This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

GPS running accuracy

Former Member over 6 years ago

Hi everyone,

Just wondering if my device is tracking as it should, as I find its accuracy way off some times when I'm running, and I don't know if it's normal (being a gps watch). When comparing to my mobile running app, the difference are of 0.16km for a track of 5.6km (measured by mobile device and google maps).

https://connect.garmin.com/modern/activity/3441437309 (using gps+galileo)

https://connect.garmin.com/modern/activity/3436153996 (this one's totally off but may be a one-time error or something, using gps only)

https://connect.garmin.com/modern/activity/3428870211 (first run, only using gps)

My problem is mainly that the track is sometimes off which makes my pace awkward, and buildings aren't skyscrapers. During the first km of my run streets are small, but the rest has big streets and clear sky (even a park).

I'll try to use gps+glonass on my next run as I've read it works better, but I'd like to have a better feeling when running as sometimes it shows a distorted average, which I find really important for my training.

Thanks for everything!

0 JSRUNNER_ over 6 years ago

moc_run - you need to make the activities public if you want us to see them :) - a difference of 0.16 doesn't sound too bad. When you say about a better feeling when running are you talking about instant pace when you are running or are you talking about afterwards when you review your training?

0 Former Member over 6 years ago

Hi JSRUNNER_,
Thanks for your answer, I made my profile public. :)
Yes, I was speaking about the instant pace, which I look to see if I'm training well. With 1 sec recording and glonass I'm loosing 10% in distance (100m every 1km), which is making a distortion in my numbers... is this normal in gps watches? (I had the idea they were more precise than mobiles, it's something I want to check so I can make the arrangements after the run...)
In fact, I believe my pace is not being calculated well as the number is too high sometimes for my running/waking speed: for example this morning my watch was saying 11 min/km walking but the mobile stated 9:45 min/km...

0 JSRUNNER_ over 6 years ago

moc_run still cant access them. Also I'm confused you say that you are losing 100M per KM but you quote a difference of .16 KM over a total of 5.6KM that's only 28M per KM which is not unreasonable

I have a few questions

Does your route on connect follow the route you took or does it have great deviations and cut off corners etc

Is your Garmin distance greater than the google maps and/or Phone distance or lesser - and how do you know which is right - the best way to be certain is to run a measured course

Also

1 sec recording won't help with instant pace it just affects the number of data points recorded not the number of GPS fixes.

GPS is not good for instant pace particularly where that pace varies - there are many threads on here and those that want to train with an accurate instant pace generally use a footpod which they calibrate.

0 Former Member over 6 years ago

Hi JSRUNNER_ sorry I didn't see that the images were locked (I pressed the locker and selected "to everyone", I believe they are unlocked now)
Firstly, thanks for your time and answers/considerations!

To answer your questions:

- When walking and selecting glonass+1sec, the distance lost during these last two days was of 100 meters for every km (walking 2 and 3 kms)

- The route has cut off corners, and the line moves as if I am alternating sidewalks along the run (which I'm certainly not doing)

- The garmin distance is less than the measured by the phone, affecting my pace (as time is the same). I know that I'm running at a greater speed because the circuit I run is 1.2km (which is similar to the number the phone adds but less on the watch)

I didn't know about the 1sec recording, just followed Garmin's advice
I wasn't aware that gps wasn't good for instant pace, I thought the calculations the phone and the watch made would be more similar than the results I got...

Again I appreciate your help and guidelines!

0 almaca over 5 years ago

Same problem with GPS accuracy of vivoactive 3 (software 7.20.0.0).

I have post my experience in that post:

https://forums.garmin.com/sports-fitness/healthandwellness/f/vivoactive-3-3-music/204400/poor-gps-accuracy-and-reproducability

I am waiting a response from Garmin.

0 Former Member over 5 years ago

Have you tried using GPS+Galileo? It'll eat more battery but Galileo as a system is improving on their accuracy. In dual-band devices they are aiming for centimeter accuracy.

0 Former Member over 5 years ago

Everyone probably understands this, but: The accuracy of GPS depends on the number of satellites the device can receive. The minimum is 3, but typically many more are needed to get good accuracy. The GPS signal is at abut 1.5GHz, which is attenuated by just about anything, including your body (also buildings, foliage, and precipitation). When your device can't see a good portion of the sky for any reason, your accuracy reduces substantially. Thus, running beside a large building or under trees is an issue. Also, the fact that your device is on your wrist is an issue. Having sweaty clothes covering your device could be an issue. (But, I think relatively thin layers of dry clothing covering the device is not an issue.)

I am actually surprised that no-one sells a GPS receiver that is incorporated into a hat. It could transmits to your watch via Bluetooth. It might substantially improve the accuracy! Making the antenna matched to the 1/2 wavelength (~10cm) would also offer a substantial improvement.

Now, consider what that would look like! Doesn't everyone want to run with a hat that has a 10cm antenna on it?

0 hhasert over 5 years ago in reply to Former Member

0 Former Member over 5 years ago in reply to hhasert

The internet sells everything

https://www.agsgis.com/Anatum-GPS-Antenna-Hat_p_12.html