GPS normal?

MiniForklift

@_marcus_ said in GPS normal?:

@theguyfromthesummit but that‘s the question. Elevation most of the time is the same than with my Apex Pro during trailrunning. But distance is not. S9P always records more km. Today Apex Pro 11,39km/504m+ vs S9P 12,14km/496m+. Of course I don‘t know which is correct. But seeing the track makes me think that S9P overestimates the distance a bit. Ok, in the end you get a better pace with S9P

My training partner has a Garmin Enduro and I use a S9P. On our hill runs of 20km(ish) and 500m or so of vert I usually finish up to 400m ahead of him whereas he gets a significant chunk of ascent…

I’m happy to read that it looks like the S9 ascent ‘issue’ is being addressed in an upcoming update; over the course of a week I reckon I must lose hundreds of metres of vert ?!

@miniforklift said in GPS normal?:

@_marcus_ said in GPS normal?:

@theguyfromthesummit but that‘s the question. Elevation most of the time is the same than with my Apex Pro during trailrunning. But distance is not. S9P always records more km. Today Apex Pro 11,39km/504m+ vs S9P 12,14km/496m+. Of course I don‘t know which is correct. But seeing the track makes me think that S9P overestimates the distance a bit. Ok, in the end you get a better pace with S9P

My training partner has a Garmin Enduro and I use a S9P. On our hill runs of 20km(ish) and 500m or so of vert I usually finish up to 400m ahead of him whereas he gets a significant chunk of ascent…

I’m happy to read that it looks like the S9 ascent ‘issue’ is being addressed in an upcoming update; over the course of a week I reckon I must lose hundreds of metres of vert ?!

And which one is closer to real value of ascent?
On my experience some Garmins if they have the continuous altimeter calibration on produce higher ascent numbers.

@andré-faria on my skimo tours we compared Fenix 6x Pro, S9B and COROS Vertix. S9B elevation was 5-10% lower than the others. But it had more kilometers.
On trailrunning elevation is close together between S9P and Apex Pro, but S9P records more kilometers and its GPX Track has more wobbling eben when I go straight up. Or with other words: Apex Pro track is smoother and more realistic.
I am curious about elevation in winter with the S9P

TheGuyFromTheSummit

@theguyfromthesummit
Here the GPS track
I followed the paths and pressed pause at the summit, so all this jumping around was during the way up or down…

isazi

@andré-faria said in GPS normal?:

And which one is closer to real value of ascent?
On my experience some Garmins if they have the continuous altimeter calibration on produce higher ascent numbers.

Always difficult to judge, it depends on terrain and the definition of ascent (that may seem straightforward, but it is not).

The S9 baro and peak have issues (currently) with hilly terrain, while they perform very well in continuous uphill/downhill. The ascent/descent algorithm is being worked on, and current testing shows very good results, without making it too sensitive.

Truth is often in the middle, e.g. running on very flat terrain in Amsterdam Suunto often gives me 0 ascent/descent, while Garmin gives me tens of meters of up and down that do not exist.

isazi

@theguyfromthesummit that looks like a pretty terrible track, and it also looks like there was no GNSS signal at all in multiple points. In my experience the S9 baro works very well in the mountains, so this is very strange. Are you on trails or climbing? Are there steep gorges around? It should really not look like that.

TheGuyFromTheSummit

@isazi
The first part of the ascent looked like photo 1.

The last view meters were steeper, but actually only one side was obstructed by the peak (a bit of scrambling)
The final meters wer exposed (photo 2)
My phone had always a nice GPS connection.

DMytro

@theguyfromthesummit sorry of off-topic, but which summit is that? Looks great!

isazi

@theguyfromthesummit it looks like I expected, the track in those conditions should not be this bad, and should not lose satellite fix

You said your watch was working better in the past. Anything changed? Do you think a hard reset (reinstalling the firmware with a clean sweep) could help?

TheGuyFromTheSummit

@isazi
It all started with the 2.16.26 update in June/ July. The old software worked fine and I didn’t change anything.
I give the hard reset a try and reinstall the firmware.

Thanks!

TheGuyFromTheSummit

@dmytro
No worries:

Path up to kleine Ochsenwand.
In picture: Steingrubenkogel and Marchreisenspitze

Tirol - Kalchkögel

Djordje S

@theguyfromthesummit I completely agree. With latest update Ive noticed too that S9 was more reliable and accurate with previous firmware.

Jamie BG

@łukasz-szmigiel
Absolutely agree with Lukasz on this.
Haven noticed this on the S7 and on the F6xPS. I noticed sometimes my distances would be slightly out and when I check my post activity maps I noticed that I the track didn’t always have me starting at the same place. So the next time I ran I waited for the affirmation of good GPS fix, and then I went into the maps. 80% of the time your icon is on your position, but the other 20% of the time it wasn’t (was within 100m), and giving that extra bit of time 30s-60s was always enough time for it the position fix to be on your exact spot.

Łukasz Szmigiel · S9PP 2.39.20

@jamie-bg I’ve seen this issue of “wait a little bit longer” on the forum that I really think a new way of informing users of GPS fix quality should be developed. Either change the threshold when the watch responds with green arrow or make an intermediate phase, like “no fix” - “ok fix” - “good fix”. You can start the workout with “ok fix” and it’ll most probably get better during first 15 minutes of workout but you can wait a little longer for it to stabilize and be good from the start.

Djordje S

@łukasz-szmigiel Per my opinion there is a space for improvement in S9 line firmware/features in terms of stability/accuracy. For now I would be satisifed if exisitng one could be “stabilsed”

Jamie BG

@łukasz-szmigiel - problem with that is people then complain that it takes too long to get a fix…In most cases on average most people don’t overly get worried with a 100m or so difference.

Can’t win either way - either complain too long - or complain not accurate enough…

Łukasz Szmigiel · S9PP 2.39.20

@jamie-bg you can. Redesign slightly the GPS icon. Make it filled as it is right now for “ok fix” and add a minuscule detail for “good fix”.

Those who don’t care won’t notice and it’ll be backwards compatible but those who see the difference and for whom it matters will know what to look for.

But it’ll never get implemented as it’ll take too long to design / develop / test while bringing too little of an improvement and we all know that

Dimitrios Kanellopoulos

@łukasz-szmigiel yup. While I am with you I agree to what you said.

DMytro

@dimitrios-kanellopoulos that’s too bad
Would’ve loved such icon.

Djordje S

@dmytro It could be done with little effort.Simple colors on existing icon.Red-bad,amber-fair,green-good.