I think you're assuming that the new measurement is less accurate than the old one. Unless one has a reference of known accuracy, I can't see how one could make that determination.
There is a bit of truth in the statement. I've been using the Band since Xmas and with the legacy firmware, there were days that I burned over 4500 calories - probably overshot the reality by fair amount. There was a thread on this forum about similar experiences by others.
After the January update, brought it down quite a bit where I never saw 4000 calories. many in 3000s but I work out 5~6 days a week and couple of long days days a week where I spend quite a bit of time outside walking.
Now with the most recent update, my sense is that it overshot with the conservatism. On a day similar to when I used to record 3500 calories, its about 800~900 calories lower. My resting HR is typically in upper 50s to lower 60s so it really looks like how it treat low HR calorie burn more conservatively than before. I have not noticed significant difference in my workout calories burned for any given activity, so I assume its the other times.
I have to agree -- in February I started to see some 2400+-calorie (burn) days that I'm certain weren't that high. Those strange highs have plummeted and now I'm seeing a more consistent 2100-2200 calories burned a day since the update. I'd have to guess at some point errors were creeping in and the update repaired the issue. It's all reverse engineering in the end though -- who really knows.
-Matt
Its quite difficult to determine what is and is not the right value without knowing age, height, weight, activity level and general health conditions as a minimum. So your values may or may not be accurate. But the general trend is consistent with what I am seeing.
Here is an example. Last Friday, I did not workout and walked 5,754 steps/2.8mi and burned 2,195 calories. On Saturday, I did not workout and walked 14,998 steps/7.3mi, or almost 3x as the day before, and burned 2223 calories - essentially the same. Both walks were outdoors (I live in SoCal). While I am not expecting 3x the calories burned, I am expecting a noticeable difference, rather than essentially the same.
Many body analyzer scales have athlete mode versus normal. Perhaps Band needs someway to differentiate sedentary lower HR versus healthier HR that does not work very hard while moving one's body under normal condition, like walking, to better characterize the calories burned. The muscles are moving just the same regardless of HR when walking...