Enough
-
@cosme.costa my first fillings about zonesense:
https://forum.suunto.com/topic/12991/zonesense-reflections -
As a remember, saying « it doesn’t work for many, based on the few people sharing bad experience, is not a clue.
Sleep tracking works perfectly fine on me, also on my wife, my friends using Suunto’s too…with SV, SRs, SR, S9pp. It’s very rare that it fails. Are my settings involved in accuracy ? I don’t know,maybe (usual sleep tracking set from 23h to 7h, auto DND also from 23h to 7h, no sp02 tracking, HRV tracking on, no notifications).
So, don’t forget that if does not work on you, even confirmed by some other users here, doesn’t mean it’s a general issue. -
@Tieutieu it works for me as well: HRV and SPO2 tracking enabled, DND enabled and disabled manually, notifications enabled generally.
One night it didn’t track in the last months. Sometimes I get an odd nap added.The tracked sleep data isn’t better or worse than my Garmin has been. About the same…
-
@2b2bff sleep tracking started working when I turned the watch to the inside of the wrist for the night. The results have been believable for weeks now. Still the results are nothing but curiosity for me. I mainly wear the watch nightly because of the wake up alarm, which is more subtle than my phone’s smog horn.
-
@cosme.costa said in Enough:
@brechtvb From what I have been reading and from my own experience, OHR, since S9PP, has improved greatly. Issues with OHR readings is not only a problem about the sensor and is more with the technology, and all the brands have them.
like OP i had a cheapo garmin (FR45) and it never had a total HR sensor failure like suunto race s does for many, many people. just yesterday on reddit: https://www.reddit.com/r/Suunto/comments/1kjn850/race_s_titanium_hr_drops/ - never mind general the general inaccuracy.
the suunto race and race s (and presumably the rest of the suunto line using the same sensors and software) specifically have bad OHR relative to other brands’ OHR. i think that should be clear if you see the amount of noise about this issue, relative to the userbase of the race.
OHR as a concept is flawed, but there is still good/bad OHR within that spectrum. suunto has bad OHR.
-
@Tieutieu As someone who has had issues with Vertical, Race and Race S tracking sleep your comments are a bit off. I have seen comments for the life cycle of the watch saying that it does not work.
I can understand as someone who has zero issue with something chiming in and telling people it is not an issue, but until it is an issue this is a common response.
If it was not an issue, why would it be on almost every update in the recent past?
Telling people what is happening to them is not real is not helpful. Including friends family, and even the neighborhood canines do not help fix a problem.
I would be curious to know what percentage of watches not working for you would qualify something as a “general issue”. I know you are trying to be helpful, but telling folks that what they are experiencing is not real is not useful and can feel like shaming. -
@Tieutieu To be clear, I love your posts and look forward to what you have to contribute, but frustrations are frustrations. This is a great place for people to find the help that they need to make their awesome watches even better.
-
@Stavrogin I completely understand your frustration, and also that my comment add maybe more to it. That’s why I shared my settings, if maybe it could also help. It’s unfortunate your sleep tracking isn’t working properly.
-
The delivered with strap was quote stretchy and the buckly actually gave me a rash (especially when sweating/exercising).
Recently i bought new strap for the suunto race s. It feels more rigid and i had the idea to do a test.
This is on a gravel bike, low speed in the end is because of di2 battery that was giving up.
Good: no cadence lockins, not bluntly not measuring
Test: Race S wrist HR <–> Suunto belt + ios suunto app
-
-
For what it’s worth, this is a follow-up to my comments 9 May 2025. After observing the same random spikes in HR measurements on a Garmin watch I thought I trusted as I saw on the Suunto Race, I anecdotally tested the Race again (very mildly).
Basically it seems like the Garmin watch I used and the Race for me provide fairly similar HR measurements. If anything, the difference appears to be smoothing. I suppose Garmin might use a different algorithm to flatten the OHR measurement provided by the operating system, at least compared to the Suunto OHR measurement output. This is a total guess based on simple observation with no motion, periodically over a few days.
I updated to the recent 2025Q1 firmware. Maybe that is a factor? Honestly I don’t recall why I thought the watch’s OHR measurement was “wrong generally 85% of the time” because it seems solid now – at least enough for me to be comfortable using it for guidance. Which is great, I really like the watch.
Note I observed the same issue with a Coros Vertix 2S. In Coros documentation I found a reference to OHR measurement accuracy in relation to cold hands, as the blood flow changes under certain conditions and this can have an impact on OHR referencing. For the record, I live in Alaska and this is totally a plausible situation for me. It was the first I read about the idea, which makes sense. Just a thought.