• Amazon.com

Hands-on: Garmin Vivosmart 4, now with Pulse Ox and Body Battery


As is customary for this time of year, it’s all about the wearables – and the last two weeks haven’t disappointed in that realm. Today we add to the pile with the new Garmin Vivosmart 4, which is, as the name implies, the fourth edition of the Vivosmart series. And approximately the 86th Vivo-branded device.

As with new Fitbit’s Charge 3 last week, there aren’t massive changes here from the previous generations. But in many ways, both companies are on a very similar track – especially when it comes to the addition of SpO2 tracking, which Garmin has added into the Vivosmart 4.  Previously that was only seen on their highest end and most expensive watch, the Fenix 5X Plus which came out back in June.

Still, that’s not the only new feature – so skip down below for the quick run-through of what’s new on this $129 wearable.

Oh, and before we do that, note that this isn’t a full in-depth review. It’s a quick hands-on based on a handful (well, wrist-full) of days of use on a media loaner device. Down the road, I’ll likely do an in-depth review, mostly because some of the new features here are actually unique on this device and worth investigating deeper.  Since it’s not a full in-depth review, I could find over time that things get either better or worse than my quick poke at it. Got it? Good. Let’s go. Onwards!

Update: In September 2019, Garmin added Connected GPS to this device via a firmware update, which means you’ll now get a GPS track of your workout as well, as long as your phone is with you. More details in my post about that here.

What’s new:


As with most Garmin device iterations you generally see features introduced on other devices filter down into lower-cost devices.  And while that’s certainly the case with some new features on the Vivosmart 4, there’s actually two cases where these features are found on no other Garmin devices.  Specifically, the newly added features of:

Body Battery: Gives you a constantly updated ‘energy’ score throughout day
Sleep Pulse Ox readings: Tracks SpO2 readings during sleep and shows on mobile app

Neither of those two new features are found on any other Garmin device (yet), no matter the cost.

Atop that, there are some differences between the Vivosmart 3 and the new Vivosmart 4 that extend beyond that. Some of these are hardware, and some of them are software updates previously seen added elsewhere in the Garmin realm. Here’s my typical bulleted style list:

– It’s smaller: The unit is now thinner (width) than before
Battery life increased: Now 7 days versus 5 days
– Far clearer screen: It’s actually quite an improvement over the Vivosmart 3
– And that’s it: Best I can tell.

Now, some other things may appear new, that came in via later updates since last year’s Vivosmart 3 launch. For example, the Vivosmart 3 got Abnormal Heart Rate Alerts this past summer, and the Vivosmart 4 carries those too.

There’s a huge list of firmware updates since then on the firmware updates page. But this list doesn’t do a great job of capturing software-based updates on the mobile app – for example with the advanced sleep tracking that they rolled out this past spring or so.  That’s covered on the Vivosmart 3 and Vivosmart 4.

If you want the hands-on summary of everything in the new Vivosmart 4, I put together this quick video that covers both the app and the unit hands-on:

When it comes to the display, the new touchscreen is definitely easier to see than the Vivosmart 3, there’s no question about that. The Vivosmart 3 always felt a bit fuzzy because of the layer of silicon over the screen.  It worked well enough, but it just wasn’t super crispy.  This is super crispy.


On the flip-side, I still find the touch responsiveness a bit cumbersome. Specifically with waking it back up again. It’s not an always-on display like some watches, so you have to either raise your wrist or tap it.  And I find that sometimes swiping through the menu (especially down), it doesn’t seem to always register my touch.  Note that you can customize the way the main dashboard page looks:

2018-08-28 16.22.43 2018-08-28 16.25.21

Of course, there’s two main new features here. First being Pulse Ox, and the second being Body Battery. I cover Pulse Ox in the next section, leaving us this moment for Body Battery.

This feature is only seen on the Vivosmart 4. And yes, I asked if it’s coming to other watches and merely got back “it’s exclusively launching on the Vivosmart 4”.  Essentially it’s like the days of the Street Fighter video-game where you had your energy level. This level tracks throughout the day based on resting HR (from the optical HR sensor), stress tracking (also from the optical HR sensor), workouts/activities, and sleep quality.  The more you do, the lower your score goes.  Get some rest, and your score goes up. The easiest way to look at it is my day yesterday:

2018-08-30 00.47.36

A couple of things above to decipher:

A) You can see I slept from roughly midnight till 8AM, my battery says I was fully ‘re-charged’ by 4AM
B) The grey icons indicated automatically tracked activities, mainly two commutes by bike
C) The green icons indicate my two actual ‘workouts’ (tracked on the Edge 1030 in fact, not this device), these were relatively short 25-minute test rides
D) The orange is the stress level measured throughout the day when not sleeping
E) The blue line is my body battery level over the course of the day

The app also outlines this in a more wordy form:

2018-08-29 17.09.09 2018-08-29 17.09.14 2018-08-29 17.09.18

Now the question is what do I think of it?  Well…I guess it’s correct in spirit.

But I’ve got a couple of concerns here.  To begin:

A) If my body battery was re-charged by 4AM, that’s…full of crap.
B) I know this because I woke up at 6AM to screaming children, and was exhausted…and went back to sleep. 4AM would have been far worse.
C) Why did my rides have virtually no effect? Was it because I did them on another device?
D) Typing this post (from 11PM on this chart) had a seemingly bigger impact than my first ride outdoors

Now, perhaps writing this post is more brain-impacting than my ride, I’m not sure. But I’m not convinced.

Now don’t misunderstand me. I do think that the overall concept of a body energy type metric is exactly what’s needed. I’ve long talked about the lack of an overarching metric that takes into account things like resting heart rate trends, and sleep, and activity and gives you truly actionable instructions. As Garmin notes in the app – the number here is very explicit in what you should be doing in terms of going out for that workout or not.  I’m just not sure I agree with how that number trended yesterday.  Or the fact that the day before when I went and did a 45-minute strength circuit workout that it basically had no meaningful impact on my body battery number (it says 57 below because I forgot to pause it after). Again, I need a bit more time here with this feature…but my initial impression is that Garmin might need some more time with it too.

2018-08-28 23.56.09 2018-08-28 23.57.34

And before I forgot, the Body Battery score is also shown on the watch itself like other widgets, just by swiping up/down. You can further configure it onto the main dashboard display if you want.


I do want to note that Body Battery is a FirstBeat feature and also found under a different name in the Suunto 3 Fitness watch. However, the implementation is quite a bit different. First, Suunto doesn’t show you any historical information on their mobile app – so that’s a biggie. Second, Suunto’s underlying data is different because they don’t have a 1-second always-on capable optical HR sensor. Instead, the Suunto sensor samples less frequently. It might not make a huge difference, but it’s worth noting.

Finally, for lack of anywhere else to stash it – here’s the stock image of all the colors. Obviously I’ve got the black one, but they’ve got a handful of other color choices. Unlike the Fitbit Charge 3, there’s no swappable bands here.

MCJT 12206 vivosmart 4 series image

With that colorful bit out of the way, let’s dive more deeply into the new Pulse Ox bits.

Pulse Ox Redux:


Pulse Ox was first introduced on the Fenix 5X Plus back in June as part of the larger Fenix 5 Plus series. At the time, only the highest end Fenix 5X Plus variant got the feature, which required additional hardware sensors. Garmin said they saw that feature as somewhat unique and also a bit of a bleeding-edge type experiment of sorts.

The main (marketing or engineering, depending on one’s view) thinking behind Pulse Ox on the Fenix 5X+ was really more focused on climbers and those at high altitudes.  The idea being one could watch blood oxygen levels there and do a better job tracking over time than typical finger sensors would do (since that process was more cumbersome to do frequently). Garmin hasn’t yet gone through any sort of FDA certification for Pulse Ox readings, though, it’s something they’ve said they’re seriously considering (I talk about that at length in my in-depth review in the Pulse Ox section, including the entire new FDA Software as a Medical Device realm).

But that story shifts a bit with the Vivosmart 4. Here, the goal is more about using the data during sleep tracking than it is about high altitude adventures. This thinking is virtually perfectly aligned to all of Fitbit’s efforts. The main difference though is that Garmin is exposing far more of the actual data, rather than Fitbit’s plan to give you general trends.

Here, let me explain how it works. First, you’ve got to enable it. And actually, you seem to have to enable it in two places and in two ways. First is enabling Pulse Ox as a feature, but then enabling it for the sleep portion.  For example, you can take a Pulse Ox (SpO2) reading at any time from the menu on the device.  You’ll go into the heart rate sub-menu (where you can get things like VO2Max as well), and then select the PulseOx icon:

Garmin-Vivosmart-4-PulseOx-2 Garmin-Vivosmart-4-PulseOx-1

When doing so it’s recommended to sit/stand/whatever still. Don’t move. Same basic advice from the Fenix 5X+.  The reading tends to take 10-20 seconds to complete.  You can see the red light illuminating my skin as well:


Taking a direct head-on picture of the red light is tricky business, but here’s a slightly different angle:


In any case, once that’s done you’ll get your reading:


I have noticed though that when it fails to get a reading it seems to go one of two ways:

A) It outright tells you it failed and to adjust fit and try again
B) It just says 94%. For example, you’ll see a few pictures in this post with 94%, that’s when it measured the nothingness of space

But that reading is kinda like a one-off. At this time it doesn’t appear to go anywhere in the mobile app as saved data for later. Fart in the wind and all that.

Instead, in order to get Pulse Ox readings to track you need to enable the sleep Pulse Ox section within the mobile app for sleep data. I made the mistake of only enabling within the first portion for my first night, and I got no Pulse Ox data.  So definitely remember to enable both parts.  It gives you an explainer of the magic you’re about to invoke:

2018-08-29 09.26.28 2018-08-30 00.05.38

Then, fast forward to the second night with it on and I’ve got this pile of data afterwards:

2018-08-30 08.10.41 2018-08-30 08.10.20 2018-08-30 08.10.06

Which, is not much data.

You’ll notice it doesn’t actually measure my SpO2 values all night long. Instead, it just picks some period in the middle of the night to do its thing (this is just barely shown above in the middle screen at the very bottom by that elongated blip around the 2AM marker). It remains to be seen how exactly Fitbit does this on the Charge 3/Ionic/Versa units.  That won’t happen till later this year sometime as part of their Sleep Scorecard beta program.

In the case of last night for me, it apparently thought I was sleeping while I was still working away, and in fact as I stood up to go get a glass of water from the fridge, it chose that exact moment to try and get its Sp02 reading. I could tell from the red light glowing into my skin. It doesn’t appear to have tried again that night.  I did get a video the day before showing how to access the setting, recorded from someone else’s screen – and that definitely shows a bit more data during the night:

2018-08-30 12.43.50 2018-08-30 12.47.39

Nonetheless, initial troubles aside (keep in mind, it starts shipping today), the idea behind monitoring this though is that the data could be used to assist people in figuring out sleep-related issues. While this data point by itself may not be valid in a medical/clinical scenario, integrating multiple data points including movement and heart rate can start to paint a much more cohesive picture than most people could ever get – short of spending lots of time and money on medical-grade sleep devices.

Ultimately of course, it’s clear that all of the fitness tech giants (Garmin, Apple, Fitbit, and likely Samsung) are working towards these same goals with likely the same metrics. All companies have at one point or another talked about ways to address the huge numbers of people impacted by sleep disturbances (sleep apnea, atrial fibrillation, allergies, asthma, etc…).

Of course, Garmin does have a slew of disclaimers on their site nonetheless about (not) using it as a medical device (with a section on Pulse Ox as well):


In the meantime, I’m interested in seeing what this data looks like longer term – and of course, where Garmin plans to expand it beyond the Vivosmart 4. Obviously, it takes additional new hardware. So it’s not something they can just turn a software switch on and enable on older units.  But given they’ve just added the feature to their least expensive optical HR device – after adding it to their most expensive device introduced this year…it stands to reason it’s going to quickly fill in on every other unit between those two points.



Overall, I like the direction Garmin is going here with both Body Battery and Pulse Ox. I’m just not convinced either is ready for prime time yet.  While it’s still early days (I’ve only had a few days worth of data), the data doesn’t always trend in the ways I’d expect it to (primarily with respect to Body Battery).  But the value of what Garmin (and I presume FirstBeat behind the scenes) is trying to convey is exactly what I’ve been asking for, for years. The idea to take in all these metrics (sleep, resting heart rate, heart rate variability, workouts, activity, stress), and give you a simple 0 to 100 number: That’s brilliant. At least in theory.

When it comes to SpO2, obviously, that’s the new industry trend.  I’m actually less concerned about inaccurate SpO2 readings at this point (it seems to be accurate when it does correctly get a reading), as just the lack of readings and logic that goes into the nighttime pieces. For example, the fact that last night it didn’t retry again when I was actually sleeping is a bit peculiar to me.  As is the fact that it doesn’t seem to save one-off Pulse Ox readings I take during the day.  I will say that I’d certainly hope that any further wearables Garmin develops would have Pulse Ox within them, given they established the price floor (this device, the least expensive optical HR unit they make), along with the price ceiling (the Fenix 5X+ – the most expensive watch this year).  Thus, not doing it on every other future device (like Fitbit is doing) would be a solid face-palm.

And while some may think I’m being slightly harsh, I’m told you should be able to pick up the Vivosmart 4 today at major retailers around the US.  So as such, it’s kinda a device that’s ready to roll with current firmware/hardware. Said differently: It’s not beta, it’s fair game.

Still, I’m looking forward to trying it out more over the next month and putting together a more complete review after they’ve perhaps addressed some bugs and more importantly that I’ve got a number of weeks of data to start looking at trending on. Perhaps there’s a learning aspect to some of the Body Battery bits, just like there is for other metrics like VO2Max and Training Status.

With that – thanks for reading!

Found This Post Useful? Support The Site!

Hopefully you found this review useful. At the end of the day, I’m an athlete just like you looking for the most detail possible on a new purchase – so my review is written from the standpoint of how I used the device. The reviews generally take a lot of hours to put together, so it’s a fair bit of work (and labor of love). As you probably noticed by looking below, I also take time to answer all the questions posted in the comments – and there’s quite a bit of detail in there as well.

If you're shopping for the Garmin Vivosmart 4 or any other accessory items, please consider using the affiliate links below! 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. Even more, if you shop with TPC (The Pro's Closet), you'll save $40 on purchases over $200 with coupon code DCRAIN40! The Pro's Closet has been a long-time partner of the site here - including sponsoring videos like my cargo bike race, as well as just being an awesome Colorado-based company full of good humans. Check them out with the links below and the DCRAIN40 coupon!

Since the Garmin Vivosmart 4 is no longer sold, I recommend looking at Garmin Vivosmart 5:

And of course – you can always sign-up to be a DCR Supporter! That gets you an ad-free DCR, access to the DCR Quarantine Corner video series packed with behind the scenes tidbits...and it also makes you awesome. And being awesome is what it’s all about!

Thanks for reading! And as always, feel free to post comments or questions in the comments section below, I’ll be happy to try and answer them as quickly as possible. And lastly, if you felt this review was useful – I always appreciate feedback in the comments below. Thanks!

Post a Comment

Your email address will not be published. Required fields are marked.
If you would like a profile picture, simply register at Gravatar, which works here on DCR and across the web.

Click here to Subscribe without commenting

Add a picture



  1. francis

    For example the fast that last night it didn’t retry again when I was actually sleeping is a bit peculiar to me

    • gingerneil

      There’s a couple in the last few paras – Ray was obviously getting stressed by that point! :)

    • It’s funny, I wrote almost everything the night before. Except the closer section. Looks like that 3ish hours of sleep is catching up.

    • mpulsiv

      Ray, you have to try to get more sleep. I have been reading your blogs for years, back when you traveled a lot. You were sleep deprived. Now, you got two peanuts. I don’t want to derail this thread but just take a minute to check out a podcast or book of “Why We Sleep” by Matthew Walker, Director of UC Berkeley’s Sleep and Neuroimaging Lab, where he explains the urgency of getting proper sleep.

  2. gingerneil

    This is an interesting device… but I’m assuming no GPS, and you have to go to the vivosport for that? An updated version of that, with GPS, would be great for my daughter. Any possibility of seeing that soon (ie before Santa packs is sleigh?) She would like a step counter to compare with her friends, and uses a tomtom multisport for running. Having something small and neat on her 9yr old wrist, that does both jobs, would be ideal. The vivoactive and forerunner 35 are a little too big for her.

    • Correct, no GPS. Not even Connected GPS like a Fitbit device either.

      You would indeed have to go to either the Vivosport for that (which seems to be on sale semi-often), or one of the smaller Forerunner units. What about the older Vivoactive (original), that’s super slim and I know people find it for crazy cheap.

    • gingerneil

      Slim, but still a little larger than I/she would like. The band approach of the fitbit/vivosmart is ideal – but the vivosport is looking a little old now.

    • cute_polarbear

      right. no gps unfortunately. been using vivosport since it first came out and overall, still the best slim ‘life style’ device, heart rate monitor, with gps so far on market it seems. (it’s priced pretty well also) i have been waiting for an update from garmin for vivosport and thought vivosmart 4 would have been a souped up version of vivospot.

    • Mangesh

      I got the vivosport for my 10 year old beginning of this year. And I am impressed with it. The vivosmart 4 seems to have some nice new features. Is there a vivosport 2 coming soon with these additional features? Probably a better upgrade to the current vivosport.

    • gingerneil

      I certainly hope so.. a vivosport 2 in time for Christmas would be ideal! :)

    • Steven Priefer

      I read it does have connected GPS now.

  3. Michael Schmidt

    I can’t believe that also version four has no replaceable band. The band will tear after one year and you could buy a new device. Good for Garmin, but sorry, not my view of resource friendly products … Wan´t replace my vivosmart HR by a vivosmart 4!

    • Yeah, was kinda surprised by that as well – especially with the trends towards swappable bands.

      It does seem though that when Garmin tried swappable bands for activity trackers in the past, it didn’t seem to do as well.

  4. Dave Lusty

    So I guess that confirms our suspicions that Fenix 5 and 5s lacked Pulse Ox to allow for a Fenix 6 with more features then. Battery was always a weak excuse when the hardware difference was a red LED replacing a green LED. Now glad I didn’t upgrade to 5 plus yet, I’ll wait for the 6!

    That said, I’m not keen on the way these companies are getting further and further away from measured metrics. Body battery isn’t something that’s measured it’s something derived from some derived things and some measurements and a lot of assumptions. Like running power and the various running dynamics metrics all this does is give yet another number to look at and absolutely nothing to back it up in terms of science or actionability. Inference is not the same as measurement and no matter what marketing say I still maintain that you can’t measure how high my foot is from the floor from a sensor on my chest. Even a sensor on my foot is making HUGE assumptions about forces from a device that’s not accurate enough to make those assumptions.

    If we knew what to measure to see how much energy a person had, we’d also be able to replace that “special sauce” to give people more energy. Literally none of that exists today, making this whole idea hokum. You may be right that the numbers will start to look more realistic in future firmware, but I completely disagree that they will ever be meaningful or related to your body in any way. It’ll be an algorithm faking a metric at the very best. Just like my best “step count” days are those sat lazing on my boat or behind the wheel of a car.

    On the sleeping problems front – the vast majority of sleep issues are down to lifestyle (or kids!). Just like the majority of weight issues are due to eating. People are paying thousands for little magic devices to help with a problem that actually needs a dose of reality and some life changes. If you eat a cheeseburger, watch TV or finish a bottle of wine just before bed every night you’ll be fat and tired. A watch won’t help with that, lack of cheeseburger, TV or wine will help with that. (yes, there are a TINY percentage of people with genuine medical issues…)

    • gingerneil

      You have to go to something like Runscribe for proper measured metrics. For Garmin run dynamics, the sensor is too far away from its trying to measure, but the Runscribes are right there measuring directly. I dont see running power being any use to me, but the RS output is very useful and definitely directly actionable.

    • Dave Lusty

      Runscribe aren’t measuring any of the metrics they report either. A ruler can measure distance, an accelerometer cannot – distance is inferred from other metrics which in this instance are not accurately measured either. Don’t fall for the marketing, a LOT of the numbers produced by this kind of technology are far from accurate or reliable. If accelerometers were as accurate as these people claim, we wouldn’t need GPS because we’d just measure everything from our point of origin. We can’t do that without military grade submarine guidance systems costing billions – even those need to be calibrated periodically and checked against other known information.

      In the case of pace, it’s close enough to reality to be useful. Now we’re looking at “body battery” we’ve entered the realm of utter nonsense. Hormones play as much a role there as expended energy and those aren’t even being considered in the calculation.

    • gingerneil

      With RS I’m referring to their core metrics where things like pitch and roll CAN be measured directly and are being done so right on each foot independently – not miles away at the waist or chest and on a single device. This enables calculation (note – not estimation) of things like pronation angles and velocity. Further, these are real world things that can be compared and verified against other real world things / observations / measures. This is where RS is different to other vendors attempting derived running dynamics.
      If you take your argument to the logical conclusion, garmin/suunto etc devices like those in the above post don’t measure *anything* directly infer them from effects (ok, maybe time aside??). Pace/distance from computation of GPS signals, heart rate by capillary blood flow, cadence by looking at the wrist movement etc.
      I agree that ‘body battery’ is a combination of lots of things. I would say that nobody should rely on one single measure for anything, but you can add this to your own ‘feel’ and then come to your own conclusions/actions.
      If I wake up feeling tired, was that because I haven’t had enough sleep, or very poor quality sleep ? That’s hard to determine through ‘feel’, but these devices may be able to offer some insight.

    • Bo

      I sort of agree on the sleeping front but there is an egde case of people like me who apneu for other reasons combined with asthma. The respiratory doctors have been very happy to get both my perceived score and the measurements out of the CPAP and off the Garmin.

      That said I am rare.

    • David Juiliano

      I use a Stryd device. Went to the track yesterday and did 2×4000 at tempo or 10 laps each with the GPS off on my 935 and Stryd reporting the distance. The stryd reported distance was 4027m on both intervals. This seems pretty close to to me as it’s higher than 99% accuracy. Check out this link.

      link to fellrnr.com

    • ekutter

      I generally agree with Gingerneil here. RS isn’t truly measuring distance, and isn’t yet as accurate as I’d like, and I’m definitely not bought into the whole running power thing. But some of the other metrics are pretty close to direct measurements, like impact and breaking forces combined with pronation and pronation velocity. I have been really surprised to find that I have a pretty strong left/right imbalance with impact even though I feel completely even. As for distance, most of what I do is trail running and I’m hoping it gets to a point where the computations RS uses are more accurate than GPS which is definitely not very accurate on many of my routes. One particular route where it was way off was running the Grand Canyon. Just look at the track and I’m making 1/4 mile detours into the cliffs.

    • Dave Lusty

      @David Juiliano yes it should be close, it’s counting how many times you did a certain movement that you told it the distance of. Now try taking that same sensor, putting it on a different person’s shoe and getting them to run up a mountain without recalibrating and see how you get on. It’s NOT measuring that distance it’s parroting back information you’ve given it with some clever processing.
      @Gingerneil there are very simple tests you can do with a smart phone to demonstrate how reliable those angles are. For instance, line up the edge of a phone with a table and get the compass on screen. Slowly move the phone back and forth and your compass will come back to the same number. Spin it quickly a few times and suddenly your table is turned 10 degrees. Beware technology that looks accurate!
      @ekutter so you’ve been told that you have an imbalance. Your body has developed this imbalance naturally, possibly because your legs are different lengths or because of some old injury. Do you think artificially correcting to make your running or cycling symetrical is a good idea? The human body isn’t symetrical and trying to force yourself to be could cause more harm than good. There is NO research in this area – your device is giving you noise that you cannot sensibly action.
      There’s a huge difference between useful measurement and derived metrics. The amount of excess noise being generated from very little actual data is staggering these days and that trend seems set to continue as Garmin and others add features and avoid researching them. I could create 30 new derived fields tomorrow from the measurements in my watch, and I could give them fancy names and some marketing person may even be able to come up with a good accompanying story as to why you need them. Would they help your fitness though? Or maybe just go for a run when you’re not aching from the last run and listen to your body.

    • gingerneil

      Hmmm… EVERYTHING we measure in running is a derivative calculated from an input of something else (or many many things!). Do you look at your GPS tracks/pace/distance and use them to manage your training load ? That data is the result of very very complex analysis of the GPS signals broadcast from space. Why would you use/rely on that more than something that is derived from a device right there on your foot?
      Run to feel and listen to your body – yes, that’s essential. But I am due a long run today as part of my Ultra training – should I just run until I *think* I am getting the training effect I want from the run, or should I use the derived distance and pace GPS measures to *GUIDE* my training put in the miles I want at the pace I am aiming for?
      Specifically on runscribe – you are hugely over simplifying. The devices do need calibrating, and that process is built into the Runscribe dashboard. Thats both for pace as your personal data and running gait is input to tune the models over time, or as a simple ‘mounted’ calibration when you clip them to your shoes to set a ‘zero’ point. Run up a mountain with GPS and you are NOT measuring that distance – you are again processing a load of data and coming up with an estimate for something else. If you are a user of GPS / sports tech (assume you are considering the general readership of DCR), then you must have some acceptance of these inherent limitations – RS etc are different in that they are using different tech, yet the same in that they are combining complex measurements to give you something useful and actionable. How you use those (or don’t) is up to you and that’s the same across the board. I find the gait analysis from Runscribe very useful (linked in with how I’m feeling), but I’m not bothered (yet) about power.

      With regards to imbalance – depends on the source and the cause. Yes – we all run differently. That’s why I don’t agree with the trend of stability etc shoes – I tried them to ‘fix’ pronation and ended up with a hip injury as you just push the body out of line somewhere else. Instead, I trained myself to go from a hard heel striker to a soft forefoot striker. This allowed my body to adapt over time and to become a much quicker and more efficient runner. Runscribe was of huge benefit during that process as it gave me detailed feedback on HOW I was actually running. So moving to forefoot became measurable and not just something I could do through feel. Same with getting my cadence up during the same process.
      Next – I would LOVE something that could measure arm swing. One arm ‘hangs’ and remains almost static (cc: Steve Way!) as I run and I am sure I lose a lot of efficiency there. I can try to swing it a bit more, but I’d like to me able to actually measure this.

    • Dave Lusty

      “Why would you use/rely on that more than something that is derived from a device right there on your foot?”

      Simple, it’s because GPS is a direct measurement. Your location is a known measure based on those signals from space. Your new location when you move is also known and accurate, allowing the difference to be reliable.
      The pod on your shoe, attached to your laces uses accelleration to guess how far your shoe has moved. It has to be calibrated for your specific type of movement and running style and speed. What this says to anyone with an inquisitive mind is that it’s not actually measuring distance at all, just doing a great job at pretending to do so. As a result, if ANYTHING is different about today’s run, the data is meaningless. Your laces may be looser, different shoes, different weight, pain in your leg, different terrain, different pace – all of these things will throw off a foot pod. With GPS it’s affected mostly by weather, but this effect is consistent accross a timespan of an activity so although you may show as 10m too far to the east, every measurement is 10m too far to the east so distance and pace are unaffected.
      Those are besides the point though, using either of those things which measure distance and time to calculate power is madness. Using heart rate to somehow determine how much energy is left in your tank is also bonkers as Ray demonstrated in his testing. One day we may be able to collect enough information to give a reasonable number, but we’re not even remotely close to that today and personally I feel if the number is meaningless we shouldn’t show it to people who don’t know any better. This is, after all, just another attempt to add more features and sell more watches. The idea that Garmin are trying to somehow improve fitness with this died off long ago. They still haven’t explained running or cycling dynamics data, instead leaving it up to rumour and inuendo to fill in the gaps with guesswork. The people trying to make themselves 50:50 while running or cycling show this perfectly – the body doesn’t want to be symetrical. Most people can easily see the harm in trying to train to be 60:40 balance because it’s “unnatural” yet think changing your natural rythm to 50:50 is perfectly fine. We’re not heading in a good direction for fitness or training here, it’s sales and data junkies pure and simple. Garmin Connect doesn’t even display any of this in a consumable and actionable way!

    • Arjen

      But GPS is not just a direct measurement. The actual measurements from the GPS are used in a calculation (a Kalman filter for example) and the filtered value is what is reported back to you. The filter is a bit slower to respond to direction changes, so even if the sky is clear overhead, corners will not be as sharp in the GPS track as they are in real life. The upside is that the GPS track will be smoother when you are moving in a straight line as direction changes arising from measurement error are somewhat discounted.

      The resulting GPS track can be pretty accurate, but that is not because every measurement is highly accurate, but rather because the algorithm is good at filtering out errors. When you zoom in on the actual data points you will notice that what you see is an approximation and not a direct measurement. Does that make the data unusable or something to not rely on? Definitely not, but you have to understand the limitations of GPS just like you have to do with something like Runscribe.

      When I run intervals, I use a footpod (calibrated via GPS) to get better pace information as the reported GPS pace varies far too much due to corners, bridges and tunnels in my neighbourhood to be useful. For long runs where the pace is more even and I don’t need to know an approximate pace at every moment GPS is more than good enough. As you noted the footpod is not measuring distance directly and is highly dependent on running style and speed, so that is definitely something to keep in mind.

      For the best results when running intervals a stopwatch and a 400m track is all you need, with feedback every 100m if you want. On a track both GPS and a footpod will fail if you look closely enough.

    • Rob

      You make some good points, if being a bit cynical.
      But the Body Battery is the future, man.
      AI which considers measured data from your sleep, fitness/action, and resting time and then estimate your energy reserves ? That’s Si-Fi man ! I want it !

      Garman MUST figure this out and issue firmware/app updates to make it work. This could be a real competitive edge. The next step will be adding in cognitive performance, and then suggesting what and how much sport to do, foods to eat, and when to go to bed…

      We are in the dark ages when it comes to brain/body optimisation.

    • Fred Wagner

      Look into the Whoop strap at whoop.com which comes closer to what you are looking for. I have been using it for 6 months now and it does a great job of measuring strain on your body, sleep, when to go to bed, your recovery levels the next morning; based on HRV, sleep, type of sleep, RHR. Fred

  5. Nathan B

    Whilst I can see why Garmin want to push people to buy more expensive versions of their hardware, I still don’t understand why they don’t offer a Phone-GPS option in this sort of device in the same way FitBit do.

    I know a lot of people that are casual runners, that wear a device such as this, and do the occasional Saturday morning parkrun etc which they record on their phone. These people aren’t going to go out and buy a £200 running watch, but by offering GPS utilisation from the phone, they’d be tying people into their software.

    Ray explicitly says… use the fitness band your friends and family use.

    If they’re all using Garmin connect… then there’s a good chance friends will too!

    • gingerneil

      I’d be interested to see how accurate the pace/distance is for runs captured on this. If they are accurate enough for that off parkrun, then I suppose you dont really need GPS.

    • David Tucker

      I’ve actually always found the Vivo line to have very accurate distance. My original Vivofit was almost always dead on and I’ve found the vivosmart 3 to be similarly accurate. If I were only a casual runner, I’d probably be good enough with it.

  6. Blackster

    Any CT Links Ray? i know the Europe site lags a little on new listings but nothing at all really surprises me. I’ll happily buy this, one reason to maximize my 645 battery life, and the 2nd to see why my sleeps so shit. I guess SPo2 would drop if i stopped breathing? Not that my wife moans about snoring, nonetheless would be nice to see what’s happening.

  7. David Tucker

    I think a few interesting things here. While I am not a fan of companies using people to ‘test’ their hardware & software, in this case getting as much real world data as soon as they can is only going to help them hone the technology more quickly. I also find the body battery information a really good step since what’s the point of using the full Garmin ecosystem if they don’t know the other devices exist?

    I notice you wear the band above your wrist. I’ve always worn my Vivosmart 3 below the wrist but I feel like that it may be more accurate where you are wearing it. Is that what you’ve found?

    • I agree that it’s going to take real-world data for SpO2 to get better. That’s in effect what Fitbit is doing by driving their Sleep Dashboard program from the Fitbit Labs division.

      As for where to wear it, I aim for about 1-3cm upwards from my wrist (towards my elbow). That’s the recommended spot.

    • PeterF

      Ray, have you ever checked whether it makes a difference when you’re wearing a sensor on the inside of your wrist/arm vs the “regular” position?

      And, completely unrelated, any word if Garmin has improved on its stair counting algorithm?

    • Chris

      seems like consumer Sp02 is currently where optical HR was five years ago. The only question I have for it is – will it move to something more valuable than just another metric that people obsess over their particular number instead of understanding what it means ?

    • Chris

      from a strictly physiological point of view, it shouldn’t. Your radial artery is located closer to the bone than your vein, and the LED is detecting variations in pressure against the arterial walls as the blood pumps through arteries. a vein actually has small ‘gates’ that allow the blood to return to the heart under low pressure. the LED ignores these, so it shouldn’t matter whether its inside or out. What does matter is how tight you have your band. Too tight, you’ll cut off circulation, artificially increasing your heart rate (less oxygen to the limb, the heart will pump more to compensate) too loose and you get sweat under the band and that can screw up the readings.

    • RE: “seems like consumer Sp02 is currently where optical HR was five years ago.”

      Sorta, but it’s also tricky.

      Unlike then, almost all these companies are looking to do something with respect to FDA approval of the SmO2 readings. Of which the certification for that is actually very clear cut.

      But as Garmin pointed out during my meeting with them way back in April about the topic: Said FDA certification for SmO2 on a medical device is actually not that high of a standard/bar.

      I linked to the test previously, but you can see you can have a heck of a lot of variance in there and still pass. Not only that, but the certification test is done with the person sitting in a chair and practically strapped down. A far cry from someone hiking on the sound of a mountain at 25,000ft, or even just rolling around their own bed at night.

      And that’s where it’s gonna get tricky. Our mental expectations are aligned to ‘medical grade’, but I think we forget that most people getting SpO2 readings are in very well controlled circumstances, and even can fail.

      I remember super clearly this past winter when my daughter had to spend a few days at the hospital and watching the SpO2 machine keep failing to track her properly (the French nurses refused to just let us disable the never ending alarms). This was what appeared to be a fairly high-end machine by a major manuf. The point being – even in a hospital setting it’s not quite perfect.

      What I’d rather see is Garmin aim to find a way to make it more clear when false-positives are occurring. By actually recording that data in a trending sort of way, it makes it easier to identify them. Which is why it’s still baffling to me that individual non-sleep-driven Pulse Ox measurements aren’t actually stored (or at least shown) anywhere.

    • Chris G

      No, that is wrong regarding SaO2 measurement in almost all important ways. Transcutaneous SaO2 measures haemoglobin saturation by comparing absorption of wo different light frequencies in capillaries. My concern with nocturnal SaO2 measures is that in a population with a low probability of disease, most ‘abnormal’ readings will be false positives leading to distress, anxiety and mostly unnecessary medical testing, at a cost to the person and society in general. I’m a respiratory and sleep physician.

    • Chris G

      Just to be clear, the ‘wrong’ bit was by Chris above DCRs comments. Though I am wary of ‘sleep disturbances due to allergies and atrial fibrillation’. Be careful not to believe the hype from the companies in an area you are not expert in…..

    • Bikeman

      Garmin doesn’t think there’s a problem with stair counting so there’s nothing to fix. If you need a more reliable stair counter, Garmin is not your answer.

    • Eli

      Seems like we need more data of SpO2 to know how to use it. Even HR data we don’t fully know everything about going by the fitbit study that came out recently:
      link to digitaltrends.com

    • Chris

      I should’ve clarified – I was discussing Opitcal HR at that point, not Sp02, relating to whether or not the HR would be different based on what side of the wrist you wore your watch on. To me, consumer Sp02 is a very niche product that likely will give more anxiety than usable data for the forseeable future.

    • Chris

      that’s optical. I can type, honest.

    • Matthew

      I’m also a doctor and completely agree. If it is measuring Ray’s sats at 93% at rest, and presumably at sea level, it seems highly unlikely that is an accurate result. Unlike heart rate your oxygen saturation levels shouldn’t vary much unless you are seriously ill! Doctors and nurses are trained not to believe a saturation monitor unless it is correctly sited and giving a sensible reading (most machines will show the pulse waveform as it helps check there is a good signal). Measuring during sleep relies on a lot more than a few random measurements. If a patient came to me with readings from a sensor like this I doubt it would really help me, but it will cause anxiety.

    • Fwiw – last night I got a number more readings than the night before. Readings very frequently in roughly 20-30 minute batches of one-minute frequency. Data all over the place between 85% and 96%. It reported the average value at 90% for the night.

      When doing measurements seated on a couch, I pretty consistently get 96% readings. When something fails (usually for a photo), I consistently get 94% readings.

      Notably, the unit really seems to be struggling with figuring out the times I’m asleep. It showed me going to sleep at 10:58PM. I actually went to sleep at 2AM. I was working quietly on my laptop in bed for those three hours, which may be why it had issues. It did correctly pickup my wake-up time at 8:32AM.

    • Matthew

      Unless you are at altitude a healthy person should usually have a saturation level in the region of 95-100%. It doesn’t even matter how fit you are. It will only be lower if you have significant lung or heart disease (plus a few other rare problems). Even someone with asthma will have normal saturation levels unless they are very ill. They can dip lower during sleep but an average of 90% would be significantly abnormal. A proper monitor usually has an accuracy of +/- 2-3%. The technology involved in a wrist sensor must be different to a medical device. They actually measure the red light passing through the tissues and the sensor is placed opposite the light source. A wrist based device must be measuring reflected light. I’ll have to do some studying!

    • Mike P

      Unrelated to the new features, but on “struggling with figuring out the times I’m asleep” I did notice on my Vivosmart 3 over the last 10 days a tendency to suddenly get it all wrong. At times even my corrections get ‘corrected’ in the wrong direction again. Suppose its the ‘magic’ on the backend, as the last device firmware update is of June 26, 2018. Potentially, with the Vivosmart 4 rollout and related adaptations things are getting wobbly a bit at the moment..

    • David Tucker

      Hah this sparked quite a discussion. I was talking this over with my wife, who is a doctor, just to see her thoughts on the potential of all of this data. We can all agree, I think, that clearly the devices themselves are not close to being accurate enough for medical purposes but she thinks there is a ton of potential here for early detection of a large number of common issues.

      I think in 5 years we could be having very different conversations about these features.

    • Chris

      As a fellow medical professional, I agree. Between people wearing it too tightly or too loosely, not cleaning their sensors properly, and the apparently sloppy tolerances of the device itself, this seems more proof of concept than actual usable diagnostic data. But , as I alluded to earlier, these are the exact conversations that we had regarding wrist-based optical HR 5 years ago.

      Matthew’s right – if your average Sp02’s were at 90% and you weren’t in Boulder, CO or the alps ? You’d likely be intubated and given a nasal cannula.

    • Data continues to be the same in terms of PulseOx. No substantial differences between them.

      Two random notes however:

      A) Some folks have looked at my Fenix 5X+ photos and wondered about lower PulseOx values there. In that case I was legit in the alps at high altitudes (10K ft+). However, the ones here with the VS4 were all at/below sea level.

      B) While PulseOx on the VS4 hasn’t materially changed for me, I will say that Body Battery has. I’m just not sure it has in the right way. But it has changed. For example last night/this morning for some unexplainable reason the kids actually slept in, so I got like 8.5hrs of sleep. Crazy amounts. Yet, my Body Battery only peaked out around 55, rather than earlier on hitting 100 on only a couple hours of sleep. On the flipside, yesterday was a long-ass day with almost no food or water and i did a 3hr ride and an hour run, plus other riding and trains and just never ending. So perhaps my body hasn’t recovered. hard to know.

    • Rob

      Really ? I have one on my S6 phone, and there are dozens of them on Amazon, and if I went to an ER and compared with the expensive hospital gear, I wouldn’t be within 1 or 2 % ?
      Seems like commodity tech at his point.

    • Rob

      Wow, I was responding to comment #40 about spo2 detectors. Comments seem to get stuck just anywhere here ! Sorry.

    • Rob

      Now I am having doubts about SPo2 tech in this band. Matt makes sense, as Chris was alluding to: a reliable light on one side of the finger or an earlobe, with a reading on the opposite side must be much more reliable.
      Also, I experienced reading around 85% last weekend after several failed attempts (with my s6 phone), and I was at the beach with all that fresh air !
      Same problem yesterday, tried wetting my finger a bit. Got my 97%. Dry skin.
      Ray ! Try some moisturizer !

  8. Dave

    Isn’t the Body Battery just a rehash of the Nike Fuel concept? How is this much different and/or improved?

    Shane Miller has commented that companies need to get better at presenting the vast amounts of data that they have on their users, to their users in a presentable format. Is a single number really enough to inform users whether to go out and train or not? As Ray saw in this article, fitness companies also seem to neglect how much of a toll strength training (which is a massive industry on its own) takes out of a person and how it may affect future cardio workouts.

    I would be interested in peoples’, Ray’s, comments and thoughts on how these companies should use and present data to people. With the advancement in AI there should be better way to make the data usable.

    • If I remember correctly, Fuel was simply calories rebranded. So fuel points were totaled over the course of the day. Versus this decreases from 100 (or another value).

      I think in many ways Body Battery actually goes to exactly what both Shane and I have been talking about. Again, at least in theory. If they can work out the kinks in it – it may be a really good first stab at having some metric to focus on. The one benefit here actually is that it’s in theory not entirely Garmin specific, but rather riding on the coattails of the Firstbeat – so Suunto is doing roughly the same thing as well. I’m sure we’ll see other watch companies that leverage Suunto also utilize it down the road.

    • Chris Ga

      Have you been looking into what the Whoop is doing? For a noob like me, it seems they are ahead of Garmin and Fitbit when it comes to body battery, restitution, recovery, etc.

    • Jim A

      I couldn’t agree with you more Chris. I am hoping that Garmin continues to head in the same direction. I’m sure they have a considerably larger database and could draw on that for some effective metrics. It would be nice to get those features without the ridiculous $30/month that Whoop currently charges.

    • Carlos Martinez

      Is Body Battery just mowing the lawn in a different direction that Whoop with “strain”?

    • Hendrik Santo

      I try out the body battery for past two days, I think it still need improvement. Because it shows me I drain energy when I was taking nap for 1 hour.

  9. Eli

    So body battery is really “Body Resources” from firstbeat which is also on the Suunto 3 Fitness:
    link to firstbeat.com

    Which looks like a newer version of:
    link to firstbeat.com

    • Eli

      I do wonder if this new metric includes the data behind the overnight recovery test:
      link to firstbeat.com

    • Good point, same feature, but rebranded.

      However, I think the differences are actually quite a bit more substantial. First, the Garmin is sampling HR at 1-second intervals 24×7, whereas the Suunto isn’t. I’m not sure that’s making a massive difference by itself, but it’s worth noting.

      Second (and far more importantly), Garmin is actually saving this data and making it accessible from the mobile app. Suunto isn’t saving the data anywhere, so you can’t trend with it or really do anything beyond the ‘now’ shown on the display of the watch.

  10. Heronimos

    Aai.. They still integrate the device with the band. And de bands are a problem with all vivosmarts. My vivosmart 3 is back to garmin for the third time within a year… This time a broken band (and a lot of discussion). The last 2 times I was on time and could return because it was clear that the band would break in a few weeks time….

    I thought they would get smart and save the expense of warranty on all those devices.

  11. injuneer25

    I’m wondering if anything has been done to address the stairs climbed? Mine constantly had me climbing 10b flights of stairs in my single story home.

    • David Tucker

      Did you ever call Garmin about it? My VS3 had an issue with the altimeter that was giving me crazy stair counts so they replaced it and it’s been fine since.

    • injuneer25

      Contacted them this morning after your suggestion. Should have a replacement soon. Thanks!

    • David Tucker

      I’m glad 8) If there’s one thing I can say about Garmin, they have excellent customer service.

  12. Markos

    So when should we expect Vivoactive 4 with these new features (SpO2 and Body Battery?

  13. Andrew

    Just a quick question. I just picked up the Fitbit Alta HR mostly because the vivosmart 3 was much larger. Do you know the width of the vivosmart4? I’d much rather this over the Fitbit assuming it’s in as slim as advertised.

    • On my wrist, sticking a ruler below the band, it’s about 11-12mm wide. I’m sure Garmin’s site has an official measurement, but that’s the unofficial DCR one for ya. :)

  14. Uwe

    Is ther a Heart rate broadcastibg function over BLE? Thanks!

  15. Hunter Marshall

    You always provide the best reviews, thank you for that. Does this device, or are you hearing of any smaller devices, that can connect to a chest strap? Currently forced into having to use one of the watch series to be able to do so, besides using the all day heart rate from the smaller devices. Thanks!

  16. Harris

    Interesting. I’ve been looking at the Whoop device which factors all stresses in and gives you a recommended recovery goal. This kind of looks like it is going down the same path, but a cheaper version. There are some great HRV apps out there, but integrating the data into a dashboard with all your other health and training data in a useable format is going to be a ‘holy grail’ I think…

    • Eli

      Whoop seems to do way more with recovery data. This just shows how much strain your body has been put in and less knowledge about how much you recovered from the strain just that if you aren’t straining you must be recovering but doesn’t actually check your recovery.

      Like the difference between Firstbeat Sports (like Whoop) and Firstbeat Athlete (like Vivosmart 4) (ignoring how Athlete doesn’t exist anymore)

  17. Jjj

    I wish they kept the always on screen like the HR i have

  18. Bene

    „But that story shifts a bit with the Vivosmart 3. Here, the goal is more about using the data “

    The Withings Pulse had SpO2 years ago. If it measured bullshiz, it also showed 94% :-)

    Garmin really should hire some withings guys. Not only for the small trackers, but especially for the Index Smart Scales connectivity

  19. José

    Garmin either needs to fix their back end integration of Garmin devices or come out with one device that captures everything. Even within the same device family, it is a problem. I had my Fenix 3 replaced, and now all old records are gone.

    Also, Garmin needs the ability to better filter/change/delete invalid data on Garmin Connect. My Vector 3 pedals are just filling my Garmin profile data with garbage. No Garmin, I’m not hitting 60k Watts. Now all my performance metrics are garbage with all my record setting Vector 3 pedals.

    • Do you have an example of a 60K watt activity on Garmin Connect? I only ask because I know they have filters in place above a few thousand watts that discards those values on GC (unless something has changed).

      As for device replacements and metrics, that’s largely what Physio TrueUp does on most devices from 2017 onwards (i.e. Fenix 5). There’s still some gaps there, but it’s pretty close. The one main gap is just configuration of devices (like settings).

      But as you can see above, despite it getting Body Battery wrong (or at least not to my liking), it did actually pull in my Edge 1030 ride data automatically.

    • José

      I’ll have to look for the 60k Watt ride, but here is a recent ride.

      Regarding Physio TrueUp, maybe that is the answer. Most of my devices are older and the lack of integration is an issue.

  20. Any thoughts as to whether Garmin will push the Body Battery feature out to the fairly current high end watches such as the Forerunner 935 and Fenix 5 series?

    • I sure hope so.

      I’d think a fair line in the sand for that feature would be Fenix 5 Series and FR935, as well as FR645 and Vivoactive 3. In other words, anything mid-range and higher the 2017 watches.

  21. Stephen

    Ray, have you gotten to sample the new versions of the Vivomove HR that released today alongside this? It looks like they’re mainly just different colors, but I really want to like this watch and was hoping maybe they made some tweaks to the screen to make it readable in sunlight.

    • I haven’t. My understanding is that it’s purely a color scheme thing. I’m not aware of it addressing the sunlight bit.

      I did hear though that they continue to be of Garmin’s best sellers.

  22. Christian ¶:-)~

    would be interesting what happens when i use Fenix 5 and Vivosmart 4 together?

    Vivosmart 4 as Primary Fitnesstracker for steps, Sleep, etc and Fenix 5 for running or Edge for cacling, will be Physio True Up works in the Background to update the Body Battery value?


    Christian ¶:-)~

    • Will

      I’ve been wondering the same thing! If I wear a Vivosmart all day, but the Fenix I already have for workouts, will it all merge together in Connect?

    • In this scenario, if you record a workout on the Vivosmart 4, you’ll have duplicate works in GC. But if you just wear it for the ride/whatever but don’t start a workout, then the data is blended nicely.

      Or, even if you just take it off and only wear the Fenix for the ride/run, then the data is also blended nicely.

      That’s mostly what I’ve done lately is just use the VS4 as my everyday unit, while doing workouts with the Fenix or Edge.

    • Will

      Hey thanks! That’s the answer I’ve not been able to find anywhere … I think I’ll finally switch from the Fitbit over to Garmin so I can use just one app …

      Thanks again!

    • Dustin

      Can you please confirm this? I believe the Vivosmart website says to leave both devices on. Do swim workouts count as well? Also, could the Vivosmart pull in third party workouts, like TrainerRoad?

    • Philippe Stein

      After a few months, could you confirm the scenario is still valid?
      I have a 935 that I use only for workouts and I’m interested in having an additional tracker. In that case that would be the Vivosmart, for sure.

  23. Rob

    Hey Ray. Will you be reviewing the Amazfit Stratos watch? I would love to see an in-depth review to make it an easier decision to buy it or go with some Garmin product ie FR 645M which costs more.

  24. Remco

    Now this is an interesting one. So if the focus is to measure spo2 especially during sleep it would be nice if it could detect snore. During snoring the breathing efficiency is reduced. This should be visible as a somewhat lower spo2 during the sleep. But it must measure more points for that. As a trick it could if snore is detected vibrate and the wearer might change its sleeping position reducing the snore.

  25. Filipe Teixeira

    As always brilliant review.
    In your opinion how does this garmin matches with the fitbit charge 3? I have the feeling that the fitbit is a way more polished device than the garmin. Am I wrong?

    Thank you

  26. Ivan

    maybe the future is patch (Kenzen and similar) or armband (SpO2 readings, sleep, and more advanced features).

  27. xcyclist

    I think it is a slap for Fenix 5+/5S+ buyers as cheap vivosmart has more cutting-edge functions than the top of the line watch. The fact that Vivosmart has smaller battery and smaller body invalidates any reason why SpO2 cannot be added to 5+/5S+.

    • Ivan

      That’s also true.

    • I agree it sucks that the entire 5+ line didn’t have Pulse Ox.

      But, it’s probably also to keep in mind timeframes here. While it was ‘only’ 2.5 months ago that Garmin rolled out the 5+, that’s quite a bit in manufacturing time-frames.

      I think it’s often easy for us to say ‘they could have added it here, or there, or anywhere’, but there’s usually a lot more that goes into it than that. For example – as I noted in the Fenix 5+ review, production there actually started mid-April, not June. So now we’re talking almost 5 months ago. They stock-piled units for two months while waiting for software to finalize (unheard of for Garmin).

      And ultimately, like any tech, things kinda have to start somewhere. There has to be a line in the sand eventually. As I said above though, I’m actually more interested in what comes going forward. I’d fully expect/hope that every watch from here forward above $129 has Pulse Ox. Otherwise, I think it’s a bit of a case of Garmin showing its colors of too-often re-using hardware in different models for too long.

    • xcyclist

      I do understand the development and supply chain challenges. On the other hand, vivosmart has those challenges too, perhaps, not to the extent of Fenix 5+ but still has them.

      It would be great if Garmin would make its device lineup clear.
      It feels like they do not plan, but take a rush measures to respond to the market.

      Why did vivosmart HR+ with GPS exist, was it a temporary solution (plug in the lineup hole?) before Vivosport has been released? Only to make next iteration of vivosmart missing GPS? So it has no clear use case within the same device subbrand as it gets and loses critical features (GPS) along the way. That is especially annoying for me as a customer as I cannot rely on the fact that if my girlfriend has Vivosmart HR+ the next Vivosmart she buys will get all the same functions being updated/improved and some new added in addition. That forces me to read the reviews again and again every time and if I do so there is always a chance I find Fitbit or other product better. It is especially important in lower cost devices as the customers are usually are not that keen to dig into details as users of more advanced devices.

      One of the clearest marketing messages iPhone/Samsung Galaxy has is that you get everything you had in the old model is better in a new one. And that is where Garmin is messing it up completely almost every year.

  28. Bruce Burkhalter

    Does this (and other bands) monitor sleep well if you only wear it at night? Interested in sleep data but don’t want to wear it all day. Or should I look at some of the dedicated sleep monitors?

    • You can definitely just wear it at night. However, Garmin recommends that you start wearing it about 1-2 hours prior, so it can more clearly find that ‘cutoff’ point between wake and sleep. Obviously, Body Battery wouldn’t be correct. But PulseOx should work just fine.

    • rabbit

      PulseOX isn’t reliable for me. If I measure 10 times in a row, I get values in a range of+- 10. Tested it also twice at doctors with a medical Ox device: both times my 5x+ > 8 off. It’s not more than a toy. If there is no medical certification for the PulseOX and the measurement get more accurate, this feature is …Don’t really understand, why people are keen on this

  29. Aleksander H

    I hope garmin intends to bring a similar update to the Vivosport line. I absolutely love the one I have as it’s the perfect blend of features and size for me.

  30. Marius

    – Does Vivosmart 4 pair with Garmins speed sensor, cadence sensor?
    – Can Vivosmart 4 be used as HRM to Fenix 3, connected via ANT+?

    • It can’t connect to any sensors, but it can broadcast out ANT+ to other units. I used it the other day for a ride. It did stop broadcasting at some point, though I haven’t dug into why (not sure if I perhaps hit a button or something).

    • Thomas

      Works fine as HRM sensor with Fenix 3 once you disable autostart for exercise tracking and gesture-based screen activation. HR for running consistently shows a ramp up of 10-15 minutes until it gets quite accurate. Even body battery seems to make sense!

  31. Jason

    Still no external HR strap connectivity? I find these wrist based HR readings are great: if you aren’t exercising. My Fitbit Charge 2 was all over the place on the treadmill last night, and I was WALKING, not bouncing around. My arms aren’t particularly hairy and I wear the device as tight as I can while remaining comfortable. Sure, I was sweating. If Garmin, a company who makes their own chest straps, don’t start integrating the ability to connect to their smaller wearables, I won’t even consider them.

  32. Alex Lee

    I love your fantastic post. I need swimming information though.
    I’m getting sick and tired of using a stupid gearfit2 pro. I’m planning on buying the new vivosmart4.

  33. Markus

    Hey , anything new on the accuracy of the “body battery” ? As Garmin states that it will take some time to “learn” you ?
    Also saw that a new firmware arrived ?


  34. David Thomas

    Is this suitable for swimming in/does it give any swim metrics?

  35. Phil A

    Looks like they added pool swim to this model.

    Was wondering if you have tried that yet?

    Also was wondering if it has a way to hit a lap button?

    On my old vivoactive HR that died I could hit that button to tell it I was resting, and then again to start the next interval.


  36. Wayne G

    My wife got her Vivosmart 4 and it seems great(she is white). I tried it on and it doesn’t work(I am black). Garmin seems to frequently not test products on people of color. Is there any rhyme or reason for this? I have a Vivosmart 3 that works, have had a Vivoactive 3 Music it worked. Tried a Fenix 5 Plus it didn’t work. Is there anyway to get Garmin to do a better job? In this day and age it’s a shame that they are not inclusive in their testing.

  37. Frank

    I got mine from CT three days ago. There is very little I like about it. First off, it looks effeminate. Second, it does a crappy job of tracking steps and measuring heart rate while walking. Third, the screen/human interface is maddeningly inconsistent. Fourth, after shipping both the Vivoactive 3 and Vivosport with the same charging cable, they reverted to inventing a new one for the VS4. I was curious about the body battery thing but, if it cannot even accurately determine my pulse rate if I am not running or at rest, I can’t trust any of the other data it gathers either.

    Of they add body battery to the 935 I’ll upgrade to that from my Fenix 3. If not, I’ll stick with what I’e got. This thing is going back.

    • Hendrik santo

      I have return my vivosmart 4, the stress level is off big time compare with my vivosmart 3 and there goes the body battery accuratecy.

  38. Niklas

    Any idea if the heart rate accuracy for cycling is improved from version 3?

  39. Rob

    GREAT review, although now I’m disappointed it’s not the slam dunk I thought it was.
    Regarding inaccurate sleep data, did you respect these conditions ?:

    Device must be up to date on software. You can use Garmin Express to manually check for and install any available updates.
    Garmin Connect Mobile must be up to date. Use your app store to check and install any available updates.
    Device must be worn at least two hours before bedtime.
    Optical heart rate sensor must be enabled while sleeping
    If you have more than one device, you must wear the one marked as your preferred activity tracker to bed.

  40. Patrick

    Does this have a smart wake up alarm? I use a smart alarm clock on my phone, but my wife is getting tired of me waking up at 4:30 for masters. I am looking for a watch/activity tracker with a vibrate smart wake up alarm and I’d prefer to stick with Garmin.

    Will you be reviewing the Withings Steel HR Sport?

  41. fl33tStA

    Hi Ray,

    do you still have updates on Body Battery, cause Fenix 5 Users see the Option of Body Battery in Garmin Connect as you can see in the attached screenshot?

    Will it come to other devices too?

    Otherwise, will it work via Physio True Up when i use Fenix 5 for running and the rest of the day vivosmart4 as favorite Tracker?


  42. Donna Sullivan

    Bought this today (November 3, 2018) and I Cannot find Anywhere as to WHY my clock face will Not accept and correct to todays date. (time is ok).
    Anyone got a clue?
    I really want accessibility to Both date and time.
    Thanks to anyone who knows their way around this one

  43. JB


    The “touch responsiveness a bit cumbersome” problem, in your experience do you think that is something Garmin could fix with a firmware update or a hardware problem?

  44. Andrew

    Any chance there’s an in depth review coming soon? Christmas is coming and I’ve been eyeing this for awhile.

  45. JB

    Excellent news, thnx :)

  46. Melinda

    Still hoping for an in-depth review of this, last-minute shopper that I am…

  47. David

    Ray, can you give a bullet point difference list between this an the vivosport? Unfortunately your product comparison tool won’t let me compare them since they are in different categories.

  48. Kevin

    How is it’s 24/7 heart rate? Also what would happen if you have multiple Garmin devices monitoring HR, will you end up with more HR data points in GC? E.g. using a 735 and the VS4 (for oximeter) at the same time for sleeping?

  49. Anyone have an opinion on how accurate the heart rate on this is? Give how narrow it is I wonder if enough light would get in to interfere … Thoughts?

  50. Alex

    I train with an Edge 1030 and am looking for a wearable to match. I want to improve the recovery time and training level estimates of the Edge by wearing something that tells the algorithm a bit more than the data it gets during the workout. Would this one be sufficient or do I need to get a Fenix 5 for that?

    Thanks for your advice, I just used it to buy a new trainer.

  51. siddharta1979

    As a recreational cyclist (no races), for my workouts I have a heart band and a wahoo bolt.
    If I were looking for other type of data, based on non-work out days (sleep, resting heart rate, etc.), where should I look?
    What I mean is perfectly summarized by your conclusion: “the idea to take in all these metrics (sleep, resting heart rate, heart rate variability, workouts, activity, stress), and give you a simple 0 to 100 number: That’s brilliant. At least in theory,”
    What would be your recommendation for the “practice”? Get this tracker? Wait? Get a different tracker, like an apple watch? For example, with your conclusion in mind, what do you use?

  52. Kristina Boseva

    Hi Ray! I’ve been very hopeful that an in depth review of the VS4 is coming for a while now (incl. HR readings and sleep comparisons). Do you use it as a daily device still or have you dropped it all together?

    • Fred Wagner

      Look at the brand new Polar Ignite which DC just reviewed. I actually tried the VS4 for a week and returned it. I found it clunky to use, the HR not a good enough overlay on a chest strap or a Polar OH1 memory optical unit, and useless in the pool. Wrist based HR monitors are inherently inaccurate for some activities and the VS4 was no exception which is why I use a chest strap or the OH1 on my goggle strap against the temple for swimming. The sleep data was disappointing vs my Oura Ring. I am excited though to try the new Polar Ignite as Polar makes one of the better optical units along with Garmin but on the higher end side of wearable.

  53. Joriene

    Thanks for this review, which IMO is very in-depth for a hands-on (especially compared to most of the ‘reviews’ on the internet that basically read like press releases).

    I started using the Vivosmart 4 a few days ago and immediately noticed that my stress levels are significantly lower than the ones measured with my last Garmin, the Vivomove HR. I have a chronic illness and had a very busy day before I started using the VS4, and my initial Body Battery immediately after I started wearing it was 42 or so? I thought that was promising, went to bed, couldn’t sleep and barely made 5h40min, but then was surprised to see that after just 3 hours of sleep (sleep that suddenly got stress readings as low as 0 as opposed to my usual 20-25 with the Vivomove HR) my Body Battery was fully restored. Exactly like your chart. And because I was dog tired the next day, I didn’t move much, which kept my Body Battery number high in the nineties and caused my Vivosmart 4 to proclaim that maybe I should use my overload of stored energy to, I don’t know, move my ass a couple of hours before bed? Ugh.

    I really like the idea of Body Battery. I think it’s a brilliant and simple metric that has the potential to truly stimulate lasting behavioral change in the sense that it’s not just telling you to move (or ‘MOVE!’ as the Vivomove HR used to command), it’s also telling you to stop moving or stressing. It could identify stressors, like that long commute with traffic jams (prompting you to finally move closer to your work or take a bike). The band even alerts you to high stress levels, so in theory, this could teach people to avoid (a second) burn-out. That’s pretty huge, but only if Body Battery is reasonably reliable.

    So my question is: did the Body Battery data get better? I’m asking because I’m considering returning this Vivosmart 4. If the idea is only awesome in theory, but the data doesn’t add up, then I’d rather use a cheap Mi Band 4 for example until such time Pulse Ox and Body Battery are more than a nice idea. Thank you!

  54. David

    Looks like they just (recently?) added connected GPS capability. (Last night’s walk did not have it. But it showed up tonight).

  55. David

    I got the V4 Firmware upgrade last week, and it says it added Cycling Activity, and Connect GPS. Anyone seen any details about what that functionality actually does. The release notes are so brief as to be useless.

  56. bogordo

    I have a FR 945 since may and a VS 4 since monday (4 days). The body battery for FR945 worked “normal”, thus, 100% at the beginning of the day and 20-30 at night.

    Now with the VS4 as primary unit i have 100-90% during all the day, so something is not fine here. Of course i’m doing the same activity as always, can’t trust this stuff :/

  57. On GARMIN web site they said:
    connects to your compatible smartphone’s GPS for accurate tracking during outdoor walks and runs .
    link ( link to buy.garmin.com )

    so they add connected GPS feature like Fitbit charge 3 ?

  58. Anton

    just to say of you plan to use that device as ANT+ sensor… don’t. When its broadcasting the HR is just a joke – you are spinning like crazy and it measures HR like you are eating a sandwich!

  59. Eflodur

    Have u a comparison with the oura ring sleeptracking and bodybattery which they call readiness.

  60. Biswajit Sil

    If paired with a smartphone for GPS connectivity, will the vivosmart4 be able to at least map my run or cycling data.
    Can we push it to Strava as well through the Garmin Connect app?

  61. Pär

    Did you do a heart rate test on this device. When I compared it with a heart rate strap it was very “smooth”. On a short exercise (2.5 km) with 3 spurts, the spurts was almost not visible. When uploading to Strava I got a training impulse of 35 from the HR strap but only 19 from the Vivosmart 4.
    I haven’t found any settings for increasing the HR reading frequently on the device so I guess it should be optimized when manually starting an activity. But looking at the differences in the Strava fitnesses tracking it’s at big notable difference coming from the devices ability to measure heart rate. Or is this very common for all wrist based devices?

    • Damo

      I bought the VS several montha ago for a cheap but very small unobtrusive, activity and HR tracker.
      Sadly, I have found the HR tracking to be extremely inaccurate once my HR gets over 120 or so.
      It was a constant disappointment to do an hour session spin cycle, pulse rate sky high, sweating bucketloads yet it shows pulse rate less than 80 and my intensity minutes shows an additional zero.

      So I decided to lash out and get a Fenix 6S pro. What a revelation this has been!
      I wore both my devices to a 30 mins spin cycle session. While the Fenix6 was showing my pulse between 130 and 160 with a peak at 172, my VS was showing pulse rate of about 60-70!
      After that session, I have put the VS back in the box and will be disposing of it.

      The VS seems to be accurate during day to day activities , but as soon as my pulse rate accelerates it stops responding properly. The only thing I can think of is, maybe the narrow band is letting light in or escaping?

    • “The only thing I can think of is, maybe the narrow band is letting light in or escaping?”

      Correct, the challenge with all small bands is light leakage. In general bands have to be more tightly strapped than watches (though, those should also be snug). But even then it can still be challenging.

  62. Anthony

    I’m looking at getting a small activity tracker, for day to day wear (prefer the band to watch design). Interested in the heart rate tracking.

    Would the Vivosmart 4 still be a good buy? Or is a vivosmart 5 likely to be released soon (had wondered if there would be anything from CES, but doesn’t look like it)


  63. Jason Wilson

    Hey Ray, thanks for all you do in providing great reviews and honest opinions of so many devices.

    My question, I recently received a beautiful watch from work and as a result, will not be wearing my F6 Pro 24/7 as I have been. In wanting to keep an accurate body battery reading I am thinking about buying the Vivosmart 4 and wearing that on the days i want to wear my analog watch and always wear my Fenix 6 for workouts and not sure which to wear to bed? Thinking the F6 has the more updated sleep app.

    Will this work?

    Thanks Ray or anyone for the feedback.

    • Patrick

      Hi Jason,

      I’m looking at the same ‘challenges’ with a recent purchase of a mechanical watch to supplement my tactix Delta :)

      I want to be able to still track Body Battery, stress levels, HR and steps. I also want to still have relevant data and metrics to factor in when getting automated daily workouts on my tactix Delta.
      Finally I want everything to cross sync with my tactix Delta so I can seamlessly switch between devices.

      After some research (Garmin.com, forums, Facebook) my conclusion as of now is, that…

      – Body Battery is kinda device specific. It will not transfer between devices – only to Garmin Connect. That’s a bummer since Garmin states a ‘calibration’ period of up to 24 hours for Body Battery metric to display ‘correct’ values. And you should preferably sleep with the device as well. Makes it hard to shuffle between devices.

      – Recommended workouts are device specific, not factoring in data from other wearables. Also bummer.

      Note: It doesn’t make a difference if both devices have Physio TrueUP (instead of ‘only’ TrueUp). It still won’t work as far as I know..

      I will buy a cheap Vivosmart 4 today to see how it plays out in IRL :)


    • Jason Wilson

      Hey Patrick, how is the VS4 working with the Body Battery syncing with your Tactix Delta?

  64. Ray Russell

    Can the vivosmart 4 broadcast HR to Garmin edge 530 and iPad running Zwift/Rouvy?

  65. bionicwheeler

    Hi Ray. Just bought a Vivosmart 4 for ‘body battery’. I also ride outdoors using an Edge 520 Plus, and indoors using Zwift. I plan to wear the Vivosmart all the time, but have a question about body battery when I ride.

    Should I wear the watch like normal while I record the ride on the Edge 520 Plus, or should I also record a ride on the Vivosmart? i.e. would the Vivosmart reduce my body battery by the same number of points, or would it be more accurate by recording a ride, rather than letting it try to work out what I’m doing (and thinking I’m stressed rather than riding for example).