Wolfson DAC confirmed for Galaxy S3!

If I receive my SGS3 before supercurio gets the output impedance of SGS3, I'll measure it myself and let you guys know. By the way, I thought I could guess, by taking advantage of GSMArena data, that the output impedance of SGS3 would be around 3~5 ohms. Since they use all the same headphone (AKG K450 I guess) through devices, we could expect lower fluctuation means lower output impedance.
 
 
1. GT-i9100 (Seems like the international version has a big output impedance; according to supercurio, it's measured around 50 ohms. Korean version has it around 10 ohms according to GoldenEars)
 

 
2. GT-N7000 (around 3.44 ohms)
 

 
3. GT-i9300
 

 
Only time will tell how much the output impedance is

Quote:

studiominal said:

If I receive my SGS3 before supercurio gets the output impedance of SGS3, I'll measure it myself and let you guys know. By the way, I thought I could guess, by taking advantage of GSMArena data, that the output impedance of SGS3 would be around 3~5 ohms. Since they use all the same headphone (AKG K450 I guess) through devices, we could expect lower fluctuation means lower output impedance.
 
 
1. GT-i9100 (Seems like the international version has a big output impedance; according to supercurio, it's measured around 50 ohms. Korean version has it around 10 ohms according to GoldenEars)
 

 
2. GT-N7000 (around 3.44 ohms)
 

 
3. GT-i9300
 

 
Only time will tell how much the output impedance is

Click to expand...

Very interesting post.
 
Supercurio tweeted me last night the he won't be able to know the output impedance till next week at least.
 
I can vouch for the Note acceptable z. The S3...i'm confused now. I'll have to try it again. What it's almost certain now is that stereo crosstalk performance is just average by all accounts and measurements.

Quote:

nztechfreak said:

Can definitely play FLAC, not sure about AIFF. 
 
Plenty of third party players that have FLAC support even for phones without native FLAC support in any case, Neutron has both FLAC and AIFF. Very powerful player, even if it's UI looks like a dogs breakfast.

Click to expand...

Hey crew sorry to jump in just like this but just thought you'd like to know what I've just confirmed with the SGS3 and its play back of AIFF files and 24bit recordings. Just to give you some back ground in what I was trying to do, I own a Rane 68 DJ mixer which allows me to record a mix as 24bit AIFF file@ 48000khz(tracks I'm recording are only 16bit WAV @ 1411kps tho ), so I threw a mix down with the plan to get it to play back through the S3, mainly so I could listen to it in the car through the aux input. So first I first tried winamp & poweramp both came up with "this file can not be played by your phone" oslt. I then got the neutron player as it said it would play back 32bit files and sure enough it played back the 24bit AIFF but with a few problems. It plays back with a crackle and also reads as 32 bit when its not, so fail on that one. I then converted the 24bit file again in Ableton to 24bit WAV @ 48000khz so in effect only changing the type of file. This time play back was fine no crackle and sounded pretty good through my pair of HD-25's. Now the only thing that dodges me out about the whole thing is when I checked the settings in Neutron "Audio Hardware" it lists Bits as 32(output16) now this just makes me think either the phone can only output at 16bits or Neutron can paly up to 32bit file but also only output at 16bits.....
 
So yea hope someone can use this info in your quest to get the best sound out of the SGS3 via a DAC

Quote:

nztechfreak said:

The output from the E7 is dramatically better to my ear, night and day difference.
 
The internal Wolfson DAC is exciting, but unfortunately without Voodoo Sound it isn't up to that much, which is why USB audio is exciting right now (it's nice to know that regardless of whether Voodoo Sound is great or not I'll have good output from the S3 and be able to drive my less sensitive headphones from it).

Click to expand...

If your following Supercurio on Twitter you may see that he thinks as I did the E7 (and even E17) is worse than the GS3, Im going from memory though so he word will be better than mine
 
Supercurio  "‪#GalaxySIII‬ DAC both give lower distortions levels than FIIO E17 USB DAC. Measurements just confirmed perception. it means using FiiO E17 USB DAC instead of built-in @wolfsonmicro ‪#WM1811‬ degrades audio quality instead of improving it. so, clearly there's no need to search for a magic line-out path on ‪#WM8994‬ ‪#WM1811‬, optimized one is HPOUT!" 
 
So cancel your orders´for a E7 guys and sit patiently for Voodoo 

In my view the problem with usb dacs is that the digital signal gets reclocked again thru USB so IMD and jitter get worse. If you stick to the internal dac it gets the data off the bus in a pure I2S format.
 
Thus the digital signal that reaches the dac inside the E17 is degraded versus the pure (not reclocked signal) that gets into the internal dac.
 
That's why for portable devices i prefer just to amp a clean line out and stick to the internal dac (Unless it's pure crap).
 
Supercurio's view is different though. He thinks that E17 is a great amp but a middle of the road dac.
 
As for the 48/24 issue i'm afraid that Android downconverts everything to 44/16 thus it's not a good idea to throw at it hi res stuff since probably it truncates to 16 and a battery powered device resampling process is not something i'd trust.
 
On iOS you can play 48/16 but if you throw to an idevice a 48/24 file it will play it but truncating to 16 thus it's not a good idea to use 48/24 on the Apple realm neither even if here you get rid of the resampling.
 
Anyway i think that 24 bits are useful for mixing and mastering or at an studio. On a portable device it's more of placebo than any real benefit.
 

Quote:

On iOS you can play 48/16 but if you throw to an idevice a 48/24 file it will play it but truncating to 16 thus it's not a good idea to use 48/24 on the Apple realm neither even if here you get rid of the resampling

Click to expand...

 
I play 24/96 wav files using Ipod touch/line out to FIIO E11/Sennheiser HD25 SP and the difference between 16/44, 16/48 is marked with the 16 bit files sounding typically CD thin and the 24 bit files sounding full bodied, if there's truncation and re-sampling going on why still the sonic differences?
 
Quote:

Anyway i think that 24 bits are useful for mixing and mastering or at an studio. On a portable device it's more of placebo than any real benefit.

Click to expand...

 
I couldn't disagree more; 24/48-96 sounds vastly superior to 16/44,48, and as written above this difference is not just confined to a home rig.
 
Given that there are a number of Linux-based media players on the market that can all output 24/96 and some 24/192, why should an android based device be limited to 16/48?

24 bit is quite meaningless for just listening, any hearing difference is placebo. In a Studio it's valuable as it provides more headroom before clipping when recording and mixing ie the technician can initially be more lazy. The same results can be obtained with 16 bit in the studio too, but requires the technician to be more alert.
 
Difference between a great and a just good recording is in how it is recorded, not in how many bits where used.
 
That said, like all headphone amps the SIII drives different headphones with varying success. I found one headphone that is very happily driven by the SIII, it's the old low end Sony MDR-V300 which in fact sounds really really good driven by the SIII. I'm listening to the new Melody Gardot album right now and it sounds like very full grown HiFi, lots of body and and no harshness. Very powerful "JBL-like" punch to the sound, not the last in neutral sounding but still very good. Just like classic JBL speakers. Of course I use Volume+ to bypass the artificial volume maxlevel for the European SIII, Supercurios Voodoo Sound can only make it even better.

Quote:

24 bit is quite meaningless for just listening, any hearing difference is placebo. In a Studio it's valuable as it provides more headroom before clipping..

Click to expand...

No it doesn't 0 db is 0 db, please look up on google.  As for placebo, well in that case I may as well sell my HD650s and Earmax pro as it's just my imagination.
 
Quote:

Difference between a great and a just good recording is in how it is recorded, not in how many bits where used.

Click to expand...

Quite true.

Quote:

griff2 said:

No it doesn't 0 db is 0 db, please look up on google.  As for placebo, well in that case I may as well sell my HD650s and Earmax pro as it's just my imagination.

Click to expand...

 
That was a cheap one, that is not the way a recording engineer works. With 16-bits you have roughly 90dB S/N to play with, if you stray too far from 0dB there will be audible noise and sampling distorsion so you don't do that. With 24-bits you have roughly 120dB S/N to play with, which is MUCH more, which means you can easily put your working maxlevel 10-20dB below 0dB and still have lower noise and sampling distorsion than with 16-bits. If you do that, it means you have at the same time gained 10-20dB headroom at the top. True 0dB is always 0dB, but if you're clever you take advantage of the added headroom below to use it as a virtual headroom above 
 
That is why 24-bits is better than 16-bits, not because it sounds better when booth are properly used.

Quote:

With 24-bits you have roughly 120dB S/N to play with, which is MUCH more..

Click to expand...

 
[size=11.0pt]Absolutely[/size], you have extra noise floor, and more levels to play with; you don't have to rely on dithering.
 
I don't like dithered sound, just as I don't like an image which has been dithered down from 24 bits per pixel to 8 bits per pixel; I can clearly see the effect.
16/44 and later 16/48 was always a compromise based the recording capacity of the video recording equipment of the time (the early 1980s). 
Anyway, enough of the lecture here's an interesting article, albeit dumbed down, by Linn which I sort of agree with.

Quote:

griff2 said:

 
[size=11.0pt]Absolutely[/size], you have extra noise floor, and more levels to play with; you don't have to rely on dithering.
 
I don't like dithered sound, just as I don't like an image which has been dithered down from 24 bits per pixel to 8 bits per pixel; I can clearly see the effect.
16/44 and later 16/48 was always a compromise based the recording capacity of the video recording equipment of the time (the early 1980s). 
Anyway, enough of the lecture here's an interesting article, albeit dumbed down, by Linn which I sort of agree with.

Click to expand...

There may be scientific, electrical and acoustical proof that there actually is a benefit with 24-bit for the end user. However, most if not all blind listening tests prove otherwise. I guess a Linn rig wouldn't sound that good if you didn't even know it was a Linn, placebo ....
 
Early DAC:s sounded not so good mostly because of primitive filtering techniques at the time, those old 80:s recordings of which some are stellar while made on 16-bit gear sound way better today when the DAC:s have evolved a lot. But the number of bits are still the same.

Well let's don't turn this S3 thread into the never ending 24 vs 16 discussion
 
Awaiting for Supercurio sound analysis even if we won't get to know the output impedance for the moment.

Quote:

There may be scientific, electrical and acoustical proof that there actually is a benefit with 24-bit for the end user. 

Click to expand...

 
There is, you answered it yourself:
 
Quote:

With 24-bits you have roughly 120dB S/N to play with, which is MUCH more..

Click to expand...

 
..than 16 bits.  And good recording engineering, or remastering, will use ALL of that, not just a small amount of it, as you also alluded to when you said that recording engineers can be sloppy.
 
Quote:

However, most if not all blind listening tests prove otherwise..

Click to expand...

 
Well, as Wikipedia would put it, citation needed..  And obviously as you are a student of Nyquist, you can therefore easily defend why 44.1 kHz is good enough at a depth of 16 bits and no other frequency or bit depth is required.  Also you've alluded to the importance of filtering so please, in your own words, explain on.
 
Quote:

I guess a Linn rig wouldn't sound that good if you didn't even know it was a Linn, placebo ....

Click to expand...

 
That applies to all Hi-Fi..
 
Back on track, I really do hope that the FIIO E17 can communicate with the S3, for me especially, at elevated bit depths and sampling frequencies.

Quote:

danny93 said:

If your following Supercurio on Twitter you may see that he thinks as I did the E7 (and even E17) is worse than the GS3, Im going from memory though so he word will be better than mine
 
Supercurio  "‪#GalaxySIII‬ DAC both give lower distortions levels than FIIO E17 USB DAC. Measurements just confirmed perception. it means using FiiO E17 USB DAC instead of built-in @wolfsonmicro ‪#WM1811‬ degrades audio quality instead of improving it. so, clearly there's no need to search for a magic line-out path on ‪#WM8994‬ ‪#WM1811‬, optimized one is HPOUT!" 
 
So cancel your orders´for a E7 guys and sit patiently for Voodoo 

Click to expand...

 
Actually he made no comment about the E7, only the E17. He also says testing was without load.
 
For me the difference on listening stock S3 Vs S3 + E7 is significant - more resolved detail, better dynamics, tighter bass.
 
If Voodoo can bring out superior potential via just headphone out I'll be ecstatic, but for now I'll be listening through the E7. I think best advice at this time is to wait for Voodoo, trial that, and if still dissatisfied try USB DACs and see how they are for you.
 
I really need to gather a bunch of people and do a blind listening test, but might be hard somewhere like Auckland to get enough people...

Need some opinions guys. I have a SGS2, and the sound quality is crap. Should i sell my SGS2 for a 3, or just buy a cowon j3 or something?