[Dreadhead]

Apparently there was a talk at RMAF proving that cables make a difference.

Stereophile: This is Music

I would be extremely interested to learn what this difference is etc. I am sure they make a difference that I can measure with an osillocope but I have a hard time believing people who can't hear .5 db shifts in response will hear the difference.

Anyway I'm hoping that they make the results a little more public than at some audiophile show. Keith O. Johnson has serious technical chops so I am interested (but he also a vested interest in the audiophile industry).

 

[SenjStevo]

Who is Keith O. Johnson? And more importantly i wonder if we will be able to see the full details of his experiment; assumptions, equipment etc...

Until we do have this information we know nothing really. You have to ask why it's taken this long for someone to come out with a study.

 

[Dreadhead]

Keith O. Johnson is one of the inventors of HDCD (among other things).

Information about Reference Recordings, "Prof" Keith O. Johnson and HDCD Recordings

To be honest I think the guy from Stereophile is most likely just not following what Mr. Johnnson was saying correctly. I'm thinking of emailing Reference Recordings and seeing if they'll ask Mr. Johnson to post it somewhere so the world can see it and evaluate it.

 

[Pio2001]

Jitter in S/Pdif cables is no news. It is measurable, and turned inaudible, and even not measurable, in the DAC.

Noise in power cables should be easy to measure. Filtered out by power supplies.

The effect of the quality of CD pressing have already been measured by Dunn, Dennis and Carson at Prism on integrated CD players (not on drive + DAC models, since they have independant power supplies). Their vast blind test done with controlled pressings coming from different factories or cutter settings showed no audible difference though it was done among people who claimed to hear the sound quality of different pressings.

Interestingly, measurments showed differences from a disc to the other, with no correlation with the cutting technique. They also showed that the measurable distorsion was amplitude modulation (caused by small variations in the power supply), and that it is very easily mistaken for jitter in spectrograms. Which in turn makes very doubtful Bruno Putzeys measurments, from Philips, of jitter on CDR, and the subsequent theory that power supply affects the clock, that in turn affects sound quality.

http://www.prismsound.com/m_r_downloads/cdinvest.pdf

 

[upstateguy]

Quote:

Originally Posted by Dreadhead /img/forum/go_quote.gif Apparently there was a talk at RMAF proving that cables make a difference. Stereophile: This is Music I would be extremely interested to learn what this difference is etc. I am sure they make a difference that I can measure with an osillocope but I have a hard time believing people who can't hear .5 db shifts in response will hear the difference. Anyway I'm hoping that they make the results a little more public than at some audiophile show. Keith O. Johnson has serious technical chops so I am interested (but he also a vested interest in the audiophile industry).

Hey Dread

It remains to be seen if any of this stuff is reproducible.

USG.

 

[digger945]

Quote:

Originally Posted by Dreadhead /img/forum/go_quote.gif I am sure they make a difference that I can measure with an osillocope...

Do you have access to a quality oscilloscope? Do you have some way to superimpose the results of two IC's?

 

[Pio2001]

Quote:

Originally Posted by digger945 /img/forum/go_quote.gif Do you have access to a quality oscilloscope? Do you have some way to superimpose the results of two IC's?

If you do not plan to reach the megahertz frequency range, a good soundcard and a recording software can beat many oscilloscopes regarding accuracy.

 

[edstrelow]

Quote:

Originally Posted by Pio2001 /img/forum/go_quote.gif Jitter in S/Pdif cables is no news. It is measurable, and turned inaudible, and even not measurable, in the DAC. Noise in power cables should be easy to measure. Filtered out by power supplies. The effect of the quality of CD pressing have already been measured by Dunn, Dennis and Carson at Prism on integrated CD players (not on drive + DAC models, since they have independant power supplies). Their vast blind test done with controlled pressings coming from different factories or cutter settings showed no audible difference though it was done among people who claimed to hear the sound quality of different pressings. Interestingly, measurments showed differences from a disc to the other, with no correlation with the cutting technique. They also showed that the measurable distorsion was amplitude modulation (caused by small variations in the power supply), and that it is very easily mistaken for jitter in spectrograms. Which in turn makes very doubtful Bruno Putzeys measurments, from Philips, of jitter on CDR, and the subsequent theory that power supply affects the clock, that in turn affects sound quality. http://www.prismsound.com/m_r_downloads/cdinvest.pdf

Not a particularly convincing study or a particularly good publication. It certainly wasn't the Journal of the Acoustical Society of America. They note that the subjects often didn't follow the protocol. They apparently were letting different users use their own players, so you don't even have one set type of equipment. At they end they had just started to test "golden-eared" listeners but didn't get very far.

You really can't prove the null hypothesis in part because the failure to show a difference can always be attributed to problems with the experiment.

Oddly at the end they discuss motor-induced intermodulations of the analog signal in "one-box" (no separate DAC?).

 

[Pio2001]

Quote:

Originally Posted by edstrelow /img/forum/go_quote.gif You really can't prove the null hypothesis in part because the failure to show a difference can always be attributed to problems with the experiment.

Yes, but since they especially chose listeners who claimed to hear these differences, it proves, at least, that the claim about differences was wrong.
This is not the same as the null hypothesis, it does not prove that all crows are black, but it shows that most people who claim to have seen white crows were wrong.

Quote:

Originally Posted by edstrelow /img/forum/go_quote.gif Oddly at the end they discuss motor-induced intermodulations of the analog signal in "one-box" (no separate DAC?).

They tell that these intermodulation are absent from the output of separate DAC.

 

[digger945]

Quote:

Originally Posted by Pio2001 /img/forum/go_quote.gif If you do not plan to reach the megahertz frequency range, a good soundcard and a recording software can beat many oscilloscopes regarding accuracy.

I have already performed this experiment. Didn't notice any differences.

 

[myinitialsaredac]

I will be doing a similar experiment focusing on interconnects, but I will be using a high resolution audio analyzer and restricting it into the audio band. Check out my experiment here - http://www.head-fi.org/forums/f133/r...tudent-368745/. I listed out in some pretty solid detail what will be going down. Should be interesting, goin down to ASU to finalize the process before I conduct experiment =D.

Dave

 

[digger945]

OK, how do you plan to record and superimpose the results of each test for each IC?
When I did it I went beyond .0001 second and I don't really know how much voltage.
I superimposed with Schope plugin for Cakewalk Pyro audio creator.

 

[myinitialsaredac]

The audio analyzer sets the two inputs on top of each other or one input on top of the original waveform (because it produces its own wave). Really a handy tool.

Dave

 

[myinitialsaredac]

Also I spoke with mike and george about superimposing them on the oscilloscope if thats what I end up using, George said that his had a simple funtion that calculated the differences between the lines and he dubbed it the "m-line" which I reference so often.

Dave

 

[digger945]

I forgot to mention, I used an 0404usb for both playback and record, and two ALO IC's, one solid silver, one copper cryo.
What is the resolution of your analyzer?
Can you tell us the make and model of the machine? In all my research, I only remember one or two analyzers that had the capability of superimposing two signals, or at least two channels on the screen simultaneously.
Are you gonna vary the test tone, ie not just 10kHz but a variety of freqs, possibly a modulated signal?