The results are in: after last week’s initial tests of the iPod mini’s battery performance, we noted that we were cautiously optimistic that Apple had quietly remedied certain issues of concern to iPod users, most importantly accurately estimating the mini’s duration of continuous playback. Mid-week, we updated our findings to reflect surprising intermediate results, and now after a series of eleven tests, we have reached several conclusions that might interest potential iPod mini buyers.
Three Types of Tests
We performed a total of three types of tests on the iPod mini, two of which were documented in our previous reviews and report, and one of which is new.
Type A consisted of a full recharge and discharge of the iPod mini’s battery without user intervention or backlighting, using “shuffle songs” mode, no equalizers, and 50% volume with headphones attached.
Type B was identical to Type A except that “shuffle songs” mode was disabled and the iPod was instead allowed to play continuously in sequence through its song list. We did not expect to see a significant difference between the iPod mini’s performance on Types A and B, because in both cases the mini was given complete control over the caching of songs, regardless of the order in which they were played back.
Our third test was an attempt to duplicate results achieved by Walter Mossberg of the Wall Street Journal, using the same testing mechanism he tried with two pre-production iPod minis some weeks ago. (Thank you, Mr. Mossberg, for contacting us to share your testing conditions.) Like our Type A test, Mr. Mossberg used shuffled playback, but he also set the iPod minis on 75% volume and turned the Vocal Booster equalizer setting on, each of which adds to the strain on the battery.
Type A Test Results
For reasons further described below, after printing our intermediate findings, we concentrated the majority of our remaining testing efforts on Type A-style tests. Initially, the iPod mini had failed to meet Apple’s estimates in this type of testing, falling below six hours on its first test and barely exceeding that on its second.
However, after each Type A test, we saw that the iPod mini’s performance improved, eventually exceeding Apple’s estimates by over an hour and a half in three successive tests. In fact, the mini eventually delivered ten hours and ten minutes of continuous playback, coming within half an hour of Walter Mossberg’s ten hour and forty minute performance.
- Test 1: iPod mini hits 10% at 5 hours playback, 0% at 5:20, dies at 5:45.
- Test 3: iPod mini hits 10% at 5:33, 0% at 6 hours, dies at 6:19.
- Test 7: iPod mini hits 10% at 8:55, 0% at 9:22, dies at 9:39.
- Test 8: iPod mini hits 10% at 8:43, 0% at 9:23, dies at 9:46.
- Test 9: iPod mini hits 10% at 9:21, 0% at 9:54, dies at 10:10.
For comparison’s sake, we ran a Type A test on a third-generation iPod that had been discharged and recharged over the course of many months, and found that it came within ten minutes of Apple’s estimated eight-hour battery life:
Test 5: 3G iPod hits 10% at 7:14, 0% at 7:29, dies at 7:50.
By Test 7, and continuously through Test 9, the mini’s battery appeared over time to outperform the third-generation iPod under similar (though not scientifically identical) Type A testing conditions. We had seen similar better-than-3G results for the mini in Type B tests, but were not sure whether the mini and 3G iPods were behaving truly differently, or whether the iPod mini’s battery was still being broken in.
Our working assumption at this point is that after approximately five discharges and recharges of the battery, the iPod mini is capable of consistently delivering power performance near or exceeding Apple’s estimates. Prior to that, we assume, it may not charge or discharge as efficiently or predictably, yielding the discrepancy in results we saw during our first four tests. We had not changed any of the testing conditions or the audio library on the iPod mini, or used the device for leisure listening during our test period. Therefore, holding everything else equal, we can only assume that the performance improvements came from an increasingly broken-in battery.
Type B Test Results
After printing our intermediate Type B findings, in which the iPod mini exceeded Apple’s estimates by approximately one or two hours, we did not continue to run further Type B tests. Though we continue to caution readers that Type B does not represent realistic usage conditions for typical iPod users, the use of smart playlists and other non-interactive forms of continuous audio playback will yield relatively impressive performance times such as these.
- Test 2: iPod mini hits 10% at 7:50, 0% at 8:49, dies at 8:58.
- Test 4: iPod mini hits 10% at 8:42, 0% at 9:40, dies at 9:56.
- Test 6: iPod 3G hits 10% at 7:10, 0% at 7:30, dies at 7:48. Even nearly fresh out of the box, the iPod mini outperformed a well-primed 3G iPod by over an hour on its first test, and over two hours on its second. We believe that battery priming is most likely the reason our comparative test of the third-generation iPod yielded results consistent with Apple’s stated estimate, and statistically almost identical to the 3G iPod’s performance in Type A testing.
We can only assume that another run of this test, slanted as much as it is towards battery conservation, would now yield results matching Walter Mossberg’s best time, given that our iPod mini’s battery has had additional burn-in time since we ran these Type B tests days ago. But instead of running this easy test, we decided to run Mr.
Mossberg’s more demanding test and see how our mini would perform.
Walter Mossberg Test Results
Our final two tests of the iPod mini duplicated the test conditions used by the Wall Street Journal’s Walter Mossberg on two pre-production iPod minis sent for review by Apple. For reference, Mr. Mossberg achieved three scores across two iPod minis: 7:46, 9:15 and 10:40, or two to nearly five hours better than our first Type A test.
We administered our tenth total test at the relative peak of the mini’s performance, following Test 9’s ten-hour, ten-minute Type A result, and we were not expecting a substantial difference when this test was run. But we were surprised.
Test 10: iPod mini hits 10% at 6:28, 0% at 7:08, and dies at 7:27.
Puzzling? Yes. We were going to call an end to our testing after ten successive runs, but we had to run one more test to see whether our results necessarily varied that dramatically from Mr. Mossberg’s findings. And they did.
Test 11: iPod mini hits 10% at 6:32, 0% at 7:04, and dies at 7:19.
When Test 10 had concluded, we wanted to believe that the iPod’s battery was merely experiencing occasional hiccups.
