GPS Accuracy Issues

Same experience. They measured less distance, which I can admit, but pace! Pace for running is important and they show pace much worse than real, while old  Forerunner 225 shows it quite good. Support says it's GPS drift, but it always drifts to slower side, which looks like software problem.

Tried different GPS settings - no luck. Software version is 4.0. Hope they'll fix it soon.

Hello, 

I got the same impression that GPS track is rather good, but distance (and therefore pace) is underestimated. 

Sunday, I ran a roughly 40km trail, and was very surprised to cross the finished line at 38.6km! 

As a matter of investigation, I tried to import my GPS track (gpx from Garmin site) to two different websites (strava and open runner), and both agreed that the track length is indeed 40km!

I really do not understand the issue here. I have a clean track (some corners cuts and wobbles, but with heavy tree cover it is normal), with the right length, but somehow the distance shown on the watch and on Garmin website is wrong! 

I tried the same thing on some other tracks I have (correct distance fonction on strava, with open runner to verify actual distance) and I always have the watch underestimating the distance even when track is fine (better on road, worse on trails), but corrected distance on strace agreeing with the manually drawn itinerary in open runner.

Seems to me that GPS accuracy is not the main issue here, but a bug somewhere in the distance calculation that is always off.

Hope this helps.

Edit : gps + Galileo, 1second recording, as it is the mode giving best results for me. Issue is more marked in other modes... 

Hi,

I have quite the same observations (except that in my case gps+galileo is worse).

My feelings about all this are mitigated... When running with other people my gps trace is not so bad compared to the others but indeed the total distance is always lower (especially on trails with a lot of ascent/descent).

Some turns are a bit shortened (which is not good), but gps drift looks much smaller when not moving or when gps signal is bad (which is good). But could the Instinct be right against all the other watches ? Quite unlikely...

Now I use 3D distance. This is a bit cheating but at least it gives me results more consistent with other people and with courses I am navigating.

I use 3D when trail running or hiking and don't consider it cheating.

Same here. My standard run of ~6 km is alway ~250 m shorter than that estimated by my girlfriends Forerunner. I did not know who was right though. Yesterday, I partipated in a 10km run. My Instinct watch gave 9.62 km. I was worried that my finishtime was perhaps misleading because the track may have been shorter. But other runnes had estimated distances of 9.9-10.1 km. I recalculated the track by afstandmeten.nl, which gives 10 km (+- 100m). 

There is definitely something wrong with the distance calculation by the instinct. I hope Garmin pays attention to this. I use GPAS+GALILEO, 1s frequency. But from the comments here, the error seems to be independent of satellite system.

I know that these comments appears like 'I thought I would have been faster than I am'. But the problem is really that this instinct gives smaller distances and paces than other watches,. This is not the only forum where this problem has been noticed. See here (https://www.reddit.com/r/Garmin/comments/c6fgrf/instinct_gps_accuracy/ and here https://www.reddit.com/r/Garmin/comments/bdw7l6/instinct_gps_issue/). I will soon do a more in depth study on comparing this wil different GPS settings (GPA, GPS+GLONASS, GPS+GALLILEI) and together with the Forerunner 235 on the same arm. This should give comparable data...

That's a good idea. I was blessed to be able to add a 945 and wear it daily now. I'm planning on a hike or trail run today and will record it by both watches and can check/compare the distances.

Ok, here goes...

Today’s hike I had the 945 on my left wrist and the Instinct on my right wrist. I turned off the wrist heart rate on the Instinct. Both watches setup for GPS-GLONASS, smart recording for the 945 and 1 sec for the Instinct. 3D Speed and 3D Distance on for both.

Within the first mile I could see the difference of about 0.05 mile, the Instinct showing 1 mile when the 945 had 1.05. Later after a few more miles it was closer to 0.055 per mile difference.

Later in the hike, in an area where the Instinct has had GPS signal problems before, I once again had signal problems, but, only associated with the Instinct as the 945 was humming right along indicating a good GPS signal.

At the end of the hike, both watches were within .03 miles of one another. The 945 showed 10.03 and the Instinct showed 10.00. I thought, just how could that be? Maybe calculated it incorrect as we went along, but maybe re-figured it at the end and came up about the same…?

NO

After up-loading both activities and looking at them… the Instinct clearly had signal problems. From the point in the hike where the signal was degraded until the end, the track was not very good. It’s the wandering off and back that made up the lost distance. I’ve never had this bad of a track from the Instinct, but, this is the first time, this time of the year that I’ve been in this heavy tree covered area with it.

The 945 did absolutely great, track wise. The wrist heart rate, not so much.

I’ll see about taking the Instinct out with me tomorrow when running and see what happens.

Here my first check. I ran with my Instinct and with my girlfriends FR235 on the same arm (see https://connect.garmin.com/modern/profile/7123ce98-1ff0-4647-a34a-dd46bac65687 , this should be a link to my two runs, profile of me (Jurrien Knibbe)). Over ~8.5 km, the instinct was 260 m short compared to the FR235. I clearly see that the instinct is cutting corners, whereas the FR235 is following the paths that I had taken through the park. Here, Instinct was with GPS+GALILEI, FR235 was with GPS+GLONASS. Next time I will run with both on GPS+GLONASS to check if its truly due the watch or due to the satellte system.

cheers Jurrien