How did they "**** up big time"? People with issues are in the minority by far, I am perfectly happy with mine. OP should return his phone if he isn't happy and move on. He of course could of added his blog to one of the other blog posts out there saying the same exact thing.
Got a tip for us?
Let us know
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
My (brief) experience with the iPhone 4 [Updated x2]
- Thread starter Rend It
- Start date
- Sort by reaction score
How did they "**** up big time"? People with issues are in the minority by far, I am perfectly happy with mine. OP should return his phone if he isn't happy and move on. He of course could of added his blog to one of the other blog posts out there saying the same exact thing.
Can I borrow your time machine sometime? There's a few investments I'd like to make.
update
I'm still waiting for Apple to send me an RMA.
In the meantime, I did some more testing of the iP4's behavior at higher signal strengths. I tried the death grip on the phone when I had signal levels (without case and without holding phone) higher than about -91 dBm. I could not cause a loss of signal strength with my hand for any of the following signal levels (in dBm):
-85, -83, -81, -77, -61
For starting values below about -91 dBm, the signal level always plummeted to -113 to -109 dBm. The threshold between the two situations (susceptible to hand and not) seems to vary. Sometimes -87 dBm was solid, other times it was lower or higher.
So, this would seem to indicate one or more effects happening:
1) There is a saturation effect, wherein whatever is causing the detuning in your hand (water, proteins) can be saturated for sufficient power levels. In other words, when the microwave power is high enough, your hand becomes "transparent" to the phone. Here's an old example of this effect with ammonia (accessible through most universities):
http://prola.aps.org/abstract/PR/v73/i9/p1053_1
The crux of what I'm referring to (the x,y scales will of course be different for different molecules and frequencies) can be seen here:
2) Another possibility is that the iP4 has some software algorithm that can adapt the matching network in the GSM/UMTS module when interference is present. However, perhaps it only works well at sufficient signal strengths. If this is true, then the problem may be ameliorated by modifying said algorithm.
The point to take home is that in addition to the weird signal-to-bar mapping, the reason for such wildly varying experiences is that above a certain power level there is no signal loss with the death grip. Hmm.
I'm still waiting for Apple to send me an RMA.
In the meantime, I did some more testing of the iP4's behavior at higher signal strengths. I tried the death grip on the phone when I had signal levels (without case and without holding phone) higher than about -91 dBm. I could not cause a loss of signal strength with my hand for any of the following signal levels (in dBm):
-85, -83, -81, -77, -61
For starting values below about -91 dBm, the signal level always plummeted to -113 to -109 dBm. The threshold between the two situations (susceptible to hand and not) seems to vary. Sometimes -87 dBm was solid, other times it was lower or higher.
So, this would seem to indicate one or more effects happening:
1) There is a saturation effect, wherein whatever is causing the detuning in your hand (water, proteins) can be saturated for sufficient power levels. In other words, when the microwave power is high enough, your hand becomes "transparent" to the phone. Here's an old example of this effect with ammonia (accessible through most universities):
http://prola.aps.org/abstract/PR/v73/i9/p1053_1
The crux of what I'm referring to (the x,y scales will of course be different for different molecules and frequencies) can be seen here:
2) Another possibility is that the iP4 has some software algorithm that can adapt the matching network in the GSM/UMTS module when interference is present. However, perhaps it only works well at sufficient signal strengths. If this is true, then the problem may be ameliorated by modifying said algorithm.
The point to take home is that in addition to the weird signal-to-bar mapping, the reason for such wildly varying experiences is that above a certain power level there is no signal loss with the death grip. Hmm.
You call that a wall of text? What is in the OP?
You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!You've no idea of what you are talking about!
very good read. can you please elaborate a little on number 2 with the software algorithm thing? what you are experiencing is very interesting. anandtech's study showed that the signal ALWAYS attenuated by upto 24dB, no matter what the starting signal strength was. so even if the phone had a max strength of -51dB (5 bars), death gripping it would cause it to fall to at most -75dB (still 5 bars). this was their explanation of why many people did not notice their signal dropping, because even after the upto 24dB drop, it was still showing 5 bars. It is strange that at -85, -83, -81, -77, -61, the signal would not drop for you. but yea, i'd like to know a little more about the number 2 effect you described below.
I did some additional testing today with a video camera and I will be posting them to YouTube soon. (I'll post here with the links).
To address your question appleg33k, those two possibilities are by no means exhaustive, just the first two that came to mind. #2 is perhaps some wishful thinking on my part, but if you look at the block diagrams for similar chips, there is a matching network (essentially an arrangement of capacitors and inductors), and the chips are controlled (usually) through a serial bus. The iP4's chip is a custom version from TriQuint (I think), so we may never know what's inside it. But, it's certainly reasonable to imagine that the matching networks can be adjusted via the serial bus.
Now, it may be that the matching network is simply fixed and that's it. No software update will help. However, if it is SW adjustable, then some massaging of the algorithm could help things - a little. We're talking about femtowatts of signal power here, which is extremely close to the noise floor in wireless communications channels.
A minor hardware fix could involve getting chips with two networks: the normal one and one that is better suited for when your hand makes contact. Inside, a switch controls which network is used.
Finally, as regards AnandTech's analysis, it's not explicitly clear that he actually observed a 24 dB drop for strong signals. His mention of it was more to explain why the bars behave as they do. I hope my videos will add some support to my earlier comments. It would be nice if more people could try the bars to dBm conversion and report back. It's not that difficult, so I'm a little surprised we aren't seeing more about it.
Stay tuned.
Sorry I couldn't resist.Videos of iP4 behavior near strong signal
Here are the links as promised:
Strong signal, no loss with hand:
http://www.youtube.com/watch?v=xjGDvjGMWDY
Weak signal, degradation with hand (for comparison):
http://www.youtube.com/watch?v=SejWN7I2u2w
Sorry for the shakes. As you can see, for strong initial signals the power remains constant in the presence of my hand. For weak signals below about -85 dBm, I observe the usual behavior seen everywhere else.
Enjoy.
Here are the links as promised:
Strong signal, no loss with hand:
http://www.youtube.com/watch?v=xjGDvjGMWDY
Weak signal, degradation with hand (for comparison):
http://www.youtube.com/watch?v=SejWN7I2u2w
Sorry for the shakes. As you can see, for strong initial signals the power remains constant in the presence of my hand. For weak signals below about -85 dBm, I observe the usual behavior seen everywhere else.
Enjoy.
Frankly, it seems more likely to me that the power measurement supplied by the chip is in error. It's not like they are using a bolometer or some other highly accurate measurement system. The measurement range of received signal strength in a chip-based receiver is usually compressed at the ends of the range. Thus at high power levels, a significant change in signal level may make only a slight change in RSSI. How well the software maps that value to absolute power at the receiver input is suspect. In short, don't treat the signal-strength indications of the phone like you would readings from your lab instruments, especially at very high or very low values.
True, but the signal range that I observed this effect over isn't confined to the upper and lower limits. And yes, the number indicated on the phone isn't necessarily an accurate estimate of the true power.
I can drive up to a tower and it reads around -51 dBm. Drive a few blocks away on the same street, it drops to about -60. Another few blocks and it's -75. At each of these positions, I can place my hand on the bare metal and the reading is unchanged. However, in an area with only -93 dBm placing my hand on the metal band causes the signal to drop 20 dB AND I lose my data connection. I'm not saying my hand has zero effect at the stronger signal powers, just that the effect is likely negligible for purposes of data transmission and call quality.
That makes perfect sense. Once some level of signal is present the bit-error rate is so low that more signal makes little or no difference. If you get close enough to the tower, you could probably wrap the phone in tinfoil and it would still work.
That makes perfect sense. Once some level of signal is present the bit-error rate is so low that more signal makes little or no difference. If you get close enough to the tower, you could probably wrap the phone in tinfoil and it would still work.
Yes, that would be another interesting test. I'd have to cut a little hole to be able to run the FCC app, though.
The point I'm trying to get across is that the iPhone 4's antenna isn't necessarily a terrible design. The AnandTech piece made it seem like you get 20-24 dB worth of loss when you death grip the phone no matter what the signal power is. But, what I've seen is that the phone doesn't care what your hand is doing as long as the signal is above a certain threshold. It's not the case that you simply stay within the fifth bar. Below that threshold, well it's a different story. Maybe it's not attenuation or detuning; perhaps your body is injecting noise into an already noisy situation. I guess only Apple's engineers may know.
On another note, after all this talk about sweat and galvanic conduction between the WiFi and cell antennas, I grabbed a multimeter and checked for continuity across the infamous gap (without holding the phone). As I suspected, they're already shorted together.