Originally posted by abacus07 using capacity ratio only makes sense if the equation looks like this: 2500(charged)-0(discharged)/2000(charged)-0(discharged). That's not how batteries work.

Perhaps I am not understanding you.

Capacity should be from standard IEC testing taking the measurement

after a standard charge (1/10C for 16 hours)

then a rest of at least 1hour (but not more than 4 hours)

and a 0.2C discharge until the battery reaches 1.0V

My regular eneloop usage in the K-x does not go down quite as low as 1.0V -

but I do a discharge at 0.2C after they have been used in the in the K-x to shutdown -

the remaining capacity as indicated by the Maha C9000 using 400mA (=0.2C) discharge

is between 0mAh to about 70mAh -

which means the K-x manages to use almost all the capacity of the regular eneloop

from 0% to within about 3.5% - averages to about 1.75% remaining capacity -

so usage average of 98.25% capacity

Taking the worst case example of 70mAh remaining capacity

(the largest remaining capacity I measured over 2 years with 29 battery changes, using 3 sets) -

(for everything else being equal) then the equation is:

(2500-70)/(2000-70) = 2430/1930 = 1.259

ie: about 125.9% this is a difference of only 0.9%

from the indicative estimate using the simple capacity ratio 2500/2000 = 125% -

which should be well within experimental/measurement error.

However since not everything is equal - this is very dependent on how well the new eneloop XX maintain their operating voltage under-load - if for some reason they drop their voltage under-load more, then the K-x may "prematurely" shut down, when there may well be remaining capacity capable of powering less battery fussy devices.