Issue with Proximity Sensor during calls - Continued v2

I just received my iPhone 4 two weeks ago and the proximity sensor on my phone is about half way between the speaker and the top of the phone. The facetime camera is round and to the left of the speaker and very clear to see. You can see the proximity sensor using a flashlight at angle to the iPhone. It is very difficult to see, it almost looks like a red smudge. If you make a call (not on speaker) and put your finger over that spot the screen should go blank immediately. I seem to be one of the lucky ones that don't have a problem with the proximity sensor...my issue is with the iPhone 4 and Microcell but that's another story.

mbassoc2003 wrote:
Using your phone on your shoulder is a big no no. You can't do it with the iP4 because of the type and position of the sensor. The 'software fix' in the current build of the beta isn't a fix. It's a patch that tells the iOS to wait slightly longer before it switches the screen back on. It's yet to be seen how effective it's gonna be, but don't hold your breath. I think they are going to do their best within the realms of trying to compensate with the software, but I think people who have the issue regularly are likely to be disappointed, as all they can do with software is delay the incremental periods between the screen flashing on and off during calls. The issue remains a user interface issue, and whilst it's not the user's fault, just as the 'death grip' issue is not the user's fault, it is still the user that creates the fault. That is something no software is going to be able to eliminate.

I respectfully disagree. I believe the issue to be that the OS is getting confused on the state of the headset/speaker phone/ear speaker bit. Some combination of actions is flipping this bit at the wrong time and the OS is re-enabling the screen thinking you've inserted the wired headset, gone to speaker phone, or BT (as it's supposed to). I'm betting it has to do with the new background audio APIs which are proving difficult for app developers to master.

In working with Apple on this, I'm focusing my testing on phones set up as new with no 3rd party apps and comparing to one with bunches of apps (especially bg audio enabled) to see if there's any kind of pattern to latch on to.

bmwhd1 wrote:

mbassoc2003 wrote:
Using your phone on your shoulder is a big no no. You can't do it with the iP4 because of the type and position of the sensor. The 'software fix' in the current build of the beta isn't a fix. It's a patch that tells the iOS to wait slightly longer before it switches the screen back on. It's yet to be seen how effective it's gonna be, but don't hold your breath. I think they are going to do their best within the realms of trying to compensate with the software, but I think people who have the issue regularly are likely to be disappointed, as all they can do with software is delay the incremental periods between the screen flashing on and off during calls. The issue remains a user interface issue, and whilst it's not the user's fault, just as the 'death grip' issue is not the user's fault, it is still the user that creates the fault. That is something no software is going to be able to eliminate.

I respectfully disagree. I believe the issue to be that the OS is getting confused on the state of the headset/speaker phone/ear speaker bit. Some combination of actions is flipping this bit at the wrong time and the OS is re-enabling the screen thinking you've inserted the wired headset, gone to speaker phone, or BT (as it's supposed to). I'm betting it has to do with the new background audio APIs which are proving difficult for app developers to master.

In working with Apple on this, I'm focusing my testing on phones set up as new with no 3rd party apps and comparing to one with bunches of apps (especially bg audio enabled) to see if there's any kind of pattern to latch on to.

^This by far makes the most sense of any hypothesis posted in relation to this issue. If so, it could be significantly more time consuming to find the root cause of the problem and fix it.

In working with Apple on this, I'm focusing my testing on phones set up as new with no 3rd party apps and comparing to one with bunches of apps (especially bg audio enabled) to see if there's any kind of pattern to latch on to.

Does this statement mean that you are in fact helping Apple out with this issue? Are you an Apple employee?

So if this is a software issue, is it a case that it is so insignificant that it does not warrant a fix? Or is it a case that the issue is so hard to fix that all the resources of a multi-billion dollar company cannot fix the issue in 10 weeks? It can only be one or the other if it is software related.

I could well believe the issue is so small and insignificant it deserbes no fix, but I cannot believe that if it is software related that Apple have not been able to fix it in ten weeks.

Or, and I'm just speculating, maybe the sensor is positioned poorly and some users cannot work with the phone. Because, if it was a software 'glitch' then we'd all have the problem. And the vast majority of people don't.

just got off the phone with Apple in US (reached using 611 on AT&T)
they offered to exchange my new iphone4. The claimed the PS was a known hardware issue.

I am hesitating a bit because I fear getting a refurbished unit with other unknown issues. They offered to repair my own phone and send it back in 3-5 days. but still i wonder ...

however, trying to wade through this thread, i dont really see much mention of whether new hardware fixed it for other folks.
I have access to 3 different iphone4 and all have the same bad PS behavior.
I still have more time on my initial 30 days, so maybe the s/w fix is the real solution. I am simply not 100% confident that there is a hardware fix that can be made without a phone redesign.

so ... what are your experiences with hardware exchanges.
did apple get a batch of bad PS and thus they can fix the problem with a simple hardware swap ?

To be clear - I am not an Apple employee nor do I play one on TV.

I am a paid member of the iPhone developer community. As such, like any other dev, I can submit defects to Apple. Sometimes Apple has already had the defect reported and closes new submissions as 'duplicate'. Other times, they reach out for additional information and suggest things to check, scenarios to test, etc.

In this case, my open defect on the prox sensor has not yet been closed or marked duplicate and Apple has reached out for more data and specifics on the testing being done to re-create the issue.

I therefore am speculating but with some insight.

I do think there is a reasonable chance this issue is software related. However, it could well be a combination of software and hardware. Besides just a bad batch of prox sensors, there's other ways hardware could be involved.

For example, let's suppose for a moment that the iOS 4 code base is behaving correctly - when on a call using the ear speaker, prox sensor input is used to darken the display and disable it when the sensor detects the phone is being held to your ear. When using any other output device (speaker phone, wired headset, BT) the OS ignores the input from the prox sensor and leaves the screen on (subject to normal sleep timeout). Let's imagine this behavior is controlled by a Boolean that is either True or False depending on which output device is being used.

Now in this scenario, the way things go south is when something other than the authorized methods toggles that Boolean to False. One way that could happen is rouge software (either Apple's or 3rd party). This would be a software problem only.

Another way this could fail would be a slight, intermittent fault within the headset connector itself. Perhaps not even a fault, maybe just a manufacturing tolerance that isn't right. This would be hardware.

Now what if both of the above are in play? Makes for a long diagnosis phase.

FWIW, I think it's a software only scenario but a **** difficult one to find/fix. If it was hardware, I think that would show in a test harness right away and a rolling change at the factory would have happened already.

Thank you for clarifying your thoughts and invovlement. What you've written is interesting for sure. People seem angry that Apple hasn't rolled out the software yet. It's clear it's a complicated issue and the fact the solution is likely rolled into the soon to be released 4.1 update adds further complication I'm sure. Plus people often forget the past, which is a good barometer of the future. And that is that everytime the iPhone has been released in the June timeframe, the first major software update doesn't typically roll out until September. If the past predicts the future, the update will be out next month. Hopefully for those who are really afflicted with the PS issue, there problems will be resolved. Who is it who said "patience is a virtue?"

@greenstem on iphone: My advise would be to return it within 30 days and wait for 4.1. All my friends here in the UK are experiencing the same issue! No point exchanging it, there is a high probability you will get a new phone with the same issue and by the time you realise it has the same issue, your 30 days are up and you are stuck! Return it, and wait, if its fixed with 4.1 buy it again. I am in the same situation, waiting. If its not fixed I am getting the new itouch until the next revision of the iphone.

Message was edited by: Daddy Pig

ask yourself what are you going to do if it is not fixed with 4.1 software and then act accordingly with your own answer. this thread is pretty lengthy and getting tired.

Daddy Pig wrote:
@greenstem on iphone: My advise would be to return it within 30 days and wait for 4.1. All my friends here in the UK are experiencing the same issue! No point exchanging it, there is a high probability you will get a new phone with the same issue and by the time you realise it has the same issue, your 30 days are up and you are stuck! Return it, and wait, if its fixed with 4.1 buy it again. I am in the same situation, waiting. If its not fixed I am getting the new itouch until the next revision of the iphone.

Message was edited by: Daddy Pig

What I don't understand is the posts from people with 4,000 plus posts to their names saying that they haven't experienced the problem and don't know anyone who has - makes no sense.

Meanwhile, I don't know anyone who doesn't have the proximity sensor issue to a greater or lesser degree (and I do know more than a handful of iP 4 users)

rbrylawski wrote:
Thank you for clarifying your thoughts and invovlement. What you've written is interesting for sure. People seem angry that Apple hasn't rolled out the software yet. It's clear it's a complicated issue and the fact the solution is likely rolled into the soon to be released 4.1 update adds further complication I'm sure. Plus people often forget the past, which is a good barometer of the future. And that is that everytime the iPhone has been released in the June timeframe, the first major software update doesn't typically roll out until September. If the past predicts the future, the update will be out next month. Hopefully for those who are really afflicted with the PS issue, there problems will be resolved. Who is it who said "patience is a virtue?"

You unrestricted in support the Apple in it's fraud. It is amassing what you suggesting. Let's Apple sell the phones that people can not even use as a phone and let all be patient for 3 and may more months when MAY BE Apple comes with some kind of miracle.

One one hand the question is what if Apple never can fix it . What then ?
On the other hand ( and it will not happened ) if Apple fixes it in 6 months ? Will Apple give then some money back to the buyers ?

I hope you read some people here explained the lifespan of the phone is about 1 year and for those who buy it now 10 months. Even if the issue fixed in 3 months from now it will be 6 months ( if so long , may be less ) before the new model comes in. But people paid the FULL price for current model.
Not even mentioning ATT subscription and contracts still binding poor buyers.

Where did you get this information that 4.1 is released next month ? Who told you that ? Also who told you that any fix is coming in 4.1 ? May be it is coming in 4.3 ? Then what ?

Everything would not be not so bad and would be understandable if Apple would ( I know it is too much to ask ) issue the statement and let all new buyers know that it is an issue with PS and everybody should think before they buy ihone4.
Instead we have one statment saying " Some users experience this and we try to address in the next release ".

And from what we hear , Apple offers to replace one bad unit for another bad unit or wait for indefinite amount of time for something nobody knows what and when.

But Meanwhile Apple continue to sell MILLIONS defective phones.

Not only selling but setting a trap when after 30 days it is next to impossible to get the money back. But even within 30 days if you have a small scratch it is also impossible . So not only a fraud in selling the defective device it is by now clear it is done knowingly by Apple but also knowingly setting up all the traps ( including but not limited to encouraging vague statements about software fix ). Everything is done by Apple to sell the defective device and minimize the people wiliness and possibility returning it for the full refund.

But absolutely nothing is relevant for you. You have ( another miracle ) special phone with absolutely no PS issue and for you Apple is the best company and everybody should be patient does not matter what.

But with all the respect since it is your position you simply having some interest having such unfair approach.

@bmwhd1 and @Goflying

What does any of this?... (background audio API's) ...have to do with anything?

also this... (confused on the state of the headset/speaker phone/ear speaker bit)

I have downgraded to my 3Gs and I am running the most current version of iOS4 on it and I have NONE of the PS issues that I had on my iPhone 4.

Please explain why this would cause problems on the iPhone 4 but not on the 3Gs

Goflying wrote:

bmwhd1 wrote:

mbassoc2003 wrote:
Using your phone on your shoulder is a big no no. You can't do it with the iP4 because of the type and position of the sensor. The 'software fix' in the current build of the beta isn't a fix. It's a patch that tells the iOS to wait slightly longer before it switches the screen back on. It's yet to be seen how effective it's gonna be, but don't hold your breath. I think they are going to do their best within the realms of trying to compensate with the software, but I think people who have the issue regularly are likely to be disappointed, as all they can do with software is delay the incremental periods between the screen flashing on and off during calls. The issue remains a user interface issue, and whilst it's not the user's fault, just as the 'death grip' issue is not the user's fault, it is still the user that creates the fault. That is something no software is going to be able to eliminate.

I respectfully disagree. I believe the issue to be that the OS is getting confused on the state of the headset/speaker phone/ear speaker bit. Some combination of actions is flipping this bit at the wrong time and the OS is re-enabling the screen thinking you've inserted the wired headset, gone to speaker phone, or BT (as it's supposed to). I'm betting it has to do with the new background audio APIs which are proving difficult for app developers to master.

In working with Apple on this, I'm focusing my testing on phones set up as new with no 3rd party apps and comparing to one with bunches of apps (especially bg audio enabled) to see if there's any kind of pattern to latch on to.

^This by far makes the most sense of any hypothesis posted in relation to this issue. If so, it could be significantly more time consuming to find the root cause of the problem and fix it.

I am new to my iPhone 4 (4.0.1) and about 50% of my calls drop leaving my phone in some random screen. I never had this issue with my 3G.

I learn (e.g., via this thread) that this is a PS (proximity sensor) issue. I can only wonder how people argue at all how this can be anything but a design fault or a series fault.

It is trivial to diagnose the problem. I can go to voicemail and play back a message. While it plays, the PS is active. So, I can move my hand near the phone and watch the screen go on and off. The toggle distance (for my phone) is about 1 inch and the latency for reaction is about 1/2s (not very exact figures, more meant to illustrate what I am going to say). Exact figures are easily measurable though and iPhone4 and iPhone3GS could be measured exactly with regard to this property (toggle distance and toggle latency).

The 4 figures (4/3GS distance/latency) and their comparison should exhibit the nature of the problem with the iPhone 4 and end the discussion forever so we can simply wait for a fix.

Are the zillions of Apple bloggers empty brains? Why is nobody doing this and actually creating content and facts? I am Falk Lumo and my blog does it for Pentax dSLRs. It is possible ...


What makes my phone drop calls is this: I phone and hold the phone a couple of mm away from my ear. Because I don't like to press anything against my ear. Obviously, when I argue on the phone, the phone moves by more than an inch and this activates the screen. Only to hit the ear some fractions of a second later which then presses, e.g., the FaceTime button. Probably one method to drop the call. The PS is not sensitive and/or fast enough.


The fix (if latency cannot be improved to a few milliseconds by a firmware fix) is still easy enough:

When the PS first disables the screen (or at the beginning of a call) start a special mode. Whenever the proximity sensor then enables the screen, start a timer (about 1.5s). While the timer plays, don't display the naked phone call control screen, but display the naked phone call control screen with a protective slider labelled "slide to control your phone call". After the timer fires off (i.e. after 1.5s) the protective slider disappears so that normal operation isn't complicated too much.

Do I really have to work for Apple before obvious solutions like the above are shipped to millions of customers who all are dropping calls? Some of the calls I dropped have been decisive business calls ...


Apple, act immediately or destroy your reputation.

I should clarify that the 1.5s delay is meant to represent a limit of typical elapses of time where a user moves forth-and-back his phone-holding hand during a conversation.

As we're not all equal, this parameter should be adjustable in system settings, possibly up to the infinite value which would eliminate the problem once and forever.