me & her

Failure Rates and the iPhone 4

Yesterday, early adopter reports began pouring into gadget blogs concerning two separate problems with Apple’s not-even-officially-released iPhone 4. The Unofficial Apple Weblog was first to report issues with the iPhone’s new Retina Display in which faint yellow spots (or sometimes yellow lines) were appearing on the screen. That issue was quickly followed by widespread reports that some iPhone 4s suffer (at least the perception of) lost reception when held in the left hand.

Engadget:

What’s more annoying than spending hours lining up for a shiny new gadget? Learning that your precious phone can’t actually connect to the network. Well, depending on how you hold it — word has it that the iPhone 4’s bottom-left corner isn’t playing nice with your skin. If you recall from the keynote, that’s where the Bluetooth / WiFi / GPS antenna meets its GSM / UMTS counterpart.

In both instances, some users have the problem, and some don’t. At this point, the “lost bars” glitch seems to be the more widespread occurrence, even though accounts about what triggers the issue and even how it manifests are pretty varied.

Walt Mossberg’s pre-release review touched on the subject:

Yet, in some places where the signal was relatively weak, the iPhone 4 showed no bars, or fewer bars than its predecessor. Apple says that this is a bug it plans to fix, and that it has to do with the way the bars are presented, not the actual ability to make a call. And, in fact, in nearly all of these cases, the iPhone 4 was able to place calls despite the lack of bars.

As it to be expected, Gizmodo has posted articles about both issues, and I’ve seen a few people surmise that perhaps this is just Gizmodo being Gizmodo. Unfortunately, too many other blogs are reporting on the issues for that to be the case. These are real issues.

Still, these early reports may not be the end of the world for Apple, or for those still hoping to nab an iPhone 4. I’ve seen claims that a failure rate of 3 to 5 percent is acceptable in consumer products. This Daily Tech article about Xbox 360 failures, for example:

Asked differently about whether or not the Xbox 360 falls into the ‘normal’ three to five percent return rate, Holmdahl said, “We don’t disclose the actual number,” and “We don’t comment on that.”

I’m not sure where that range of percentages comes from, and haven’t located any study to corroborate the author’s claim, but looking at various articles about the failure rate of gaming consoles from Nintendo, Sony, and Microsoft, it certainly seems reasonable.

For example, the same Daily Tech article cites an astonishing 33% failure rate for the Xbox 360. 1/3 of every Xbox sold experiences a critical failure requiring a replacement unit. That number could even be as high as 54%, according to a blog posting on the Seattle PI website:

In fact, a Game Informer survey of 5,000 readers found that the Xbox 360 has an astounding 54.2 percent failure rate. That means 54.2 percent of Xbox 360 consoles fail in one way or another.

I’m willing to accept the 3-5% range because the same article touches on the failure rate of Sony and Nintendo’s consoles, as well:

That’s well above the reported failure rates of Sony’s PlayStation 3 (10.6 percent) and Nintendo’s Wii (6.8 percent).

So, if Apple sells 600,000 iPhones, they’ll need to get to 300,000 (or 200,000) defective devices before they match Microsoft’s undeniably shoddy record, 60,000 before they match Sony’s reported rate, or 15,000 before they even go beyond what is considered an “acceptable” failure rate. Furthermore, it’s worth noting that Microsoft’s extreme failure rates haven’t ever deterred consumers from purchasing Xbox 360s. What the industry deems “acceptable” doesn’t seem to correlate in any meaningful way to consumer purchasing habits.

So: 15,000 defective iPhones would be a big number, and it would certainly lead to a lot of angry customers, but it’s (apparently) not unreasonable. 

It’s also worth mentioning that this is the first completely redesigned iPhone in a couple years: This launch is likely to see a larger fail rate than (say) the 3GS because the 3GS didn’t really deviate from the hardware design of the 3G, which itself wasn’t a radical departure from the 1st generation iPhone. Apple’s manufacturing partners had likely worked out most of the glitches.

The iPhone 4 redesign not only involves completely new materials, but in the case of the Retina Display, it involves a completely new process in which the screen and the LCD are fused together. It’s also the first iPhone in which the antenna is a structural component of the device itself. (Indeed, there is some speculation that this could be the cause of the reception issue. If that turns out to be the case, the iPhone 4 will have to be seen as a monumental failure of pre-release quality control.)

Apple’s biggest and most immediate problem is going to be bad PR: For whatever reason, there are a lot of people who want to see Apple fail, and fail spectacularly. Consumers, for their part, don’t know anything about failure rates; a defective product is a defective product. Someone whose first experience with an iPhone is a splotchy yellow screen or an alarming reception issue won’t care (or won’t know) that the HTC EVO 4G and the Droid Incredible are suffering from embarrassing screen issues as well.

For better or for worse, Apple is the Goliath of the smart phone industry, and there are a lot of Davids who will be thrilled to play up even an acceptable failure rate, while glossing over the HTC display issues. (HTC glossed over the issues, why shouldn’t everyone else?) Couple that with the fact that consumers are more likely to report problems with new devices than satisfaction, and that blogs are more likely to collect and promote horror stories than success stories (this is where concerns about Gizmodo’s integrity become more valid) and this could become a massive PR nightmare for Apple, given that the iPhone 4 was already suffering from an unusually glitchy launch.

If, as Mossberg suggests, the reception issue can be solved via a software fix, Apple needs to post that fix sooner, rather than later. Today would be good—yesterday would have been better.

My guess is that the yellow splotches will eventually be tied to an isolated glitch in the manufacturing process; probably when the glass is fused to the LCD. Whatever the cause, so long as this issue remains relatively rare, and so long as Apple has replacement units ready, I don’t see this turning into much more than a hiccup for those unfortunate enough to be saddled with an affected unit.

Within a few days, I suspect Apple will release a software update for the iOS which will purport to correct the reception issue, at which point some will claim that the issue has been fixed while others will claim that they’re still seeing the issue. Supporters will assert that those people are simply confusing AT&T’s legendary reception issues with the (now fixed) iOS software issue and detractors will gleefully assert that this is the beginning of the end of Apple’s second coming. In support of the latter crowd, Gizmodo’s Jesus Diaz will probably post a new article about Apple’s latest monumental failure, at which point his readers will claim that Apple sucks and that everyone should should have bought the more open, flash-enabled HTC EVO 4G instead of being a stupid fanboy and drinking the Steve Jobs kool-aid.

If all goes well, the fervor will then slowly die down and Apple will go on to sell a lot of iPhones to a lot of stupid fanboys including, in all likelihood, Jesus Diaz.

Somewhat related:

For the last month or so, I’ve been battling a unique reception issue on my iPhone 3GS. Everything works as expected until I lose service in an area with poor coverage. (My cubicle, for example.) When that happens my iPhone doesn’t seem to want to automatically reconnect to AT&T’s voice network, even if I move to an area where I normally get good reception. The data network works (I can browse the internet on 3G) but I can’t make calls and the status indicator reflects zero bars. Once I reboot, reception returns to normal. After a trip to an AT&T outlet and then an Apple Store (and after attempting the suggested fixes: A new SIM card and a hard restore without using my backup) an Apple retail employee finally relented and replaced my 3GS. Unfortunately, my new 3GS is exhibiting similar symptoms. I suspect there’s a software conflict and my experience is enough to lead me to believe that the iPhone 4 reception issue could, as Apple suggests, be a software problem, or even tied to a bug in a specific application that a significant number of people have installed.