Quote:



If I understand the set-up correctly, the process of converting the analog signal to digital would effect both signals in the same manner. So in order for it cause problems with this test, the converting to .wav would have to filter/color the signal in the exact opposite way that the power cable would enhance it provided the quality of the recording is sufficient in the first place (which it should be).

Quote:



Not sure what you mean ? The specified PC audio device looks pretty good from a technical perspective, what characteristics are you referring to ?

Actually I am not sure from the OP whether the power cable was used on the DVD, the DAC or on the PC ?


Quote:



Perhaps the OP could answer that ?

I looked at these files in Audacity, and I have to say they are not identical. It *isnt* the length or the starting point, once you trim them and align them they are still not the same. The differences are admittedly *miniscule* and invisible unless you zoom down to the 0.001 second level but they are not identical as such, there are differences in energy levels. The audibility is questionable but there are measurable differences.

Whether these differences are attributable to the cables is another matter. I would like to see some replication, perhaps these differences are at the level of variations you would expect from any two trials, given that a voltage supply from the mains can vary by as much as 10% from time to time it may be as simple as that, more data would be helpful.

Also Audacity, while very nice and a personal favourite is pretty crude by professional standards, this may also be a limiting factor.

I'm going to try and answer this leaving out as much signal processing terms as possible, but I may not be able to refrain.

The only way information would be lost in the conversion is if one of the power cables would be able to introduce signal content above half of the sampling frequency (nyquist frequency). Then the filters in the ADC would remove it so it would not alias to another frequency it should not be. but as he was using a CD to play the sound clip you are limited to a bandwidth of 22050Hz, so even sampling at 44.1K (samples a second) should not be a problem, let alone the fact most sound cards can go up to 96.2K nowadays.

Also another place where information could possibly be changed is in the load that is presented by the ADC to the DAC. Depending on the circuit design it could present a different load than your amplifier or processor that you would normally hook it up to, thereby possibly changing the response of the DAC.

All these exceptions aside, the relative comparison of the cables would be valid still. And I applaud SamNOISE for doing this experiment, just proves alot of things me and a few of my audiophile friends on campus have known for some time.

Quote:




This is a very good point. It is difficult to know what the error in the experiment is, especially with only 1 test case.

Quote:



I was going to do the same thing this morning (with GoldWave rather than Audacity, but the end result is the same) but didn't have time. I'm glad someone took that step.

What I would like to see is two different .wav files (or lossless files) created with the same cable to see if the types of differences that you are observing happen even with the same cable.

Now we're talkin'. You have to be careful not to assume the conclusion in the design. You have to assume there might be a difference caused by a different power cord and then ask what would allow it to be detected if so or prevent it from being evident, on the other hand. It's the latter I'm asking about: whether the quality of the interconnects could mask any difference while still acting the same both ways. Likewise the reception to the card or card characteristics and a/d conversion. You have to be sure the setup would reveal any difference that IS there. Then, if and when it doesn't, it really means something.

Quote:



Let me see if I have this straight. Your suggestion is that power cables may make audible differences but the differences may be masked by other variables such as interconnects and or capture cards.

So you would want to repeat the experiment with a large number of different interconnects and or capture cards.

My question was somewhat different. I wonder if the differences I observed were natural random variation. If such random variation could be replicated then you need to derive the limits of variation for cable 1 then do the same for cable 2. Until you can do this you cannot proceed as you need to adjust for this.

Obfuscation on parade. No difference is no difference.

See ya
Steve

Quote:



You could do either or both of 1) try a variety of cable sets and cards, but a large number would seem unnecessary and make it impossible which I wouldn't want to see happen, or 2) get recommendations from those who believe they hear differences in power cords of which interconnects and cards are most conducive to detecting these differences and use one of those. For cables this would probably be some higher end brand like Virtual Dynamics Master or higher, Purist Dominus, Synergistic Tesla or some such. Likewise for cards, though for all I know you may be using the best already.

The difference is one will rip/tear/break easier, but you can spend less money replacing the cable every ten years than buying a huge unbreakable one

Quote:



except of course that that files ARE different on the small scales, according to nick_charles and my own observations.

Now we can argue about how much of that is random fluctuations in the equipment (margin of error), and it would be nice to have 2 or 3 files from each power cord to get a better idea, but clearly your above statement is false and misleading.

Do any of the folks that sell (or review) power cables ever make false or misleading statements. <eye roll>

I think Bigshot might have been better off including the word "significant" in his post, but he is still dead on: obfuscation is obfuscation. Call the duck whatever you want, but if it quacks, swims and flies it is probably still a duck.

But what do I know, I only have 13 posts, right?

Quote:



no, because this time one file is slightly longer.

Quote:



The important word in this is "probably". It is proof that is being sought so that doesn't quite do it. But it appears close and a few more efforts to rule out other remaining possibilities might be all it takes. How in the world is that "obfuscation"? Try to take off you combative and partisan blinders long enough to notice that I am not trying to defeat this experiment but to see if it can be perfected and rendered convincing and irrefutable to all. Some of you must be interested in more than victory and jeering at those of different opinions, mustn't you?