GPS Accuracy

I've read his tests. He also wrote one should use Stryd with stock 100 calibration which would result my pace being 2% off. So while some of it may be good advice it may not apply to all.

I've read his tests. He also wrote one should use Stryd with stock 100 calibration which would result my pace being 2% off. So while some of it may be good advice it may not apply to all.

Your correct for sure and I'm not saying that Fellrnr is correct in all his statements, he's probably not and also human ;)
But for sure: his conclusions fits mine and he seems to have a deeper knowledge of the "whys" then I do.
I found reading his thoughts to be helpful for me.

Btw: I was out for a new run today and once again GPS accuracy is almost spot on, without Glonass. In fact better then ever when Glonass was enabled. It's enough for me to be satisfied with whatever Fellrnr or others say :)

Could just be that GLONASS works better in some areas. For me it definitely doesn't seem to hurt performance. So far would seem to even slightly improve it.

Today I did a run which is 10048m map plotted. My F5 recorded 10044m.

So far I've been very satisfied with the GPS performance. Biggest error I've seen so far is some 30-40m off. This still means well under 0.5% error.

5S recording extra distance

Feeling Frustrated!!! I have been using the 5S for a month now and have found the GPS accuracy is sub par compared to my 920XT. Specifically, the 5S records roughly 1 tenth of a mile extra on every mile, and that adds up rather quickly. My rides/runs are all at the beach (flat, strait, no trees, and no buildings higher than 4 stories). So getting a good GPS signal isn't (or shouldn't) be an issue. I have been using the 920XT on the bike simultaneously and that's how I noticed the issue. My bike route happens to have mile markers on the road, so on my last ride I hit the split button on each watch to record the difference. Sure- I might hit the button a little early or late each time, but it shouldn't be a whole tenth of a mile off. You can see the results in the Connect files available at the below links (FYI- I'm using GPS Only on both watches). It looks to me like I'm not alone with this problem. Has anyone heard anything from Garmin on this issue? Any chance its a SW thing that they will be able to fix?
https://connect.garmin.com/modern/activity/1813943648
https://connect.garmin.com/modern/activity/1813944937

PS- I don't really care as much about creating pretty tracks as I do about the actual distance/speed accuracy.

Feeling Frustrated!!! I have been using the 5S for a month now and have found the GPS accuracy is sub par compared to my 920XT. Specifically, the 5S records roughly 1 tenth of a mile extra on every mile, and that adds up rather quickly.

Are you using GPS for speed and distance or the bike sensor.
As it is recording cadence, I guess you could get distance from speed sensor on the bike.
In that case you need to check if both watches have configured exactly same wheel size.
Looking on the GPS accuracy of your watch I can't imagine that you have serious issues with GPS.
BTW: owning a 935 as well after F3 GPS accuracy with trail running and reading similar complains on F5.

Are you using GPS for speed and distance or the bike sensor.
As it is recording cadence, I guess you could get distance from speed sensor on the bike.
In that case you need to check if both watches have configured exactly same wheel size.
Looking on the GPS accuracy of your watch I can't imagine that you have serious issues with GPS.
BTW: owning a 935 as well after F3 GPS accuracy with trail running and reading similar complains on F5.

I'm using the GPS to record speed and distance on both watches. I plan to do the same measured test/route on my next run wearing both watches but sadly I'm currently nursing a sprained ankle.

Noob question here.

I recently ran a 52 mile race in northern Wyoming and wanted to compare this year's results to last year's. I exported the two tracks to GPX files, looked at them in different viewers and noticed that the total distance of the race varied depending on the viewer I was using.

I am confused about the various distances I got in different viewers! Can someone explain? What distance is correct? I thought a GPX file would always yield the same distance.

Last year I was using a Suunto Ambit 3 Run:
- watch distance 51.9 miles
- Google earth (gpx) 50.3 miles
- GPS Track Reader (gpx) 52.4 miles
(4% variation)

This year I used a Garmin 5s:
- watch distance 52.1 miles
- Google earth (gpx) 50.0 miles
- GPS Track Reader (gpx) 52.6 miles
(5% variation)

As I type this I wonder if Google earth flattens things. There was a lot of vertical accent (1.2 miles) and descent (2.2 miles) over the course.

Both watches were roughly equally accurate: The traces were never more than about 5 meters from the trail when I could see the trail on the satellite image.

Feeling Frustrated!!! I have been using the 5S for a month now and have found the GPS accuracy is sub par compared to my 920XT. Specifically, the 5S records roughly 1 tenth of a mile extra on every mile, and that adds up rather quickly. My rides/runs are all at the beach (flat, strait, no trees, and no buildings higher than 4 stories). So getting a good GPS signal isn't (or shouldn't) be an issue. I have been using the 920XT on the bike simultaneously and that's how I noticed the issue. My bike route happens to have mile markers on the road, so on my last ride I hit the split button on each watch to record the difference. Sure- I might hit the button a little early or late each time, but it shouldn't be a whole tenth of a mile off. You can see the results in the Connect files available at the below links (FYI- I'm using GPS Only on both watches). It looks to me like I'm not alone with this problem. Has anyone heard anything from Garmin on this issue? Any chance its a SW thing that they will be able to fix?
https://connect.garmin.com/modern/activity/1813943648
https://connect.garmin.com/modern/activity/1813944937

PS- I don't really care as much about creating pretty tracks as I do about the actual distance/speed accuracy.

I still don't have a resolution, but I thought I'd provide an update on the response I have gotten from GARMIN thus far as well as some other test results.

First up, below is the email response I got from Garmin (along with a recommendation to use "Smart" data recording- but the Fenix 5S is already set up to Smart Recording)
"Looking over the information, the track information seems nearly identical, My theory is that your Fenix 5s may have every second recording enabled. When this setting is enabled the device lays several more track points in an effort to help you collect more specific data. However, as a trade off if the device records this often, it is much more prone to be affected by what we refer to as GPS drift.

GPS Drift is something that can affect every device, but tends to show more on modern devices with high sensitivity receivers and fast recording settings (such as every second recording). Connect does its best to cut the extra unneeded data points, but it may be that the distance held the inaccuracy.

The two GPS tracks seem nearly identical, so we will need to take a deeper look at the original files, but the unit itself does not seem to be malfunctioning. "

Next up was a curious find. I did a run/walk (3' walk/ 1'run) on the same marked road and I wore both watches (one on each arm). This time they matched each other for the whole course almost exactly. So perhaps there's something about the speed factor. Total 6 mile route was off by less than 3/100ths.

I've been testing the F5 vs FR235 for a couple of months and have posted in other threads about it. The FR235 almost always has a smoother track and smoother pace. The F5 is way more erratic. I'm not sure I can buy into Garmin's explanation, but there's no doubt (in my mind at least) that that F5 is more susceptible to "GPS drift" than the FR235.

I changed the F5 to smart recording last night for the fun of it, and the track and pacing were no better. In fact, they kind of look worse.

I'm so disheartened with the GPS in this device. It's a beautiful watch otherwise.