Suunto vertical ascent/ descent totally incorrect

[freeheeler]

@maszop
we’ve had different issues with the S9B that I am really really happy Suunto got rid of them

[maszop]

@freeheeler I understand progress and new features, but breaking things that work well is unacceptable.

[isazi]

@maszop the Baro was not working well at all, in fact it was much worse than new watches, being not only susceptible to wind but also rain and water. And the Ambit had the hole on the bottom of the watch, getting crazy if you did sweat.

Edit: not to minimize the issue, I had it today for example, 35 meters of ascent when it should have been 0, because of code yellow winds

[maszop]

@isazi I used 9 Baro for a long time and then 9PP and Vertical. The same routes, similar weather, many times the same crossings. My results are clear. 9 Baro calculated the total elevation gain correctly in most cases. Newer ones almost always add at least some 500-1000m for routes of 15-20km distance and 1500-2500m of elevation gain.

Edit:
One from many examples:
9Baro, bad weather - 18.06km, 1860m total ascent;
9Baro, good weather (slightly shorter version) - 1740m total ascent (Garmin Instinct, the same hike - 1708m ascent);
9PP, good weather - 18.19 km, 1763m total ascent;
Vertical bad weather - 17.99km, 2302m total ascent (9Baro with the person I hiked with - the same bad weather - 1840m ascent).

[Brad_Olwin]

@maszop said in Suunto vertical ascent/ descent totally incorrect:

@isazi I used 9 Baro for a long time and then 9PP and Vertical. The same routes, similar weather, many times the same crossings. My results are clear. 9 Baro calculated the total elevation gain correctly in most cases. Newer ones almost always add at least some 500-1000m for routes of 15-20km distance and 1500-2500m of elevation gain.

This is your use case, not a general phenomenon so I disagree. It is very windy where I live and very, very rare to have ascent/descent errors.
Have you tried cleaning the sensor area? Have you performed soft or hard reset? My experience is with2 different S9PP, 2 Race watches and 2 Vertical watches.

[pilleus]

@Brad_Olwin said in Suunto vertical ascent/ descent totally incorrect:

so I disagree

+1

I never had problems with a Suunto watch and wrong ascent/descent data. Okay, the Spartans record a little less than the actual watches, but only if there are long flat passages with only a little up and down.

[maszop]

@Brad_Olwin Do you hike in the mountains with many steep climbs, without trails, i.e. relatively slowly? You’ve previously written about skiing and cross-country running, and these are completely different activities that may not have these problems. The problem is a lot of erroneous altitude measurements that are too fast - pulsating altitude readings that do not correspond to the ascents or descent.

These sample routes given earlier are approximately 18 km, 1800 m ascents and approximately 8-9 hours long.
With these types of routes, the error is usually 500-1000 m more than it should be.

[freeheeler]

@maszop
I know what you mean and I saw this kind of measurements when road cycling. it is the activity where you point the sensor directly into the wind.
it doesn’t happen if you slightly cover the sensor holes.

I have put my watches in many different situations and cycling headwind is the worst. actually it is the only bad situation in my experience. there’s a workaround for it to put the watch on the handlebar.
so in my opinion it is the best compromise across all kind of sports.

[maszop]

I may be wrong, but I think it could be easily solved using the same principle as calculating distance based on GPS data.
To simplify, the position is recorded using GNSS every second, which gives 32,400 measurements for 9 hours. If all these points were connected with a line, the same absurd results would be obtained. This does not happen because before calculating the distance, the route is simplified and fewer points are used.
By the same principle, it would be possible to remove such noise from redundant measurements. It seems that these measurements are still too many for “slower” sports.

[pilleus]

@freeheeler said in Suunto vertical ascent/ descent totally incorrect:

it is the activity where you point the sensor directly into the wind.

With the Vertical it is impossible to have the sensors in the wind when mounted on a handle bar, because the openings of the sensor point towards the driver and not forward (below the Sunnto logo).

Or are these not the openings for the sensor?

[freeheeler]

@pilleus
absolutely correct.
I usually wear my watch on the wrist, even for cycling and mtb. I do put watches on the handle bar for testing purposes. depending on the speeds, position of my hand, headwinds, gloves, jacket sleeves etc. there can be differences in total ascent and they are visible on the graph similar as in @maszop ‘s graph.

[pilleus]

@freeheeler

I never had issues with a Suunto or other watch while cycling. Neither on the handle bar nor on the wrist. And I am on the bike every day and often have 500 to 1000 meters of ascent.

An easy way to check it is in Strava the auto correction or a gpx viewer.

I’m glad to have no problems with ascent/descent.

[freeheeler]

@pilleus
do you have a bike with dropper bars?

[pilleus]

@freeheeler

No, why should I torture myself? If I want to go fast, I ride a velomobile or a fast recumbent bike …

[freeheeler]

@pilleus
you made me notice that I wasn’t precise enough. on the roadbike the hand position is often different to any other bike handlebar and hence the position of the sensor holes…

[pilleus]

@freeheeler

Yes, but I think it doesn’t matter. On my recumbent bike and trike I often let my arm hang down and the sensor holes are directly in the wind. No negative impact so far for the reading of the barometric pressure.

[pilleus]

Strong winds can affect the air pressure around the sensor, which can result in an incorrect barometric altitude reading. Typically, an increase in wind speed results in lower air pressure, while a decrease in wind speed can result in higher air pressure.

If I have the severe weather alarm switched on in my Suunto watches, then warning messages should constantly come during the changes if the wind so often affects the barometric altitude measurement.

Has anyone noticed this yet?

[freeheeler]

@pilleus
yes and no.
yes, I have received several storm alarms since I have my Vertical.
no, it never happened to me due to winds, because it needs a certain pressure drop over a specific time

[Swaddy61]

@freeheeler Ditto - Several storm alarms with SV when there have been storms coming, and none at any other time.

[maszop]

@pilleus In my case, i.e. for activities such as hiking, mountaineering, trekking, what you write about usually does not matter that much (directly) because the current altitude is more or less correct (we are talking about mountainous terrain, so some inaccuracy does not matter), I can can accept them even in strong winds.

The problem is that there are too many of these results, all with some small error because this is how the barometric altimeter works, which add up, giving a completely incorrect total result.

With little wind, these deviations are not large enough to significantly affect the overall results. In stronger winds it is very visible - sometimes over 1000 m additional climbs.

Instead of fighting the wind, which is impossible anyway, it is better to simply introduce a filter that would remove the noise of too many results.

To start with, the simplest filter to remove some unnecessary altimeter’s records, and then over time it can be improved by introducing more sophisticated algorithms.
This could be done as an additional option to be turned on during or after activity (because it would certainly reduce the accuracy of the measurements somewhat), when it is clearly visible to the naked eye that the elevation sums are completely wrong.