Showing as awake between sleep stage transitions
planned (soon)
Randell
Merged in a post:
Weird Awake Graphs in Bevel After Switching to Helio Strap
S
Simen Eliassen
I recently stopped wearing my Apple Watch at night and switched to the Helio Strap. Sleep is now coming into Apple Health via the Zepp app.
Total sleep duration and sleep stages (REM, Core/Light, Deep, Awake) look identical across Zepp, Apple Health, and Bevel.
But Bevel shows a super fragmented “awake” graph with tons of tiny wake spikes. This didn’t happen when I slept with my Apple Watch.
S
Simen Eliassen
Thank you for the prompt reply, Randell!
I only see three awake windows for the example I provided. It matches the sleep stage graph in Apple Health.
Randell
Simen Eliassen: I see. yea, we have an existing ticket for this issue. Let me merge this to that ticket, while we work on a fix for this.
S
Simen Eliassen
Randell, thank you! I appreciate how quick you responded to my report. Keep up the good work!
Randell
Hi Simen Eliassen, Thanks for reporting this. We might already be aware of this issue and are currently investigating it.
To help us confirm, could you check if these inaccurate "awake" stages are the ones logged to Apple Health with a 0–1 second interval?
To check this, go to Apple Health -> Sleep, scroll down to the bottom, tap "Show all data" -> tap on the sleep awake stages to check the timestamps.
A
Austen
I think the cause of this bug might be that some apps write data to Apple Health in a way that separates the boundary between adjacent sleep stages — for example, 00:00–1:00 REM, 1:01–2:00 deep sleep. Bevel then seems to interpret that one-minute gap as “awake.” This should be quite easy to reproduce and fix, so I’m a bit surprised it hasn’t been identified yet. Or have you already looked into this and confirmed the issue lies elsewhere? It just doesn’t quite add up to me.
Randell
Austen: Thanks for reporting this. I’d like to confirm if this is the same issue mentioned on this ticket.
Could you please check your Apple Health data for the “Awake” stages shown in Bevel and see if they’re actually 0–1 second intervals or gaps in the data? It’s possible that Bevel is mistakenly interpreting these as “Awake” stages, a known bug that we’re currently aware of and working to fix.
You should be able to find it by going to Apple Health -> Sleep -> Scroll down to the bottom, tap "Show all data" check the timestamps in there. (Please refer to my screenshot as an example).
I want to make sure that this is the same issue first.
A
Austen
Randell I've already confirmed it.
A
Austen
Same problem using Zepp.
Randell
Merged in a post:
False micro-awakenings in imported sleep data
Daniel
Hello,
I noticed a recurring issue when syncing sleep data from Pillow. The Bevel app consistently displays multiple short wake periods (“micro-awakenings”) throughout the night that were not actually recorded by Pillow.
As you can see in the attached comparison (Pillow vs. Bevel), Pillow reports minimal wake time, while Bevel shows many short wake episodes that didn’t occur.
Please review the sleep data import or analysis algorithm to ensure accurate phase mapping. Let me know if you need further screenshots or logs.
Thank you!
Randell
Hi Daniel, This is an issue that we are aware of, We are still working out on how filter these small/micro you are seeing. Let me merge this to an existing ticket for you.
Randell
Merged in a post:
Filter out short sleep wakeups as movement
Aaron Oneal
I use an Oura ring instead of my Apple Watch to sleep and it records movement as a 0-2 minute wake up. Bevel takes these sleep data entries literally and reports every night that I’ve had terrible sleep with a dozen wake ups when it’s really just moving my arm or rolling over. I would really like to use Bevel but I can’t without a way to filter out these movement entries or set a threshold for what Bevel considers a wake up.
Randell
Hi Aaron Oneal, Can you please send a screenshot of your Sleep chart in Bevel, Apple Health, and Oura? I'd like to check and compare if it shows the same chart in Apple Health as Bevel does.
Aaron Oneal
Hi Randell, it's hard to discern looking at one day, but the trend I notice is that Oura appears to record "movement" as a < 1 minute wakeup and they don't count it against sleep efficiency scores whereas Bevel appears to. I don't expect both apps to interpret exactly the same, but these are drastically different assessments of efficiency which seems due to the extra movement noise or micro-wakeups recorded by the ring. Since Apple Health doesn't distinguish between movement and full wakeups, I understand this might be a bit of a hack on Oura's part to insert the records. But if there is no industry standard for how these are handled, having some kind of filtering control to dampen the extra noise and bring these closer into alignment would make the Bevel experience more accurate when using Oura hardware from what I can tell.
Aaron Oneal
Randell, any thoughts on filtering out these movement spikes recorded by Oura? It really throws off the results in Bevel.
Randell
Aaron Oneal: Hi Aaron, we might have a similar ticket that is already planned to fix this. However, the sleep data source on that issue is Pillow, I can merge it to that so that the team can take a look at this issue as well for the Oura users once the team works on the ticket.
Grey
marked this post as
planned (soon)
Load More
→