Hi DextrosePLZ

For sleep data, have you tried clicking on the "Activity Summary" tab in the Diary section?

Avispon

Really like this thread with support from Polar. Was in the market for a V800, then the 910XT after DC’s review, now the Polar is back in scope again.

The device is not measuring calories burned. It can only measure motion and heart rate. From those it can predict calories burned. However, there will be lots of person to person variability. For example, heart rate generally increases linearly with work rate, but it can be significantly elevated with no increase in work rate at all. In the heat, any heart rate based system will over predict calories burned. A motion based system will have even more problems because many motions will not be detected by the measuring system. Non exercise activities of thermogenesis (NEAT - Google it) can account for significant calories. I can't imagine that anybody's activity tracker can stand up to the type of validation you mention, not yet.

---

Simplify, Train, Live

I have a question about the future open water swim metrics. Is Polar aiming for 910XT-ish data? So route and distance with about 10% margin of error? Or will the V800 be more precise?

Boxer Briefs (I like Adidas Climalite, and my other favorite brand is "Whateverthehellisclean")


1) Sleep data is in the Flow app, or in the Diary (day) at the bottom.
2) That's a very good point & question.. We have it on our priority list, though I don't have a timeline on that.. Something to note: the web team is not the same team working on the feature additions above.
3) Yes

1) In our tests it remains "awesome" - and *most* of the feedback we're getting from the market is in line with our experience (of course with anything a few on the other end). Ray still has a very early beta - so a production unit is on the way to him now.
2) We haven't tested it enough with an optical sensor to give you a good answer yet..But I'll come back to this.

On your opinions. 1) We're following mostly all standard Bluetooth Smart profiles - what we've found is some third party sensors not fully complying with the specifications. But we're investigating, and hopefully bringing some light to some of the sensor manufacturers and how they're implementing the profile. 2) It mostly has everything a Loop does, except for the daily goal. But we've heard some good feedback, it's why conversations like this on forums like Slowtwitch are so important.

About swim metrics. At the watches current point, would work for OWS if placed in your swim cap? (so it never goes under water)

I don't see why not..

Then for the before mentioned October release, what is being added in? I can see pool swimming, but will it also change so that you can accurately track OWS whilst on the wrist?

Its been widely reported that the Polar v800 does not work with bluetooth certified sensors.

Assuming that the v800 is Bluetooth certified then the bluetooth certification process failed for either all the other sensors, or for the V800 or both. Either way it suggests that being BTLE certified is meaningless when it comes to interoperability, which is sorta the point of a standard.

Now the question - in keeping with the rest on this thread. Whats your favourite colour?

Thanks for the insights. I'm really happy to see a statement about user experience and how important it is.

As someone who is both a triathlete and a programmer, I'm curious where Polar does their software development? And whether there might be opportunities for some enthusiastic slowtwitchers to contribute to the success of the V800 / V650 / future offerings (had seen some opening on careers page a handful of years ago but nothing now).

Yes, October update to swim metrics include indoor pool metrics, and updated OWS smoothing. The quality of all measurement is important.

We do all of our programming(and really everything) in house. Our HQ is in Finland. We don't have any vacancies that I know of right now on the programming team - we've actually added quite a bit of resources to these projects already. But surely it wouldn't hurt to send me some info so you're on our radar. Shoot me a message and I'll be happy to route. This page is also updated with vacancies: http://www.polar.com/...polar/jobs/vacancies

I know all that, but would like to get the thoughts from a Polar person regarding their validation testing.

---

_____________________
Don't forget to attack!

I am not sure that you are asking the right question. I don't see this as a hardware issue at all. It is a mathematical modeling issue. You should be asking what model are they using to predict calorie expenditure and if it is based on heart rate, or the accelerometer, or both. For example, if it is heart rate, perhaps they are using the regression formula from this study that is widely used:

http://www.ncbi.nlm.nih.gov/pubmed/15966347

In that case, the question is validation of the regression formula in the original study, not the Polar device. We already know that Polar can measure heart rate.

---

Simplify, Train, Live

Like I said, I'm looking for Polar on this regarding their validation work. Although I appreciate your expertise if we start a discussion or debate there's no way I'll get a response from Polar regarding my specific questions.

---

_____________________
Don't forget to attack!

Good question. Two part answer

• Calorie expenditure for training - all based on heart rate (and user information), using algorithms we've researched and validated for the better part of 25 years. The algorithms are proprietary but published studies are available online (Owncal Validation)
• Calorie expenditure for activity is based on activity (and user information) using the accelerometer. We've internally validated our algorithms, and a few third party studies are happening right now

In terms of sleep measurement. You do not have to press a button. That's automatic (and also based on movement).

Operational question: when I have it in "triathlon" mode - I hit the start button to begin; , then the back button to move to T1, then the start button to begin bike; then the back button to move to t2; then the start button to move to run. what do I hit at the end of the run to complete the race? if I hit the back button again... it appears it is moving into a t3 mode or something.

At the end of a race, press the back button once to pause - immediately press and hold the back button for 3 seconds to exit.

Predicting calorie expenditure from HR is pretty solid for heart rates between 100ish and 160ish. The problem is when people are not really exercising and HR can be all over the place. So getting an exercise calorie expenditure is generally no problem and can be done reasonably accurately with HR and I assume that Polar would be an industry leader in this area. I remain unconvinced about doing this for whole day energy expenditure. I am more unconvinced when it comes to accelerometers. I think for certain predictable exercise modes like running, the challenge is not too bad. But other modes are going to be a problem. However, this is not a Polar challenge. All activity monitors currently have this issue (https://minds.wisconsin.edu/handle/1793/66292). However, I am open minded and if validation studies say differently, I am willing to change my mind.

---

Simplify, Train, Live

Completely.

We'll be publishing documents at some point soon on validation of Polar activity based calorie expenditure. I'll note to repost in this thread once available.

FWIW: A decade ago I dropped 25 pounds over the course of 25 weeks, meticulously tracking all foods I ate but using only rudimentary calorie expenditure estimates (ie: cycling 500 calories/hour, running 100 cal./mile, walking 80 cal./mile, etc).

At the end of those 25 weeks, the calorie in/out balance was within 3% of the predicted 3500 calorie deficit per pound needed for weight loss.

In other words, for me, there would've been no practical gain in having a more accurate tally of my calories expended.

So, in theory, I have one of these headed my way in November: BSX Insight

Any chance you guys will support this? Please :D

Technical question:

The intended upgrades/added metrics and what have you - how will that work?
Like a firmware upgrade where the end user will download and install the software/patch on the unit or wil the units have to go back to Polar for the upgrades?

I'm asking because the early RC3 GPS units (read: mine) lacked GPS based elevation support in the unit itself and the functionalty was added in later iterations.
No biggie for me because the webservice provides elevation data from Google Maps for later analysis and such I just couldn't be arsed to send in the unit for upgrade/replacement.
For the V800 however, the upgrades appear substantial and I would hate to plunk down a fair amount of $$$ (or €€€ for that matter) just to have to send the unit back to Polar a coule of times to get the full current functionality.

NOT knocking Polar or the RC3 GPS in any way here, just curious...

---

___________________


...compensating my lack of fitness with inadequate strength and endurance...

It's done through software on your computer

http://www.polar.com/...l_English/manual.pdf

Much obliged, kind Sir

---

___________________


...compensating my lack of fitness with inadequate strength and endurance...