Body Fat % on Index S2 Smart Scale Seems Inaccurate part2

I just tested this out and it seems to have gotten worse in terms of accuracy. And I still have to adjust my age and activity level to get a body fat % that's close to accurate. Maybe it'll be more consistent though. 

My Tanita reads 13.6%. Before the upgrade S2 was reading 16.5%. After the upgrade S2 jumped to 22%. After I updated the body composition correction factor to 13.6%, the S2 now says I'm back to 16.4%. I'm not sure what the update accomplished - but improved accuracy definitely wasn't part of the update. Perhaps Garmin could add a daily physic reading feature if the S2 is just going to make things up.

Yea, i don't know what the point of that update was. 

Question, what is the "body composition correction factor"? is that something you can change? I'm just fiddling with age and activity level to get the body fat to be at least close to accurate.

At least you can use your S2. The update effed up its Wi-Fi connection. Now syncing is impossible. Nothing works. lol

Garmin-Kevin Can you explain what was changed to improve the body fat % reading? I'm not seeing any improvement and still have to tweak it by changing my age and activity level.

Ugh, that stinks. Sorry to hear that.

https://forums.garmin.com/sports-fitness/healthandwellness/f/index-s2-smart-scale/321992/index-s2-does-not-sync-settings-after-upgrade-to-3-00/1561755#1561755

Not everyone is affected, but we're pretty much effed. 

As for body comp correction factor: You can now enter calibration values in the Index 2's settings (GC).

Btw. I'm very curious about the solution here, since updates are delivered via Wi-Fi. 

this morning it gave me 8.1% so thats pretty close to the 8% it "knows" as fact, yet its crazy that it immediately gives a variance and others are still getting bigger variances.

30.9kg muscle, 5.1kg bone and 67.1% water so its still allocating most of the reduction in fat to water and a bit to muscle and bone. 

statistics suggest my bone mass is likely 2.5-3kg so it is interesting for garmin to say twice that, especially since we believe these scales to be heavily based on statistics. i think it is clear that 5.1kg is not possible

water% is plausible but this is highly variable

adding together the composition components gives a total mass 24% higher than the actual but thats not unusual since muscle contains a lot of water so it can be counted twice 

FWIW my 8%BF is very conservative - i pulled out the old scales i used to use and they give me 6.6% which i have reason to believe is reasonably accurate. 

at least they are now starting to use "estimate" in their wording