Fenix 5 Plus - My thoughts

This is not the first, and won’t be the last post of GPS accuracy, and while I have my own struggles with Garmin (reliability, being able to use the watch off-grid etc.) I think there are users who demand perhaps too much: GPS watches rely on a relatively weak signal that can ‘bounce off’ or get distorted by structures. Houses, buildings, roads etc. Humidity and climate, for example rain or snow, also have an impact. Finally there is the small size of the watch that’s being moved constantly, and perhaps impacted by clothing. There’s a lot of potential interference that the watch has to filter out. 

Mobile phones have more data sources besides the GPS (mobile phone towers, wifi networks) that send stronger signals and are less prone to interference. Plus phones have bigger and stronger antennas and tend to move less than watches. It’s possible to shield a GPS antenna with your body, but quite unlikely to do the same for a phone. I think that’s pretty much a given, and unnecessary for Garmin to point out.

Perhaps when Garmin releases a watch with a build-in mobile chip the accuracy in urban areas will notably increase, but otherwise we’ll only see small incremental improvements in accuracy as the algorithms that filter interference and calculate position are getting better, but yes, they’re limited by physics  

Having measured the distance of the run with a superior device (phone) why insist on repeatedly doing it with an inferior device (Garmin Fenix)? If you know the distance then time and heart rate are valuable on their own, no? 

I can’t comment on the bezel. 

Thanks for your reply Leee. Obviously I did expect too much and maybe if the watch was half the price I wouldn't be so annoyed.

I like looking at my pace (average) to gauge my performance and with distances being incorrect means that the pace will also be off. 

If I was looking at times, then I can buy a cheap Casio and use the stop watch. Sure if you are going to spend £x00 on monitors and pods it may work great, but I'm not serious enough to warrant that.

I wrongly thought that a Garmin device would be more accurate than my phone

Almost no dedicated GPS device is going to be better at tracking then your phone is. The reason for this is that phones use a combination of GPS signal and cell tower triangulation to determine location, and this tends to be far more accurate than GPS alone.

GPS is of a frequency which tends to be bounce off solid objects such as solid buildings or large rocks which results in extremely skewed location data when youre running in the city or hiking through a valley with large rock walls around you. That said, Garmin could certainly do a lot better in omitting what is clearly invalid data (i.e. using the activity type and step count to determine that you haven't magically teleported half a block away within two steps when running).

It must be made out of pretty soft ali when it should be made out of something hardend like they use on some ali wheel nuts.

It's stainless on the non-sapphire version, and yes, it should have been treated.

No the phone does not navigate and get accuracy by cell phone network, in the way you think. You get your satellite information, and you know where WiFi pointds are, but that does NOTHING for accuracy. The phone is larger, might have better GPS hardware and more space for a better antenna. 

Did you try also the combination GPS + GLONASS to see if it performed better?

Assisted GPS does use extra info for accuracy, bit for first fix and initial calculations:

"Standalone GPS provides first position in approximately 30–40 seconds. A standalone GPS needs orbital information of the satellites to calculate the current position. The data rate of the satellite signal is only 50 bit/s, so downloading orbital information like ephemerides and the almanac directly from satellites typically takes a long time, and if the satellite signals are lost during the acquisition of this information, it is discarded and the standalone system has to start from scratch. In A-GPS, the network operator deploys an A-GPS server, a cache server for GPS data. These A-GPS servers download the orbital information from the satellite and store it in the database. An A-GPS-capable device can connect to these servers and download this information using mobile-network radio bearers such as GSM, CDMA, WCDMA, LTE or even using other radio bearers such as Wi-Fi. Usually the data rate of these bearers is high, hence downloading orbital information takes less time."

I can see that my pace records a drastic drop every time I turn a corner.

How wide a corner and how tight a corner are you turning, because that's pretty much what will happen. Inertia alone means you need to decelerate slightly to turn corners.

"I feel that Garmin should make people aware that their devices aren't any more accurate than a phone. There is simply no point having a Garmin if it isn't accurate in my mind"

For many others it is though. Post an activity here and we can examine it. Is it urban/rural? Is it largely straight, or many turns. Mapping a route via push pins will only give you a rough guide to distance. You  pointed out that you've got a 200m+ deviation from the phones alone. You have yourself in the mindset the phones are "right" but are they? Mapped by push pins, you will tend to mark corners as 90* turns (for example) whereas in real life we all run a circular path around corners as we swing around. Over a few K those savings would add up...

If the GPS tries to overtly smooth the data, it will run the distance short. You can often see that with comparing Garmin and Strava data.

Given GPS's inherent inaccuracy's what you are posting sounds like a combination of the phones inaccuracy's and the watches inaccuracies. Pick up and work to it, phone or watch. At least that way the uncertainties will be consistent. 

Run the same run with about six devices, and see the spread of distances over time. Uncertainties are inherent over distance with GPS. But my Fenix measured 65.97k for a 66K ultra. I'll take that level of accuracy for a consumer device with an aerial the size of a large coin.

Look on the CIQ store for datafields that do smoothed pace over time, for example https://apps.garmin.com/en-US/apps/28fdec7f-3807-4d24-b85a-4c86b40501f4

This averages the pace over a set period of time (for example 6 seconds) that helps smooth out the jumps in pace as the Garmin is calculating distance etc for a watch that is moving in quite a complex arc.

I see many of my friends with Garmin watches finishing marathons with +-20 meters than the original distance. That is extremely accurate in my opinion. I experience the same thing with my own Garmin Fenix 5.