hmmm, strange... really no complaints here. Not saying it's the best ever, but surely better then what I see in your links. Because 9 out of 10 times I can't post anyway due to the forum I'm going to keep it short, but 2 things:
1. Do you have 1-sec recording on? If I see your links it looks more like smart-recording? Which combined with high speed and short turns might cause issues 2. Do you wait a proper time before you start? Ie, get the green gps-light and wait a small bit more?
These 2 can largely affect the tracks and positioning...
Otherwise there might be an issue with the watch itself, I run in a city area as well although no huge amount of large/high buildings here, but it performs just fine for me.
That waterfront segment seems to have huge problems generally. The guy with the CR did not actually run the whole route and the one in 2nd had an initial speed spike straight across the water. The bloke in 3rd actually ran it OK. However all the traces I looked at had some issues. A look on Street View suggests you are going round fairly high buildings on tight turns - never going to give the best results. It is obviously disappointing if your GPS traces aren't perfect in places where you run regularly but think this is quite an extreme example.
Hello, I can see it exactly same way as Nicoloco, I'm very happy about features, but GPS accuracy is very disapointing. Just ligt tree cover and instant pace goes immediately 30 or more off way. When clear view to sky, performance is great, but overall nothing much. Even thinking of returning FR935 / For that money I was expecting better results.
I'm running with the Forerunner 935 since one month. Before I had a Forerunner 310XT and a Fenix 3. With the 310XT and the F3 the tracks were sometimes buckled, but the calculated distance was very accurate. (At 10km competitions the deviation was less than 1%.)
The tracks of the FR935 are looking well on the map. But the calculated distance is generally 1,5 to 2 percent shorter than on the old watches and also shorter than theoretical distance.
You can see it on the following tracks. The normal distance is 10,0km. The distances with the FR935 were shown also with 10,0 and 10,05km on Garmin Connect. But the tracks were 200m longer (100m by a dertour and 100m at the end). As Comparison you will find old tracks from the 310XT and F3.
I know that the GPS accuracy at consumer products is +-2%, but the shown track on the map is ok. Only the calculated distance is too short. (The track with 1s recording was the shortest track.)
If distance and pace accuracy is very important to you, I'd recommend the Stryd footpod. I don't run on trails, and in my experience it is EXTREMELY accurate with the default calibration (and no auto calibration). It has always been within 0.5% of the plotted distance (and never over the distance) and instant pace is definitely usable. It's been so good that I've turned off GLONASS and 1 second recording. Even so the GPS tracks have been better then the F3 and the 910 before that.
@Olu.F.: Thanks for your answer. Sure, the Stryd footpod is well, but it's also very expensive.
I'm wondering, because the tracks from the old wachtes and the new forerunner are looking well on the map. Only the calculated distance is short. I know this from my old Fenix 3 with the first firmwares. Tracks were often short, but Garmin corrected it by updates.
If distance and pace accuracy is very important to you, I'd recommend the Stryd footpod. I don't run on trails, and in my experience it is EXTREMELY accurate with the default calibration (and no auto calibration). It has always been within 0.5% of the plotted distance (and never over the distance) and instant pace is definitely usable. It's been so good that I've turned off GLONASS and 1 second recording. Even so the GPS tracks have been better then the F3 and the 910 before that.
I've been eyeing off Stryd, but $$ has prevented me currently. Have you any experience with Garmins Footpod? for mostly distance tracking?
Garmin foot pod gives a decent enough pace but will certainly need calibration to make that meaningful and then only so within certain pace ranges. Unless you do all your running under trees or in other areas of poor GPS I would just use the watch for distance.
I'm running with the Forerunner 935 since one month. Before I had a Forerunner 310XT and a Fenix 3. With the 310XT and the F3 the tracks were sometimes buckled, but the calculated distance was very accurate. (At 10km competitions the deviation was less than 1%.)
The tracks of the FR935 are looking well on the map. But the calculated distance is generally 1,5 to 2 percent shorter than on the old watches and also shorter than theoretical distance.
You can see it on the following tracks. The normal distance is 10,0km. The distances with the FR935 were shown also with 10,0 and 10,05km on Garmin Connect. But the tracks were 200m longer (100m by a dertour and 100m at the end). As Comparison you will find old tracks from the 310XT and F3.
I know that the GPS accuracy at consumer products is +-2%, but the shown track on the map is ok. Only the calculated distance is too short. (The track with 1s recording was the shortest track.)
i hear u... i'm also running with my forerunner 935 for almost a month now, previously was using fenix2, and have tracked once with both device and found out that for every 1km tracked on my fenix2 the FR935 lost 0.01km, yeah people will say you only did one comparison workout running on both device, but i'm so used to my Sunday long run route with my fenix2 that i will know when will it give me a nudge on what km, did a few more long run on FR935 with gps+glonass and every second tracking result are the same, also ran on my usual weekday jogging track as shown below, both f2 and fr935 recorded 12.21km (fr935 setting was at gps+glonass & every second tracking) respectively, but on mygpsfiles it show the fr935 actually tracked 12.3km. Apart from the distance tracking problem, i must admit that FR935 seem to track more cleanly then my fenix2 but I do hope that in time Garmin can fix the distance tracking via firmware updates.
Garmin foot pod gives a decent enough pace but will certainly need calibration to make that meaningful and then only so within certain pace ranges. Unless you do all your running under trees or in other areas of poor GPS I would just use the watch for distance.
Unfortunately I do a lot of trail running of late, much prefer it than footpath running...So a pod probably would be better for me
