"There's an awful lot of hoopla about that iPhone antenna." Why yes, there is. And while there's much to criticize about Apple's response, we're glad to see they've stopped pretending the problem doesn't exist.
Apple spent most of the time last week defending the iPhone 4's antenna, using both data and rhetoric that was impressive in its scope.
Apple's primary defense is that every phone's reception sucks when you hold it. The company line from the very beginning: "Gripping any mobile phone will result in some attenuation of its antenna performance, with certain places being worse than others depending on the placement of the antennas. This is a fact of life for every wireless phone."
Apple showed this "fact of life" using three non-iPhones in particular: the BlackBerry Bold 9700, HTC Droid Eris and Samsung's Omnia 2. The phones are death gripped; the bars plummet.
It's true, you can readily demonstrate attenuation on a lot of phones. Just the handful of phones at my place, where I have mediocre coverage, show that. Gripping the Nexus One on the bottom causes a bar to drop; the BlackBerry Bold 9700 drops three bars, precisely as demoed by Apple; neither the Droid X nor the Evo drop any bars held with a normal death. (You have to hold the very top of the Droid X or Evo to knock off a bar.) The iPhone 4, upgraded to 4.0.1, touched in the single weak spot, drops 2 bars.
The difference is that the other phones, with the exception of the Nexus One, require a grip—your hand has to wrap around where the antenna is located to cause noticeable attenuation, while touching the iPhone 4 at a single point disrupts the signal. (As RIM and Nokia were quick to reply when Apple singled them out.)
Which is exactly what Ryan Block asked Apple's Bob Mansfield about at the post-conference Q&A:
Q: How does touching the corner with a single finger seem to cause this issue? It's not just a grip, it can just happen by touching a single finger.
A: Your body is a pretty effective signal absorber. When you make contact with that phone, its performance in contact with you is less than its freespace performance. It's a way to attenuate the signal by some amount.
Then why would cases, which you wrap your hand around, be a working solution to the problem? As an engineer told us, the fact that bumpers are the solution shows that the real problem has to do with skin contact with the iPhone 4's antenna—not attenuation simply from signals passing through your body. Rather, he suggests, it seems like the antenna's impedance is changing as a result of the skin contact, so it's possible that a thin electrically non-conductive coating could solve the problem. (Perhaps this is why Steve promised to re-evaluate the issue in September.) That's why the other phones don't have a signal loss from attenuation that's quite as profound.
It's also why Steve Jobs offered his own "pet theory": that not enough cases have been sold for the iPhone 4. That seems a little silly, as the iPhone 3G and 3GS didn't have a single weak point, affected by skin contact.
The iPhone 4 is using "a very advanced new antenna system." In fact, Steve says it "is a more advanced antenna than ever has shipped on a smartphone before." And, later he says, "looking at the data, we don't think we have a problem." What data is that?
Well, they've spent $100 million on a very fancy facility with 17 different antenna characterization chambers to test phones.
Only 0.55 percent of all iPhone 4 users have called AppleCare about the antenna or reception. (This includes Genius Bar visits, apparently.) That's around 16,500 complaints, out of 3 million iPhones sold. "Historically for us, this is not a large number," Steve says. (What is a large number then? A normal number? He does not say.)
For comparison, HTC says that the Droid Eris—one of the phones Apple used to show that attenuation screws every phone—has fielded reception or antenna complaints from just 0.016 percent of all users. (It's possible that it's harder to find where to complain about the Eris.) Oh, and the return rate for the iPhone 4 is just 1.7 percent, versus 6 percent for the iPhone 3GS. Happy campers!
But then there's this: "I can tell you since we're being totally transparent, that even though we believe the iPhone 4 antenna is superior to that of the iPhone 3GS, I must report to you that the iPhone 4 drops more calls per 100 than iPhone 3GS." The rate of iPhone 4 dropped calls is less than 1 additional dropped call per 100 than the iPhone 3GS.
That number is fuzzier than it seems, too. Converting that to a percentage, less than 1 additional percent could be anywhere from .01 percent to .99 percent. Presumably, Steve would use the smallest number possible—something like less than one half of one percent. So let's round up to one percent. Adding another percent to 100,000 dropped calls is an additional 1,000 dropped calls. Or, put a more damning way, and flipping back to the way Steve framed it, if the iPhone 3GS dropped 1 out of every 100 calls, the iPhone 4 could drop up to 1.9 out of every 100 calls—it's possible we're talking a nearly double dropped call rate.
Still, it's a pretty curious thing. Shouldn't the iPhone 4's "very advanced antenna" result in fewer dropped calls, not more, especially if it's superior?
If you want to boil it all down, Steve says, "The heart of the problem is: Smartphones have weak spots. We made ours extremely visible. Some took advantage of that to demonstrate it. It was very easily demonstrable. We screwed up on displaying too many bars and made that demonstration more theatrical than it needed to be."
Apple says they've been incorrectly displaying the bars, so they've fixed it. "We screwed up on our algorithm...Our choice is to put the correct algorithm in, which we've just done." It's true, they have corrected a real issue, as Anandtech beautifully shows. The iPhone's signal bars painted a much rosier picture of the signal situation than reality, and the compressed dynamic range of the bars made them only slightly more accurate than licking your finger and sticking it in the wind. They're now far more meaningful and evenly distributed indicators of signal than they've been for the last two years. But why didn't Apple do this sooner, if it was so terribly wrong?
Well, originally it made the iPhone's reception look better, that's why. It stretches credulity to super-spandex levels to think it slipped by Apple's engineers for years, undetected. (After all, they're making the most advanced antennas in any smartphone, with $100 million testing facilities. They didn't notice the iPhone used a juiced signal indicator?) The flipside of pumping the bars, what came back to bite Apple, was that the signal drops looked worse than they actually were.
There's a weird subtext about transparency and victimization through all of this, too. A sense of, "See what happens when we're transparent?" Apple's not known for being very open, so it's interesting that when Steve talks about the pains they're taking to be transparent more than a couple times—"we went through a lot of trouble to put this beautiful line in the steel" to mark the weak spot—it's interspersed with comments like the one above and "When people are criticizing us, we take it really personally. Maybe we should have a wall of PR people keeping us away from all that." As a journalist, this is laughable: Apple has one of the most opaque and impregnable walls of PR people in the industry. Apple's silence, enforced by PR staff who will clam up at the slightest hint of someone going off-message, are legendary.
That's why you'll never get a straight-up apology from Apple. They're smart enough to know that would open the door to a class action suit and too proud to throw themselves on their sword over a single irritating-if-minor issue. It's just not their style and never will be while Jobs is in charge.
See all those numbers? It's a little problem. We work hard. We love you. We make great stuff. We're giving out free cases. Stop talking about this. Steve says, "We've been working really, really hard for the last 22 days to try to understand what the real problem is, so that when we solve it, we actually solve it, rather than just putting a bandaid on it."
This elides the fact that cases are a bandaid. Almost literally—it's a piece of material affixed at top a gaping hole in the phone. There is a problem when people's skin comes into contact with the weak spot on the phone. On every single iPhone 4.
So! There is a problem with a particular point on the iPhone 4's external antenna. When you touch that single point with bare skin, you lose signal strength—around 24dB, to precise. The old method of displaying signal bars did made it look worse, because of the way it compressed the dynamic range. The free cases will help—significantly, according to Anandtech's numbers, but it's not a permanent fix. The vast majority of the time, it does offer better reception.
Apple's done the right thing, for now, by fixing the iPhone's signal display, which might be the best of any smartphone—even if it's a little ridiculous that this is what it took for that to happen. And the free cases do ameliorate the antenna problem, even if they don't fix its cause. Maybe, as Jean-Louis Gassee suggested, it should've been framed as a tradeoff from the very beginning. Informed consumers, radical transparency, crisis averted. Everyone would've talked about how much they love their iPhone anyway from the very beginning.
The iPhone 4 is a fantastic phone—among the best yet made—and the antenna issue may dent its crown, but doesn't dethrone it.
Illustration: Nikki Cook