I couldn't begin to tell you why this is what they had me try, but so far it seems to have worked for my issue. I have had a few minutes log in the exercise ring, without using the Workout app, which is what I was after. We'll see if it keeps working for a longer time period.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
Heart rate updates after 1.01
- Thread starter Pagemakers
- Start date
- Sort by reaction score
I couldn't begin to tell you why this is what they had me try, but so far it seems to have worked for my issue. I have had a few minutes log in the exercise ring, without using the Workout app, which is what I was after. We'll see if it keeps working for a longer time period.
Wow, this is an interesting experiment. Good work!! I agree with your conclusion... I'm finding it very hard to trust the activity data following the update. Apple MUST address these questions marks surrounding accuracy soon... I wonder if we'll get any watch OS update info at WWDC. I'm guessing we'll get more of the 'lets pretend everything is fine' approach.
I agree with you. Accuracy isn't that important for me. But consistency is. As an analogy, let's say the watch was able to measure the calories of the food you eat and one day it says the slice of pizza is 50 calories and the next day it says a similar slice of pizza is 400 calories. This is the same instrument telling you wildly different results of basically the same thing. So if you're trying to watch your calories, how do you know what to believe? I'd rather have it be less accurate but at least be consistent with itself. That way you could gauge whether you are making progress or not. As it is now, it is completely dependent on the Workout app which to me makes it not useful at tracking calories when the workout app is not engaged and you are just going about your life - something Apple advertises this does.
GDMORNOT...My test lasted 10 minutes. If it lasted for 2 hours the calories would be out by the hundreds. If I did that daily at the end of the week they would be out by the thousands. Tens of thousands by the end of the month.
Get it now?
Get it now?
I'm thinking a lot of people here are over estimating the HR monitor use...
I have a wrist tattoo that doesn't play nicely with the HR monitor at times, so most of the time I have the HR monitor off, and guess what, it still tracks exercise and move. So if they used the HR monitor as much as a lot of you guys are saying, it would hardly be/ not tracking anything.
It was 10 minutes apart previously and still recorded move and exercise, I'm willing to bet the HR monitor very rarely played a part.
It's 99% the gyroscope and accelerometer doing the work.
I have a wrist tattoo that doesn't play nicely with the HR monitor at times, so most of the time I have the HR monitor off, and guess what, it still tracks exercise and move. So if they used the HR monitor as much as a lot of you guys are saying, it would hardly be/ not tracking anything.
It was 10 minutes apart previously and still recorded move and exercise, I'm willing to bet the HR monitor very rarely played a part.
It's 99% the gyroscope and accelerometer doing the work.
Does it really not move at all? I just got home from school around 4. I took a screenshot of my activity just from being in school. I'm on 1.0.1 and my exercise ring was filled up 17 minutes...
My watch seems to register exercise perfectly; it definitely requires a very brisk walk to register exercise. It's not easy to get, and it's not supposed to be. But it is definitely possible, with my watch at least. I haven't used the workout app yet.
Does it really not move at all? I just got home from school around 4. I took a screenshot of my activity just from being in school. I'm on 1.0.1 and my exercise ring was filled up 17 minutes...
My watch seems to register exercise perfectly; it definitely requires a very brisk walk to register exercise. It's not easy to get, and it's not supposed to be. But it is definitely possible, with my watch at least. I haven't used the workout app yet.
Mine wasn't moving AT ALL but now it seems to be back to normal after unpairing and repairing...
That's very interesting. Maybe I'll try it too. So you unpaired the watch, then when repairing, did you restore from backup or set up as new?
I'm still sitting on my 1.0 watch that arrived last Friday? I was considering holding off updating until 1.0.2 or WWDC, but after encountering a few minor bugs I decided that I might as well join the 1.0.1 crowd. This is my second watch, first was SGS 42 on 1.0.1.
Anything worth testing before I update?
Anything worth testing before I update?
Last edited:
Does it really not move at all? I just got home from school around 4. I took a screenshot of my activity just from being in school. I'm on 1.0.1 and my exercise ring was filled up 17 minutes...
My watch seems to register exercise perfectly; it definitely requires a very brisk walk to register exercise. It's not easy to get, and it's not supposed to be. But it is definitely possible, with my watch at least. I haven't used the workout app yet.
It will still register excercise but it's about half the rate of doing the same excercise with the workout app running on your watch.
You may as well just pluck a figure out of the air.
This surprises me not at all. The watch running the activity app is taking your HR consistently, whereas the watch not running the workout is taking your pulse when you tell it to, and inferring your calorie based on only a handful of readings and your physical motion. Moreover the watch running the app is using an algorithm specifically for the elliptical to calc your calorie burn, while the other watch is using a more general calc.
ANY data for your calorie burn, from any device/machine outside of professional-level equipment, should be taken with a pinch of salt.
But they both update the same fitness rings. There lies the problem!
The rings update at different rates while doing exactly the same activities.
The rings update at different rates while doing exactly the same activities.
Yeah, because the watch only knows what it knows. Without the constant HR, it's guessing, and it's guessing wrong (the fewer data points it has, the more likely it is to be wrong). It would have been wrong even if it took your HR every 10 minutes, it just probably would have been less off. It would be most accurate if it could take constant HR the entire time you wore it, but that would be a battery killer. It would also be more accurate if you told it what you were doing all the time, but no one wants to do that, LOL.
It's not perfect, and it's not going to be (in this iteration). It was somewhat more accurate before, no doubt. But even if Apple changed it back, right now the tech just doesn't exist for it to give you an accurate and consistent representation of your burn without you helping it out a little bit.
When I was taking a course about computer repairs, one of the things they told us was that a very high percentage of "problems" are corrected just by turning the computer off and then back on, followed by disconnecting peripherals and reconnecting (I've also found that just powering down isn't enough, but literally unplugging works - go figure). They even told us to send the client out of the room when doing this so they wouldn't know you really didn't do anything special.
So I suspect that is a big part of the reason for their instructions on the watch. But it is darn frustrating, especially when it turns out that isn't the problem. And on the iphone, they'd tell me to set it up as "new" and not use a backup to restore because the backup may have the same flaw. Great - except then I'd lose all my game scores! And since I'm #1 in the world on some, that just isn't acceptable!
Flur - Again I disagree. During my 10 minute test I checked my heartbeat on the workout watch and via glances on the non workout watch. The heartbeats we're the same as you'd expect.
I did this 3 times. IE every 3 minutes or so. My heart rate would have changed that differently in the 10 minutes yet the results were vastly different.
I did this 3 times. IE every 3 minutes or so. My heart rate would have changed that differently in the 10 minutes yet the results were vastly different.
Yep, I explained for that upthread - The watch running the workout was using an algorithm specifically for the elliptical to calculate your burn, because that's what you told it you were doing. The other watch (the blind watch) had less information to go on - not only fewer heartbeats, but also no knowledge of what the rest of your body was doing while your wrist was moving back and forth. All it could do was make a guess at what it was you were doing. In this case, its guess resulted in a lower estimated calorie count.
Basically, give the watch as much information as you can.
Yes please. Do a couple different workouts as similar as you can with a Workout app set and without it. Compare calorie count and exercise ring credit with each. I would like to see if the 1.0 software is better at figuring out your activity level without the need to go to the Workout app. Some people say it is better in 1.0.0 and you can just put it on and forget about it and trust that it will get all your calories almost as good as if you set up dedicated workouts.
I get what all you are saying about giving it more information so it can do a better job. And if we were talking +/-20% with the values, I might agree with you. But we're talking 100+% differences in some instances. For example, I did a 20 minute elliptical (87 calories without the Workout app, 301 with it). Seems like too big of a difference to me for a watch that they advertise can pick up on your activities in the background without you needing to do anything. And if the 1.0.0 software CAN do this without being so far apart, then clearly this is just a programming difference and not an inability of the watch to passively monitor activity.
I'm beginning to be on board with flur here. His logic makes sense to me. It does seem logical that if the watch takes 3 HR readings 10 min apart in a half hour, it could assume you were doing "something" constant to keep it there. But how is the watch to know? Maybe you decided to a 30 second wind sprint 3 times in a half hour, every 10 minutes. I would think that would be less calories than a 30 minute run.
Maybe that every 10 min HR had very little to do with the "extra" exercise people were seeing in 1.0. It's possible that in addition to the HR change, Apple decided to "tweak" their algorithms for the accelerometer and gyroscope that were giving folks that extra exercise. If that's the case, they could gradually "refine" those algorithms over subsequent releases to get closer to what it did originally, to the point where the Apple programmers feel the results are what they should be (but not 1.0 numbers). I'm sure they'd never admit to this, though. That's something they could do and still maintain their stance on not reverting to the 1.0 HR version.
It's definitely possible that they changed the algorithms. It's also possible that they're not using the HR at all for determining exercise when not using the workout app. That would certainly explain millerrh's elliptical discrepancy - the arm movement on the elliptical is not terribly different from one you might make when walking (excluding the different positioning), so without the HR data it may have looked like a walk to the watch.
It's not surprising the using elliptical machines confuses the watch. I'd be surprised if it could accurately track anything other than walking / running without being told what you're up to. I'd also be surprised if the fitness app can even track elliptical activity accurately without you feeding in the resistance level you're using.
I'd love a Watch OS update during WWDC but – by going off past experiences with big updates – I imagine there might be something mentioned but that it's probably not quite ready and that we'll have to wait a few more weeks. Let's hope I'm wrong!
I highly expect apple to announce a few things for the watch's O.SI'd love a Watch OS update during WWDC but – by going off past experiences with big updates – I imagine there might be something mentioned but that it's probably not quite ready and that we'll have to wait a few more weeks. Let's hope I'm wrong!
1- Opening it up more for Dev.. And in true Apple fashion, giving them just enough slack to make apps that run on the watch itself. But as far as fully letting them get access to features and stuff.... Lol, well, we saw how long it took for them to give anyone anything for the iPhone
2- Some new smaller features. Not sure what, but something new
3- Updates to iron more bugs out
Mine seems "stuck" at crediting me two minutes of exercise per day for the housework even if one day I'm at it for several hours and the next day just two hours. That is after not crediting me at all for a couple of weeks.
I'm having to unpair and re-pair as well as rebooting phone and watch for other issues on a regular basis like about every week. I've not had the time or inclination to do this for almost two weeks now so I'm getting whatever exercise glitch there might be as well as failing to get haptic taps for some notifications and not having the phone feature work normally on the watch. My phone has never worked as well as my husband's, with odd freezes and lag and occasional battery drain, so it could be a problem with my iPhone and not the watch. Wish I knew.
But I've shaken off the winter lead in my butt. I am more fit and active as a result of pushing myself to meet calorie goals so there is at least that.
did they fix i ?Same problem here