which is exactly what I would have done if I had had them!
Okay then, np. That test answers exactly 1 question: What is the Break_in capacity of 38 months stored blister Glitter? Beyond that, nothing can be said, stated, concluded. Not even compared since there is no direct reference. No point can be made.
I also get it that one would not want to do more than 2 runs: for 8 eneloops and 1 mh9k, it takes 2 runs because 1 mh9k has 4 slots. And 1 run takes 39h+ (C>D>C = (16+1+5+1+16)hrs). So you're investing 78hrs and still are far away from having gained valid data with which a point could be made on the forum.
For the record, a more meaningful test would be:
1) Taking the Glitter out of the blister at the time of purchase, 38 months ago, recondition them, and finish by performing 1 Break_in test. (a second Break_in test at that point in time would give the exact same mAh result, so 1 Break_in is enough.). The values gotten out from that Break_in test are valid values, e.g. G1234:1999/1998/2002/1997mAh, G5678:1998/2001/1997/1998mAh.
2) Then storing for 38 months.
3) Finally repeating the procedure: taking the stored Glitter out from your storage container, recondition them, and finish by performing 1 Break_in test. (a second Break_in test will give the same mAh result, you'll see.). The values gotten out from that Break_in test are directly comparable to the former test, say G1234:1991/1989/1990/1987mAh, G5678:1985/1988/1987/1990mAh.
For this test, of course, the cells must have the same slot position: G15 always in #1, G26 in #2, G37 in #3, and G48 in #4. Why? Because the 4 slots of the RL mh9k charger differ slightly in performance, this phenomenon is called tolerance.
Such a test would tell us something about max capacity before, max capacity after, max spread before, max spread after, actual loss of max capacity (per cell). And since the two data sets have reference character, it would now be
legit to perform simple statistical analysis techniques on them, like averaging, standard deviation, etc.; the results of the calculations wouldn't be meaningless. (If you perform data analysis on an invalid data set, the result will have no meaning whatsoever. Totally pointless.)
Yes, different strokes for different folks, i agree. I think i made my point, so i'll leave it at that. Merri Xmas y'all.