Apple Collected Over 18,000 Hours of Health and Fitness Data for Apple Watch

[69Mustang]

That's actually not quite what people say.

They generally meant "first doesn't mean anything unless you also do it right".

Apple's 64-bit chip was done right and fast.

No, I think they/we meant/mean first doesn't mean anything. Whether it's right or wrong, fast or slow, being first doesn't win any prizes, trophies, or brownie points.

Apple did a great job with their 64-bit chip. That it was the first is about as relevant as who made the first smartphone or mp3 player or haptic feedback tech for a laptop. Not relevant at all.

 

[japanime]

10,000 is a large sample size. Are you thinking about random samples needed to establish trends in larger groups? If so, you're barking up the wrong tree.

Apple is gathering the data needed to calibrate the Apple Watch across a wide variety of exercises, environments, and bodies. The part of the video I saw they were gathering data on calories burned. They don't need large random samples.
They'd need to get the Watch sensor readings vs. actual calories burned and adjust (or even help create) their exercise models based on that. 10,000 would cover a lot of cases if you were systematic (and Apple certainly looks systematic there.)

Calibration is an important aspect of developing any measuring system. 18000 hours is a very large sample for a system that really isn't all that important. A heart monitoring system in a hospital is important. A fitness gadget is not.

Most of my grad school engineering studies revolved around statistics, so I'll give it a shot.

It all depends on what you're trying to do with the data. Are you trying to collect a sample that is representative of the whole population or merely a subset? Then it's much more important that you have an evenly distributed sample that reflects the overall population than it is that you have lots and lots of data points.

You might be surprised by how few samplings you need to accurately reflect your target population.

Finally, it would probably be most useful not to use data from hundreds of thousands of different people over the course of two years. One of the most useful types of datasets is the type that tracks the same people over time. This is called a panel dataset.

Thanks to all of you for your replies to my question. That's what I love about these forums — if ask a valid question, you'll get some great answers!

 

[Arndroid]

Still overpriced and still ugly. That being said, I may pick one up for the ease of use and compatibility in the iPhone. But the Moto 360 sure looks good. I may need to compare.

The moto 270 amirite?

 

[Winni]

Try to imagine Samsung running 10,000 1–2 hour test sessions in a secret fitness lab over two years. Not very believable, is it? Neither the extreme pursuit of quality and usefulness, nor being years AHEAD of the industry instead of following 2 months behind the latest rumor about Apple...

I don't know if/when I'll want an Apple Watch, but I do know Apple goes way beyond the halfway spaghetti-on-the-wall product development practiced by the competition.

Right...
http://www.samsung.com/global/business/healthcare/healthcare

Never let facts get into your way. A company that produces real medical products surely doesn't test anything...

 

[Jakexb]

Incredibly excessive testing for a heart rate monitor. They were clearly hoping to release an oxygen monitor *at least*, given all the respiratory testing gear.

Really looking forward Apple Watch v2

 

[AppleScruff1]

Pretty cool, but expected. As a tech company as prestigious as Apple, you have to test it and perfect it otherwise your reputation will suffer.

Solid homework. Nothing surprising here..... reminds me of that glimpse into Apple's wireless/cellular testing lab they revealed after the ruckus about the so called "antenna gate"..... God, I hate that word.....

Testing and perfecting before release, like Apple Maps and iOS 7 and 8?

 

[whiteboytrash]

Most treadmills i have been on have a chart that shows what your maximum heart rate should be for various age brackets. Generally, the older you are the lower your max heart rate should be. I've never seen any other kind of measurement referenced. Like another person mentioned... when i watch cycling on tv... i've only ever heard them talk about how many watts they are generating. But i've no idea if thats because they have no accurate way of measuring their heart rate while riding the bike, or if its just because they don't think its useful. For me measuring watts generated would be like saying "i can bench press 200 pounds, and you can only bench press 150... therefore i'm getting a better workout than you." If someone is a small, fragile person.... they could be capable of generating far less watts, yet still be in far better shape than you.

Watts when cycling is the be all and end all of cycling metrics. Heart rate means nothing. You can measure heart very easily with a strap on a bike. Even when in a large group with cross talk on Bluetooth. Watts becomes very important when correlated with body weight. Cyclists attempt to drop body weight but improve power. You climb faster with less weight and more power. It will be expressed as watts per kilogram, ie 3.22w p/kg. Doping is so prevalent in cycling because its so effective with weight loss. 65kg's and producing the power or a a 100kg man makes a huge difference when going uphill for hour or more.

If you look at pro cycling these days they're all painfully thing and using AICAR for weight loss and power, micro dosing on EPO for recovery.

If they can get lactate testing into a watch in the next few years that will kill it. But not sure how you'll do a blood test with a watch!

 

[benlukes]

But only health data between the hours of 9am and 5pm, the rest of the time the phone has to be charged lol

Battery life + apple = biggest tech joke ever

Somewhat agree. If the watch can't monitor you while you sleep you aren't getting the full health picture. What's worse, reading Apple's information on the watch carefully seems to imply that your heart rate is only monitored when you use the fitness app or heartbeat "glance", which would be utterly horrible. Hopefully they only write it this way to assuage privacy concerns, and there is a way to turn on constant monitoring.

Also...if you can't wear the watch during the night that destroys another key feature besides sleep monitoring: silent alarm! Sleep, fitness, and taptic alarm are the three main reasons I would want a smart watch.

----------

I will admit, to me this really just says:

Our device cannot accurately monitor anything, like the proper medical expensive equipment here can.

So, we are collecting data, so that when our device reads something (all that it can, given it's just a cheap consumer device with a monitoring device that costs a few dollars)

We can then use this poor data, and try and match it up with this accurate data and give you some rough idea on the watch.

Oh, and "please note any health data displayed by the watch is to be used as a guide only"

We need something that is a close proxy to the face masks, because no one wants to wear a face mask when they exercise. I'm not saying this is it, but there has to be another option. If it is only 97% accurate, then so be it. It would be great if you could calibrate the watch to your own body to improve the accuracy. Anyway, even heart rate data alone is very useful. Athletes use heart rate monitors all the time as training aids. As long as the heart rate data is accurate, I'm happy.

Is that an actual quote from something relevant?

 

[TechGod]

Testing and perfecting before release, like Apple Maps and iOS 7 and 8?

But they don't do that for iPhone, iPad, MAC OS, IOS, etc…

Wish they spent that much attention on IOS.

Except Apple did not when it came to, say, Maps, or the iPhone 4 antenna. It's just an assumption on your part. But Apple isn't alone here. Garmin, one of the leaders in GPS, and GPS watches, put out the FR620 watch last year. At launch it was notoriously inaccurate. People were posting GPS trackings show them running in the middle of a building. Pretty sloppy.

So maybe it's expected, but companies don't always go the extra mile so to speak to get it right or move the tech ahead. It's impressive what Apple is doing here. It demonstrates they want to be a fitness leader. It's probably why Nike nixed it's own hardware. It' wasn't doing nearly the research Apple has done.

To what end all the research makes the Apple Watch a better product remains to be seen. LED heart rate monitors in other watches are inaccurate. I'm most interested to see if Apple has solved that problem.

So all of you KNOW for a fact that Apple didn't test their software?

I know the launch of iOS 7 and 8 was pretty bad, however the public testing betas have helped Apple(at least in my experience) iron out a lot of the bugs.

 

[OtherJesus]

I heard Samsung has 18,001 hours collected.

 

[benlukes]

I love the message below the video saying "For those outside the USA, here is a low quality version".

Not entirely sure what is implied by this, but I'm guessing it is for those with a poor connection? If so, why not just say click here if you have a slow download rate.

Probably because the ABC video isn't available outside of the US. Just a guess.

Given that most developed countries seem to have much faster internet speeds for a lot less money, I don't think the folks who posted that would say it for any other reason.

 

[SteveBlobs]

Um, why exactly did she conduct the interview from a treadmill?

 

[AppleScruff1]

So all of you KNOW for a fact that Apple didn't test their software?

I know the launch of iOS 7 and 8 was pretty bad, however the public testing betas have helped Apple(at least in my experience) iron out a lot of the bugs.

I'm sure they tested the software and it was far from perfect before it was released. You made the comment about Apple testing and perfecting before release, I gave a couple of examples that showed they really didn't do a very good job at it.

 

[FlatCable]

You'd be 100% wrong. Unless you think he's been trolling the forum for 5 years (member since 2010). I'll give you this, for a new member of the forum, you are more than willing to go out on a limb and be completely wrong. That's something.

Just to point here that I don't care.

Apple uses aspiration marketing better than any company I can think of. You can take products from Apple, MS, Samsung, Google, or whomever and achieve the exact same things. Apple is just hella better at making the association between your activity and their product. Along with making a quality product, it's one of the reasons Apple is so successful.

Every company does marketing. I don't see why Apple shouldn't do marketing for themselves, as for others, do marketing against Apple all the time...

How many articles about the Apple Watch alone that you want that are plain wrong, and not just innacurate, but prejudicial against the W... do you think that this is by change?

But marketing alone doesn't matter. Ask Samsung.

Pretty much all brands say up to X hours. As with any product, including Apple's, MMV based on usage. Apple realizes this. That's why Tim said people would most likely need to charge every night.

Normally they point in "days"...

Now, did you check up the apple document: https://www.apple.com/watch/battery.html

They could say "2 days", and don't even lie, with their 48 hour battery life... and that's constantly checking the time (which can be a notification) every 15 mins. Even on your sleep. And the funny thing is that it's what some watches are only good for...

Why don't they do that? Simple, their costumers know them, and we know that they give us realistic times. That means they don't have anything to hide.

Good to see you've taken it upon yourself to be arbiter of who Apple is for.

Sorry, if you want to be fooled, there's no use in looking to Apple, just buy Android, they are always awesome in the specs.

Welcome to MR. Long, long ago we came to the conclusion that "first" means absolutely nothing; tantamount to getting the first post in a thread.

It's not the first 64 big CPU, not even the first 64 bit ARM ever...

It's just to show that just because other companies manufacture the CHIP, it's an Apple proprietary chip.

 

[gavroche]

If they can get lactate testing into a watch in the next few years that will kill it. But not sure how you'll do a blood test with a watch!

Ha... a needle that randomly jabs into your wrist!
Cycling would be one of my favorite sports, if it wasn't for the drug/disqualification issue. It's one of the only sports i watch that i can watch, see someone win.. and then find out months or years later that the person i watched win, is no longer called the winner. Kinda cheapens the viewing experience, since at the end of the race, you can't even be sure if the winner will remain the winner. And half the time, when they disqualify someone for a drug issue, they give the win to the runner up, who most likely also had drug issues. There should be a limitation on how long after a race someone can be disqualified.

 

[Trius]

If it were journalism and not promotion, they would have asked how the best way of monitoring how hard you're working out (which the guy from Apple said was the face mask) is related to the Apple watch, which measures heart rate, not respiration.

It's collaborative data. He says in the video the face mask helps them determine more accurately how many calories you've burned.

 

[Trius]

I will admit, to me this really just says:

Our device cannot accurately monitor anything, like the proper medical expensive equipment here can.

So, we are collecting data, so that when our device reads something (all that it can, given it's just a cheap consumer device with a monitoring device that costs a few dollars)

We can then use this poor data, and try and match it up with this accurate data and give you some rough idea on the watch.

Oh, and "please note any health data displayed by the watch is to be used as a guide only"

Your thinking about this is a little backwards. They're using machines that have already been developed, tested and proven to monitor certain aspects of health and using the data to calibrate this new product.

Them using this equipment to fine tune data collected from the watch does not diminish the efficacy of the watch's ability to monitor anything.

 

[mmartinezs]

Good to know

 

[FlatCable]

I'm wondering.

Do you actually buy this story as it's being presented to us?

A secret building, full of camera photogenic fit employees, all working out on gym equipment for 2 years that looks brand new, all totally unaware as to what they were doing there?

Does not any of this look or sound in the slightest bit fishy?

OMG, so much wrong in your comment.

Photogenic fit employees? Not all of them are photogenic, fit, or either, plus, after all that exercise, you're bound to be fit.

And what's wrong with the equip? Looks like my gym's equip.

 

[H2SO4]

I heard Samsung has 18,001 hours collected.

They have likely so much more medical research carried out than Apple it just isn’t funny.
Post 104……...

 

[pdaholic]

Today my coworker introduced me to his Galaxy Gear S watch. That thing is pretty impressive! Has GPS and a SIM card. I know I'm supposed to pledge allegiance to the Apple flag and all, but wow.

 

[OtherJesus]

They have likely so much more medical research carried out than Apple it just isn’t funny.
Post 104……...

Ok, I was wrong. They have 18,002

 

[H2SO4]

Ok, I was wrong. They have 18,002

LOL, chortle, guffaw guffaw.

----------

Today my coworker introduced me to his Galaxy Gear S watch. That thing is pretty impressive! Has GPS and a SIM card. I know I'm supposed to pledge allegiance to the Apple flag and all, but wow.

Equally nice to look at in a different way. Having said that unless Samsung can come up with something phenomenal, my other Apple purchases would make AppleWatch a better buy even with it’s inevitable manufacturer imposed limitations.

 

[ZipZap]

There is a big difference in how much research should/could go into an entirely new product type, than updates to an existing product. Unless you are proposing that Apple should not update their iOS on an annual basis (since you seem to be suggesting that they should put two years of research into that as well).

----------



Good to know that you consider the watch nonsense. Unfortunately, someone else chimed in before you, and called iOS nonsense, so Apple already chose to accept their arbitrary opinion and give up on it.

No there isn't. Apple should put the same devotion into fixing the existing issues. Patch after patch yet issues persist. Oh and in many cases we wait months and months for the fix that fixes nothing.

Right now the watch is nonsense. It's has to prove itself useful. At $349 I don't think it makes sense. Do you realize how small 42 mm is? Do you know you carry an iPhone that runs apps and tells the time? Right now, nonsense seems appropriate. But it's your money. Waste it as you see fit.

 

[Tech198]

Blue??? Why would they turn blue?

they wouldn't, they'd turn red.