Nike + iPod Sport Kit Accuracy (Read 987 times)

I have been running some accuracy tests on my new Nike + iPod Sport Kit. My initial results can be found here: iPod Accuracy. I want to collect a lot more data, and have a bunch of different tests I want to run, so stay tuned... Btw, if you have done any type of accuracy testing on the iPod, I'd love to hear about it.

I have been running some accuracy tests on my new Nike + iPod Sport Kit. My initial results can be found here: iPod Accuracy. I want to collect a lot more data, and have a bunch of different tests I want to run, so stay tuned... Btw, if you have done any type of accuracy testing on the iPod, I'd love to hear about it.

I've been extremely - and surprisingly - pleased with the accuracy so far. With a couple exceptions, it seems to work as well as my GPS unit, which cost roughly 10 times as much. I only calibrated it at 400m; I wasn't even aware there was a 1600m option. I suspect you're right that you'll get more accurate results that way. While I have definitely not been as scientific about it as you have, here's a few things I noticed: * Uncalibrated, it seemed accurate only to about one-tenth mile per mile (which is just about what they claimed. It pretty consistently told me that 1 mile was actually 1.1 mile. * After calibration (only at 400m), it was absolutely accurate on the same track, at 1 mile and 2 mile distances, to within 2 or 3 paces. (Which might even be better than the GPS, to tell the truth. On a curved track, the GPS can cut corners sometimes.) * I messed around with the calibration, and it seemed that accuracy is markedly improved if you calibrate both the walking and running. I initially did only the running calibration, and it seems (and this is just perception here, no science involved) much less accurate; once I'd calibrated both, it improved significantly. So the lesson is - do both. * Here's the one big limitation: it seems clear that the calibration is MOST accurate for paces near the pace you used when you calibrated it. I was smart enough to guess this ahead of time, so I calibrated it right at my normal easy pace, what I do most of my running at (around 8:45/mile). So when I do most of my runs, it is perfectly accurate. It told me my 7 mile run was 7.04 ... and to tell the truth, since my measurements are coming from GPS or Google maps, I don't know EXACTLY how far it is ... so for all I know, the iPod/Nike thing is giving me the best measurement. Either way, it's definitely close enough. * But at tempo run pace, when my pace drops to a couple minutes per mile faster than easy run pace ... the accuracy ain't so good. Again, perception not science ... but I'm pretty sure that at faster paces, it's telling me 2 miles is 2.1 miles. Not a huge error, but enough to make it less useful for my tempo runs. * It also seems less accurate on mixed-pace runs, fartleks or trail runs. * The pace per mile never seems particularly accurate, although I've never seen any other device that gets pace right, either. HOWEVER, it seems that this improves dramatically when pace is kept really steady (and closer to the calibration pace). I ran with my lady today, and she runs slow and steady - really steady - and I was amazed that the pod kept reporting the very same pace. When I use it on my own runs, it'll tell me 12:00 miles one time and then 15 seconds later tell me 6:00 miles. Today, it stayed within a 10 second range the whole time. Too much information? Well, there you go - initial findings and perceptions. I could definitely be wrong about some of it (as in, maybe my tempo course actually is a little longer than I think it is!). But the bottom line is that it's pretty damn accurate - and for the price, nothing comes close. In fact, if you're running on a course with a lot of turns ... it might be more accurate than a GPS. And of course you don't get that annoying satellite disconnect problem. All in all, not too shabby. Plus, I dig it when Paula Radcliffe tells me what a stud I am.

I don't believe the accuracy improves after calibration.  Calibration is really just a fudge factor.  As you said above, the biggest factor for accuracy is stride length and pace.  I have found mine to be somewhat accurate.