This discussion has been locked.

You can no longer post new replies to this discussion. If you have a question you can start a new discussion

24/7 HR Sampling

I thought I'll put that topic in an own thread. Many people have the perception that Garmin reduced the HR sampling frquency as of FW 3.20. And to many this topic seems very important, yet so much 24/7 HR was one of the main reasons to get that watch.

Discussion about this started in the FW 3.20 thread around here: https://forums.garmin.com/showthread.php?337137-FR235-Firmware-updates-v3-20-and-v2-30-quot-Sensor-Hub-quot-(2015-12-09)&p=770858#post770858

My last post there about this was: https://forums.garmin.com/showthread.php?337137-FR235-Firmware-updates-v3-20-and-v2-30-quot-Sensor-Hub-quot-(2015-12-09)&p=771434#post771434

I will post my upcoming findings about this topic here from now on.

The first "result" I might have found: When moving around continiously the HR is read every 10 minutes.

From the last two nights of sleep I would guess so far, that when not moving much HR is being read about twice an hour. I suppose Garmin has implemented some algorythm that reduce "time to next sample" based on movements/steps being registered.

Note that this is just what I see my watch is doing, me registering time and step count when a reading happens.

After I stopped moving around, the next reading came 20 mins later, having made 60 steps in between.

Please feel welcome to post your own findings, graphs etc about allday HR here...

0 Former Member over 9 years ago

There seems to be an implication here that outside "activities", the heart rate is used to calculate calories. I didn't think it was or can anyone verify that is the case in either the FR235 or Fitbit HR? If so, how does an FR230 or FR630 work and what is the basis of the burn calculation at less than 60% of max HR where that is more than 90% of the time?

0 R_Tellis over 9 years ago

That's a great response Paul.

How many calories per day would you think the lower sampling rate might account for, and would it be enough to make a difference in your daily choices?

As far as Garmin is concerned this is a running watch with activity tracking features so if some aspect of activity tracking needs to be reduced to improve it as a running watch that's the direction they'll take.

0 philippe.berini over 9 years ago

There's been the question what advantages a 24/7 HR monitoring brings along.... Besides maybe better calories estimation RHR tracking seems to be the main feature for most.

What about the maximum HR per day that is reported in Connect? Isn't this a bit useless? Maybe if the HR from activities would be excluded you could see what your highest "effort" in a day was. But don't really see the use of this.

0 timgrose over 9 years ago

Knowing your max HR is crucial for the VO2 Max, Recovery Advisor and Training Effect algorithms and by implication calories in actual activities too.

Probably more useful than RHR TBH so you can get a view on how often you really push things. Well assuming you can believe the value of course...

Would be interesting to know many extra calories the algorithm thinks you consume just by getting up a walking around a fair bit during the day with, one would think, only moderately raised HR for short periods. I find it hard to believe that this adds on masses of calories?

0 philippe.berini over 9 years ago

Well yes, but thats the maxHR reported in the activity that is important, I agree.

But I was talking about the one reported with the all day HR... And also the one on the watch in the HR widget, maxHR for the last 4 hours... this is probably very useless.

I researched the internet a bit, for my height/weight 1 Km of walking should burn about 60 calories, so it's not that much. I don't know if that number contains Bmr but for the time of a 1 Km walk it's not much either.

Since I don't track calories intake I don't care that much ( still would like to see more realistic numbers in GC for allday steps).

0 Former Member over 9 years ago

For 4 of the 10 days that I have had 24/7 HRM operating I have a "peak" that repeats exactly at the same level several times a day (per attached). This peak number may be different between the days and does not represent an actual increase of HR. Anyone know what it is?

0 Former Member over 9 years ago

As per a previous post in another thread:

OK, here is my answer and it mainly relates to calorie burn. 9 months ago I bought a Fitbit HR to track steps and calories. This information is fed into myfitnessPal where I record food eaten. By targeting a certain calories deficit I have managed to lose 25 Kg since then and I am now nearly at a healthy weight. If you look at the attached example of a Fitbit trace you can see it is fairly granular as it samples every 5 seconds. If I get up and walk around for a bit you can see the HR rises and calorie burn increases. It is a motivator and genuinely useful health tool. For overnight you can see there are changes in HR. It may be only a few BPM but this info can be used for RHR. Now if you compare this to a Garmin 235 chart then its the other end of the scale. A 5 minute activity could be missed if sampling is low and RHR could be a few BPM out.

So why did I buy a Garmin in the first place? Well the Fitbit is not that accurate in certain activities and can take a long time to lock onto the HR. You could say the Garmin is having these issues also as perhaps do all Optical HR monitors, BUT the unique selling point is the ability to attach a strap when you want to such as interval training. I.e. you have the best of both worlds.

However a (apparently) very low sampling rate could be making HR and calorie burn inaccurate outside of activity recordings making it a poor activity tracker. For those people that see this watch as a step up from their Fitbit Surge and HR with better optical tracking and ability to pair a strap then the low HR sampling is a major disappointment, especially considering the cost. If there was an alternative right now then I think a lot more would be returned. Unfortunately it appears that Garmin Customer service don't seem to "get it" with regards to the tracking. Unfortunately in the UK there is normally no option to try before you buy. If you use it you can't return it.

Of course there are a lot of people who are just data geeks and like to monitor things like this! :-)

Hope that helps?

To you this is a running watch. To others the 24*7 tracking is just as if not more important. You may find this hard to believe but I don't actually run outdoors, I prefer to cycle for which I use my Edge 1000. My intention was to use (When Father Christmas delivers my watch on the 25th) the 235 for fitness tracking, indoor running, stationary bike and general cross training with the option to wear a band when optical HR wasn't working too well (Weights/interval training). I would like everything to be in one place. I.e. GC, but the potential inaccuracy in the tracking has me (and it seems lots of others) concerned. With this watch Garmin has the potential for a flexible multipurpose watch with good running features. They just need to listen to its users regarding the tracking element.

I'm with Paul. I had a Fitbit Charge HR and Suunto Ambit 2S and the 235 seemed the perfect opportunity to combine the best of those 2 devices.

Whilst the Fitbit is a little unsophisticated, it pretty much does what it says it will do, which is flash those little green devil eyes every second of the day.

I'm not a committed runner with age and injuries getting the better of me, and I work from home, and travel on business. So, having GENUINE 24/7 HRM is a good guide to what I'm burning up. I'm not bothered about calorie intake counting as I eat pretty healthily, but getting an idea on what I'm chewing up helps my peace of mind in keeping my general activity good, and also helps keep weight etc in check.

Plus, it's great when I do go for a run etc, so it's a true all-round device, if they got to grips with the HRM issue.

At the end of the day, they market the device as a running, activity tracker, built in HRM, sleep monitor, so if they want to play with the likes of Fitbit, they have to get their act together.

Otherwise, it's not fit-for-purpose...:rolleyes:

0 philippe.berini over 9 years ago

Ian provided me with some fit files from the MONITOR folder as he is still on 3.10 FW.

What I see there in comparison to my 3.20 monitoring files is the following:

the maximum time between records is also 15 minutes

However

there are less consecutive records with the same HR value, if there are, it's mostly two of them, while I had 2-6 of them
mostly when there's a change in heart rate there are times between readings ranging from 1 minute to those max 15 minutes, while I had almost only 15 minute cycles throughout the night (and I was not the whole night being in deep sleep without movement)
or there are generally just shorter frequencies being recorded

So, the slowest reading frequency has not changed. But the sensor seems to kick in less (due to movements?) and also those many consecutive numbers I get still make me suspicious that somethings not right.

0 Former Member over 9 years ago

24/7 questions

Based on a review of my HRM data of today I have a couple of questions that I hope somebody is able to answer.

1) HR has a flatline from 3:29 to 6:59 AM (yes, I was asleep...). The recorded HR was....66. I don't know if this is related to the "66 issue", however what I noticed was that there was not a single data point between 3:29 and 6:59. That's 3,5 hours between the samples. Has anyone seen similar low sampling frequency?

2) This afternoon I was doing an hour or so of various work-outs. I was using a chest strap and the recorded HR levels and sampling frequency seem fine. After taking a shower I put on the chest strap again (to check a different HRM issue...). I've just noticed that there is now an hour between samples. Two latest samples are from 9.38 and 10.38 PM (CET). This leaves me with two questions:
- Is the sampling rate the same regardless of whether you use the internal HRM or a chest strap (meaning that SR is only effected by activity level and whether or not an activity is started on the watch)?
- During the two time stamps I was cleaning up the kitchen, taking out the garbage and doing some laundry (incl. climbing the basement stairs twice). Hardly work-out, but not complete inactivity though. Is an SR of one hour at this activity level in line with what you get from your devices?

I am running FW 3.20.

0 philippe.berini over 9 years ago

Based on a review of my HRM data of today I have a couple of questions that I hope somebody is able to answer.

1) HR has a flatline from 3:29 to 6:59 AM (yes, I was asleep...). The recorded HR was....66. I don't know if this is related to the "66 issue", however what I noticed was that there was not a single data point between 3:29 and 6:59. That's 3,5 hours between the samples. Has anyone seen similar low sampling frequency?

When I had no datapoints last week, this meant it was stuck at 66, at least for me it was like that. And no data points here mean, there is no line drawn in the HR graph. You might want to install the newest Sensor (WHR) fw that was released yesterday, this should fix the 66 bug.

2) This afternoon I was doing an hour or so of various work-outs. I was using a chest strap and the recorded HR levels and sampling frequency seem fine. After taking a shower I put on the chest strap again (to check a different HRM issue...). I've just noticed that there is now an hour between samples. Two latest samples are from 9.38 and 10.38 PM (CET). This leaves me with two questions:
- Is the sampling rate the same regardless of whether you use the internal HRM or a chest strap (meaning that SR is only effected by activity level and whether or not an activity is started on the watch)?
- During the two time stamps I was cleaning up the kitchen, taking out the garbage and doing some laundry (incl. climbing the basement stairs twice). Hardly work-out, but not complete inactivity though. Is an SR of one hour at this activity level in line with what you get from your devices?

I am running FW 3.20.

Im not exactly sure what your question is. But I have never tried using a chest strap for daily activity, not sure this even works. Could be that the chest strap only works for activities/workouts. But even with the optical sensor, there is something "wrong" with the daily sampling, just go ahead and read some posts above...