S2 Body Fat not accurate

StarostinAA said:

Kevin, any news about new firmware? May be new algorithms for calculating fat and etc

This product is abandoned. It was abandoned the moment it was released. I suspect the reason is that the hardware design itself has a fatal flaw, which they only realized after having already passed the point of no return manufacturing a huge run of them. Which means it will never work correctly regardless of firmware patches, and that's why we've only seen literally 1 firmware patch in two years, which didn't even make any attempt at fixing any of the problems users have been reporting.

Our best hope at this point is that Index 3 will have a new design, without any fatal hardware flaws, and that it will be better supported going forward. That is, if any consumers are still interested in buying a Garmin scale anymore by that point.

Exactly. I promptly returned mine for a refund when I saw the wacky body fat fluctuations of 10 percentage points in one day AND these threads.

I suspect that the fatal flaw is that it’s all just smoke and mirrors. 

Just wanted to add some additional thoughts on this.  I have been using the scale for over a year now and have tracked results through a spreadsheet every week and correlated that to the particular activity I was doing.  Within 2 months of using the Index 2 I had a Dexa scan which showed the scales were calculating BF% around 4% higher than the Dexa scan.  Started on a calorie deficit / high cardio routine and lost around 2 stone in 6 months and BF% reduced by 4-5% on the Index 2. I then moved to weight training and lean bulking and during a routine private healthcheck a couple of months in had my BF measured again and it was 13%.  This compared to 18% on the Index 2 so the variance had increased by 1% or so.  I have subsequently gained more weight through muscle built and my weight has risen by around 4-5 lbs in that time and the Index 2 in recording some increase in BF% but also increase in lean mass but more weighted to lean mass than BF which would be in line with what I can see in the mirror.  So my thoughts are that the scales do record changes or trends in a way but the baseline algorithm based on age is wrong in the first place.  I am 43 so I guess typical male population would show higher BF% at that age group hence my Index 2 baseline being too high by around 4-5%.  In summary I'm happy with where I am but knowing what I know now having spent nearly 2 years on a fitness journey that these scales do not give you an accurate BF% but they do show a trend of sorts.  If I had my time over would I buy a £10 set of scales to record weight and look in the mirror to judge progress - Absolutely.  Would I buy an Index S3 if released - Not a chance.  I would think a simple fix would be to just give a way to more advanced users to enter a known baseline from a Dexa scan, calipers, BF estimate so the scales build that into their calculations and give a more accurate figure.  For 75% of the population I would guess the scales are fairly accurate but for those that deviate from the norm they need a way to manipulate the numbers for a more accurate picture.

Exactly. That's what we have been trying to show here. If it's an algorithm error based on the age(average population data) how it's possible not to have found a fix for more than a year.

And a big error because if Garmin knows that their customer base is not the average population but the one who usually practice sport then why the algorithm is based on average people BF(depending on the age)

Or you are like me where the trend nor the baseline is correct. I’m not 22%, I’m 14%, and it’s impossible to go from 24% to 17% overnight. (I returned mine. If everyone did the same, it would be an effective way to force Garmin’s hand).

I just returned my scale, after wringing my hands about it since my first post a few weeks ago; thankfully, Amazon is pretty good about this stuff. Going back to tape measure and my old 14 euro LIDL super scale. Shame. I've loved Garmin since the army, when all our GPS was Garmin, and I dished out for a Tactix Charlie back in the day. But I just can't accept such poor responses about x product. 

Garmin-Kevin any update on this abandoned product?

Kevin, since this will probably take a long time to fix and probably require a new product launch ... would you guys consider checking integration with some other Smart Scales out there to Garmin Connect? At the end, the accurate numbers matter for all other Garmin features on Connect. 

Thanks for your update @garmin-kevin. Glad to hear that a solution is being sought. Many suggestions have already been made. I still believe in an adjustment in the algorithm. Slightly less sensitive and adapt it to the Garmin Community with a heavier weighting for activity level. I have to put it on 10 to get close, but I only run 5-6 times a week. But certainly, no more than 15 hours (level 10)! I think adjusting something there would help us. But maybe, it is something else? I think the active Garmin community is much more active than average. Averages work less well for us/me.