Altimeter recordings while running

[Tadas Linge]

@Brad_Olwin hello,
so i have completed these 3 trial runs after hard reset. each 20 km, almost the same route. first run and the third one was same direction and second one opposite direction. unfortunately with each run altimeter readings were getting worse. and weather condition with each run was getting better. so i am totally confused here and i think i will give them to check the watch. all data in attached PDF: Suunto altitude after hard reset.pdf

[Mff73]

@Tadas-Linge
Looking at your pdf. Are you saying that altitude difference about 10m is something wrong? .

[Mauerwegler]

@Łukasz-Szmigiel said in Altimeter recordings while running:

All Suunto watches from S-series that have barometer have FusedAlti (S9B onwards).

SSU too:

[Tadas Linge]

@Mff73 hello, not exactly. What i see, that watch fresh after hard reset is showing really good results, but then whith each use it is getting worse.

[Mff73]

@Tadas-Linge
Worse? average dif from 2m to 7m? .

[Tadas Linge]

@Mff73 hello. Yes, so this is just average, but you also need to see point by point, because in the first run there is no deviation above 10, then in second one already one such deviations and in third one there are qoote few 12s and 13s… Dont you think it is tendency? Of course i will continue fill this table, but this is what i see now.

[Iceman1]

@Tadas-Linge I have tried a couple of times to set the elevation far off from the real elevation. Then I let fusedalti correct everything. When I have one this I have a nice elevation graph in the end, starting and ending within1 meter or so.

When I dont adjust, and the elevation from the start is within 10-15 meters off from the real elevation, then that difference is kept, and is not touched by fusedalti.
I understand 10-15 m is not much, but it is very noticable on flat runs.

[isazi]

@Iceman1 because the GPS vertical error is large-ish, calibration is only performed if deviation is substantial. If the deviation is within GPS error, the watch cannot know which one between altimeter and GPS is correct (tie-breaking with only 2 sensors is kinda impossible).

[Iceman1]

@isazi yes I have understood this is the way it works. I just posted a possible way to “trick” the watch, If you like.
Instead of graphs like this starting at -7 m and ending at 5m

I get graph like this starting at 7 m and ending at 8m

[Redbeard_ranger]

@Iceman1 Very interesting trick, How much do you off set the altitude by incorrectly? like 1000m?

[Mff73]

@Redbeard_ranger
with 1000m dif, you will for sure see the job of fusedalti

@Iceman1
second graph looks better, but it is also another day, so, maybe, this second day could have been better, even without the altitude offset trick.
But it opens the door to check “longer” and see. If with big initial offset, fusedalti is able to catch a “better” altitude fix than just letting it alone, it would be interesting to know. Probably this situation is already tested, but i will personnaly try

[Iceman1]

@Mff73 You are correct that its not really possible to compare different days. It would be interesting If somebody with two watches could try to test them parallell

[Iceman1]

@Redbeard_ranger I have just set it around 50-60 m off.

[Tadas Linge]

@Iceman1 hello, actually this is what i was thinking to do but never done… we are out with family for long weekend so i have done couple of runs in different location. first thing what i have done i set up my actual elevation…:D and one day was beautiful day so readings of altimeter was really good. next day after the run i had +22meters at the finish which was the same point for start. so i have checked barometer and it had huge drop, so it kinda explains, but still i expect that watch evaluates such things during GPS activity. So my wife was going for the run same day, i have checked her Forerunner 255 and her altimeter was off +40 meters before run, because she does not care about it and never adjust. when she came back and i have checked her altitude graph and deviation was 2 meters for the start and finish point… this is what i was getting when were using Fenix 7x, never had to think about calibration or anything, watch was doing everything during GPS activity. with basically no deviation, therefore now it is getting more and more annoying…

[Tadas Linge]

@Iceman1 Hello, today i have tried this trick. instead of 45m i have manually entered 160meters, but watch did not corrected anything:

[Iceman1]

@Tadas-Linge So your real elevation at at start and finish is 45 meter? Thats not good at all!

[Dimitrios Kanellopoulos]

@Tadas-Linge is your starting elevation 45?

[Dimitrios Kanellopoulos]

@Tadas-Linge and the start to end diff is like 5-10m if you are bothered by this idk what to say

[Iceman1]

@Dimitrios-Kanellopoulos but 120 meters off is ok? I wouldnt be happy with that…

[Tadas Linge]

@Iceman1 yes