Suunto Spartan Sport GPS drift

drBubo

Hi,

I’m new to this forum and I’m fairly new to Suunto (and other) GPS watches. I can see here other posts related to the GPS accuracy of these devices, so I wonder whether you (with a lot more experience with these gadgets) have seen similar large drifts of your GPS tracks I show you below, and whether you have any explanation on that.

Please find the red arrow around the middle of the screenshot. To the right, there is a real loop. To the left, however, it just looks as a loop, in reality it should be the same track both ways. You see that a deviation between the tracks corresponding to the different directions with a maximum of 100m (shown by the yellow ruler) slowly builds up and then slowly disappears. My problems are that (1) a location error of 100m is huge, way larger than the expected accuracy of any GPS device, (2) I have no idea what factors can cause such drifts slowly building up and then disappearing.

The track was recorded in Hungary, during a run in a forest with partial or sometimes full tree cover last autumn. The device was a Suunto Spartan Sport, purchased not much earlier, with “Best” GPS accuracy, synchronised with my phone before start. The watch was on my right hand, not covered by clothes. Glonass was enabled. Since then, I disabled Glonass and haven’t seen such huge drifts, so the problem is probably solved. But I’m still interested whether you have seen GPS drifts at this level, and, whether you have any idea how to explain it.

To make things more complicated, I have already experienced similar (or even larger) drifts with my previous handheld GPS device, a Bad Elf GPS Pro 2200, which has no Glonass support.

Markus Gietzen

@drBubo I’ve had seen this a lot with the Spartan series and GLONASS enabled. My Garmin watches also showed similar (not to that extrem extent though) effects when GLONASS was enabled. Therefore I always keep GLONASS off. I never experienced any benefit of using it but I’m not running in big cities but only on trails.

Dimitrios Kanellopoulos

I had the same on Spartans with Glonass

jthomi

Same here, and I disabled Glonass, after that no more drifts.

A Former User

@Dimitrios-Kanellopoulos what di you recommend for Greece regarding glonass?

Dimitrios Kanellopoulos

@Yannis-Belouris that is a very good question and also depends on the Model.
For S9 I have not seen drifts but have seen straight lines.

At this point for the S9 I cannot commend as I am running a new FW that should be released, and thus I might say something stupid.

However, my experience in Greece + Spartans was good with Glonass. It improved trails quite a lot. Without Glonass I would get a bit more inconsistent tracks.
But to be honest I have not tested it a lot in Greece and I have also seen the offsets there.

IMO regarding Spartans it’s a bit of risk. It can improve the track/pace but it can add an offset.

Also to add that these systems get updates from time to time, so some old behaviour can be invalidated.

I have a rule of thumb:
If the conditions are hard eg Zagori marathon -> Glonass
If the conditions are easy to hard -> Athens marathon -> GPS only

cosme.costa

@Dimitrios-Kanellopoulos said in Suunto Spartan Sport GPS drift:

At this point for the S9 I cannot commend as I am running a new FW that should be released, and thus I might say something stupid.

OFF TOPIC: New software?? , for the Spartans too??

Returning to the topic, with my previous Ambit 3 Vertical with GLONASS ON the results had a big offset the first run after the activation or when the satellites weren’t updated the same day, regarding accuracy I didn’t see any improvement (at least in my area). In my SSU I haven’t bothered to activate GLONASS.

Dimitrios Kanellopoulos

@cosmecosta SGEE or AGPS update and soaking before starting an activity is a must for good tracks imo.

@cosmecosta all our watches do get the same updates

cosme.costa

@Dimitrios-Kanellopoulos said in Suunto Spartan Sport GPS drift:

@cosmecosta SGEE or AGPS update and soaking before starting an activity is a must for good tracks imo.

@cosmecosta all our watches do get the same updates

Soaking the watch? With water? What’s the reason behind? I didn’t knew…

Updates = Good!!

Dimitrios Kanellopoulos

@cosmecosta Like waiting a bit even after you get the Green arrow that shows that the position is known. Example:

Watch found the position (green arrow) you press start -> enter the forest -> watch loses position due to inadequate satellite info.

Markus Gietzen

Yes, I can only second that. IMHO current watches (regardless of the vendor) have a tendency to give you GO a bit too early. Currently I’m using my Fenix again rather than my S9 but what I do is: I select a data-page with a pace field: you will be suprised that you see some “speed” displayed altough you are not moving. So, I just keep standing at the start point until it stays solid at “0:00” or “–:–”. It doesn’t take that long but it’s actually remarkable later than the “GO” from the watch. But no panic: it doesn’t take as long as it took in the old times with let’s say a Garmin 305.

For me that’s important because I live close to the forest and thus I immediality enter wooded trails and don’t leave them for longer time. So it’s hard for the watch to catch up if I started too early.

Dimitrios Kanellopoulos

@Markus-Gietzen yup exactly.
And that GO is when x amount of satellites are found. That number can drop if entering a forest toooooo soon. Eg from 6 to 3 that is way critical.

drBubo

Hi,

Thanks for all the replies. There seem to be a wide consensus that it is better to turn off GLONASS. This is a practical solution. On the other hand, I do not understand, why it works this way. Both the good old Navsat GPS and the GLONASS system are capable of providing location information with the precision of about 10m, or better. Independently of each other. How is it possible, that the combined use of the two systems results in systematic errors in the range of 100m?

May I ask other (probably too technical) questions related to the GPS accuracy?

(1) Can Suunto watches use the GPS correction signals (WAAS in the USA, EGNOS in Europe) for better accuracy? To receive them, you might need more initialisation time, and better reception than what you need for a simple GPS lock. If they can, is there any way to know when during a workout these signals are/were used?

(2) What is the optimal position of the watch for best GPS signal reception? In other words, where is the GPS antenna in these watches, e.g., in my Spartan Sport? I know that in the case of the Ambit, the antenna is in the bump on the lower part of the watch. This allows for very good signal reception, but makes a clear preference to wear it on the left hand. Is there a similar asymmetry with the other models?

(3) What kind of data is sent to the watch when syncing with the mobile app during the “Optimizing GPS performance” step? Just the satellite positions, helping in a faster GPS lock, or does it contain some correction data as well? In other words, if I haven’t synchronised my watch for, say, a week, but I allow plenty of time for collecting GPS signals, will my position be just as accurate as after a proper sync with the app?

Dimitrios Kanellopoulos

1-> I dont know
2-> The antenna is circular, I am quite sure it’s under the bezel. For improving accuracy running or having the “screen” of the watch looking at the sky can help.
3-> AFAIK its sending SGEE https://electronics.stackexchange.com/questions/130256/sgee-vs-cgee-vs-a-gps

Mff73

@Dimitrios-Kanellopoulos said in Suunto Spartan Sport GPS drift:

https://electronics.stackexchange.com/questions/130256/sgee-vs-cgee-vs-a-gps

Very interesting.