Samsung Galaxy Watch 4: An Accurate SLEEP Tracker?
ฝัง
- เผยแพร่เมื่อ 16 ก.ค. 2024
- 🔴 Affiliate* link: geni.us/GalaxyWatch4
🟢 Instagram: / quantified_scientist
🔵 Newsletter: www.robterhorst.com
🟡 Twitter: @QuantifiedRob
* Affiliate (paid) links support the channel by earning me a small commission on any sale, and it doesn't cost you any more! My opinions are completely my own, and this content is not sponsored.
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
👇 $30 off any new WHOOP membership, affiliate* (paid) link, supports the channel!👇
join.whoop.com/QuantifiedScientist ($30 off any new WHOOP membership)
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
👇Amazon Affiliate* (paid) links. Supports the channel, doesn't cost you any more!👇
Huawei Watch GT 3: geni.us/Huawei_Watch_GT3
Apple Watch 41 mm: geni.us/AppleWatch7_41
Apple Watch 45 mm: geni.us/AppleWatch7
Garmin Vivomove Sport: geni.us/GarminVivomoveSport
Garmin Epix 2: geni.us/Garmin_Epix_2
Garmin Fenix 7: geni.us/Garmin_Fenix_7
Garmin Venu 2 Plus: geni.us/GarminVenu2Plus
Garmin Venu 2: geni.us/Garmin_Venu_2
Garmin Venu 2S: geni.us/Garmin_Venu_2S
Garmin Venu 2 Plus: geni.us/GarminVenu2Plus
Fitbit Charge 5: geni.us/FitbitCharge5
Fitbit Luxe: geni.us/Fitbit_Luxe
Fitbit Charge 4: geni.us/wjAfjU
Fitbit Sense: geni.us/CSfYQE8
Fitbit Inspire 2: geni.us/iZHlb
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
📸My Gear* (paid links, supports the channel and doesn’t cost you any more)📸
Camera body (main): geni.us/SonyA6600_body
Camera lense (main): geni.us/sony_sel_18-135
Camera (B-roll): geni.us/CanonEosM50MarkII
Light: geni.us/Godox_VL150
Softbox: geni.us/NiceFotoSoftbox
Light stand: geni.us/WalimexLightstand
▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬▬
Intro:
If you wear two Galaxy Watch 4’s at the same time, will they measure the same sleep stages? You would at least hope so, but you might be surprised by some of the findings in this video. If the two watches are not consistent, that means that either the raw measurements themselves or the algorithm underneath is not very stable. That is what I will check in this video for the sleep tracking of the Galaxy Watch 4. For 6 nights, I wore one Galaxy Watch 4 on my left arm and one on my right arm, both running the same firmware. I also wore this EEG headband that can actually measure my brainwaves, which provides relatively reliable sleep tracking. Now, the fact that I wore one watch on the left and one on the right hand could introduce some differences in the movement data both watches record during sleep. However, again, this is something that the algorithm should be able to deal with. If not, this means it is inherently unreliable at estimating your sleep stages. Let’s see if this is the case!
Conclusions:
Based on these preliminary results, I do not expect the actual sleep tracking to have changed with the new firmware, or I at least do not expect it to have improved. Now, I just tested it two nights, and also one one of the watches was not the “classic” version of the watch, which has the same sensors and software, but a slightly different design and weight. So, further testing is needed to make 100% sure nothing has changed, but as I said, I do not see any evidence of significant improvements.
Overall the sleep tracking of the Galaxy Watch 4 is mediocre at best. The most telling for me is the fact that two watches worn at the same time give such different results. I think that as it stands, the main relatively reliable use-case of the sleep tracking functionality of the Galaxy Watch 4 is tracking your total time spent in bed. However, I wouldn’t really trust the tracking of the individual sleep stages. If that is important to you, Whoop, Fitbit and the Withings Sleep analyzer are better options. I’ll be testing many more devices, and I am also working on ranking the different wearables I’ve tested so far in terms of accuracy for heart rate and sleep. If you would like to see those videos and more, consider subscribing. I also did a complete review of the Galaxy Watch 4, which is linked up here. It also helps the channel, and of course makes me feel happy, if you like this video and leave a comment, but that is of course totally up to you! See you in the next video and have a great day!
Timestamps:
00:00 Short Intro
00:47 Sleep Stages
04:37 Sleep Start/End
05:43 EEG Test
08:54 Other Devices
09:30 Software Update
12:49 Recommendations - วิทยาศาสตร์และเทคโนโลยี
I really appreciate that you are the only tech youtuber that actually goes in depth and tests the sleep tracking accuracy. Keep it up!
Ya! All only show what functions devices have. And this is the only channel where we can see the quality of the functions
You've become probably my most favorite TH-cam channel nowadays - the amount of thought, work and genuine quality that goes into these videos is unparalleled. As exciting as all these gadgets are, they are only as useful as the data they provide. Without (reasonable) accuracy, the health data is very limited in value. So thank you so much for testing the accuracy in such an in-depth and reliable way, you're helping us make better choices :)
First, thank you for your extensive tests, finally someone with actual numbers, rather than impressions. If you're so inclined, I decided that the GW4 sleep tracking wasn't up to what I thought it should be, so I purchased an App that changed everything. I wonder if you would consider tracking the GW4 using an app "Sleep as Android". Completely different results, that seem somewhat more realistic, ,but that's just a gut feeling, and it would be nice to see some scientific data.
Cool idea !
I have the same impression!
Ouch, that REM detection level gives me nightmares. So much for getting the Classic. Thank you Rob for the thoroughness. Hopefully Samsung's listening. You deserve a million subscribers, which would hopefully make the companies listen more carefully and invest more in accuracy. Most impressive delivering such recently conducted research. BTW, someone used a transparent silicon pad underneath the watch to significantly (anecdotally, not statistically) improve the sensor accuracy. Looking forward to the flight data!
The best nightmares are indeed in REM sleep.
@@chch242 Yes, you got it.
Yours are the greatest videos about smartwatch/bands on all YT! Thanks so much for your work.
Sounds about right. The deep sleep readings on my Galaxy Classic 4 are awful. My deep sleep score is consistently in the range of 1%-8% (occasionally 0% and highest was 11%). My whoop strap 4.0 consistently reads 20%-24% deep sleep. REM fluctuates wildly but seems to track more closely to the whoop strap. Shame, as it's a great overall watch, but very inaccurate heart monitor and sleep tracker. Thanks for continuing to provide all this data man!
I’ve been waiting very long for a review like this! Thanks!😄👍
Bedankt voor al je onderzoek! Dankzij jouw video's ben ik van de Samsung GW4 terug gegaan naar Fitbit. Had vanaf het begin al niet veel vertrouwen in de Samsung GW4 als het ging om slaap, hart en andere fitness tracking. Fitbit is wat dat betreft toch echt wel de betere optie. Ik kijk weer uit naar je volgende video's!
Dankjewel Andres! Blij dat mijn tests je iets geholpen hebben😊. Een hele mooie dag verder!
Great video. I'm glad you're out there doing actual comparisons. I worry about the quality control with these smart devices, and it's good to see independent third parties holding the manufacturers' feet to the fire.
Such an informative video, thank you so much for taking the time to put all of this data together for us Watch 4 users! I got my watch 4 in december and I've used it for sleep tracking every single night since (quite literally every single night) and I've never really taken the data seriously aside from the fall asleep/wake up times. I find that the watch is way too sensitive when it comes to awake time during the night, and I presume that it considers arm/wrist movements during REM sleep as being awake. Additionally, prior to the update my deep sleep was always ridiculously low, but followed the pattern I saw in this video of deep sleep generally happening in the first half of the total sleep time. After the update though, while my deep sleep time has increased significantly, the times it says I'm in deep sleep breaks away from this pattern, and its more sporatic throughout the night. I'm by no means a sleep expert and maybe this is normal when it comes to sleep cycles, but I find this change odd.
At the end of the day more work definitely needs to be done on Samsungs sleep tracking algorithm. Im really hoping you do more videos on this in the future! Would love to see a more comprehensive test on the changes that were made after the recent update? Because even though you said in the video you don't think the algorithm changed much, my sleep graphs are looking quite different when it comes to my deep sleep scores. The overall deep sleep time my watch was recording went from 10-15 minutes to around an hour.
Hi John! Thanks for sharing your experience in such detail, and I will hopefully take another look at the sleep tracking in a while!!
Thank you for the review. The only channel i found that provides such approach and data analysis. Is there any plans to test Polar watches? Thanks.
Hi Quantified Scientist, thanks for your videos, i really appreciate it. by the way do you have recommended smart watch that has accurate ECG? thanks in advance
A great in detail video thanks! I'm glad to have seen it as it was a watch I was considering buying. For me, sleep tracking is the most important to be accurate feature for me but slso important would be if it could do ecgs and detect arrythmias. Have any of the watches you've tested scored high on both of the above? Just about to watch some more of your videos but it would be great if I could be pointed in the right direction. Oh I should add, very important, I am a chronic insomniac so it takes some time to fall asleep with frequent awakenings, early rising (3/4am then maybe a nap around 7am Sam :-)
I'd like to point out one thing regarding your testing in the past, using two devices to measure against each other. But using impedance measurements of any can could potentially throw off measurements. I am by no means a scientist but I do know that when measuring resistance in an electrical circuit it can be thrown off by additional electrical signals and resistances, like if you were to hold two wire ends along with the leads and measure the resistance the meter would also measure your body's resistance in parallel with the circuit. I'm not saying I'm right. Just wondering if that was a consideration in testing or would it need to be? I know it uses light for heart rate but who's to say it's not using the other sensors as well? Just a thought..
I bought the Fitbit Sense and the Withings Sleep based on your videos. I also bought the Garmin Forerunner 45 because it was on sale. I used all 3 devices every night for a few months now. My conclusion is that Fitbit and Withings have good agreement on my sleep quality. But, the Garmin is good enough since it measured my total sleep time pretty accurately. It appears that, for me, the total sleep time most accurately predicts how I will perform the next day. So, when I wake up earlier than planned, I look at the total sleep time, and if that is too short, I force myself to get an extra sleep cycle or two.
If I were to do this all over again, I would probably use the Withings to track sleep since I don't have to wear anything on my wrist. I will use the Garmin for running, since the Fitbit performed badly for me as a running watch.
Hi Watcher World! Really happy you enjoy my content! Also, happy that my reviews helped you make some decisions😄😊
Congrats on 100k! : )
do any companies come back to you to
1) object the results
2) ask you how to improve the watches
3) how samsung can purchase fitbit and have the best phone and watch.
your reviews are brilliant and so easy to understand
I would be very interested also!
Fitbit is already bought by Google lol
I was one of the people who thought the last update had been improved the algorithm because my deep sleep time increase a lot, but no.
Thank you one more time for this clearly and excellent review. I really appreciate how you listen to your subscribers. Cheers.
Your videos are great! Do you have the heart rate chart for the devices tested like you have here in the Other Devices section in this video?
Nice as always. Can you please also test ecg functionality e.g. on charge 5 fitbit?
I can't wait for your review into the other biometric sensors on this watch.
Tremendous job. Thank you!
Thanks Dominik, this is highly appreciated!
HI, could you compare two EGG devices and see how accurate are they? I'm a little bit curioused. Thank.
Thank you very much Rob.
Thanks James for commenting!!😄😊
Great work as always
Thanks so much Ryan!!😄😊
Thank you for video. Why wouldn't you wear both watches on the same hand to avoid measurement discrepancy due to a different movement of right and left hands?
Thank you for the video. What about galaxy watch 5. And is it better to track my sleep with Samsung health or the sleep cycle app?
Great job as usual 👏
Thanks so much Georges!!😄😊
What can be the hypothesis regarding this differences between these two identical watches?
Can the choice of the arm, or the position on the arm be a possibility?
Great job btw! You made me buy a Huawei watch fit ;)
This channel deserves millions of subs and views
Hi Hax! Hopefully we can get there at some point😉
Thanks for the informative video. Can you also do a blood oxygen level accuracy test, maybe include both galaxy watch 4 and watch 5.
Thanks for the content.
Thanks Zerogenome!
what might play a role here is the difference between the right and left sides as sources. Measuring blood pressure with the same device on both arms can also provide different information. Whether that makes a difference with sleeep measurement I do not know but it is something to explore. I have the G watch 4 Classic. Might there be yet another difference between the 2 G-4 watches?
Which watch would you recommend for best sleep and intense workout tracking and with music playback, for $250
If I already avoid using the watch when sleep due to no feel confortable, now I don't even bother to try. thanks for the work,.
hi! I need help
I have a Samsung watch i tracked my sleep
with continuous heart rate on but i can't seem to view my heart rate during my sleep....i can't see any graph showing my hr during sleep how do i get that
Great video, something that it would have been nice to have in the video is on which wrist you were wearing which watch. As the ai algorithm is probably mostly trained on left arm data. It would have been nice to see if there was a small difference in accuracy between the left and right arm data.
To be fair, I was expecting similar results, the main information about sleep being brain activity and can't be traced by a watch. So an accuracy between 30% to 50% does not shock me that much, or at least I would not expect more. Something basic to be able to have a rough idea but not much more.
Did you actually set that you wore the watch on left hand on the set up prompt? Before the tests
All your videos are much interesting, but this goes to another level... Wish it would be possible to do this for all the tested devices (particularly Garmin ones for personal interest :D ) I do have the impression that i have nights where my Fenix 6 nails my actual sleep, and others when fails miserably... Would be interesting to test 2 at same time
Thanks so much for the compliment! It is indeed interesting to test two watches at the same time . Lets see for how many watches I can test this!
I have the galaxy s4 classic and appreciate the update. I knew the old algorithm was trash as it said I had under 20min of deep sleep every night. After the update I'm over 1 hour of deep sleep every night. So they did tweak the algo. I think the best health feature is the constant hr/stress monitoring and the pretty good customization.
What if they just randomly added 30 minutes to everyone's sleep-tracking just to make you believe that they fixed their algorithm?
Hi Eric! Thanks for the input! I indeed had to add the new results rather last minute😊, so it is worth checking out for sure
@@stb3987 making changes to the algorithm to do that would probably be harder to do than just fixing the algorithm. George Castanza style
Did you wear these on the same arm? One on each? I wonder if the watches sensors detect movement and uses the data to determine the sleep stages. I wonder thought, if you wore one one each arm... wouldn't each arm move differently?
Thank you for the extensive tests, I really appreciate your effort. Just a suggestion that you may consider.
I personally think that wearing smart watch with different hands will affect the result.
Generally, right-handed people wear their watch at their left hand and vice versa. So, the research from Samsung was most probably designed in such way. This explained the difference in the sleep tracking results of both watches on different hands.
May I suggest you to show the which watch is wore on which hand and which is your dominant hand? Then compare them to the Dreem2(ECG). I think this will be really interesting to look into the difference in that way.
For my personal experience, the update makes the sleep tracking more reliable. Before the update, I barely have any deep sleep. Now, it is much better.
Hi! Thanks for the suggestion! Happy to hear that the sleep tracking seems to have improved for you😊
Curious if the oxygen sensor is accurate during sleep. I got a low number three times in sleep and it's concerning.
Question do you think the information is not the same because you move one arm more than the other in your sleep? Also, thank you for making these video it is helpful.
Your in-depth analysis led me away from these much more expensive devices and I ended up just getting a Fitbit Inspire 2 (after my Versa 2 completely died), even with the Sense on a deep discount (-35%) right now. Yes the other bands have more features but if the accuracy is bad what is the point of a tracker? It's frustrating that even though Google bought Fitbit, they haven't released a Fitbit device with Wear OS. The Sense has nice features but a lot of the data is paywalled by Fitbit Premium, and the apps are pretty useless (ie. music apps - I never used them on the Versa 2 in 2+ years). I'm hoping Samsung improves their tracking algos or Garmin improves sleep tracking on their newer devices and will look to maybe upgrade then. Thanks again!
Hey rob, have you gone in-depth on the body battery algorithm by garmin? That metric is weird imo. It roughly represents how I feel waking up but their sleep tracking obviously isn't great.
Did you adjust the settings so the watch knows on which arm it is being worn? To right and left arm?
Can you do the same test but with both watches on the same arm?
Amazon has recently release their Amazfit, can you test it to see how it stands up to the rest of them?
Very nice analysis 😁👍 A little constructive Critique for comparing two watches :
I think , maybe the algorithm could take meassurong from the movement of the dominant Arm to differentiate better between REM and deep Sleep. So if possible, wear them on one arm if that doesnt restrict Blood Flow too much 💉🧛😅. (My idea is, that the dominant arm moves more or that the preferred position in bed could have influence on the movement meassuring)
I plan on doing a comp between Garmin Vivomove Sport and Venu2 S , to see if the smaller Venu 2 performs better 😅
Btw: Not that your friends from the Netherlands are DC Rainmaker and DesFit 🤣🤔
Ray is American (living in the Netherlands though), and DesFit is American living in the US.
@@chch242 Oh, thanks for the clearification. I wondered that ive seen DesFit in one of Rainmakers videos 😅 Maybe he was on vacation xD
@@residentCJ yes, Des travelled to do a collaboration with Ray when doing the overview of 2021 watches.
I always watch these right through, although in this case I'm not even interested in sleep tracking. Nobody should buy a HR and sleep tracking watch without seeing QS's meticulous reviews. The problem for his business model is that most of the manufacturers and distributors probably don't enjoy his fearless reporting!
Hi William! Thanks so much 😄😊. Indeed, manufacturers might be less inclined to lend me their products, unless they really trust them, which I hope they do😄
Are you going to do a Dreem Band 2 review?
I would think you would want to wear the watches next to each other on the same arm for a more accurate comparison??
Please compare galaxy watch fit 2 with watch 4. What does the difference it makes with the price.
If you move one arm more than the other during sleep this should affect the readings ?
Why wouldn't have had them both on the same arm?
My completely amateur results have found the Galaxy Watch 4 to be the most accurate device I have (also have an oura ring and garmin venu 2), though I feel like that is faint praise. I cannot vouch for sleep phase detection, of course, but after going through a period of insomnia (mainly early waking and middle of the night waking) I found the Galaxy Watch was best able to detect total sleep hours and early awakenings. Of course there is some error on that because I am relying on my own guesstimates of what time I fall asleep and remembering wakings, but in general I found it was the best of the bunch at detecting when I was awake vs when I was laying quietly in bed. I have found my Garmin to be the worst and basically never detects me being awake, and the Oura ring is somewhere in the middle of the two in accuracy. But again I'm only looking at generally how accurate they seem to be at detecting when I fall asleep and when I wake up and not anything with sleep phases.
I've compared results from my Fitbit Charge 2, Versa 2, and the Samsung GW4 - for me the biggest difference is that Samsung puts more info the light sleep bucket than Fitbit. However, Fitbit frequently fails to record sleep stages at all, effectively just telling me when i fell asleep and woke up.
For me, Fitbit Charge 3 seems to track my sleep fairly accurately and has no trouble with bathroom breaks.
I've just got a new GW4 to replace it since the screen has stopped working on the Fitbit C3 and my first night it stopped tracking my sleep after a 5am bathroom visit despite me going back to sleep immediately after for 3 more hours.
Nice idea, could you do it with huawei watches
My biggest issue is it saying I'm asleep until 0945, when I'd already been up and about for 2 hours..... and very active. But it wouldn't register me being awake until I physically used my watch
A suggestion/wish: I assume that you track HR during different workouts when the corresponding workout is activated on a watch. Another interesting test would be - how a watch detects HR variation during day-by-day life. One of the ways to emulate this - do some workout where hands are fixed without activation a sport mode on the watch and check if/how well/how fast it detects HR changes and if/how fast it alarms about crossing configured threshold.
Hi Gregory! Interesting suggestion 😊. I wonder if it will realise I am doing sport still, or if it will indeed be confused
@The Quantified Scientist
I bought a week ago a Samsung Galaxy Watch 4 but i still believe that my Fitbit Sense is better in Sleep Tracking.
I hope soon to see a Fitbit Sense 2 with Rounded Screen and have More Better Sensors with More functions and Internal Storage.
Hi Pavlos! Indeed, I do believe that the Fitbit will be better than the Samsung for sleep tracking for sure😄😊
@@TheQuantifiedScientist I have 5 Health - Smart Watches from different companies (Fitbit, Garmin, Samsung, Huawei, Xiaomi) and for Sleep Tracking looks that the best for now is Fitbit and specific Sense.
The only problem that i have with Fitbit Sense is that is broken 3 times in 1 Year.
So that's why i bought a New One Smart Watch.
I am looking for New Fitbit Sense 2 with Bigger Rounded Screen.
:)
Is this watch still a mess in the HR monitoring department or did it get any better?
I think this mainly shows the accuracy of wearing on the right vs left. It may be affected by your sleep position too.
Hi Glay! To me, if right vs left is inconsistent, that is an issue. Any algorithm should not rely to heavily on measurements that are different between both hands. I have tried the same with Fitbit, and this had agreements between hands between 80% and 90%, and a >70% agreement with polysomnography and EEG.
I've watched all of your sleep tests. Why is the graph that compares the different models always changing? On some of your graphs the Woop4 is the best and at other times you have the Fitbit line showing to be better than the Woop4? This is very confusing. I hope that you are not changing the graphs depending on which product offers you more of an incentive to sell them. Please explain better which products are the most accurate for sleep monitoring.
But wt if the other hand moves while one don't 🙄, anyway do the test again when the new one UI out 🙂
Very interesting comparison. You used the galaxy watch 3 as well. But that video you said the 3 was very accurate. Can you compare the 3 to the 4.
I wonder if the gw5 improved this.
What about the sleep tracking on the Google nest hub? That's another device you have not tested before.
Hey fellow scientist. A better possibly more comprehensive comparison should have swapped hands half way through your test and done measurements again. If blood pressure is a factor in the collection or even temp detection (I would have no idea what is under their hoods or how they work here) your dominant hand might shift the data. So basically much of the deviation could be due to dominant hand.... Let me know your thoughts. (You could easily complete the data by grabbing some more measurements...)
Ya know, at times I feel like the Samsung watch classic is pretty accurate, sometime I feel like it it's completely off! 2-3 times a week it seems to be spot in with sleep time and score. As in, I know when I fell asleep and the score aligns with how tired or rested I feel. Then 1-2 times a week it's completely off! It will tell me I fell asleep 2+ hours after I fell asleep, it's crazy
Please tell us about body composition of galaxy watch 4. I wanna buy it only because of this option but l have heard it is not accurate. What should I do. Please help us!! Your channel is what I always looking for! Thanks alot
Bit confused....how can they predict sleep?
Could just be placebo, but my sleep tracking has had a marked change since my GW4 received the update. I get a lot more deep sleep tracked, which historically seems more accurate for me! Would love a full follow-up!
Same for me.
Are you not getting enough deep sleep? or was your chart considered normal
Can plz do the same testing by wearing two apple watches... please
Hi Jay! Hopefully I can do that at some point 😊👍
Hi! I have GW4 Classic and for me, it seemed like sleep stages and generally tracking is quite consistent and accurate on my watch. However, upon new FW update that came, it kinda went to s**t, where it started showing a little bit too much deep sleep and it's not as consistent for me as before. So, update made it worse for me. I know I don't have EEG device, but for me it was good enough and I could read the graph (and it was also similar to your EEG graphs), but now I can't.
Also, the awake moments are mentioned in the details of sleep stages, it detects awake moments also when you change position or move a little more. Those moments, which you don't even remember, because you were really just asleep.
Hi Tomas! Thanks for your input, and interesting to hear how it worked for you. Sorry to hear that it showed some issues
This video is very good. I just wanted to ask you: did you also set the arm and button layout in the settings? They recently added this so the watch stays accurate no matter where you wear it, at least they claimed that
The new Redmi Smart Band Pro looks really promising. Please do a video on that!
Awesome, so sumsung gives basically random results for sleep tracking. Can you do same with one of the fitbit units? As they deemed as one of the best trackers.
Hi Denis! I have done some of this in the past, and I think it is even included in some of my very early videos. From what I remember two Fitbits (not even the same model) generally agreed between 80-90%. However, new testing is needed to look at current firmware
Hi there, i found your youtube chanel and i'm loving it. Can you please recommend me a sport band under 50 euros for cardio workouts ( so good heart rate tracking ) ? Thanks
Hi! For under 50 euros it is still quite a struggle to find really good watches... There are some OK watches out there. However, I still need to quantify the differences between watches to get a better feel for the more subtle differences. You could check out my review on the Huawei Watch Fit, I don't remember all the details, but I think it was quite OK
Unfortunately I agree with your results. My Galaxy Watch 4 is terrible for sleep data, and not great for fitness data either. I plan to sell it since its past the return window.
Hi! Sorry to hear that... Hopefully someone else is mostly interested in the smartwatch features and will take it of your hands😊
Mine got worse after the update, heartrate and oxy. is all over the place now.
Great review. My galaxy watch 4 has the worst sleep tracking I've ever had. It detects about 3 minutes of deep sleep a night. I purchased a new watch and it was the same. This effects your sleep score making also tracking useless. Samsung common!
Hi Shep! Thanks for sharing your experience!
I was dismayed to see how bad the G4 was. My biggest issue is not knowing if you are awake or not. If you choose to lay in bed and watch YT videos, it assumes you are still asleep. If a person was bed ridden the G4 would say you are Snow White. Totally worthless. Likewise, for exercise....when biking, 1/2 hour at a moderate rate the watch gave me half the calories burned compared to the bike, and only a third the calories of my carb tracking program. I also checked other web sites for calorie consumption and found most estimates to be between what the bike showed and my carb app. If the G4 is worthless for sleep and calories burned, I might as well be wearing a $10 Timex.
Differ with your judgements of which product is better. Engineers know that test results will have differences. There are openly declared acceptable tolerance for these differences.
Acceptable are 1, 2, 5, 10 or 15 per cent deviations, in the engineering sense. Engineers state the deviation tolerance.
The quality control of the final products from the factory does not normally allows public circulation of items outside these tolerances of quality. However, some brands release versions of less performance for the products that fail these tolerances. For example, Intel, AMD, and other manufacturers do this regularly.
Engineering and scientific testing do not just rely on the single sample, for final judgements. Biologically and sociologically, the acceptable deviations are also stated.
In your final comparisons of the many brands, the brands should be clustered as similar, in groups. Not as individuals, one only brand, compared to another brand. Otherwise, a brave but first analysis of many brands.
thanks for the detailed review. why there is literally no Apple watches on the graph in "other devices" part of the video?
Hi! good question. Because Apple does not track REM or deep sleep, but just "asleep" and "awake"
@@TheQuantifiedScientist thanks for the reply
Best ever !
Thanks so much!!
the watch is great for functionality with Samsung devices. lots of useful compatibility features if you have a Samsung phone. However when it comes to the accuracy of the data it offers, it kinda looks like they just threw together a half baked algorithm and shipped it out hoping nobody would notice. I hope Samsung improves their health data accuracy in the future.
I want to review more on Huawei gt3 I think it is the best so far
You should have swapped the watched to the opposite arms and wore for the same period
hi Rob 🏋️
Hi Rob!💪🏼😊
Wow, you start to make fantasy Rob to compare two same brand 🥳💯. There is a new watch on the market xiomi watch S1. if you test it and make video İ will be very happy. Thanks alot....
If you can a Garmin fenix 6 test would be great
Hi Carlo! Thanks so much for commenting! I understand you are interested in the Fenix 6. However, given that the Fenix 7 was already released, I do not think it will make the priority list, given the many watches that I would like to test. I hope some of my other videos will provide you value though.
Pog review.
I am a little disappointed that you did not catch both watches with both hands ;) :D
Question
Did you guys ever broke your watch during sleep?
Which watch is the best for sleep tracking?
Hi Mateusz! Fitbits, Whoop and the Withings Sleep Analyzer are amongst the best. I show an overview somewhere in the second half of this video I believe😊
Dude, thank you so much for the information ... really helpful!
Were these tests done before or after the update that supposedly improved sleep tracking?
@Carrot Cruncher @The Quantified Scientist Said both watches are running the same firmware, but he also said he doesn't expect tracking to change with the new firmware. That leads me to believe he's using the old firmware. And for such a scientific approach, you'd think assuming anything would be a blunder, but maybe @The Quantified Scientist has a good reason for thinking nothing will change with the new firmware. Regardless, I'd like to know why.