Missing 400ish.

10m mtb: 650 for me, others had 1221 and 1372. I'm minus 600.

13m mtb: I had 1155, everyone else had 1501, 1530 and 1464. Again missing 400.

I too am continuing to have significant underreporting of the elevation of climbs while biking on known routes. I believe some of the deviations appear in more windy conditions. When I'm going at a high rate of speed or into the wind up a gradual incline the watch fails to register the gradual climb. I've also had several mountain bike climbs (I'm slow) on non-windy days where the watch again under reports elevation by several hundred feet on a 1500+ ft climbs. .

The clock it is not accurate. It is going up with 2-5 minutes so I need to sync with GPS for the clock to be accurate. The problem is that after a period of time, the clock go up with 2-5 minutes.