Most times I buy or receive products I’ve got a mental expectation of how that product will probably turn out. And most times, it’s correct. Sometimes a product hits it out of the park well beyond my expectations (the original Scosche Rhythm+ for example), and sometimes a product that you expect will do well craps all over itself (Wahoo ELEMNT MINI Bike Computer). But for the most part, things are often easy to predict.
So, when I ordered the newly released Beddit 3.5 made by Apple – I figured it’d be a slight upgrade to the previous Beddit device I’d bought years ago. The fact that they included the .5 in the product name was a giveaway to being a slight refresh.
But I was wrong.
Ever so wrong.
Without question, this is the most disappointing Apple product I’ve ever bought. Given I buy an excessive amount of Apple gear each year, this was actually completely unexpected for me.
And it’s not because it doesn’t do what it says. It actually sorta does that.
Rather, it’s because they took away almost all useful features from previous versions, and made the product as close to pointless as you can get. For a product that’s supposed to be all about data – they’ve neutered it and made it more gimmicky than a cereal box activity tracker. And all at twice the price of their nearest competitor.
Though, I suppose I should tell you what’s in the box. But, before I do that, if you want this whole review in a tidy video format, than look no further than the play button below:
Ok, onto the box.
What’s in the box:
I had joked on Twitter about the box for the Beddit device being void of any information. A harmless offhand joke, but turns out that was actually as honest as Apple could be about what the product does (almost nothing).
Though the bottom of the device does at least list compatibility requirements. Note that Android is no longer supported.
Nonetheless, inside the box you’ll find exactly three things:
With some descriptive text, you’ve got:
A) A very simple manual
B) The Beddit device with attached USB charging cable
C) USB wall adapter for your region
And that’s it. Here’s a quick look at the manual, which simply shows you where to place the unit:
And then a closer look at the expertly coiled cable. I wish I could coil cable so perfectly.
The sensor piece meanwhile has both a black fabric side and a white fabric side. The black side has some sort of non-stick like surface on it. This is actually an improvement over earlier Beddit versions, and is much cleaner. And that’s all there is for the unboxing piece.
The Basics:
The first thing you’ll need to do is place the unit on the bed and ensure you’ve got a power outlet or USB port nearby. If you use a USB port aside from the charger they give you, then ensure it has enough amperage to operate it. The unit comes with a 1A charging adapter, and on the inside of any wall outlet/USB port it’ll list how many amps. As long as it’s 1AMP+, you’re good to go! It’s probably best to not plug it in yet, wait till you’ve got it situated on the bed.
So go ahead and lay it out on your bed. In the case of Beddit it’ll go just under the fitted sheet (else you’ll get all tangled up in it if you placed it atop the sheet). This is different than some systems that can actually go below the mattress entirely (like Withings or EmFit).
You can see here the black side facing down. That’s the stickier of the two sides:
Being below only a singular sheet, you might notice it. It doesn’t bother me any, though, The Girl found it a bit annoying at first. That annoyance dissipated about 10 minutes later.
The unit comes with a pretty darn long USB cable, so you can route it however need be to your wall outlet or USB port. And you can pretty easily make the entire thing invisible as well, assuming your bed sheets pass a Martha Stewart styling test.
Once that’s done it’s into the phone app (Beddit 3.5, not Beddit 3.0 app) to get it added and paired up. The entire process only takes a few seconds. Note that if you’ve got the Beddit 3.0 installed as I did, it’ll constantly warn you that the world is going to end. It does this every time you open the Beddit 3.5 app until you rid your phone of the 3.0 app. Don’t do this, because once you do you can never get that data back again.
I suppose in some ways the ratings seen in the Apple App Store should have been an indication of what’s to come. But I’m an optimistic person most days, and hoped my experience would fair better.
As part of the setup you can do a quick test to see if things are working. I’d have as much fun as possible with this portion of the setup, because you’ll never again get this much data.
And then you’re ready to sleep. However, one thing you’ll need to be sure of is that the app is open in the background of your phone somewhere, and that your phone is nearby. Since this relies on Bluetooth Smart to send the data to your phone, the phone must be within range. Also, it MUST be plugged in. I actually discovered this last night when I forgot to plug my phone in, and thus, have no data for last night. This shows the weakness of the system in that it can’t cache any sleep data at all on the device.
If you want to want to specify a wake-up time, you can do that. No, it’s not an alarm, Beddit removed that functionality in this version. Instead, it merely does some simple math to figure out what time you should go to bed.
From a sleep starting standpoint there’s nothing you need to press or start (assuming the app is somewhere in the background), it just works. Go to sleep, and wake up and get data:
Note that there is the option to enable snoring detection. That’s down in the settings. The benefit to that feature is that it makes it super-easy to see whether or not the app is working in the background, as it puts up a small red audio warning/notice in the upper left corner of iOS:
However, the downside to that feature is that if you want to use your phone for something like music to fall asleep to, you can’t do that since it’s using the microphone to detect the snoring.
As for the data, we’ll go through the two tabs that have data. As you’ll see, there’s not much here. The first tab is the daily summary of your sleep that prior night. This will auto-populate in the morning after you wake up. It does NOT show any real-time stats while sleeping.
The upper portion shows you the time you went to bed, and whether or not you met your goal bedtime (which will illuminate the bed green). Next to it is how long you slept. It does not show anywhere in the app what time you woke up (easily). That’s annoying.
The middle portion shows a brief analysis of your night graphically. I actually hate this little graph because you can’t zoom in or look at any further detail, a downgrade from past Beddit versions. You also can’t see clearly exactly what time you supposedly woke up, because the numbers/data are small. Additionally, it’ll list time awake, efficiency, breathing rate, and room temperature and humidity. These are probably some of the more neat stats of the device. But you can’t look at how these changed over the course of the night.
And finally, the bottom portion shows your heart rate. It’ll list the highest and lowest, as well as the average. Further, it’ll show you a tiny little non-zoomable chart of your heart rate throughout the night. You can also place a numerical grade on your sleep. Pressing that ‘Learn more about your results’ button does nothing more than send you off to a generic support page on what I just described.
Here’s another sample day to look at:
So what about the second tab – trends? This is it:
Listed in 7/30/90 day increments it simply plots the key metrics noted above. And that’s it. None of these graphs are zoomable or exportable, and none of them can be applied to other durations. You can’t tap on them or open them to get additional detail, or compare metrics against one another.
If you have a shorter night of sleep, you’ll get one of a whopping two different suggestions as a notification popover the next morning:
– “Try taking a short walk during the day to keep your energy level up”
– “To help you stay focused today, try to get outside a few times for some fresh air”
Seriously, that’s all it ever offers you. It has the creativity of a light switch.
Note that Beddit does transmit data to Apple Health, assuming you’ve enabled it in the settings or during setup. Speaking of which, here’s the settings. It’s also where you can set your bedtime and sleep time goals.
And here’s what the data looks like in Apple Health (HealthKit):
The quality of the underlying data is actually somewhat interesting here. Beddit appears to log heart rate and respiratory rate every 5 minutes (whether it’s an average of that 5 minutes or a single data point isn’t clear). And then it appears to truncate your sleep into various chunks, that are generally directly adjacent to each other. Still, kinda an interesting approach.
Still, the key takeaway for me though are the major features removed from past versions of Beddit (3.0 and prior). Specifically:
– No more sleep score (a cumulate total of numerous metrics into a singular nightly score) – entirely removed
– No more extended details on each of the metrics (you could expand out every metric and dive into the details)
– No more explainers in the app for different metrics (you could expand out a metric and get clarification)
– No more cloud sync (you could sync all the data to a Beddit cloud platform)
– No more deep vs light sleep functionality (entirely removed)
– No more restless sleep tracking
– No more Android support (gonezo)
– No more ability to specify/set notes (you could add tags/notes to any night of sleep)
– No more smart alarms (you used to be able to set an alarm based on sleep state)
It’s like they went out of their way to make a far less capable product. And in some ways, it’s hard on the side of a blank box to explain the lack of detail. The inability to zoom or compare or do anything with any of the data in the app. All of that’s been dumbed down.
To me, that might have all been acceptable had the unit been priced lower. For example, the Withings Sleep sensor is $79. That’s reasonable, and despite the fact that it has more detail than the Beddit 3.5, it’s half the price. With Beddit being twice that price at $149, it’s hard to justify that pricing – even using the Apple pricing logic.
While doing a bit of research/validation on old features, I found this line from a MacRumors review of the previous Beddit 3.0 version:
“In its current incarnation, the Beddit Sleep Monitor is a decent sleep tracker. With Apple’s tweaking, I think it could be a lot better. In my experience, it’s collecting all the data it needs, but with some fine tuning, that data could be better interpreted and more valuable.”
Unfortunately, the exact opposite happened. Apple somehow made it worse and took away much of the detail that would have been helpful to those facing sleep-related challenges. On the bright side though, they saved ink on the side of the box.
Comparative Data:
Validating sleep trackers is always tricky. There’s elements that are easy to verify (when you fell asleep and when you woke up), but there are also elements that are trickier to verify (sleep movement for example). And then there’s those in the middle – like heart rate during sleep. One might think getting accurate HR during sleep would be easy, and in some ways it is. But it’s also tricky.
If one were to simply wear a typical chest strap to bed, that might give you data. But it’s likely it wouldn’t be super accurate. Chest straps do best with some amount of sweat or gooey stuff to help the electrodes. In a dry setting, it’s tricky. So instead, I used optical HR sensors in watches. In my experience these do fairly well in nighttime scenarios (assuming said nighttime duration doesn’t include high-intensity activity).
In all, I’ve got the following comparative data sources during my test period:
1) Beddit 3.5 (this review)
2) Polar Vantage M optical HR sensor
3) COROS APEX optical HR sensor
4) Withings Aura Sleep Sensor HR data
5) EmFit QA Sleep System HR data
All of those devices measure at least the following:
A) Time I went to bed and/or started sleeping
B) Time I woke up and/or stopped sleeping
C) Total time sleeping
D) Average HR (and sometimes more detail)
Most of the devices give additional detail as well that I’ll dive into.
I’m not going to go through two weeks worth of daily comparisons, because frankly that’s about as fun as picking lint off of a porcupine. Instead, I’m just going to pick a few random days and look at the attributes of those days. Let’s look at January 2nd, a couple of days ago. Here’s the set of data from Beddit 3.5:
And for comparison, here’s the set of data from the Withings Sleep sensor:
Then there’s light and noise detection from the Withings Aura, as well as temperature displayed:
It also includes a full nightly sleep state graph that I can look at sleep states at any point in time:
Then, here’s the data from the Polar Vantage M watch (with optical HR sensor) – which shows sleep state over time (the Beddit doesn’t).
And with the Polar unit I can zoom in on that graph and as well as adjust the fell asleep/wake up times in case they are inaccurate:
And I can also zoom in and look at the detail of my heart rate at any point in time:
And here’s the data from the COROS APEX watch (with optical HR sensor). It clearly has the least scrollable interface, but does contain most of the same data. And you even get sleep states on the graph if you hold down (screenshot to the right).
Now, let’s look at a quick summary from a data comparison standpoint for that night.
So here’s some general trends for this night:
Heart Rate: Virtually everyone gets the averages/min/max pretty close
Time I went to sleep: Except for Withings, these are all within 20 minutes. My actual time I fell asleep was about 12:55AM, so pretty close to most.
Time I woke up: I woke up at 7:15AM, so most units were within 3-4 minutes of that. The Beddit doesn’t actually specify a time you woke up, and there isn’t enough granularity on the chart to see when that is exactly.
Time Asleep: With the exception of the Withings, all units are within 11 minutes of each other. In most cases, this is the metric you’re likely looking at.
Time Awake: Again, all units are pretty close here as well, between 13 and 26 minutes total.
It wasn’t until I had captured all these screenshots and consolidated the data that I noticed the Withings was the outlier on the initial ‘fell asleep’ time (and thus, duration too). So I went back and looked at the data for the past week to see if it was always off or not. For all other days the Withings and Beddit (and even Polar and COROS) agreed within +/- 6 minutes. So that’s pretty darn close. Figures I’d screencap the ones that weren’t. Sigh.
As for the EmFit? Well, it captured beautiful data last night. And in theory I’d have finally had all five systems concurrently working. But alas, it wasn’t meant to be. While the EmFit captured beautiful data (seen below), the Beddit was the one who crapped the bed. Because I stupidly didn’t plug in my phone (since it was at nearly 100%), the Beddit doesn’t capture any data at all. It’s ultimately a bizarre limitation. Certainly the unit could have recorded that data and sent it on later when a connection was made (save for snoring data, which requires the phone’s microphone).
Other Options:
When I look at other sleep trackers out there, they fall into two basic categories:
A) Mostly wrist-based wearables
B) Dedicated sleep devices
For this purpose, I’m ignoring medical grade devices/setups. Instead, purely talking consumer stuff here. Even the cheapest $50 Fitbit or Garmin Vivofit will track your sleep. As well almost any $20 generic activity tracker. And honestly, most of them do a fairly good job at detecting the time you went to sleep, how often you moved, and then when you woke up. The same core information Apple’s Beddit provides, except usually with lots more detail and plotting.
Where you pay a bit more is when you want heart rate data included in that. I actually think resting heart rate data provides the *most* useful metric of how well your body is handling life on a day to day basis. I’ve talked about that extensively here. To get 24×7 HR readings (and thus sleep heart rate readings), you’re roughly in the $120+ category for name brand devices from Garmin and Fitbit. And again, all of those track your heart rate just fine.
So what about sleep-specific devices? Well, I’ve tested a few over the years. The two I have the most experience with are EmFit and Withings Sleep sensor. I have the original Withings Aura sleep sensor, rather than the newest one that connects to your smartphone. But my understanding is the underlying sensor and data is the same (just one uses your phone instead of a bedside clock).
I’ve placed most of the Withings data throughout this post, so you can see the user interface pretty easily. It’s clean and simple, and you can click on any graph and get more details than I’ve screenshot – so that’s good too. Also, if you have their activity trackers, then it integrates with that as well.
Both the Withings and EmFit track far more data than Beddit 3.5 does (previous Beddit versions gave additional details).
The EmFit unit takes the cake on the most detailed data out there from a consumer device that I’ve tried. In fact, it’s overwhelmingly detailed. But I actually have a hard time recommending EmFit QS unless you have a specific sleep condition that you want tons of data on, or if you’re just a data geek. There’s no smartphone app for it so you have to use a web browser for everything, and setup is cumbersome at best. It does have way more data than everyone else by a massive long shot, all of which you can slice and dice and dig into until the cows come home. Beautifully data-driven clunky, but clunky nonetheless.
As you can see, there’s an astounding amount of data in there, including real-time monitoring from anywhere on the planet and all. But the lack of an app makes it rather clunky to use. And at $229USD, it’s a fair bit more expensive.
As you can see, it’s kinda a case of trying to decide what you want in (sleep) life and then picking which option makes the most sense. At the same time, as wearable devices start to consider sleep as part of the overall equation of fatigue/stress/training/etc on your body, then having a singular bed-focused device may not make as much sense if there’s no integration to the remainder of your day.
Wrap-Up:
Having both worked in one of the largest tech companies out there for over a decade and having worked in mergers and acquisitions prior to that, I fairly well understand that when a tech company gets acquired, sometimes funky things happen. Usually the result falls into roughly one of two buckets:
A) Acquired company’s products get discontinued, IP/features get consumed into other products/etc, while employees either get reassigned or let go
B) Acquired company’s products get expanded, with new functionality and usually better results
There’s of course always outliers, but those are the basic buckets for how tech acquisitions go. But in Apple’s case they went a different direction:
C) Take existing product, gut it of most features, sprinkle caramelized Apple dust on it, and then try and sell it as a new/better version
At this point some of you might say “Well, it’s an Apple product, they can get away with it.”
Except: It’s not.
At least not in branding. To the average consumer there’s no indication it’s an Apple product. It’s no different than any other 3rd party product sold in a typical Apple retail store. It’s got the mandatory mostly-white box (per all products sold in Apple retail stores), but otherwise the box says Beddit. Thus, it lacks the legitimacy carry-through of a true Apple branded product like an iPhone or AirPods.
Now – to be clear, if this was a $79 product like its competition (the Withings Sleep sensor), then I’d actually have far fewer concerns about it and probably be happy with it. But it’s not. It’s a $150 product. Just like it was prior to them gutting it.
As for functionality – sure, there is something to be said about the simplicity of it. But I think they took away one of the biggest selling points of before: The sleep score. That helped to quantify what you did that night by taking multiple metrics together and giving you suggestions on what to improve. Additionally, if they were aiming for simplicity they likely wouldn’t have removed the explainers in the app – making it harder to understand and quantify. Again, here’s all the things they removed (they added zero features except making the strap softer):
– No more sleep score (a cumulate total of numerous metrics into a singular nightly score) – entirely removed
– No more extended details on each of the metrics (you could expand out every metric and dive into the details)
– No more explainers in the app for different metrics (you could expand out a metric and get clarification)
– No more cloud sync and data storage/backup (you could sync all the data to a Beddit cloud platform)
– No more deep vs light sleep functionality, graphs or otherwise (entirely removed)
– No more restless sleep tracking
– No more Android support (also gonezo)
– No more ability to specify/set notes (you could add tags/notes to any night of sleep to assist in remembering what happened)
– No more smart alarms (you used to be able to set a dynamic alarm based on sleep state)
Frankly, I don’t know what Beddit was thinking. Perhaps they were just hoping that they’d get mainstream tech reviewers who would review it because it’s Apple and therefore all is great (without ever having tried other products, or even the previous Beddit product). Or perhaps they don’t actually want people to buy it. Or perhaps they just got drunk one night and decided to throw detail to the wind. Who knows.
What I do know is that I fear for a version 4.0, since at this rate it’ll just be an empty page when you open the app – and a box that’s totally blank on all sides. But that’d take courage.
FOUND THIS POST USEFUL? SUPPORT THE SITE!
Hopefully, you found this post useful. The website is really a labor of love, so please consider becoming a DC RAINMAKER Supporter. This gets you an ad-free experience, and access to our (mostly) bi-monthly behind-the-scenes video series of “Shed Talkin’”.
Support DCRainMaker - Shop on Amazon
Otherwise, perhaps consider using the below link if shopping on Amazon. As an Amazon Associate, I earn from qualifying purchases. It doesn’t cost you anything extra, but your purchases help support this website a lot. It could simply be buying toilet paper, or this pizza oven we use and love.
Thank you for this review! I had some Apple gift cards to use and was looking at Beddit and wondered if it would be better than using my Apple Watch and the AutoSleep app which I’ve been using since it came out. I’ve been happy with the watch and app for sleep tracking, but sometimes I want to take the watch off and when it is hot I get irritations if I don’t take the watch off.
Now I know not to waste my gift cards!
Yeah, I’d spend that money on something else. Anything else really.
Great review, as always. Considered ordering this when it got announced, but ordered an Oura ring instead. Wondering if I should have gone with the EmFit instead. Any chance you’ll be looking at the Oura ring at some point? A lot of the reviews out there seem suspiciously schilly and would enjoy hearing your take on it.
Yeah, I haven’t tried out Oura yet. I’m not really a big ring person in general, so wearing another isn’t too much my cup of tea.
I do agree with you that many (albeit not all) of the reviews of it are a bit less then ideal.
Not a ring person either but I would really like seeing your review of that…
One thing to note WRT the Oura Ring is that there’s no reason to wear it during the day. It does not track HR 24/7 – only during sleep – so the only thing you’d get is number of steps and the like, all things that the Oura app can import from, for example, Apple Health.
Hi,
nice review…as always.
I`m using Fenix 5x+ for collecting my sleep data and for me it does the job done just fine. I just have one request for you…would you reveal the secret of your deep sleep quantity….you are always getting from 2-3 hours of it, while I can hardly get around 1,5 hours and mostly below 1 hour (from 30 to 60 min most of the nights).
And I`m usualy sleeping from 7 – 8 hours per night.
thanks
You’re assuming it is accurately measuring both states.
link to huffpost.com
link to sleep.cs.brown.edu
Need an EEG sensor
yes I think it is…I don`t know if this is the right term (deep sleep), but I can tell by how I feel on that following day how much “deep sleep” I had…with more than 60 min I feel great, but with more than 90 min I`m indestructible…that`s a great day for a race…:)
It’s my understanding now that cell phone shouldn’t be left on the charger past 100% as it can degrade the battery. It weird the Apple of all people would encourage you to have your phone charging all night with their recent battery problems.
That’s not true at all. Your phone won’t allow itself to be “overcharged” – and modern batteries don’t care about recharges from 99% vs 20%.
In 2003 this may have been true. Phones are a little more intelligent now.
I started using the SleepScore app on my iPhone. It doesn’t have any hardware and uses the microphone and speakers to detect sleep. It has the some of the same basic data (HR, sleep start/end, etc.) as well as a sleep score. I have no idea how accurate it is. Seems to be based on some decent university research, though. It does have the same drawback that it needs to be plugged in.
So far it is interesting to use. I’m not sure what to do with the data. But I figure if I’m more aware of my sleep patterns (good and bad), I will try do a better job going to bed earlier, no caffeine, etc.
As always, thanks for the great review.
They removed Android support and deleted the app from the play store awhile ago. The android version was always limited after their version 1 product and didn’t get lots of the useful functionality like auto start and stopping to record sleep.
Too many sleep trackers are just tracking movement and heart rate so making guessing of sleep state. Not too many have EEG sensors to collect extra info like brain waves (think the defunct Zeo headband) I think its just Dreem and Philips Smart Sleep
Another great review Ray, to me it appears Apple seems to of made a decision to be superficial with everything, Activity & Health just don’t have any depth in them at all. I really don’t understand the benefits of sleep trackers full stop, it isn’t like I can go to bed and then tell myself to have 3 hours of deep sleep, I fail to see how they can help, if anything they just make you more paranoid… oh I didn’t get any deep sleep last night, must get double tonight LOL
The Dreem headband does do cognitive behavior therapy and uses data from the sensors to help with that
Kinda off topic but I noticed that some of the devices capture breathing rate. I’ve often wondered why other devices don’t capture this especially training/workout devices. I thought it might be just as interesting as heart rate or oxygen saturation or cadence or even more interesting than things like temperature or steps…
Any thoughts on why no one is doing this? It looks the date CAN be recorded so why don’t we see it? ?
Respiration rate seems like a valid fitness metric to track in relation to HR and fatigue.
It could be a function of a chest/belly strap or some other wearable tech.
It could even track estimated volume exchange.
The kit would include an inflatable test bag/device to measure your maximum lung capacity. Next calibrate the app by entering that figure, then mount your bike, assume a normal riding position, fully exhale for minimum measurements, then fully inhale (belly then chest) for maximum measurements.
Now we’ll have more charts to stare at after a workout — Respiration Dynamics ;->
A chest strap or two electrodes would be required to monitor breathing.
I can see real potential in the pre-diagnosis of sleep apnea (enough to bring it to the physician) with the combination of HR (widely available), respiratory rate and pulseox (as available in the Garmin Fenix 5XPlus).
Note: Actual equipment to diagnose sleep apnea also measures other things like multi-lead heart rythm (ECG), brain wave activity (EEG) and air flow.
Okay? But it’s not really apple, is it? I don’t see the logo anywhere in the box or product
It doesn’t sound like you read the post, as I discussed these very items.
As noted, it’s on the side panel of the box where it says it’s made by Apple. And then on Beddit’s site, any support is handled 100% by Apple and redirects you to Apple’s site. Also, the app itself says the company is Apple.
I read most of it and looked for several minutes for the info, but I’m sorry I haven’t found it. I’m only speaking as an industrial designer, this product looks completely different from all the apple products, in terms of hardware, connectors, packaging, branding, illustrations, color use and graphic design. They are obviuosly supporting a third party product. It doesn’t seem to be “made” or designed by them. This absolutely does not excuse them from offering such a terrible product, lending their name, leading people to trust them
link to beddit.com
scroll to the bottom:
Copyright © 2019 Apple Inc. All rights reserved.
link to beddit.com
Look in the middle:
Beddit is now part of Apple. Contact Apple for support.
Look at the picture of the bottom of the box. Notice Beddit is a trademark of Apple?
Apple bought Beddit in early 2017. Thats what Ray was talking about in the wrap up
Thank you, both
….. But it didn’t set fire to the bed!
I see little point in buying retail diagnostic products unless the product has been medically verified or assessed. The only sleep data I take any notice of anymore is heart rate. I look at the range low to high. If mine sits in a 10- point range from 35-45 approx I’ve always felt good, the rest of the sleep data is just crap ( in my opinion ) and I take no notice of it .. Garmin Vivoactive 3
thank you for that, saved me time and money. Maybe apple will build on the functionality over time?
inclusion of snoring time might create a few relationship problems (or maybe even solve them)
re emfit, you say “But the lack of an app makes it rather clunky to use. ” that is true but you can put a URL shortcut on your smartphone and it kinda becomes app-like and easy to access the sleep data.
RE: EmFit – That’s true, though, I find it clunky still. It requires re-auth every time for me, and some screens like the real-time recording screen for example render poorly.
As for snooring, it’s actually interesting and something I want to dig into a bit more. Neither I or The Girl snore…unless we’re sick or sometimes very rarely if the wine bottle was too generous. It’s also a bit interesting to me that it would only detect snooring for say 15 minutes on a given night. Not saying that’s inaccurate, but would be semi-interesting to to a concurrent audio recording on the side and see if that’s indeed the case.
re-auth. me too. seems to be only when I have switched devices for viewing the data eg once I use the smartphone then I have to re-auth the PC. Otherwise, the PC is ok from one session to the next and logs straight in.
One possible rationale for removing things like the sleep score, deep vs light sleep, and restless sleep tracking could be a lack of a meaningful basis for tracking those. I have never used a sleep tracking device and have no knowledge of their algorithms, but I am generally skeptical of the current fad toward amalgamating complex data into a single number. Anybody can come up with a magic formula that combines a bunch of numbers together. The hard part is actually doing so to achieve a meaningful metric, and making sure that the metric is robust to the uncertainties present in the data.
If the uncertainty on time awake is a factor of 2 (13-26 min across devices in your sample data table), you either will have a metric that fluctuates wildly (if time awake is given a large weight) or that damps out large variations in time awake (if it has a small weight or a strong damping function). Again, my comments are really more general than these specific devices. For all I know there could be lots of careful studies behind the sleep tracking algorithms and composite metrics that I’m not aware of, and particular device(s) may be more accurate than the full data spread in your table suggests (e.g. your comments about the Withings data). But without serious validation, these devices could just as easily be computing a largely meaningless number from a bunch of highly uncertain “measurements.” And if so, perhaps Apple disabled those rather than knowingly having it report nonsense.
These sorts of issues appear to be the problem with running power, where they are trying to figure out out to combine all the variables into a single number. They may be able to set some constraints on the models based on controlled lab conditions, but the fact that many companies claim to have validated in the lab and yet give such different results indicates that the full models still have a lot of ambiguity.
I 100% concur with you. That was my first thought, when I read about all the omissions: Maybe they did it because they don’t trust in that flood of data themselves.
Heck, my Polar Vantage sometimes tracks my sleep despite my not even wearing it. My 935 constantly has me sleep without interruptions like a baby, even though I generally awake numerous times due to our actual baby next to our bed.
I am very skeptical about all those metrics those devices gather, and plot. So much so that I only care about HR by now. My RHR is the only parameter (if tracked somewhat correctly) I take notice of to follow trends in recovery, training, and general well-being.
“Maybe they did it because they don’t trust in that flood of data themselves.” Yep ! maybe. In that case Apple are being honest by removing the gumf.
Going to sleep HRV and waking HRV are interesting to track. Most often HRV seems to trend inversely to rHR
…about 10 minutes later. Very subtle.
Great honest, review. Appreciate it.
I’ve used Beddit 2/3 as well as loads of other trackers over the years. Here’s my take on the Apple/Beddit situation.
I really think Apple have massively miscalculated the opportunity to enter the sleep technology market with this acquisition.
Sure, buying Beddit at the time made a lot of sense. Beddit had a nice brand. Minimalist, good product design etc – fits in really nicely with Apple’s brand ethos.
However……. the sleep tech market is one of the most dynamic in recent years, and the technology surrounding sleep sensors, algorithms, has already moved on from the tech that Beddit are famous for.
Hence, Apple has acquired a company that makes a simple mattress (ballistocardiography) sensor. This is not a new technology (it’s been around for over 100 years > link to cs.tut.fi) Beddit has nothing particularly unique about its IP either. The only real value to Apple would have the Beddit data set of existing sleep recordings, and even that it of little value now.
Today, the market is crowded with identical mattress sensing tech to Beddit (Withings, Emfit, Beautyrest, Eight, a dozen Chinese knock off companies).
But the real future in sleep tracking is pointing to other areas that offer more accuracy, efficacy and metrics ie:
– EEG (Dreem is leading the way here)
– AHI/sp02 tracking – a huge trend at the moment with lots of crossover consumer/medical devices like Beddr, EverSleep, 02Vibe, Go2Sleep
– wireless sleep tracking – again, a big trend with SleepScore Max, kardian.com etc
In short, Apple/Beddit are massively behind the curve in terms of what’s happening right now in sleep sensing technology.
Apple made a bet, but the bet didn’t pan out.
I don’t see the Beddit story as having a happy ending.
@Jeff.
we shall see what volumes a pretty product and good brand name can sell.
Some of the sleep tech you list is interesting. I can’t see much of it being there for the long term either. In my mind only BCG, wrist-based and ring-based technologies will work for a mass market as those tech do not intrude on the life of the person wearing them. Although, perhaps, niche exceptions will survive that focus on a particular issue eg sleep apnea
As well as getting correct source data these technologies also have to correctly interpret it to offer ACTIONABLE INSIGHT.
Define “correct”
Massively miscalculated?
It was a low 7-figure acquisition: a rounding-error-level acquihire for Apple and a fire sale for the founders/investors. Beddit failed. And Apple probably won’t turn it around.
Beddit’s biggest problem was their reliability. I have 5 years of data from Beddit devices (gen 1,2,3), excellent and actionable insight for a serious athlete… But I had paid for a Beddit once, and have had at least 6 or 7 devices during that time, all exchanged for free after failing in various ways within the warranty period.
When Apple killed the cloud service (incl. HRV and HR data) and bricked the devices for Android users they were fair enough to completely refund full purchase price (from 2013 for me), so effectively I’ve come out +/- 0 on Beddit financially. Obviously that’s an appalling business model.
Emfit has been my device of choice now, and has proven to be reliable and useful for tracking what an athlete actually needs.
the5krunner, lots of those devices just measure heart rate, breathing rate, and movement. Those don’t really know your sleep state so can’t really do as much. I really hope the long term isn’t limited by that limited ability to collect data.
@Daniel, agreed on QS.EMFIT with BCG readings. I’ve used it for a couple of years and am yet to see it bettered. I would accept that their data capture can be very good. I still have a nagging doubt about their interpretation of it eg sleep stages. although i tend to accept EMFIT as probably being “true”. Have you used their ALERT service? (I haven’t and would be interested to know if it is any good)
@Eli yup. but where does the tech go? Maybe there is some fancy crash-helmet-like brain wave measuring device that is super accurate at measuring ‘whatever’. but only about 500 people worldwide would ever use it. so yes it probably comes back to capturing that data you list and then correctly interpreting it. (I guess there are body temp and other things that could be sensed via a watch/ring and I would love to see more products/research on those areas).
Fitbit finally enabled the SpO2 sensor during sleep and with the addition of the sensor data, their new sleep score feature (only accessible on the web) is very good.
I find the accuracy to be good on my Ionic too.
If only they could FIX their resting heart rate calculation. I am sitting on the couch right now and my heart rate is 6 beats lower than what Fitbit thinks my resting heart rate is.
Sigh.
I bet this crap will be soon discontinued forever.
Anyway, some people fooled by Apple acquisition will buy it, adding a tiny drop to Apple’s shameless revenues.
I still don’t get why should a multi-billionaire company should spend any money in a company/product they hardly believe in…
Thanks for the review. I was looking forward to buying one of these for some time. Now I will wait. Anecdotally my experience with Apple is when they remove functionality, it isn’t because they have not thought it through. Often functionality comes back, or is replaced with something better down the road. Doesn’t help being out in the interim wilderness. Let’s see what happens with 4.0
Try ouraring.com way better sleep tracking and lovely minimalistic approach for the sensors.
With all of these sleep trackers, especially the ones that are in the bed or pillow, I’m interested in how they claim the user has gone to bed/sleep.
I used Hello Sense a while ago (company failed) and they were telling me I was asleep when I was just sitting in my bed reading, or that I would have an “agitated” sleep if I was doing some other kind of activity in bed
Thanks for this review Ray, on the back of this I got the Withing’s Sleep which I’m really impressed with!
Thanks for the detailed review. Really enjoyed your methodology. There’s still one unanswered question for me: Which sleep tracker (if any) would you actually recommend? (Withings?)
Excellent review
Thanks for the review. Are you sure that Apple isn’t just being more honest with this release?
Hi,
I not tested that device, but one thing is clear for me.
All motion sensor based devices are wrong when identify “sleep stages” based on the motion. If you don’t have sleep disorder, you will not move at all. This was published some decades ago by US Airforce studies on young healthy peoples (fighter pilots).
Also I experienced it myself. I used the Sleep as an Android mobile App, for detecting my sleep phases. The App displayed for me deep, light and REM stages. Later came up that I have obstructive apnea and I got CPAP treatment. After the adecvate treatment my hypoxic events disappeared. Like wisely, the App displayed only a long line of “deep sleep”:
So plane truth is that by motion detector (actigraphy) healthy people can establish only following parameters:
– bed time
– sleep time
– sleep interruption
Peoples with sleep disorders will see a chaotic activity with the simple meaning that you have a problem.
We must consider, that sleep disorders are extremely frequent. The prevalence of sleep disorders is ~30% of the population.
Sleep phases my be extracted from HRV monitoring as we heard it in Helsinki at the HRV Summit.
Best Regards:
Dr. György Báthori PhD, MD
Thanks for the great and detailed review! Well written!
Yes, Its very strange. I had Aura before, was very happy about it. Then after Nokia bought Whitings they discontinued Aura, as with most of Whitings products, I guess it was your strategy A) :-) However the owner bought recently Whitings back from Nokia, almost bankrupt, so it slowly getting back on track.
Anyway, then I started to look forward to Beddit, and was waiting for the new, if any, version of Beddit instead. I thought either apple discontinue it, or they make some super product with super detailed tracking, Apple Watch integration, top of the line analysis, maybe even Homekit support (Like, when you wake up the good morning scene turns on, or if you step up in the night the night light turns on). it could have been a killer product. it could easily be worth more, like $200. But then this happen… An almost empty product, void of all functionality…
No, its really strange. Seems like Apple have no strategy at all when it comes to acquisitions. they just buy and put the company in a drawer in case they need it in the future. Or if it is some patent they want, for later integration in Apple Watch !?
Big companies clearly destroys small innovative companies.
Thanks for great review. Apple did to Finnish Beddit same as Finnish Nokia to French Withings. Bought the best small company in genre and destroyed the product by updating the superior App to useless. Interesting to see how the biggest front line tech companies are unable to adapt the greatness of the best of the bedt niche companies they have desided to acquire because of their greatness.
I like deeply review of the products I will buy. Keeping Searching is main homework I used to do. Thanks
this is really such a good website dude
example.com
[url=https://www.example.com]example.com[/url]
Affordable Housing Projects In Bangalore . These housing projects are being aggressively promoted now a days as they serve as a good source of profit to the builders. For those who have read the article will understand what I am talking about.
افضل متخصصة بروتين معالج بالرياض
Houzingam