I've been putting off this review, because it doesn't thrill me to warn even a subset of people away from a popular product. But that's exactly what I have to do - in short, although the Nike+iPod Sport Kit can be a fun addition for anyone who runs with an iPod or wants a bit more encouragement to run, competitive runners shouldn't bother with it. It simply isn't worthwhile as a training aid for anyone who values ease-of-use, lap counter features, and distance and pace accuracy.
First, some background. I'm a competitive runner. I train with the High Noon Athletic Club at Cornell four or five days per week, and I like to run about 40 miles a week. At least one of those days will be a speed workout and another will be a long run. In the summer and fall, I often race on weekends, at distances ranging from 5 kilometers to 10 miles, including numerous trail races. In the winter, I run indoor track, concentrating on the mile and 2 mile races and their metric counterparts, the 1500 and 3000 meters. Although I'm sufficiently grown up that I don't need to win to feel good about myself, I do push hard, and one of my happiest moments in 2006 came in December when I broke my 21-year-old high school personal record for the 2 mile by 2 seconds. For that race, I was planning lap splits to the second and taking into account the 18 meter extension beyond 16 laps. For me, running is a hobby, a social activity, great exercise, and a needed mental break, but for what I do, accuracy also matters highly.
What I'm trying to convey is how, as a competitive runner, my opinions are akin to those of a professional graphic designer faced with a program that claims to bring illustration to everyone.
Design Problems -- When Apple and Nike announced the Nike+iPod Sport Kit last May (see "Grab Your iPod and Run," 2006-05-29), I was excited at the chance to combine my running with my interest in Apple technology. The first negative came upon reading that Nike shoes were required, since I don't wear Nikes, and serious runners don't risk injury by switching brands or even models lightly. But I quickly learned that users had discovered many ways to attach the sensor to other shoes, so I purchased the $30 kit (I already had an iPod nano, the other requirement that would add at least $150 to the cost of getting started).
When the Nike+iPod Sport Kit arrived, I eagerly unwrapped it, read the instructions, and then stopped, flummoxed, since although I could figure out a way to stick the sensor into my Mizuno Wave Inspire running shoes, I couldn't figure out what to do with the iPod nano itself. I didn't want to hold the expensive little gadget in my hand for the 45 to 60 minutes I normally run, and all the standard ways of attaching it to one's body seemed to put it in a pocket or on an armband. Running shorts don't have pockets, and both pockets and armbands hide the iPod display and controls.
That matters because I never run with earbuds or headphones of any sort and thus wouldn't hear the iPod's audible feedback. Most of my runs are with my High Noon friends, where listening to the iPod would simply be rude. But even when I run by myself, I don't wear them. I don't like the feel of things in my ears, I don't like wires dangling around my body, and it's dangerous to be on the road without being able to hear cars or dogs reliably (there's a reason headphones are banned in many races). So I need a visible interface, and I also need easy access to the controls for pausing the workout when I'm stopped at an intersection or if someone has to stop to re-tie a shoe.
And, of course, I wanted something to protect the iPod nano from sweat and rain. Rain doesn't get in the way of running, and I can't necessarily predict when it will start raining in the middle of a run.
I was blocked - I couldn't figure out how I'd use the Nike+iPod Sport Kit.
Marware to the Rescue -- Help soon appeared in the form of advice from iPod expert Dan Frakes, who covers the iPod world for Playlist. I had asked Dan privately if there were any wristbands for the iPod nano while it had the Nike+iPod receiver attached, and though he had initially answered negatively, he alerted me when Marware announced the $30 Sportsuit Relay, which seemed to answer all my problems.
The Sportsuit Relay provides a neoprene and rubber wristband case that almost completely covers the iPod nano with the Nike+iPod receiver attached (there's a small hole for the earbuds, but I'd only worry about water getting in during a torrential downpour), along with a little Sportsuit Sensor pouch that attaches the Nike+iPod sensor to one's shoelaces with Velcro. Marware also provides extensions for the wristband case if you wish to wear it on your upper arm. It's a good product, and I recommend it for anyone who wants a water-resistant case for an iPod nano with the Nike+iPod receiver attached.
My main problem with the wristband is, unfortunately, out of Marware's control. The iPod nano is long and thin, and fits reasonably on my wrist. But the screen is deucedly hard to read in that orientation, and there's no way to rotate the iPod display 90 degrees.
Calibration and Testing -- The Nike+iPod Sport Kit uses an accelerometer in the sensor to detect foot motion, and although it can be used out of the box, Apple suggests calibrating it by running a known distance while in calibration mode. According to Apple's Nike+iPod Technical FAQ, the accelerometer works by determining the amount of time your foot spends on the ground, which Apple claims is directly related to your pace. The question is, if you calibrate the Nike+iPod at one pace, does it work accurately at other paces?
To test this, I first calibrated it by running an easy 400 meters (.25 miles) at about 7:30 mile pace. The workout for the day was 5 repetitions of 600 meters (.375 miles) at 4:40 per mile pace. For each of the 600s, the iPod instead reported the distance at either .31 or .32 miles, which is about 15 percent off. On the easy 400s I ran for rest in between each 600, the iPod was accurate, presumably since my pace was essentially the same as the calibration pace. Apparently, running at different paces can hurt the Nike+iPod's accuracy, though there's no way of knowing whether that's because my stride length was longer, my stride rate was slightly increased, or my gait was different at that pace.
My track test was actually somewhat unfair, because the Nike+iPod interface on the iPod is worthless for a track workout anyway. Unlike nearly every digital watch aimed at runners and even the separate stopwatch feature built into recent iPods, the interface lacks a lap counter, forcing you either to start and stop your workout in between intervals, or to stop, reset, and start a new workout for each lap. That's just too hard, and although the interface is easy to understand, the iPod's controls are difficult to feel and press accurately while running, especially when hot, sweaty, and trembling with exertion.
I also took the iPod out on the road for an 8 mile out-and-back run at a moderate pace. I couldn't recalibrate it for the moderate pace first, not having a measured track nearby, but I was more interested in whether or not terrain would make a difference, since this particular run was largely uphill on the way out and downhill on the way back. Alas, the Nike+iPod failed this test spectacularly; recording 3.53 miles on the way out and 1.86 miles on the way back for a total of 5.39 miles. Ignoring the fact that it was 33 percent off the actual distance, since it may have been poorly calibrated at that point, the out-and-back distances should have been the same.
In and around these and other tests that confirmed my results, I ran with the Nike+iPod fairly regularly, comparing its results to known distances and to the reports from my Garmin Forerunner 201 GPS. (Yes, I was quite the sight with all those electronics strapped to my wrists.) In general, the Nike+iPod varied quite a bit, but since my pace for any given run also varies, often within the run, there was no way of predicting the results.
(As Paul Lightfoot reported in "Running with a Garmin Forerunner GPS," 2005-07-25, the Garmin Forerunner 201 is spotty; if it gets a good satellite lock for an entire run, it can be highly accurate, but if it has trouble - presumably due to cloud cover, satellite positions, or the color of your shorts - its accuracy drops precipitously. But the device is easy to read, includes a good lap counter, and is waterproof. Garmin also claims its newer models receive satellite signals better, and the $375 Forerunner 305 has a wireless heart rate monitor and an optional $100 accelerometer-based Foot Pod that reportedly kicks in automatically for indoor training or times when the satellites aren't aligning.)
Since pace accuracy depends on accurate distance measurement, the Nike+iPod pace numbers are also suspect. For instance, if I ran 600 meters in 1:47 (a 4:45 pace), and the iPod recorded the distance as .32 miles instead of .375 miles, the pace thus changes to a radically different 5:34. Confusingly, although the Nike+iPod manual never says this, while a workout is taking place, the pace reported is your current pace, not the pace for the workout so far. Thanks to TidBITS reader Mark Wickens for alerting me to this fact; I had been quite bothered that the reported pace seemed inaccurate during workouts.
More Testing Needed? Since initial Web publication of this article, I've received several comments about why my Nike+iPod provided inaccurate results. Mike Donato, an Australian computer consultant who's also an athletics coach, a champion rogainer, and a Green Beret (so I'm not going to disagree with him in person!), passed on the results of his testing, which put the Nike+iPod at about 99 percent accurate when used with the proper Nike shoes, and only about 75 percent accurate when used with Marware's sensor pouch. I can easily accept that the Nike+iPod would require calibration for accurate results with the Marware sensor pouch, but I can't see how using the Nike shoes would eliminate the problems I encountered. After all, when used at the calibration pace with the Marware sensor pouch, the Nike+iPod was entirely accurate on a flat course. Unfortunately, since I've had bad experiences with Nike shoes in the past, I'm uninterested in spending $100 or more for a pair of shoes that I wouldn't wear past a few test runs.
Mike Banks of Grantwood Technology also wrote to recommend his company's Shoe Pouch, a competing product that costs $6. Mike said that he had heard other criticisms of inaccuracies introduced by using the Marware sensor pouch, which he speculated might be due either to movement of the sensor on the shoe (attaching it firmly requires attention) or to improper alignment of the sensor (in comparison to how it was designed to fit in the Nike shoes). Mike believes that the Shoe Pouch solves these problems and will provide better accuracy. I'll test it and report back on my findings. Subsequent discussions with Mike Donato also pointed toward inaccuracy caused by looseness - specifically a comment about the Marware sensor pouch in a roundup of hacks for using the Nike+iPod Sport Kit with shoes from other manufacturers.
When I asked if they had received reports of inaccuracy, Lee Goldring of Marware told me that all the complaints Marware had received about the accuracy of the Nike+iPod when used with the Sportsuit Sensor pouch were related to incorrect orientation; as soon as the user placed the sensor in the pouch with the orange side down, as indicated in an insert that ships in the pouch and on Marware's Web site, the accuracy issues went away. I had been using the sensor in the correct orientation, so that wasn't my problem.
Clearly, I'll have to try attaching the sensor to my shoe more firmly; perhaps I'll try digging a hole in an older pair of shoes as well. In a short out-and-back run after reading all these comments, with the sensor pouch firmly attached, I got exactly the same distances, though I was running at the calibration pace on a flat course.
It's worth noting that even Apple admits to accuracy problems in various places. A support article on Apple's Web site makes clear the importance of consistent pace during calibration, saying, "If you run or walk with an inconsistent pace or travel a distance that is over or under the expected calibration distance, you may get unexpected results (such as incorrect distance of a workout)." Plus, the manual reads, "Even after calibrating, the accuracy of distance measurements may vary depending on gait, running surface, incline, or temperature." Fine, but all those things vary for serious runners.
Fine for Fitness Running -- Don't misunderstand my criticisms here. The Nike+iPod Sport Kit does essentially what it says it will do. It provides some data about how far and how fast you run, and it can help you run for a specific (if not terribly accurate) distance and for a specific amount of time. You can upload your data to a Nike Web site whenever you sync with iTunes, and many people have found it encouraging to be able to compare their workouts, monthly distances, and race times with their online friends. And yes, if you have trouble being alone with your thoughts and the environment around you while running, you can listen to music or podcasts.
In the end, anything that encourages more people to run is a good thing, and particularly when coupled with Marware's Sportsuit Relay, the Nike+iPod can do just that. But if you're a competitive runner looking for an easy-to-use training aid that provides accurate distance and pace measurements (particularly without forcing you to switch shoe models), you'll want to look elsewhere. And even if my criticisms of the Nike+iPod's accuracy turn out to be testing error, its awkward physical design and lack of a lap counter render it less useful than other devices.