To my ears they all sound very similar if not identical.  I think the bitrate gives them away.
A=B=C=D=E=F=G as far as audible differences.  The bitrates are different making them all different technically.
 
If anyone is curious the files are safe.

I converted all to wav and didn't look at them before. I thought E may have had a balance issue but I wasn't listening for that.

I closed my eyes and put Foobar on shuffle.

This is an an interesting test and in the REAL world it may show that the Emperor indeed has no clothes.
 
 
HOWEVER   hard-core subjectivists can come up with an "out" -  "Well of course no one could hear the difference all we could hear was the A-to-D artifacts of the E-Mu 0204, which masked the undoubtedly huge differences in the DACs!"
 
I don't think that argument is valid, but "they" will think it is.
 
I'm building an A/B/X  box which will compare 2 level-matched DACs in a blind, computer-controlled random test; I hope to (someday!) have this rig at a Can Jam set up for anyone to come over and take the test controlled by the laptop with results to be collected by the computer and announced on Head-Fi.  Of course, the subjectivists could claim that passing the line-level audio through the rhodium-plated relay contacts negates the test, or that the interconnects I will be using are masking the sound or......  on and on and on.
 
PUTTING THEIR MONEY WHERE THEIR MOUTH IS: Here on Head-Fi I once challenged ANYONE to be able to correctly identify the difference of a Beta 22 amplifier going through  1 meter of silver wire and 1 meter of copper wire in an A/B test; If they are correct in  six or more trials out of ten, I give them $500 of my money and $500 of their money going to me if not.  WHAT A SHOCK  no one replied.
 
All that said, I have my own plans to do an A/B listening test with a couple of DACs that I have, one of my "non scientific" tests - so-called because for reasons of ease and speed I am not going to do a blind A/B comparison at this time. Blind testing will wait until I master the code writing needed to control my relay A/B box from a PC.   (Too many projects, too little time)
 
I have a few DACs because I have a few different rooms with systems, each one needs a DAC....  but so far my impression is that the Musical Fidelity Vdac mk 1  that I have is a little "softer" sounding, maybe even a little "tubey" compared to other DACs I have-  I don't know if this judgement will hold up under A/B  or blind A/B testing; it may just be a prejudice that comes from reading reviews and so on.
 
I suspect that one actual difference that might be heard between DACs is a little roll-off in the treble on some DACs.  44.1k DACs  do differ a bit in flatness of their high frequency response, and that may be fairly easy to hear if you're young enough to have good hearing above 15 kHz.  (Which leaves me out!   hahahah!)

BY THE WAY,  how is the E-Mu 0204??
 
 
I just bought an E-Mu 0204, I want to use it to do PC-based audio measurements. It hasn't arrived yet.
 
 
The line inputs of the 0204 are rated to have lower distortion and noise than most sound cards- the problem with using a sound card for audio measurement is that many sound cards just don't have good buffer amp stages on their line inputs, although their outputs may be relatively clean.  When you are measuring amplifier THD and so on, the THD of the line input matters.
 
I'd be curious to hear what people think of the A-to-D quality of the 0204.

goodvibes said:

Don't look at my rank first.:bigsmile_face:   Perhaps we should PM them to skamp to remove another avenue of influence.LOL

Click to expand...

Use the spoiler tag:

milosz said:

BY THE WAY,  how is the E-Mu 0204??


I just bought an E-Mu 0204, I want to use it to do PC-based audio measurements. It hasn't arrived yet.

Click to expand...

It measures very well. Also, you're going to need a cable like this, with one end terminated with two mono 1/4" jacks. The jack with the red ring is the right channel jack.

Quote:

Originally Posted by milosz /img/forum/go_quote.gif
 
HOWEVER   hard-core subjectivists can come up with an "out" -  "Well of course no one could hear the difference all we could hear was the A-to-D artifacts of the E-Mu 0204, which masked the undoubtedly huge differences in the DACs!"

Click to expand...

 
That is why the original file is included as well. That reduces the list of excuses to:
- the sample track is not revealing enough, is badly mastered, the listener is not familiar with it, etc.
- the listener's equipment is not revealing enough (this, and the first one are very popular)
- the ADC has a perfect "synergy" with the DAC, and corrects its flaws
- the test does not fully reproduce real listening conditions (e.g. the DAC may perform differently depending on what it drives, or the ADC may filter out ultrasonic or RF noise from the DAC that could be demodulated by the amplifier or headphones and result in audible artifacts)

I have been wondering about number 2/3 on your list.. i am not concerned about 1, my opinion is the listener doesn't have to know the song to hear the differences it should be intrinsically clear or reasonably relevant... and if they knew the song that might make the test a lot more inaccurate.
 
how do we know exactly if our gear isnt up to spec or if the gear is auto correcting therefore the differences are not readily avalible?

2. is most often used to dismiss the negative results of others ("yes, of course you cannot hear it on your cheap ear buds and onboard audio, but that does not mean the difference does not exist"); audiophiles are of course free to use whatever expensive equipment they have, but they usually simply refuse to participate in these tests. The "golden ears" argument also falls into this group.
 
3. is statistically very unlikely to hide large errors, one would not expect a DAC to have 10% distortion, for example, and then the ADC to almost perfectly un-distort it so that the result becomes less than 0.001%. Not to mention some errors like noise are not even reversible. It also does not make much sense to deliberately design hardware that way, I doubt it would be really cheaper than making both the DAC and ADC perform well in the first place; it would also easily be exposed when using a combination of DAC and ADC from different manufacturers. I have seen conspiracy theories that the Windows drivers of sound cards "cheat" to achieve better RMAA results, but when the cards perform just as well on Linux with open source drivers, these theories are debunked. Limited and partial "synergy" can occur, but not to an extent to make a major difference, and of course more often than not the ADC just degrades the sound even more, therefore making it overall easier (but in practice not unlikely still impossible) to hear a difference.
One potential problem is recording a sound card DAC with the ADC on the same card; since both use the same clock, low frequency jitter from the clock will be eliminated. Therefore, two separate devices should preferably be used, but then ground loops need to be avoided. Another possibility is running the DAC and ADC at the same sample rate, and the imaging of the DAC filter being aliased back into the audio band by the ADC filter, and correcting the frequency response to some extent (as an example, think of an upsample-downsample loop that simply duplicates samples first, and then removes samples; even though each conversion is of poor quality separately, the loop has perfect lossless overall performance).

jake120 said:

how do we know exactly if our gear isnt up to spec or if the gear is auto correcting therefore the differences are not readily avalible?

Click to expand...

If your gear isn't up to spec, and if that's your excuse for not hearing any difference, then it stands to reason that you shouldn't claim hearing any difference elsewhere to begin with. And I don't think anyone would blame you for that (except maybe self-important idiots who like to mock everyone's inability to hear very subtle differences).

Also, ADC's don't "correct" anything. The best they can do is capture the signal as faithfully as possible. At worst, they'll add some distortion, but the likelihood that that distortion just happens to "correct" any flaws (like a FR curve that's exactly the inverse of the DAC's), is ridiculously small. And keep in mind that we're talking about the EMU 0204 USB here, which measures extremely well and suffers from very little distortion.

stv014 said:

2. is most often used to dismiss the negative results of others ("yes, of course you cannot hear it on your cheap ear buds and onboard audio, but that does not mean the difference does not exist");

Click to expand...

Ah, yes, I didn't read it that way. They can say whatever they want, I'm still going to need proof of their claims. If the differences are so obvious with their unobtainium gear, proving it should be fairly easy.

stv014 said:

audiophiles are of course free to use whatever expensive equipment they have, but they usually simply refuse to participate in these tests.

Click to expand...

Indeed. And what's their excuse? They have better things to do? Like posting their claims 10 times a day? :rolleyes:

If this is correct...
 
http://www.ethanwiner.com/audiophoolery.html
 
Then it would mean that all dac's of decent quality are beyond perceptable levels of distortion, jitter etc...
 
However it might be the analog stage of the Dacs and other factors giving different dacs different signatures...

Quote:

skamp said:

Indeed. And what's their excuse?

Click to expand...

 
In most cases when I offered to audiophiles to create similar loopback based ABX tests, they simply ignored it entirely, and just kept repeating their claims as facts.

nicholars said:

However it might be the analog stage of the Dacs and other factors giving different dacs different signatures...

Click to expand...

Feel free to identify any one of my samples, or even just ABX any two samples

The people who claimed that some of my gear had very distinct "sound signatures" aren't here, unfortunately, and they've made it very clear that they will ignore this thread altogether.