or Connect
Head-Fi.org › Forums › Equipment Forums › Sound Science › Audiophile cables, an interesting question.
New Posts  All Forums:Forum Nav:

Audiophile cables, an interesting question. - Page 40

post #586 of 1186
Quote:
Originally Posted by x838nwy View Post

 It's be interesting to do the test between two models of cables from the same company (same length). Sadly I don't have those right now frown.gif
 

Yes, I would say that would be the simplest test that would most easily show differences. On the basis that it's best to only change one variable at a time, your mind is less likely to be diverted by different "house sounds", which (at these objectively tiny difference levels) are as likely to confuse as to clarify.  

 

FWIW, my non-scientific impression on stock cables is that they are generally quite neutral. No particular house sound. Just evenly mediocre across the board.

post #587 of 1186
Quote:
Originally Posted by castleofargh View Post
 


 cable length does make a difference, I've read some complicated stuff about it from people who seemed to know what they were talking about. but ofc the cable they spoke of was a 100m long ^_^. 

 

Exactly, while length makes a difference if we're talking hundreds of meters, between 1m and 1.5m there is no difference whatsoever that can be measured to have an impact on human audibility.

 

The best test to do is just to try a standard RCA cable, 10usd max, vs a "hi-end" cable. If you go by what cable believers say, then that's where the biggest difference should be audible (though it won't be). It'll be interesting to see the results of his test.


Edited by elmoe - 5/4/14 at 1:25am
post #588 of 1186

Taken to the extreme with everything else remaining equal, longer RCA analog interconnects:

 

a] Some lengths will be better interference antennas than other lengths.

b] Longer cables will have poorer leakage current Signal to Noise ratios.

c] Will have greater total capacitance which may cause the circuit to oscillate. 

post #589 of 1186

TBH the only thing I'm really wondering about is if it is possible to reduce the stereo separation(so for cables conducting analog obviously) by making a cable in a really dumb way?

like making huge efforts to have more inductance, and keeping right and left as close as possible.

 

it's something I've been wondering for some times (also can we split a rock with lightning? and how does my ethernet cable sounds like?)

post #590 of 1186

'stereo separation' or crosstalk. Well we can run 4 different audio signals down a 300 foot Cat5 cable and the crosstalk is still in the -100 dB range.

 

It takes lots & lots of crosstalk to be audible when listening to music.

post #591 of 1186

Okay guys, results are in.

 

First things first, I made a mistake with the length of the Nordost Blue Heaven. I checked my cables and they're actually 1.5m so that's the same length as the cheapo cable I'm testing against.

 

Before we go into the numbers, I need to say that this is as much as test of cables as a test of me actually noticing things. After about 1/3 of the way through the second session, there's just one thing I needed to look for and it was actually fairly straightforward. In the song Royals by Lorde, there's a sort of rhythmic sound that's like 3  snaps in succession. On the cheap cable, they're a little more lumped together so the sound more like one continuous sound. It takes some time to get it in my head, but I think as a result, the last 4 results of each sessions are all correct. Dunno, may be I'm imagining things again… if you put enough monkeys in front of enough type writers for an infinite length of time, the complete works of Shakespere is probably an inevitability.

 

Results are:

Session 1: Y Y Y N N N Y N Y Y = 6/10

Session 2: N Y Y N Y N Y Y Y Y = 7/10

Session 3: N N Y Y N N Y Y Y Y = 6/10

 

Total = 19/30 = 63.33% => not statistically significant, as such it cannot be said that I can reliably tell the difference between the two pairs of cables.

 

That's not to say that there is no difference. To my ears there is. Apart from the above, the cello on All Through the Night (Yo-Yo Ma, Goat rodeo sessions) sounds a hair more natural and real on one of the two also. But it's a very small difference.

 

An important point I take from this is that while I'm happy with what the statistics say, having experienced the test first hand, I fear there's more to it than my simple setup can answer. You see, it's like if I show you two red tiles. The two are slightly different in that one is very slightly (but still visibly) darker than the other. Placed side by side, you'd clearly be able to tell. But if you're shown only one tile at a time and are asked to say whether it's the darker or the lighter, it wouldn't be so obvious. I feel it'd be better if I could compare what I hear currently against the same song played through either cable. Say if I could have three imaginary buttons, one to hear it played through the mystery cable, one for having it played through cable A and another through cable B and my job is to match what I'm listening to, to either A or B. May something similar exist with such facilities.

 

Now some might say that had I connected the cables directly, as opposed to through the SYS, it'd be more noticeable or whatever, but I'm wouldn't say that's likely. If we try this test on better gear, who knows. Had I tried it with an even more extravagant cable? I cannot say for certain but it is likely that the results would be similar. If anyone has these uber interconnects they're happy to part with I'd be happy to check for them. 

 

Anyway, to conclude - there does not appear to be any difference that can be reliably identified (by me) between electrical  analogue signals going through these two pairs of analogue interconnects between a DAC and a headphone amplifier.

 

Considering their price differences - roughly x50 - it appears to be one of the less effective ways to improve a hifi system. I have to stress, however, that I do not claim to be an expert with the best ears or possess a vast experience in high fidelity. These results are based on my gear and my ears. It may be different for others, I don't know.

post #592 of 1186

Considering though, that in the cheapo cable vs hi-end one, the results are roughly the same as your "mid-end" vs "hi-end" comparison earlier, it's safe to say that there is absolutely no audible difference between any of the cables used. I don't understand why you persist in saying there is a difference to your ears. If there was, you'd be able to tell. I suggest you do this:

 

Cut a sample from a song from which you have confidence in being able to tell the difference (you can use Audacity for this, its freeware and easy to use), the cello on All Through The Night for example. Do another test. You should be able to get a result above 80% at least. If not, then it's expectation bias and there really isn't any difference.

post #593 of 1186
I'm afraid that you seem to be particularly susceptible to expectation bias, x838nwy. You should sit down when this is all over and think about how to not let expectation get the better of you, even when the facts are clear. It would be like swimming upstream to have to constantly deal with that.
post #594 of 1186

Also, as an interesting note, you don't need to score close to 100% to show statistical significance. An ABX or A/B test can have a positive outcome even if you only get the correct answer 51% of the time. The closer your outcome is to 50% though, the more trials are needed to show that the result is unlikely to be purely chance driven. In your case, the sum of your last three trials is 19/30. While this is actually somewhat unlikely from chance alone, it isn't nearly unlikely enough to be considered statistically significant (if I did the math right, the chances of getting at least 19/30 correct by pure chance are around 14%). If you continue to get ~63% correct through many more trials (which is not an expected outcome based on the current data, I should add), the probability of a 63% success rate occurring by chance alone drops below 5% once you have successfully achieved 32/50 correct (so a 32/50 would be considered a possible positive result for an ABX test). If you tighten up the required P value to 1% (since 5% still means that by chance alone, 1 in 20 random trials will achieve that kind of significance), you would need a score of 62/98 to show a significant result.
 


Edited by cjl - 5/4/14 at 12:05pm
post #595 of 1186

I commend you x838nwy for doing this.  You are experiencing what at least many of us have.  You just know, feel it in your gut, they aren't the same.  Then with a blind test, well you feel differences are small, but you just know they are there.  Yet the test results indicate there is no audible difference likely. 

 

I have tried to conceive of a test like the two red colored tiles you speak about.  My first idea was separate cables on each channel listening over headphones.  But that doesn't work.  When differences I artificially induce in one channel become quite perceptible in such a concurrent test they are already larger differences than I detect with an ABX in Foobar. 

 

Also, though you have the subjective experience of seeing two tiles 'at the same time side by side' your eyes actually switch rapidly from one to the other, and do some trickery so you think you are looking at both.  Using software that could instantly switch from one color to other on my computer screen I found I could more easily pick two very close colors than if I had them side by side.  The key is instant (in my case 1/60th of a second) switching.  Putting in a dead time of 5 seconds lowered the sensitivity of color discrimination for me.

 

Finally, the logic of using music for your test seems good.  It has been known for some time smaller differences can be detected with proper test signals in blind testing.  For your cable test I would suggest pink noise.  At not too high a level mind you.  This way you have a constancy of signal character, can rapidly switch between and make your choices.  Just a few seconds before switching makes the test less time consuming to do a fair number of trials.  I find it less tiresome in practice than using music myself.  This is if you wish to give it one more go. 

 

It takes time to turn loose of one idea and get comfortable with another one. You had the guts to actually try this out.  I have found few people willing to do such a thing.  Most just bluster about how obvious it is and never quite commit themselves to testing the proposition cables sound different.  Or attack the methodology if they do and get results they don't like.

 

As you get comfortable with the idea cables are generally transparent and fully interchangeable it is somewhat freeing.  You can use whatever inexpensive cable is at hand with no worries.  It is a part of the audio world that is cut and dried.  Leaves more money and time to focus on things that really matter. 


Edited by esldude - 5/4/14 at 12:03pm
post #596 of 1186
Quote:
Originally Posted by x838nwy View Post

Okay guys, results are in.

First things first, I made a mistake with the length of the Nordost Blue Heaven. I checked my cables and they're actually 1.5m so that's the same length as the cheapo cable I'm testing against.

Before we go into the numbers, I need to say that this is as much as test of cables as a test of me actually noticing things. After about 1/3 of the way through the second session, there's just one thing I needed to look for and it was actually fairly straightforward. In the song Royals by Lorde, there's a sort of rhythmic sound that's like 3  snaps in succession. On the cheap cable, they're a little more lumped together so the sound more like one continuous sound. It takes some time to get it in my head, but I think as a result, the last 4 results of each sessions are all correct. Dunno, may be I'm imagining things again… if you put enough monkeys in front of enough type writers for an infinite length of time, the complete works of Shakespere is probably an inevitability.

Results are:
Session 1: Y Y Y N N N Y N Y Y = 6/10
Session 2: N Y Y N Y N Y Y Y Y = 7/10
Session 3: N N Y Y N N Y Y Y Y = 6/10

Total = 19/30 = 63.33% => not statistically significant, as such it cannot be said that I can reliably tell the difference between the two pairs of cables.

That's not to say that there is no difference. To my ears there is. Apart from the above, the cello on All Through the Night (Yo-Yo Ma, Goat rodeo sessions) sounds a hair more natural and real on one of the two also. But it's a very small difference.

An important point I take from this is that while I'm happy with what the statistics say, having experienced the test first hand, I fear there's more to it than my simple setup can answer. You see, it's like if I show you two red tiles. The two are slightly different in that one is very slightly (but still visibly) darker than the other. Placed side by side, you'd clearly be able to tell. But if you're shown only one tile at a time and are asked to say whether it's the darker or the lighter, it wouldn't be so obvious. I feel it'd be better if I could compare what I hear currently against the same song played through either cable. Say if I could have three imaginary buttons, one to hear it played through the mystery cable, one for having it played through cable A and another through cable B and my job is to match what I'm listening to, to either A or B. May something similar exist with such facilities.

Now some might say that had I connected the cables directly, as opposed to through the SYS, it'd be more noticeable or whatever, but I'm wouldn't say that's likely. If we try this test on better gear, who knows. Had I tried it with an even more extravagant cable? I cannot say for certain but it is likely that the results would be similar. If anyone has these uber interconnects they're happy to part with I'd be happy to check for them. 

Anyway, to conclude - there does not appear to be any difference that can be reliably identified (by me) between electrical  analogue signals going through these two pairs of analogue interconnects between a DAC and a headphone amplifier.

Considering their price differences - roughly x50 - it appears to be one of the less effective ways to improve a hifi system. I have to stress, however, that I do not claim to be an expert with the best ears or possess a vast experience in high fidelity. These results are based on my gear and my ears. It may be different for others, I don't know.

I think your effort doing real testing is great. I appreciate the effort and that you share your results here. Certainly, more trials would help to solidify your results one way or the other!

Cheers
Edited by ab initio - 5/4/14 at 12:14pm
post #597 of 1186
Quote:
Originally Posted by elmoe View Post
 

Considering though, that in the cheapo cable vs hi-end one, the results are roughly the same as your "mid-end" vs "hi-end" comparison earlier, it's safe to say that there is absolutely no audible difference between any of the cables used.* I don't understand why you persist in saying there is a difference to your ears. If there was, you'd be able to tell. I suggest you do this:

 

Cut a sample from a song from which you have confidence in being able to tell the difference (you can use Audacity for this, its freeware and easy to use), the cello on All Through The Night for example. Do another test. You should be able to get a result above 80% at least. If not, then it's expectation bias and there really isn't any difference.

 

*No. I think x838nwy is correct in being as cautious as he is.
There are strong indications that there is no difference, but the tests are in no way conclusive.


Edited by limpidglitch - 5/4/14 at 5:29pm
post #598 of 1186

If you go to that much trouble and still can't discern a definite improvement, I think it's time to chalk it up to not mattering and move on to frying bigger fish. The whole point is to improve the sound of your stereo system, not to become a testing engineer.

post #599 of 1186
Quote:
Originally Posted by bigshot View Post
 

If you go to that much trouble and still can't discern a definite improvement, I think it's time to chalk it up to not mattering and move on to frying bigger fish. The whole point is to improve the sound of your stereo system, not to become a testing engineer.


+1 on that idea.     

post #600 of 1186
Quote:
Originally Posted by bigshot View Post
 

If you go to that much trouble and still can't discern a definite improvement, I think it's time to chalk it up to not mattering and move on to frying bigger fish. The whole point is to improve the sound of your stereo system, not to become a testing engineer.


Which is pretty much what x838nwy concluded with:

"Anyway, to conclude - there does not appear to be any difference that can be reliably identified (by me) between electrical  analogue signals going through these two pairs of analogue interconnects between a DAC and a headphone amplifier.
Considering their price differences - roughly x50 - it appears to be one of the less effective ways to improve a hifi system."

And I'd agree with that, but the bombastic tone Is doing none of us any good.
It's not out of kindness for the cable believers, I just want a little decorum and respect for the process.
 

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › Audiophile cables, an interesting question.