If the tracking was not terminated, then we need to see the logs (Left ≡ menu → (?) Support → Report). This cannot be solved over a screenshot of the graph preview. Is the full graph showing the alarm icon? Did you dismiss the alarm, or snooze it?
Warning: I am on vacation right now, my next working day is Monday 17th. I am not sure if Petr will have enough time to reply to all emails. So please, be patient.
Yes I dismiss the alarm
It’s ok enjoy your vacation. We’ll talk then 
Ideally send the logs now, because there is only a limited history available in the logs (with tracking data, it has only one or two days of the history). I will get back to the logs when I can. Thanks
BETA version jas the last rerelease 13th March, the built number is 230368, the version 20240220. The new permissions messed up the backup feature, and it had to be fixed ASAP.
The Release notes have been updated, so you still have all changes listed.
The BETA channel is constantly changing, and some updates are re-released (so we can get the fix faster to the public).
If this makes you worry, you can opt-out from the BETA and keep the standard release, which is stable and updates only once per month, and re-releases are really rare.
Just don’t know when is beta or not 
BETA versions have a β symbol on the icon.
Understood. How to know difference between rem and light sleep? Both have same color ( light blue )
Light sleep reaches 50%, while REM reaches 75% (deep sleep 25%, awakes 100%).
Ehh yeah im left more confused.
If I could just know how much time I was on each stage would help me. Like other apps register them.
You see awake duration in the Efficiency chart, and deep sleep in the “deep sleep” chart (duration and ratio).
Without EEG data, you cannot tell the brain’s activity. REM phase cannot be directly derived from the movement activity.
REM is estimated by the highly advanced algorithms trained on the data sets from sleep labs. If you are interested, we explained this here.
But this estimation works on a different basis, than light vs deep sleep.
That is why we always show, that REM is estimation (“~REM” on the graph). We do not want to deceive our users, that all the sleep phases are measured the same way.
That is why we do not show the REM numbers as other apps boldly do.
Well other apps are not saying that’s a set stone Rem score. Nor is deep sleep, light sleep, any kind of sleep data produced by any app, any watch, for more advanced it is, it will always be an estimate not a fact. The only way to get an 100% results/score is by a proper polysomnography ( PSG ). DON’T take me wrong but by you saying if you put Rem score it might deceive user going by that assessment then you are indeed deceiving users by giving us light/deep sleep HR stats and scores because of what I said prior not 1 of those are real score but a guesstimate. But if you don’t want to give us a guesstimate Rem score it’s fine, easy to get on other apps, but don’t put that bold statement because it holds no water whatsoever 
Btw your link doesn’t work.
I have just tested the link and it opened the page I wanted to share. What page does it show to you? It worked for me on all the internet browsers I tried (Edge, Firefox, Chrome, Safari, DuckDuckGo, Brave).
Our message is, that the actigraphy method is great for light and deep sleep. REM sleep is estimated by a different method - which may have some limitations.
Other apps show the data the same way, so the users might get the idea, that they are measured the same way.
Or did you immediately recognize that REM is not measured the same way when looking at the data in other apps?
We are open about this - REM is not estimated the same way, and we show this in the app. I am confused why this is called “bold statement”.
If you think putting a guestimate on Rem sleep is deceiving users, then by all means take out light sleep and deep sleep because no matter how tight your algorithm or the rings or watch or whatever you can have that measures those 2 aspect is going to be as far from the truth as earth is from Mars. The only test that will give a non-biased score is a PSG nothing else.
On the same note, if Rem is not going to provide any score or ratio, why on earth put it in the app? For good looks? I mean c’mon now.
But if it makes you guys feel good about having a feature that is rather useless by all means carry on, is your stuff at the end of the day.
REM is not measured the same way - that is true. In our app, and all sleep-tracking apps in general.
We do not show REM exact numbers, as other apps do - that is also true.
And you are absolutely right. The only method to provide 100% accurate data about your sleep is the PSG and nothing else.
The actigraphy method we use is the only method, that can take data from phone/wearable/ring sensors, and estimate sleep phases from those data. This method has been proven to be accurate. It is used by all sleep tracking apps (except Sleep Cycle).
If this level of accuracy is not enough for you, and you consider the data provided by the actigraphy method as a guesstimate, you will need to get the PSG machine to get more accurate data.
It is terribly expensive, and not comfortable to use at all (with all those sensors attached and all those wires), but it is the only 100% method.
We have REM in the app, because the algorithms can estimate the REM phase with accuracy high enough to match the rest of the data. But that does not mean the REM phase was estimated the same way.
Answer this, and be completely honest about this: when you were looking at the data in other apps, did you think “Oh yes, there might be something different about deep sleep and light+REM data”?
No, I never got to that conclusion not even close BUT the guestimate was at least a bit of a good data to have and then base it ( along all the data ) to see how it correlated to my awakenings ( too tired, exhausted, fine, pumped, MOOD, cognitive abilities etc )
No I don’t need a PSG first it’s illogical to think using one having to attach all head cables and chest stomach, etc.
Also you keep pointing out you tested the app in a lab ( I guess besides a PSG test ) I’m not here to call names but I HIGHLY doubt that’s 100% factual that the app performed almost identical to a PSG.
You don’t need to prove anything to me. There’s nothing that’ll change my mind and believe an app can act almost identical to a PSG. That will be the day people will stop fighting insurance to get approval for a PSG study and just rely on a ring or watch and an app to measure their sleep, heck even people with sleep apnea will just get the app record their sleep and say, here I need a CPAP machine.
P.S I have HIGH central sleep apnea at 47AHI. Guess how many events the app has told me I’ve had when not on my CPAP? 0. So even the spo2 is flawed, it’s less than a guesstimate.
That’s why I’ve pestered you a lot about the O2 not showing in my graph etc, I never told you about my apnea before but had to now to show you it all is a guesstimate at BEST. Even light sleep, deep sleep ( no way on earth I can have all the deep sleep it shows having all my apnea events. That means for every hour I would get disturbed or awake 5 or 6 times. You do the math in 8 hours and I’ll show you an 8 hour graph so you can see it nowhere close registers an “estimate” score.
So that lab test you said you tested the app at… 
The sleep lab tested, how accurate our results are when compared to PSG results.
If the current awake setup is not sensitive enough in your environment, you can increase the sensitivity in the settings.
Breath disturbances rely on the data provided by the SpO2 sensor. Could you please share the graph from Sleep, where you have 47 dips, but the app did not show any RDI? If it happened recently, also send the logs to our support.
I said my AHI is 47 when not using a CPAP. Can be a couple points lower but not much. Can’t show you no graph since the app says I have 0 dips which is impossible to begin with. Again you mentioned the lab PSG being equal or close to the app. Just by it not recording a single dip on my sleep I can certainly say that comparison is completely fraught and or biased.
Not much else I can add. No nothing in my room is waking me up that many times, sounds and noises if there were any which again there are not, can make you wake up gasping for air.
Anyway I’ll use the app for what is worth. But can rely much on it when it won’t even record dips.
Even though is not close to the number of dips, it recorded 9x more than SAA and 02 score resembles better the issue. That’s all I have. You take what can help you
Is this the only source of SpO2 data for Sleep?
If yes, then 11 drops per 11 hours and 34 minutes tracking duration, should translate as 11/~11.5.
So approximately 1 dip per hour, RDI = 1.
I meant the graph, that shows the dips on the SpO2 graph section, but it does not show RDI value for the same graph in the pie chart section.
