Can you hear a difference between the top analog preamp and a straight wire? Listening test

[Sokel]

Don't know what is different, but I think I had nulls around -80 db. Or maybe my memory from yesterday is bad.

That was the right channel,that's the left:



And the PK Metric:

 

[Adis]

You can do this. I assume busted is boosted. That is about where I hear it as well with 40 db of gain added. Now having done this with many devices you usually get some weird noises and very unbalanced FR in the residual. This is notable for being very clean and even when boosted with extra gain. Noise and distortion levels had to be really low for this to be the result.

Does it count? NO! that would be cheating.

Well, I don't know of any difference other than the subtraction result.
And yes, I meant "boosted".

 

[pma]

Well that is what I did after I listened, then ABX'd it and then examined it with Deltawave. Boosting the difference file with 40 db of gain you faintly hear the music. I didn't do a straight inversion like in Audacity.

Properly done (32 bit float) straight inversion and subtraction results in this:

Amplified then of 40dB makes this:

It is almost unbelievable but the result is similar to the original file and there is no noise when you listen to it. All the difference is due to about sub microsecond delay in the preamplifier path (the files were recorded simultaneously - L/R - with simultaneous sampling). So, the difference is completely linear - no distortion. Linear and inaudible (in the presence of the original, not if listened as an amplified subtraction), far below human ear threshold of audibility.

 

[voodooless]

Well that is what I did after I listened

I know you know , but I'm not sure @Adis knows, too.

 

[DWPress]

Do you have an 8ch version?

 

[pma]

@voodooless : BTW your guess was correct, for the experienced member like you it can be told from a little bit higher noise bottom level above 22kHz.

 

[voodooless]

@voodooless : BTW your guess was correct, for the experienced member like you it can be told from a little bit higher noise bottom level above 22kHz.

I saw that, but for me, the tell was the phase difference near 0 Hz. The pre is either AC coupled or has a DC servo (it looks like the latter in your schematic).

 

[Blumlein 88]

I saw that, but for me, the tell was the phase difference near 0 Hz. The pre is either AC coupled or has a DC servo (it looks like the latter in your schematic).

More cheating on a listening test.

 

[wwenze]

Regarding noise, I wondered a few hours back, if the environmental air is noisy and the wire is sufficiently long, the preamp may be able to have less noise than the wire if the active preamp is able to suck up the noise at the load side. This assumes the preamp has a lower output impedance than the original source.

So just put the wire next to an SMPS. Call it cheater countermeasure.

And because you had to use 2 of those cables to connect to the pre-amp that bottle neck weighs double making the files indistinguishable.
The 2 cables removed all of the goodness the pre-amp brings.

Next time you need to use 2x a Nordost Odin 2 cable for this test for the test otherwise it is a waste of time.

To properly compare "preamp vs wire", I suggest we need to make a wire with a female connector on each end, so it functionally recreates a preamp including needing 2 more cables.

 

[pma]

To properly compare "preamp vs wire", I suggest we need to make a wire with a female connector on each end, so it functionally recreates a preamp including needing 2 more cables.

Good comments . The "wire" is a XLR - XLR cable in fact. Looking forward for more similar comments .

 

[pma]

More cheating on a listening test.

Yeah . However, I am still waiting for the first ABX result with p <= 0.05 from at least 16 trials. I guess that all supporting means like FFT analyzers are allowed .

@voodooless : DC servo. Contrary to Hypex and Purifi guys, I do not like for the amp to amplify DC, 0Hz component.

 

[JeremyFife]

Lovely little piece of music ... can't tell the difference between samples though

 

[danadam]

Don't know what is different, but I think I had nulls around -80 db. Or maybe my memory from yesterday is bad.

I get -60-something without matching (just "Show" button) and -80-something with matching ("Match" button).

That was the right channel,that's the left:

These are mono files. Weird that there is a difference in correlated null and jitter shown in the status at the bottom.

I get the same values both for L and R but different than yours

(@pkane ?)

 

[Sokel]

I get -60-something without matching (just "Show" button) and -80-something with matching ("Match" button).


These are mono files. Weird that there is a difference in correlated null and jitter shown in the status at the bottom.

I get the same values both for L and R but different than yours
View attachment 373086
(@pkane ?)

Maybe the order of comparing?
That's weird.

 

[Sokel]

I get -60-something without matching (just "Show" button) and -80-something with matching ("Match" button).


These are mono files. Weird that there is a difference in correlated null and jitter shown in the status at the bottom.

I get the same values both for L and R but different than yours
View attachment 373086
(@pkane ?)

Just to make sure,check my settings in case I have messed up with something:

 

[danadam]

Just to make sure,check my settings in case I have messed up with something:

View attachment 373087

In "General" I have "Resample" and "Dither" disabled (but I don't think that matters here) but I also have "Measure simple waveform" disabled.

 

[Sokel]

In "General" I have "Resample" and "Dither" disabled (but I don't think that matters here) but I also have "Measure simple waveform" disabled.

Ok,let me try with it for L and R with your settings:



It gets more weird I think but at least I have the same null as @Blumlein 88

 

[pma]

The only trusted audio software to me is Adobe Audition. Audacity is too simple, DW has bugs.

 

[voodooless]

Audacity is too simple

Never mind the privacy concerns...

 

[pkane]

Don't know what is different, but I think I had nulls around -80 db. Or maybe my memory from yesterday is bad.

Null of -80dB:

There is a fairly loud pop just past the 41 seconds mark:

No issues with phase or frequency response: