Bug: watch battery level only reported as whole number

The 235 battery percentage is only exact if the accuracy of the measurement is 0.1%. Perhaps Garmin have worked out their measurement is +/- 0.5% and so now only give 1% accurate results.

CW

The 235 battery percentage is only exact if the accuracy of the measurement is 0.1%. Perhaps Garmin have worked out their measurement is +/- 0.5% and so now only give 1% accurate results.

CW

That certainly could be true. Garmin doesn't even give the percentage on the default a watch face.

It's just I'm recording the hourly percentage drain for activities and that first decimal place makes a difference statistically, even if the accuracy of an individual measurement is larger.

It's a different behaviour from the other watches with basically the same firmware. So could be an "update", or could be an oversite/bug. Worth bringing it to Garmin's attention, even if this level of detail doesn't really make a difference for most.

You would be better placed to open a discussion with Garmin Support directly as well. These are primarily peer-to-peer forums.

Actually, the best place to ask about this is in the Connect IQ Developer's forum. Garmin folks are very active there. IIRC, in early 23x FW, it was the same (wasn't passing a decimal component). This could be a CIQ bug (the CIQ VM) and if not the Garmin CIQ folks there can pass it on to the FW folks.

https://forums.garmin.com/forumdisplay.php?479-Connect-IQ

While most CIQ watchfaces don't display the decimal part and only the integer part, it's often used for "rounding", so that if the battery is 99.987654, it's displayed as 100% instead of 99%, while 99.4321 is displayed as 99%, for example.

I had a look at the battery values in the FIT file through the FIT File Repair Tool for a recent run with my 735 and it does indeed only have integer values (see attached image)
Mind you I checked a FIT file from my 235 and 630 and they were integers too so not sure why you are seeing it differently.

Other than curiosity, not sure there is much for us to worry about here however! My iPhone or laptop only integer battery levels as well and TBH that is more than good enough for me.

Of more interest is why you think battery drain on the 735 is "uneven and a bit erratic"?

The drain in the my graph looks linear enough allowing for the integer values. Are you seeing this differently?

I had a look at the battery values in the FIT file through the FIT File Repair Tool for a recent run with my 735 and it does indeed only have integer values (see attached image)
Mind you I checked a FIT file from my 235 and 630 and they were integers too so not sure why you are seeing it differently.

Interesting. I don't know about within the .fit file of an activity, but if i use e.g. the battery gauge widget to mark the start and finish of an activity i usually get a value to one decimal place (e.g. XX.4% with the the 235 - but only XX.0% with the 735XT)

Other than curiosity, not sure there is much for us to worry about here however! My iPhone or laptop only integer battery levels as well and TBH that is more than good enough for me.

It does make a difference when you're trying to get a sense of the drain due to activities, but i agree, it's nothing terrible. But as said above, if it is only reporting whole numbers then rounding might be an issue, potentially a bug.

Of more interest is why you think battery drain on the 735 is "uneven and a bit erratic"?

The drain in the my graph looks linear enough allowing for the integer values. Are you seeing this differently?

An example: Did a run yesterday and measured 16.0% drain, with the battery at 75% when finished. Didn't do anything special after this but checked the % again 15-20 minutes later and it was at 73%. So somehow i lost 2% in that time; normally it drains 2% after 7 hours of sleep.

Now if the rounding is all screwed up it may have only lost a percent (plus a fraction) ... who knows. But after extensive testing of two 235 watches those were all perfectly predictable and consistent. The same tests on my 735XT have been more erratic, although if you're not measuring you probably wouldn't even notice (e.g. suddenly dropping 2%, but then taking twice as long to drop the next 2%).

My assumption is that they're still refining the battery drain algorithm. But i thought i'd bring i up anyway. After all, one of the selling points of the 735XT is its 14hr battery life, 3hrs more than the 235. I'd like to believe they don't just make these things up. ;)

After all, one of the selling points of the 735XT is its 14hr battery life, 3hrs more than the 235. I'd like to believe they don't just make these things up. ;)

In any case, it's quite possibly a false assumption that battery power consumption – specifically, calculated from the value of the battery remaining metric as reported by an application, as opposed to the physical phenomenon – is linear. The objective test of whether a FR735XT delivers the advertised battery life[sup]†[/sup] is whether the watch, after a full recharge of its battery, powers itself off and/or stops working after 14‑hours of tracking an activity for which the GPS receiver and optical HR sensor enabled.

Whether the battery power remaining is reported as 50%, 49.5%, 54% or 32% is irrelevant to the validity of the 14-hour battery life claim, especially if Garmin makes no assertion that battery consumption is linear over those 14‑hours. Yes, as a consumer I understand it may undermine user confidence and is no help to athletes who want to carefully manage their activity tracking usage of the device, but that again is irrelevant to the validity of Garmin's claim.

[sup]†[/sup]I note that Garmin carefully qualified its battery life claim as “up to 14‑hours”, so actually delivering only 12‑hours in a particular instance of activity tracking does not logically invalidate the claim, whereas delivering 16‑hours is more apt to call the claim into question.

Whether the battery power remaining is reported as 50%, 49.5%, 54% or 32% is irrelevant to the validity of the 14-hour battery life claim, especially if Garmin makes no assertion that battery consumption is linear over those 14‑hours. Yes, as a consumer I understand it may undermine user confidence and is no help to athletes who want to carefully manage their activity tracking usage of the device, but that again is irrelevant to the validity of Garmin's claim.[/size]

I appreciate that you're trying to contribute here. But please don't turn this into another thread where you miss/over-intrepret/selectively-intprepret the point (or one point amongst many, ignoring the rest), side-tracking the discussion and making heads explode in frustration across the globe. Thanks.

darrencroton - As I said in an earlier post, this seems to be a bug in how CIQ sees the battery level. It might be the FW only passing the integer part, or how its being converted in the CIQ VM.

Make sure to report it in the CIQ developer's forum, of it's it's not something you wrote, use "Contact Developer" to let the developer know so they can report it, as the Garmin CIQ people are very active in the developer forum, and can narrow it down. All devices except the 735 include the decimal part, and in the early days of the 23x, this was also a problem with those too.

Make sure to report it in the CIQ developer's forum, of it's it's not something you wrote, use "Contact Developer" to let the developer know so they can report it, as the Garmin CIQ people are very active in the developer forum, and can narrow it down. All devices except the 735 include the decimal part, and in the early days of the 23x, this was also a problem with those too.

Thanks Jim. Posted here. Hopefully someone can shed some light.