lhf wrote:
That doesn't seem to indicate it is necessarily a software issue. I'm going to wait until a fix is confirmed before I go back to iPhone 4.
Of course it's software. Many people—myself included—have come to that conclusion in recent days. And anyone who has the intellectual curiosity to conduct a simple experiment could also prove it to themselves that it's just buggy software. That's what I did when I made this video (
http://www.youtube.com/watch?v=dX7EPVDVEEg ), and you can repeat it for yourself.
Here's all you need to do:
1. Get the phone on an active call, or listening to a long voice mail, so that the proximity sensor is active.
2. Lay the phone flat on a table facing up.
3. Pick a point 0.75 to 1.0 inches above the phone's earpiece, and call that "Point A"
4. Place you hand directly over the phone at Point A, without touching the phone first. The screen will go blank (as it's supposed to).
5. Remove your hand, and repeat step 4 as many times as you want to prove to yourself that the phone consistently reacts to your hand's presence.
6. Now, cover the proximity sensor area by touching the glass directly with your hand, and watch the screen go blank, as expected.
7. Move your hand directly to Point A from there. You should go straight up from touching the phone to your previous "hovering" posiiton. The screen will come on, even though your hand is in the exact same place (Point A) that turned the display OFF before.
8. Lower your hand to touch the sensor again, and watch the screen go off. Raise it to Point A again, and watch the screen become active again. Repeat as necessary to see how repeatable it is.
9. Continue to go back and forth between the two modes described above until you're satisfied that it's predictable behavior.
The very fact that the phone responds differently to the exact same hand (or head) position depending on where you were previously is a clear sign that it's a software bug. Hardware doesn't care about what
was, only what
is. The software bug is the only thing that could make the activation distance "stick" at zero whenever there is direct contact, and then make the phone temporarily fail to recognize a small gap. After that, moving entirely away from the phone "resets" the sensor, and it is then ready to respond to a small gap.
This behavior agrees with the experiences of myself and many others who have scrutinized the proximity sensor problem. I think what happens is that at some point during a call, the phone is placed so that the proximity sensor is directly against the skin. Then, after that, if it's moved back to a point where it's looking into the ear again (maybe a half-inch gap), it won't shut the screen off (like in step 8 above). I've found that if I consciously put the phone in a position where it's looking into my ear a bit, and I'm careful not to move it higher so that my outer ear can't press against the glass in front of the sensor, the phone is happy to keep the screen off indefinitely.
Since the "normal" operation of the phone has the sensor looking into the ear a little bit (that small gap), some people never have it happen to them. That's because they only move the phone in a range that keeps a little space next to the sensor. They happen to use their phone in such a way that they never get skin-on-sensor contact, so they never get their sensor "stuck" on zero. And maybe some people always keep their prox sensor against some skin, and the fact that it's "stuck" at zero isn't a problem. But many people end up going back and forth a little...they happen to naturally hold their phone in a way where it's bound to see a small gap one moment, and direct contact the next. As demonstrated above, the software that controls the sensor HATES that.
It can definitely be fixed with software, and Apple will do that as soon as they have the time to properly test the fix. I'm usually a pessimist, but I'm confident in that right now.
