iPhone 5 sound quality

[Saraugie]

OOO! I get to make a "I hate" list! That's fun!

WOW.......I'm easy I don't like equipment that does not sound good and some crappy MP3 files.

 

[Harry Manback]

I'm easy too - I do like the sound from my Iphone 5c.  Couldn't be happier with it.  I've had iphone 3, 4s and 5c.  All sound great to me.  I just wish that the eq (don't use it much) was built  into the music app and not in a separate location under the settings.

 

[cjl]

He measured:
-0.01dB @ 20Hz and -0.08dB @ 20kHz into a 200 Ohm load, is that not as near as damn it ruler flat?
4.5 Ohm output impedance.
Less than 0.01% THD 20 Hz - 15 kHz.
106.8 dB playing zeros A weighted output noise
116 dB idle A weighted output noise

Go look at his results.

What do your measurements show?

4.5 ohms is higher than I would normally like to see, and would be potentially audible with some headphones (though those specs would be audibly transparent with the majority of headphones out there). For example, look at the Westone 2 IEMs. They have an impedance of around 20 ohms below 800Hz, but >50 ohms above 1kHz (and >100 ohms from 6-9 kHz). This means that if they are played from a source with 4.5 ohms of output impedance, low frequencies (<800Hz) will be attenuated by around 2dB, frequencies above 1kHz will only be attenuated by 0.75dB, and frequencies between 6-9kHz will have less than half a dB of attenuation. This effectively acts as a 1-1.75dB boost to the upper treble, which is potentially audible.
 
Alternatively, another IEM that might have an audible impact from this is the Shure SE535. It has a nominal impedance across most frequencies of about 25 ohms, which would result in an overall attenuation of about 1.5dB across most of the frequency range. However, the SE535s impedance drops down to around 10 ohms at 6kHz, which would cause a 3.23dB drop at this frequency. In addition, the impedance goes up to around 50 ohms at around 2kHz, which (as I mentioned above) results in an attenuation of 0.75dB or so. In other words, when used on a source impedance of 4.5 ohms, the SE535 will have frequencies in the 6kHz range attenuated by around 2.5dB relative to frequencies around 2kHz, which should be a noticeable attenuation in the upper treble.
 
Admittedly, these are fairly subtle changes, but they aren't inaudible - they should definitely be distinguishable in a proper double blind test. Interestingly, if I'm remembering from the last couple pages correctly, the person who was claiming a large change from the 4 to the 5 was using JH13 Pros, which could definitely have a very frequency-dependent impedance curve as well as a low minimum impedance (since that tends to be characteristic of BA IEMs, especially multi-driver BA IEMs). If that is the case (I can't find an impedance curve for the JH13s in a bit of quick googling), and the output impedance of the iPhone 4 and the 5 are significantly different, that could indeed cause an audible difference through those IEMs (even though most headphones would not exhibit any audible difference at all, from the specs I've seen).

 

[Tympan]

 
4.5 ohms is higher than I would normally like to see, and would be potentially audible with some headphones (though those specs would be audibly transparent with the majority of headphones out there). For example, look at the Westone 2 IEMs. They have an impedance of around 20 ohms below 800Hz, but >50 ohms above 1kHz (and >100 ohms from 6-9 kHz). This means that if they are played from a source with 4.5 ohms of output impedance, low frequencies (<800Hz) will be attenuated by around 2dB, frequencies above 1kHz will only be attenuated by 0.75dB, and frequencies between 6-9kHz will have less than half a dB of attenuation. This effectively acts as a 1-1.75dB boost to the upper treble, which is potentially audible.
 
Alternatively, another IEM that might have an audible impact from this is the Shure SE535. It has a nominal impedance across most frequencies of about 25 ohms, which would result in an overall attenuation of about 1.5dB across most of the frequency range. However, the SE535s impedance drops down to around 10 ohms at 6kHz, which would cause a 3.23dB drop at this frequency. In addition, the impedance goes up to around 50 ohms at around 2kHz, which (as I mentioned above) results in an attenuation of 0.75dB or so. In other words, when used on a source impedance of 4.5 ohms, the SE535 will have frequencies in the 6kHz range attenuated by around 2.5dB relative to frequencies around 2kHz, which should be a noticeable attenuation in the upper treble.
 
Admittedly, these are fairly subtle changes, but they aren't inaudible - they should definitely be distinguishable in a proper double blind test. Interestingly, if I'm remembering from the last couple pages correctly, the person who was claiming a large change from the 4 to the 5 was using JH13 Pros, which could definitely have a very frequency-dependent impedance curve as well as a low minimum impedance (since that tends to be characteristic of BA IEMs, especially multi-driver BA IEMs). If that is the case (I can't find an impedance curve for the JH13s in a bit of quick googling), and the output impedance of the iPhone 4 and the 5 are significantly different, that could indeed cause an audible difference through those IEMs (even though most headphones would not exhibit any audible difference at all, from the specs I've seen).

Excellent post, thank you

 

[Roly1650]

4.5 ohms is higher than I would normally like to see, and would be potentially audible with some headphones (though those specs would be audibly transparent with the majority of headphones out there). For example, look at the Westone 2 IEMs. They have an impedance of around 20 ohms below 800Hz, but >50 ohms above 1kHz (and >100 ohms from 6-9 kHz). This means that if they are played from a source with 4.5 ohms of output impedance, low frequencies (<800Hz) will be attenuated by around 2dB, frequencies above 1kHz will only be attenuated by 0.75dB, and frequencies between 6-9kHz will have less than half a dB of attenuation. This effectively acts as a 1-1.75dB boost to the upper treble, which is potentially audible.

Alternatively, another IEM that might have an audible impact from this is the Shure SE535. It has a nominal impedance across most frequencies of about 25 ohms, which would result in an overall attenuation of about 1.5dB across most of the frequency range. However, the SE535s impedance drops down to around 10 ohms at 6kHz, which would cause a 3.23dB drop at this frequency. In addition, the impedance goes up to around 50 ohms at around 2kHz, which (as I mentioned above) results in an attenuation of 0.75dB or so. In other words, when used on a source impedance of 4.5 ohms, the SE535 will have frequencies in the 6kHz range attenuated by around 2.5dB relative to frequencies around 2kHz, which should be a noticeable attenuation in the upper treble.

Admittedly, these are fairly subtle changes, but they aren't inaudible - they should definitely be distinguishable in a proper double blind test. Interestingly, if I'm remembering from the last couple pages correctly, the person who was claiming a large change from the 4 to the 5 was using JH13 Pros, which could definitely have a very frequency-dependent impedance curve as well as a low minimum impedance (since that tends to be characteristic of BA IEMs, especially multi-driver BA IEMs). If that is the case (I can't find an impedance curve for the JH13s in a bit of quick googling), and the output impedance of the iPhone 4 and the 5 are significantly different, that could indeed cause an audible difference through those IEMs (even though most headphones would not exhibit any audible difference at all, from the specs I've seen).

You make good points about performance with low impedance headphones, but isn't the most telling phrase "these specs would be audibly transparent with the majority of headphones out there"? I think this is the main point that myself and BIGSHOT having been trying to make, to counter some of the nonsensical statements about the supposed lack of idevice audio performance. One poster characterized the sound quality of the iPhone 5 headlamp as "pathetic", which is patently not true.
I agree these effects are likely to be subtle, while accepting that the output impedance could be lower, to cater for a minority of headphones.
My two cents.....

 

[bigshot]

And again, an impedance mismatch between the iPhone and a few models of IEMs says absolutely nothing about the sound quality of the iPhone itself. I can plug lousy earbuds into it and it won't sound good either. The quality of sound of a DAP is determined with line out, not with heaphones, because the headphones are always going to be the weak link.

 

[cjl]

You make good points about performance with low impedance headphones, but isn't the most telling phrase "these specs would be audibly transparent with the majority of headphones out there"? I think this is the main point that myself and BIGSHOT having been trying to make, to counter some of the nonsensical statements about the supposed lack of idevice audio performance. One poster characterized the sound quality of the iPhone 5 headlamp as "pathetic", which is patently not true.
I agree these effects are likely to be subtle, while accepting that the output impedance could be lower, to cater for a minority of headphones.
My two cents.....

Certainly, and I don't mean to imply that there are glaring audible problems with the output from an iPhone or anything like that. All I wanted to do with that post was highlight the fact that based on the specs mentioned for the iPhone 5, it is not quite audibly transparent with all headphones (and the headphones it could cause slight problems with are sensitive, portable IEMs, which are reasonable headphones to be using with a portable device like an iPhone). With the vast majority of headphones that the iPhone 5 is capable of driving (including many sensitive IEMs, as long as they don't have a strongly frequency dependent impedance and don't rely on electrical damping of the driver), there will be no audible problems with that output: for all intents and purposes with most headphones, the iPhone 5 is a perfect source. That having been said, I don't want to gloss over a problem (however minor) a device has that causes it to deviate on occasion from perfection, especially when other similarly-priced devices are available with 1 ohm or less of output impedance.

 

[cjl]

And again, an impedance mismatch between the iPhone and a few models of IEMs says absolutely nothing about the sound quality of the iPhone itself. I can plug lousy earbuds into it and it won't sound good either. The quality of sound of a DAP is determined with line out, not with heaphones, because the headphones are always going to be the weak link.

I disagree - the output of the iPhone is clearly meant to work with headphones, not just as a line out, and the fact that it causes audible boosts or attenuations at certain frequencies with some headphones that are not terribly esoteric or unusual designs is a problem. I agree with you that the headphones are always going to be the weak link in the system, but the player/amplifier should be designed such that the signal it provides to the headphones is identical (within the limits of audibility) to the signal recorded in the music. With a significantly nonzero output impedance, this is not the case - if you were to measure the electrical signal at the output of the iPhone with the headphones mentioned above connected, you would find that the frequency-dependent attenuation would be present in the signal the phone was providing to the headphones, and not an artifact of the headphones themselves.
 
Now, if the vast majority of headphones on the market did not have any frequency dependent impedance, and this was only the case with a few uncommon, esoteric audiophile models, I could understand the designers not caring about catering to such a small crowd. Similarly, if it were designed as a preamp, and its output would always be fed into a high-impedance input, it wouldn't matter at all. However, a relatively low, frequency-dependent impedance is common with many other headphones and most speakers as well, and as a result, an amplifier or output designed to drive headphones or speakers should take this into consideration and have an impedance an order of magnitude or more below the minimum impedance of any headphones or speakers it will ever need to drive. This also helps with electrical damping of the drivers, although that is not really a problem with IEMs to my knowledge (it can be a problem with some full sized headphones, but 4 ohms should be sufficient to provide reasonable electrical damping for the models I am familiar with, and it would only become a problem in this regard with an output impedance well over 10 ohms).
 
Besides, if you only ever measure how a device performs into a high-impedance load, you'll get distortion, noise, and frequency response measurements that look dramatically better than you would ever see driving a realistic load, and as a result, those measurements would not tell you anything at all about how the device really performs when driving real speakers or headphones. A lot of mediocre designs look perfectly fine when their output is hooked up to a several thousand ohm load, only to fall flat on their face when asked to source more current to drive a low impedance (Note: I'm not saying this is the case with the iPhone - by all the measurements I've seen, it does just fine driving a low-impedance load, aside from the potential slight frequency attenuation or boost based on the headphone's impedance curve).

 

[cjl]

Oh, and for what it's worth, I think you would have a very difficult time distinguishing a 1-2dB frequency-dependent boost or attenuation without a near-instant level matched switch, even in the treble where your ears are very sensitive. You could probably distinguish it in an ABX with a good fast switching setup, but with any significant time for your audio memory to degrade, I think it would be exceedingly difficult to tell any difference at all, even in the worst cases I could find which I highlighted above. I don't mean to imply that there's a major problem with the iPhone 5 as a source, nor do I mean to imply that it would sound dramatically different than any other reasonably competently designed audio source. However, as an engineer, I do like to nitpick...

 

[Roly1650]

Consider the nit truly picked

 

[cjl]

Consider the nit truly picked

Then my job here is done.
 

 

[Tympan]

  Then my job here is done.
 

Thanks for your contribution, I'm hearing what you're saying too. I'm done with this thread as well 

 

[bigshot]

  I disagree - the output of the iPhone is clearly meant to work with headphones, not just as a line out

 
Usually, you measure each component separately. Otherwise, there would be a different spec for every player with every headphone model out there. Anyone shopping for headphones should be aware of their impedance peccadilloes and should take that into consideration when deciding what to buy. But that is a matter of function, not performance. I wouldn't buy a CD player and expect it to play SACDs, and I wouldn't buy a DAP and expect it to work equally well with every headphone in the world. If it works with most of them that gives me plenty of options. (My iPhone 4 works perfect without amping with my Oppo PM-1s.)

 

[bigshot]

  Oh, and for what it's worth, I think you would have a very difficult time distinguishing a 1-2dB frequency-dependent boost or attenuation without a near-instant level matched switch

 
In music, I doubt I'd hear it at all. That small of a deviation doesn't matter. I listen with human ears.