New SMART barometric algorithm

maszop

From software update announcement:

Initial threshold reduced to 2 m (previously 3 m)
Meter-by-meter measurement after the initial threshold
New air/water filtering (wind, rain, humidity, friction)
Elimination of false ascent and descent spikes
Improved altimeter accuracy and stability

In very strong winds, the problem still persists, but the error is much smaller. In this case, it was about 250m too high. According to the Suunto Vertical, 1185m – it should be around 900-950m.

I’m generally satisfied; previously, my ascent totals were often over 1000m higher than the actual total. But there’s still room for improvement.

It can be clearly visible at the upper part of the mountain, where the wind was definitely the strongest, that the graph is very jagged and hence the inflated results.

sky-runner

@maszop said in New SMART barometric algorithm:

Meter-by-meter measurement after the initial threshold

By the way, I find it to be foot by foot rather than meter by meter when the watch is configured in imperial units. One foot = 30.5 cm.

maszop

Today I had the opportunity to test it more thoroughly in strong winds. The problem persists, and its scale is still significant. Suunto reports 2,118 meters of ascent, instead of the actual about 1,500 meters.

I’m not convinced by the constant responses that it’s impossible to calculate accurately in windy conditions, because the Garmin Fenix 8, in identical conditions, indicated 1,600 meters, which is already acceptable accuracy.

VoiGAS

Also there should be the ‘FusedAlti’ thing that helps in such cases. But as FusedTrack is gone, maybe thats also the case with FusedAlti?

wakarimasen

Walked to the summit of Pen y Fan today. The summit is 886m and my Race read 887 at the top, without prior calibration. Not too bad

maszop

@wakarimasen This has nothing to do with measuring altitude itself — that’s never been a problem with Suunto.

Look at the top part of the altitude graph, and you’ll see a jagged line with multiple little, sharp peaks (caused by strong wind) that add up, giving absurd ascent and descent totals.
It’s a bit better since the last update, but the results are still absurd and essentially useless for any analysis.

Audaxjoe

@wakarimasen I believe he means accent. Suunto is normally spot on for the exact altitude.

As a cyclist I’ll go out with a group and the accent totals can vary greatly even if people have the exact model.