[skamp]

I am curious as to how the "original" sample was created.

Nothing special: I extracted a 30 seconds sample from a 16 bit / 44.1 kHz track that was ripped from a CD.

 

[Hi Rez]

Quote:

Nothing special: I extracted a 30 seconds sample from a 16 bit / 44.1 kHz track that was ripped from a CD.

More specifically, what utility did you use to extract the sample?

 

[skamp]

Audacity.

Spoiler

I assume someone will claim that Audacity is somehow broken, and thus this test is invalid.

 

[nick_charles]

Spoiler: Warning%3A%20Spoiler!

I tried listening tests and could not distinguish between any of them. Then I used Audacity to do a spectrum analysis on each. I exported the data as text and imported it into Excel, then I graphed the FR in Excel for each DAC I had 1023 data points. All 6 lines were different but they were all also the same (give or take) until you hit about 19 - 20K, at which point the signal is at -70db that is none of the DACs showed anything that could be regarded as a different signature in frequency terms up to 19K. Taking the min and max energy levels at each frequency and the differences between them (absolute values)  and charting them showed that the difference between quietest and loudest did not reach 1db until 11K (-53db). the differences between the traces below 20K could easily be ascribed to variance in the recording (AD process) and level matching from the sources.

 

[Hi Rez]

So six of the samples were recorded with the EMU 0204 USB, and all seven were edited with Audacity.  Did you use the identical Audacity preference settings for all edits?
 
Seems it would be difficult to unequivocally rule out potential effects from the EMU 0204 USB or Audacity.

 

[chewy4]

Quote:

So six of the samples were recorded with the EMU 0204 USB, and all seven were edited with Audacity.  Did you use the identical Audacity preference settings for all edits?
 
Seems it would be difficult to unequivocally rule out potential effects from the EMU 0204 USB or Audacity.

Why would that be difficult?

 

[Hi Rez]

Quote:

Why would that be difficult?

How would one conclusively test it?

 

[chewy4]

Quote:

How would one conclusively test it?

As for making sure Audacity isn't adding any effects... Just don't turn on any effects...
 
As for the recording device(this goes for Audacity as well), if you can't pick out the original sample it's obviously not a problem.

 

[skamp]

Seems it would be difficult to unequivocally rule out potential effects from the EMU 0204 USB or Audacity.

You're welcome to provide ABX logs demonstrating that the recordings are audibly different.

 

[Hi Rez]

Quote:

As for making sure Audacity isn't adding any effects... Just don't turn on any effects...
 
As for the recording device(this goes for Audacity as well), if you can't pick out the original sample it's obviously not a problem.

With any digital editor, even adjusting the volume levels requires bit manipulation and dithering.
 
But the original sample was edited with Audacity....

 

[nick_charles]

Quote:

You're welcome to provide ABX logs demonstrating that the samples are audibly different.

 
 
The possible (devils advocate)  problem is that the recording process may obscure differences. We assume that the recording device/software is accurate to 16 bits but even so if the differences between the analog outputs are very slightly different the quantization process has inherent (low order bit) errors and could obscure small differences. My take on this is that if real differences between two DACs are obscured by a 16 bit AD process then they are not differences worth worrying about. But the critics can say "of course I can hear the difference between two nominally flat 16 bits DACs but not when there is a 2nd AD stage"  - I went through all of this when i did my cable tests...

 

[chewy4]

Quote:

With any digital editor, even adjusting the volume levels requires bit manipulation and dithering.
 
But the original sample was edited with Audacity....

Why would he even adjust anything like that?
 
And why would it matter if the original sample had any alterations, if the others were recorded from it being played?
 
Just to be clear, you are worried about inconsistencies causing audible differences that shouldn't be there, right?

 

[nick_charles]

Quote:

With any digital editor, even adjusting the volume levels requires bit manipulation and dithering.
 
But the original sample was edited with Audacity....

 
 
In another thread in this subforum we saw some evidence that even converting a 24 bit file to 16 bits with the worst possible type of dither and truncation did not do any audibly detectable damage...

 

[Hi Rez]

Quote:

Why would he even adjust anything like that?
 
And why would it matter if the original sample had any alterations, if the others were recorded from it being played?
 
Just to be clear, you are worried about inconsistencies causing audible differences that shouldn't be there, right?

skamp matched the volume levels of all the recordings so differing volume levels would not influence the test.
 
Hypothetically, If the "sound" of the original sample could have been influenced by the editor, and then all the recordings made from it were then similarly influenced by the same editor.  Not saying it happened, just asking how one can rule it out.
 
Actually I am not worried about audible differences that shouldn't be there.  In my first post I stated that I didn't hear any....

 

[skamp]

I think you should worry about the influence of solar flares on my electronic gear, as well.