As an owner of an iPhone 4S that just got an offer to swap it for a Galaxy S3 i took a little while to convert the accurate measurements taken by the helpful and skilled guys of Goldenears.net
 
This is just a thread meant for reference aimed to guys that take into account objective data.
 
The purpose is not to start the typical flame war of deaf objectivists vs subjectivists with gifted ears that know better.
 
Here's the performance driving a 33 Ohm resistive load (Note: the lower the better)
 
 
                            Noise           THD        IMD+ Noise     Crosstalk    Output Impedance
 
iPhone 4S             - 95.5          -89.89          -85.19             -75.9               1.97
Galaxy S3             - 96.0          -77.72          -73.97             -53.3               3.45
 
What science says is (basically) what my ears heard when tried both side by side without knowing these figures.
 
iPhone sounds more open and uncongested. Plus resolution is step above (Subjectively the decay of percussive instruments in the iPhone is pure joy)
 
And what these measurements don't tell is that out of the box output power of the iPhone is far higher (Dunno if Supercurio could uncap the S3 at some point)
 
Thus now i'm just weighing in if could live with the dip in sound quality in exchange for 4.8 inches of screen state. It's hard letting go of aural bliss you know !

wow thats actually considerably better across the board. I'm a hybrid sub-objectivist or 'realist' the iphones and ipads of late actually do have pretty excellent objective and subjective performance, despite how fashionable it is to bash their sound quality.
 
thats pretty crappy crosstalk on the S3 … either they are using a really bad mini jack socket, or their power supply and grounding is pretty sub-par. the S3 wont be so excellent with multidriver IEMs with that output impedance either.
 
got any numbers with reactive loads?
 
the only reason I would and have considered getting an android of some sort, would be the USB digital output; which appears now possible with jailbreak anyway.

I'm kind of a mixed aficionado as well. I consider numbers and perception. But usually what measures good sounds good and what sounds good measures good. No contradictions at all.
 
Samsung devices crosstalk is always just average (Galaxy S, Galaxy S2, Galaxy S3 or Galaxy Note).
 
From all the objective data i've come across all i can conclude is that in Apple smartphones' sound quality ranks higher than it does in the competition. May be they don't forget that music brought them here.
 
And talking about wolfson i think they are kind of stagnated (at least in the handheld realm). From the iPod 5.5 to present day they just have got rid of the bass roll off but nothing else really (Some manufacturers even have not been able to get rid of the bass roll off as of 2012...cough Cowon
 
Unfortunately i don't have data for a reactive load.

indeed, I strive for good numbers, but also dont forget that when it comes down to it, all we have as human beings is our subjective response to the stimuli. Whether this response is reflected in the objective performance, or simply a trick of the mind is all there is and either is just as valid and real to the 'observer' as the other.
 
the bass rolloff has nothing to do with the dac, more the damping factor of the output stage. clearly cowon still plays it 'safe' by including some sort of resonance damping output resistor to protect the amplifier from cable and/or transducer capacitance, while apple implementations either include this resistance in the feedback loop, or simply provide a more robust output stage
 
I wish the low power AKM codecs (like the AK4480 as used in the fostex HP-P1) were more popular, they are producing some excellent reasonably priced and easy to implement power efficient dacs.

Guilty. The bass roll off is related to the amp instead of the DAC. Thanks for the correction.
 

yup, the more recent idevices are DC coupled ie. no output caps; so this helps keep bass response faithful as well.

Is that the International S3 (I9300) or one of the US Variants(I535 for Verizon, I747 for AT&T or T999 for T-Mobile?) 

You know that just the international version has the Wolfson dac right? GSM Arena reports Noise -90.3dB / Crosstalk -92.6dB for the International and Noise -82.3dB / Crosstalk -78.4dB for the AT&T one, so even if their measurements are wrong, seems that they sure are very different SQ-wise. 

crosstalk, as I said, (past a certain level) has nothing to do with the dac and wolfson make crap dacs just like everyone else, thats even if the dac chip itself by itself meant anything
 
the GSM arena page mentions nothing of 'test' method or loading and simply looks like its pulled straight from marketing material, thus meaningless. It may in turn be just pulled from the datasheet for the dac used, not a real world measurement in a real design with a real load

well i guess i missed at making my question understandable. 

i don't have a S3, but i know that the int. ver. has a different dac from the us version, the site (gsm arena) even gives different noise ratios for each version so i imagine that there is some significant difference between each version, and i was wondering which one was being discussed and tested on the topic. 

i do believe you Qusp, not trying at all to say that you are wrong at all, it's just that i'm thinking of getting an int S3 but it's quite expensive and i'm taking my time to read more about it and i'm a bit curious about it. if even the INT ver. is that bad, i will just stop looking this cellphone and maybe just wait for the iphone 5 to drop (so i could by a cheap iphone 4 )

oh yeah I get that OK, no worries. sure they'll be different, but it shouldnt have a significant effect on the noise or crosstalk, as both of those factors are more down to the amp section. I just dont think numbers like those offered at gsm arena are very useful, its very rare for manufacturers to give real measured and loaded test results and even less common to find it on sites like that, where audio performance is just one of the factors, not even a major one.

Quote:

elfary said:

As an owner of an iPhone 4S that just got an offer to swap it for a Galaxy S3 i took a little while to convert the accurate measurements taken by the helpful and skilled guys of Goldenears.net
 
This is just a thread meant for reference aimed to guys that take into account objective data.
 
The purpose is not to start the typical flame war of deaf objectivists vs subjectivists with gifted ears that know better.
 
Here's the performance driving a 33 Ohm resistive load (Note: the lower the better)
 
 
                            Noise           THD        IMD+ Noise     Crosstalk    Output Impedance
 
iPhone 4S             - 95.5          -89.89          -85.19             -75.9               1.97
Galaxy S3             - 96.0          -77.72          -73.97             -53.3               3.45
 
What science says is (basically) what my ears heard when tried both side by side without knowing these figures.
 
iPhone sounds more open and uncongested. Plus resolution is step above (Subjectively the decay of percussive instruments in the iPhone is pure joy)
 
And what these measurements don't tell is that out of the box output power of the iPhone is far higher (Dunno if Supercurio could uncap the S3 at some point)
 
Thus now i'm just weighing in if could live with the dip in sound quality in exchange for 4.8 inches of screen state. It's hard letting go of aural bliss you know !

Click to expand...

Ive seen the iphone 4 measure below 1 ohm for output impedance, That said I agree with these measurements. The SGS3 cant touch the iphone4 or 4s.

I love my Apple products for what they do. And I, for one, believe the numbers in the OP. iPhone has a better sound “resolution” (for lack of a simple term). Yet, I still prefer various models of Samsung for music playback. I heard an S3 today and I like it. Perhaps it simply has a better “color” to my ears.
 
When I got my first iPhone (the first gen in 2007) I compared the video playback quality to my Dell Axim X51V. By the numbers, the X51V was superior with a picture resolution of 640X480 to the iPhone's 480X320. However, the iPhone looked better, in fact, a lot better. The iPhone had better color depth and a darker black level. The Axim's higher resolution was rendered meaningless.
 
BTW, I can't get the bass to show up on any Apple DAP I've tried. Am I doing something wrong? They all have big midrange, so-so highs, and weak bass. It's disproportionate no matter the headphone used. The tone is off a bit as well.
 
My sister has the 4S. My sister-in-law has the 4,  and I still have the original iPhone plus an iPod Touch 3G. In a blind sound test, I couldn't tell which was which (of the 4 units) using Apple's standard music player. And yes, my hearing is great. I just took a battery of tests in June.
 
Most of the aftermarket players (that I've tried) sound better than Apple's own music player. Apple's player gets (by far) the best battery life though. Would that be due to better software encoding or weaker music processing? I don't know where the Apple music player is represented in the numbers. But, if there was a Poweramp for iPhone, I would be all over it.
 

I'm a very satisfied S3 owner, that said, these measurements only validate what my ears heard when A/B'ing the 4S Vs the S3: iPhone is noticeably better.
 
Anyways, I'm bypassing the internal DAC with my S3 by taking advantage of it's support for USB audio. Sound is being handled by the HeadAmp Pico USB DAC/amp I got this week, and the 4S sound quality pales in comparison to what I'm getting now. I would expect Samsung to bring USB audio support to all their high-end phones from now on, and it's coming more widely to modded Android devices on Jelly Bean now too, so I'll never again be reliant on the manufacturers DAC choice and implementation for my sound quality. Bliss.

elfary said:

As an owner of an iPhone 4S that just got an offer to swap it for a Galaxy S3 i took a little while to convert the accurate measurements taken by the helpful and skilled guys of Goldenears.net
 
This is just a thread meant for reference aimed to guys that take into account objective data.
 
The purpose is not to start the typical flame war of deaf objectivists vs subjectivists with gifted ears that know better.
 
Here's the performance driving a 33 Ohm resistive load (Note: the lower the better)
 
 
                            Noise           THD        IMD+ Noise     Crosstalk    Output Impedance
 
iPhone 4S             - 95.5          -89.89          -85.19             -75.9               1.97
Galaxy S3             - 96.0          -77.72          -73.97             -53.3               3.45
 
What science says is (basically) what my ears heard when tried both side by side without knowing these figures.
 
iPhone sounds more open and uncongested. Plus resolution is step above (Subjectively the decay of percussive instruments in the iPhone is pure joy)
 
And what these measurements don't tell is that out of the box output power of the iPhone is far higher (Dunno if Supercurio could uncap the S3 at some point)
 
Thus now i'm just weighing in if could live with the dip in sound quality in exchange for 4.8 inches of screen state. It's hard letting go of aural bliss you know !

Click to expand...

Are you referring to the International or the American version of the galaxy s 3?

Sent from my ADR6425LVW using Tapatalk 2

It's the performance of the 3G international version. Not the US variant.