Slashdot Mirror
Apple To Issue a 'Fix' For iPhone 4 Reception Perception
Lisandro and several other readers let us know that Apple has just released a statement addressing the signal issues a lot of users are having with their iPhone 4. They claim to have discovered the cause for the drop in bars, which is "both simple and surprising" — a wrong formula used to calculate how many bars are displayed for a given signal strength. "Upon investigation, we were stunned to find that the formula we use to calculate how many bars of signal strength to display is totally wrong. ... we sometimes display 4 bars when we should be displaying as few as 2 bars. Users observing a drop of several bars when they grip their iPhone in a certain way are most likely in an area with very weak signal strength, but they don’t know it because we are erroneously displaying 4 or 5 bars. Their big drop in bars is because their high bars were never real in the first place. ... We will issue a free software update within a few weeks that incorporates the corrected formula. Since this mistake has been present since the original iPhone, this software update will also be available for the iPhone 3GS and iPhone 3G." Wired notes that there is still a signal drop when the iPhone 4 is gripped in particular ways.
This isn't a fix for the attenuation caused by touching part of the antenna, it's a fix for a longstanding software issue that makes it harder to manage. The issue's still there, and if you're seeing lower signal or slower speeds on your iPhone 4 than your previous iPhone, the patch won't fix that.
No kidding!!! What do you say at this point?
It's actually the other way around, but never mind.
Anyway, I'm not sure what to make of this. On one hand, this will certainly earn a fair amount of ridicule as it sounds like redefining reality to what Apple wants it to be. A fix to the Reality Distortion Field, so to say.
OTOH, I've had some experience with sensors, and there's sometimes ambiguity to how the signals should be evaluated/presented. I'd guess that a logarithmic scale is a better fit for the relationship of absolute signal strength and perceived quality than a linear one. If they previously used a linear scale, this update might be appropriate.
This doesn't change the fact that the signal strength changes with how you hold the phone. If the change manifests itself only in fewer bars, everything will be alright. If actual call quality or reliability is affected, this change won't do anything for that
Fleur de Sel
There's a serious analysis here, with some extra commentary by OS News here.
First off I totally beleive this is possible. Very often my non-apple phones flicker between 4 bars and no bars.
But what is remarkable about this is that according to Pogue, apple designed the phone to find the "best" cell connection not the strongest. Apparently there is a difference. Naively I could appreciate that a tower that oscillates between 4 and 0 is worse than a steady 2.
Thus is it surprising that given they paid attention to that kind of detail they would get the actual formula wrong.
My guess is that the formula used to pick the cell tower is distinct from formula used to drive the display. Or they did something like add a variance bias to the mean to represent steady weak towers as having more bars.
In any event, assuming their explanation is correct, it does seem to jibe with their other public statements insisting that there is no actual problem, just a perceived one and that all cell phones do this to some extent.
Some drink at the fountain of knowledge. Others just gargle.
OK, funny funny. But just to be a debugging Nazi, the "min" and "max" functions are used backwards in the code above. You have to "max" with 0 and "min" with 5, otherwise you just force the value to one extreme. It makes sense if you think about it.
It's anything but simple.
To be fair, I'm more careful with code that I'm not writing as a joke.
With digital phones, it's not necessarily signal strength (cell phone parlance = RSSI, Received Signal Strength Indication) you want to display. You want to include some measure of the signal/noise ratio, since that is a better indicator of a phone's ability to communicate.
And, in fact, that's what most cell phones do. For CDMA, it's usually some combination of Ec/I0 with signal strength, and for GSM it's some combo of reciprocal bit error rate combined with signal strength. Often time averaged and perhaps peak reading. So it's not just a simple "x dB RSSI = y bars" calculation. The combined metric is called SQE (Signal Quality Estimation).
"National Security is the chief cause of national insecurity." - Celine's First Law
Anandtech posted a review of the iPhone 4 the other day and they have a break down of the signal strength in dBm compared to how many bars are displayed. The specific page is here.
Basically it looks like there is a huge range for what is considered five bars, and a small range for the remaining four bars.
It's Defective by Design (TM)
For once, Apple copied Microsoft.
Check the consumer protection legislation in your area. It probably has something along the lines of products needing to be fit for the purpose for which they were purchased for a reasonable length of time, taking into account the price paid.
There are two types of warranties: Legal and Conventional.
Conventional Warranty (limit your rights): We warrant foo for 1 year (no warranty on batteries, screens, keyboards, accessories, etc).
Legal Warranty: Fitness for purpose for which it was purchased, taking into account price paid, etc. In other words, you paid $3k for that big-screen tv and it croaked 1 day after your conventional 1-year warranty expired? You can still use the legal warranty via small claims court. They can't hide behind the limitations of the conventional warranty - the conventional warranty is in addition to the legal warranty (which makes extended warranties stupid purchases).
If you are going to bring up the anandtech study, you may also want to mention that the article states when signal quality is low the iPhone 4 is much better at keeping calls alive:
From my day of testing, I've determined that the iPhone 4 performs much better than the 3GS in situations where signal is very low, at -113 dBm (1 bar). Previously, dropping this low all but guaranteed that calls would drop, fail to be placed, and data would no longer be transacted at all. I can honestly say that I've never held onto so many calls and data simultaneously on 1 bar at -113 dBm as I have with the iPhone 4...In reality, reporting based on SNR makes a lot more sense, since I couldn't make calls drop driving around an entire day cupping the phone, despite being at -113 dBm (1 bar) most of the time.
I've not had dropped call issues from the iPhone4. That's not to say you will never have a dropped call, this is after all AT&T we're talking about here. But I have had much better results in making and keeping calls compared to the older phone, so people who are holding off buying an iPhone worried about dropped calls are doing themselves a disservice.
For me, tightly gripping has more of an impact on data speed than calls - and even then, it doesn't always affect data speed. But it's again a worthwhile upgrade, because the phone has better latency and so network use feels more responsive as per DaringFireball.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
I believe they are currently using a non-linear scale. I don't know the exact relationship, but it sure looked logarithmic at a quick glance. '5 bars' was covered by ~50% of the 'usable' signal reading. From what I understand, this is reasonably representative (perhaps slightly optimistic -- "look we have more bars" marketing strikes again?) of actual performance, so if they do too much tweaking, it'll be much less representative of real quality drops/increases. Plus that doesn't solve the fact that (some) people are actually quantifying this as a drop in data speeds or calls dropping when held in certain ways -- and not solely in 'number of bars'.
Here's the link with numbers and more info.
Funny, but sadly you swapped min and max (and forgot closing parenthesis) :
max(5, log(signal_strengh_dB)) >= 5
therefore
min(0, max(5, log(signal_strengh_dB))) == 0
min(3, max(5, log(signal_strengh_dB))) == 3
Swapping min and max :
min(5, log(signal_strengh_dB)) <= 5
0 <= max(0, min(5, log(signal_strengh_dB))) <= 5
3 <= max(3, min(5, log(signal_strengh_dB))) <= 5
I have discovered a truly marvelous proof of killer sig, which this margin is too narrow to contain.
There is a reference to it here:
http://www.appleinsider.com/articles/10/06/28/nokia_pokes_at_iphone_4_death_grip_gets_called_on_it.html
Jumpstart the tartan drive.
Man up and log in.
I don't understand the logic behind this suggestion. Could you explain it to me?
Slashdot contains, as one of its greatest features, a meta-moderation system. With a run-of-the-mill, provide-an-email sign up system, you too can log into the site and participate. Your comments will be tied to that account, and the moderators can judge the relative worth of your contributions. Further their efforts are judged by the meta system, which determines which moderators get which points.
Another facet of this service is identity, friend/foe, and post history. All of these things can be used to help sort out a poster's relative worth to you, the reader.
The Anonymous Coward circumvents this system in multiple ways. Presumably they do so in order to get their comment across without participating in the system.
'Coward' is implicit. You're afraid to be identified, for one reason or another. Should you work for Apple, please do participate on slashdot under an anonymized user name. This way I can foe you and move on...
BESIDES it doesn't matter - a dropped call is a dropped call. If the iPhone 4 drops calls in areas where other phones (like the iPhone 3) worked perfectly, then the problem...
The iPhone4 does not drop calls in areas where other phones work perfectly. Instead of relying on your vast Apple Hater Rage to inform your opinion, why not try reading a carefully done study instead:
From my day of testing, I've determined that the iPhone 4 performs much better than the 3GS in situations where signal is very low, at -113 dBm (1 bar). Previously, dropping this low all but guaranteed that calls would drop, fail to be placed, and data would no longer be transacted at all. I can honestly say that I've never held onto so many calls and data simultaneously on 1 bar at -113 dBm as I have with the iPhone 4...In reality, reporting based on SNR makes a lot more sense, since I couldn't make calls drop driving around an entire day cupping the phone, despite being at -113 dBm (1 bar) most of the time.
The reality is that the iPhone 4 is BETTER at making and receiving calls. This is my own experience, and of the other people I know who have the phone - and also borne out by actual testing.
As Apple said, if people think there is a problem they can just return the phones, no questions asked. But I don't think Apple will see many takers.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
If you read Apple's statement you'll see that they are changing the display as per AT&T's suggested standard:
"To fix this, we are adopting AT&T’s recently recommended formula for calculating how many bars to display for a given signal strength."
That makes the previous behavior unlikely to be due to some secret cabalistic plan of AT&T's.
How exactly do YOU hold your phone? I just naturally picked up my Droid and held it up to my ear and I pick it up with my left hand with my thumb extended along the length of the left-hand side of the phone and two or three fingers wrapped around the back holding the right hand side of the phone.
Yes, that's how I pick up my iPhone too. That causes no issues in everyday use. Yes, even though you are slightly touching the metal band it two places.
What does start to cause issues, is if you then squeeze REALLY hard and wrap your fingers around the front more. Then you are blocking the signal a lot more. Then you can sometimes (not all the time) create the bridge across the antennas that reduces signal quality.
Thus the term, death grip.
"There is more worth loving than we have strength to love." - Brian Jay Stanley
I can have a trendy, overpriced device
The iPhone costs the same[1] as any other (new) Android smartphone. Less, even, depending on how you calculate it. Now there is the fact that the data plan is no longer unlimited, but I dunno, 2GB is more than twice the volume of the most mobile data I've ever used in a month (even with tethering). But then, I operate with an ad and flash blocker, so I'd imagine others' results would vary.
[1] AnandTech: Apple's iPhone 4: Thoroughly Reviewed
The secret to creativity is knowing how to hide your sources. - Albert Einstein
Yes and no. The issue had to do with the scale used. From AppleInsider:
"Apple's iOS allots nearly 40 percent of reception levels to five bars, from -51 dB to -91 dB. But the distance from four bars to one bar of reception is much less, from -91 dB to -113 dB."
Unless you live under a rock, or have never cracked open ANY cell phone manual, you would know that they all recommend avoiding the lower left side where the antenna is located. The problem with Apple's configuration is that they used 5 bars for -51 to -91 db, which in itself is fine, since any call with those signal strengths would be very stable.
The problem has to do with dropping signal strength when you cover the antenna. Although you couldn't cause any phone to lose 63 db from grabbing it, getting a 20 db drop in signal strength isn't all that uncommon. If someone was in the weaker 5 bar range (say -90 db), which is still a decent signal, the phone would display 5 bars. When you covered the antenna, you could potentially get a 20 db drop, bringing you down to -110 db, which is a very poor signal. It gave the impression that someone went from 5 bars to no bars, or even no service (no service being very possible with -110 db).
There are no standards defining what dB is represented by the number of bars. It is totally up to the hardware manufacturer. As to whether or not this was a mistake, who can say? The haters will claim it's a marketing ploy, the lovers will claim it's a fair way to rate the signal. I'll probably fall in between.
What this fix will do is better represent the number of bars by signal strength in a more linear fashion, so that -90 might only be three or two bars, and covering the antenna would drop you to 1 or no bars. Although any phone might make a call at 1 bar, you should also expect that you might drop it with such a poor signal.
Frankly, I think most of the posts in these and other forums are full of 'noise' from people who hate Apple, from Droid fans, to people who just dislike them for their closed systems. It's an ideal way for them to take shots by spreading FUD and making the issue appear far worse than it actually is. Unfortunately, it also makes it much more difficult for people to judge the scope of the problem, which in itself is a win for Droid from a marketing perspective. I do know that I haven't dropped a single call on the iP4 regardless of the bar display. I suspect most fall into the same category. Reviews of the phone also back up the same. Dropped calls have actually been reduced under the iP4, regardless of the number of bars displayed. Anandtech actually has a nice write up and some more thorough testing of signal strength. What they found is that the exposed antenna's do make the iPhone 4 more susceptible to interference, but not so much that it would affect people with decent signal strength. It actually performs better than a droid for signal loss when it's in a cover:
http://www.anandtech.com/show/3794/the-iphone-4-review/2
Standing Wave Ratio is a measure of how much of the signal that you trying to fire out of an antenna is being reflected back.
A low SWR is an indication that most of your transmitting power is actually being delivered to your antenna.
I find it so hard to believe that anyone would have designed a hand held radio that lets the antenna come in contact with the human that's holding it.
Its bad enough to have capacitive coupling (Where the plastic casing is thin enough to allow the signal to be transfered to the user).
If you think about it most hand held radios (walky-talkies) have their antenna's in the end of a stick. Its only as phones have become small that this stick has become small too and as such produces concentrated RF emissions right next to your head. As a result manufacturers mostly place the gsm antenna at the bottom of the phone.
Most antennas have a hot end (usually the tip) its high voltage low current and a cold end (usually the base) low voltage high current.
Is it purely down to detuning or just a plane short circuit?
I've seen demo's on youtube that show an actual call being dropped, no amount of s-meter recalibration is going to fix that.
It would be interesting to see if the part of the side antenna that is (so say) shorting is the hot end or the cold end. If its the hot end then the problem may well be caused by the bluetooth or wifi radio actually swamping the gsm radio receiver. This could be fixed in software by detecting that the phone is being held (somehow) and turning off the wifi or bluetooth whilst in a call. This could be made to come into play only in a poor signal area.
(sorry to those of you in the know. Being a HTC user I'm not sure which side the wifi / bluetooth is on)
If its the cold end that is shorting then the best answer is a bumper. I'm sure it won't take long for these to be appearing very cheaply in ebay.
I can tell you that your entire post is horseshit. How do I know? Actual testing with actual numbers provided by Anandtech. Specifically, this page.
The bars do not convey an accurate sense of the phone's ability to make a call/transfer data. What a lot of people are seeing is going from 4 or 5 bars to 1 or none, but the difference in signal quality isn't from "super awesome to poor" (as a 5 bar to 1 bar drop implies) it's really from "not so good to poor" (which would be more accurately conveyed by a 3 bar to 1 bar drop).