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 21

post #301 of 1186
Quote:
Originally Posted by bigshot View Post

To test cables, the first thing to do would be to do a double blind test to determine whether there is a difference between them at all. Two cables, randomly sampled, as long a listening period as the test subject wants, on their own home equipment if they want, enough trials to eliminate chance.

This has been done many times. The result is always the same. You won't want to hear the result.

 

That's more like it. Links please. The ones I can find seem to only loosely focus on cables and more or silly comparisons like car battery jumper cables vs. 10k cables.

post #302 of 1186
Check out the myths thread. There have to be a few in there.
post #303 of 1186
Quote:
Originally Posted by x838nwy View Post
 

I think A but I am willing to look into the possibility of B. Sadly those who believe B seems to only want things wrapped in a 'double blind' wrapping. So my suggestion would be for them to try for themselves seeing that they won't accept reasonable alternatives.

Do you not understand what 'double blind' means?

To test if something is audible, you must do a listening test. Double blind simply means that the effect of extraneous variables is removed, which is mandatory for any rigorous result.

 

Nobody cares if you can measure a difference that nobody can hear.

 

Quote:
So, do you mean to say it's all bs? If one takes these results as gospel, then cables DO sound different.

The effect of cables is known and well-measured. Nobody is saying cables have zero effect. In longer runs, the type of cable is important.

In that first test, it shows that even for 3 metre lengths, even the worst cable was within at worst 0.5 dB of the best performers at extreme frequencies (20khz). This is likely to be far below the threshold of audibility.

The cable designed by real engineers for digital (high frequency) transmission, which costs something like $2 per metre, did better than the 'audiophile' cables designed by quacks.

 

Effectively this means that if you are using very long cables, you may need to spend a few more dollars per metre to buy a cable designed for real purposes by real engineers. This is the exception and most people can happily use jumper cables and hear no difference.

 

The second example shows that there are differences between amplifiers; however amplifiers are not cables.

post #304 of 1186
Quote:
Originally Posted by higbvuyb View Post
 

Do you not understand what 'double blind' means?

To test if something is audible, you must do a listening test. Double blind simply means that the effect of extraneous variables is removed, which is mandatory for any rigorous result.

 

Nobody cares if you can measure a difference that nobody can hear.

 

The effect of cables is known and well-measured. Nobody is saying cables have zero effect. In longer runs, the type of cable is important.

In that first test, it shows that even for 3 metre lengths, even the worst cable was within at worst 0.5 dB of the best performers at extreme frequencies (20khz). This is likely to be far below the threshold of audibility.

The cable designed by real engineers for digital (high frequency) transmission, which costs something like $2 per metre, did better than the 'audiophile' cables designed by quacks.

 

Effectively this means that if you are using very long cables, you may need to spend a few more dollars per metre to buy a cable designed for real purposes by real engineers. This is the exception and most people can happily use jumper cables and hear no difference.

 

The second example shows that there are differences between amplifiers; however amplifiers are not cables.

 

I don't think you've been reading, but I understand. It's all a little dry. It is understood that extraneous variables must be removed. But does the knowledge/belief that there is no difference before test count? In fact, knowing that they're testing for differences in cables sets up all sorts of biases in people's minds. Some will try to look for things that might not exist, while others may become blind to whatever might be there. Other factors such as those others that I have mentioned must also be considered and dealt with properly also. My general reservations include the fact that the finer points of our hobby is a bit of an acquired taste and requires developed senses and some are more acute at picking up these things than others. This can play a big factor on results. For example, if one expects a bunch of people to be the indicator to whether cables make a difference, do we know if the same group of people would agree on another test in the same environment but with  changes that are actually more pronounced and deliberate? To me it's like measuring something without first calibrating your measuring equipment. Anyway...

 

And I was under the impression that a lot of people are saying that cables have no effect as long as they are functional. (We ARE including wire hangers, after all.) So they do have an effect on the sq now? That's good to hear, thank you.

 

0.5dB difference is in audible? It's not all that far below audibility. http://www.audiocheck.net/blindtests_level.php?lvl=0.5 some people can go better than that. It's amazing.

 

I don't think I've said anywhere that cost=performance. It's an interesting result, nonetheless. It sort of shows that that particular $2/meter cable sounds "better" than the 'audiophile cable'. So if we call the $2/m cable an audiophile cable, would that change things? The point here is whether or not cables can make a system sound different or not (i'm refraining from 'better' here). The $2/m cable seems to say that it does.

 

So at the magic 3 m, it somehow matters. As I do not see anything intrinsically magical about the number 3,  i assume the same factors whatever they are that matter at 3 m remain important at 10, 19 or 1 meter, correct? May be at a lower magnitude but still there, may be? This kindda goes against the "all cables are the same and if you think they're different you're imagining things" posts I've been reading.

 

You know, I'm sort of with you on this one, though. Between my gungnir and mjolnir I use the schitt cables at 6" and I can't think of anything that will better them cos 6" of decent cabling should at least do less damage (if it does at all, granted) than anything 1 m long, no matter how amazing. So up to a point, if you go extremely short, things probably don't matter. I think there's a wireworld youtube clip where the guy talks about this. But I don't think this discussion has any limits on length however.

 

You seem to have respect for proper engineering of things and so do I. Given that, I'm sure you also know that there is no such thing as "the best" design in engineering. That $2/m cable I'm sure can be made better too. If the task is taken up by someone who works for a company that makes audio cables and he/she ends up improving on it and sells it for $20/m, would that be somehow a scientific impossibility suddenly?

 

Finally, I included the second link to which you refer as an example of how a dbx test may not be so successful at times. Or at least sometimes results are not entirely reliable.

post #305 of 1186
Quote:
Originally Posted by x838nwy View Post

 

I don't think you've been reading, but I understand. It's all a little dry. It is understood that extraneous variables must be removed. But does the knowledge/belief that there is no difference before test count? In fact, knowing that they're testing for differences in cables sets up all sorts of biases in people's minds. Some will try to look for things that might not exist, while others may become blind to whatever might be there. Other factors such as those others that I have mentioned must also be considered and dealt with properly also. My general reservations include the fact that the finer points of our hobby is a bit of an acquired taste and requires developed senses and some are more acute at picking up these things than others. This can play a big factor on results. For example, if one expects a bunch of people to be the indicator to whether cables make a difference, do we know if the same group of people would agree on another test in the same environment but with  changes that are actually more pronounced and deliberate? To me it's like measuring something without first calibrating your measuring equipment. Anyway...

It's already been stated that you cannot prove the null hypothesis. However cable enthusiasts and sellers who claim to be able to hear an obvious difference should be able to tell in a DBT. Unless they're being fooled by placebo. It is only common sense to treat these claims with suspicion, let alone spend a thousand dollars on them.

 

And I was under the impression that a lot of people are saying that cables have no effect as long as they are functional. (We ARE including wire hangers, after all.) So they do have an effect on the sq now? That's good to hear, thank you.

 

The 'simple' answer is 'no, there is no difference'. However a simplification is by definition different to the truth.

 

0.5dB difference is in audible? It's not all that far below audibility. http://www.audiocheck.net/blindtests_level.php?lvl=0.5 some people can go better than that. It's amazing.

A 0.5dB absolute level difference, perhaps. But what we're dealing with here is something like 0.2dB down at 10kHz, and 0.5dB down at 20 kHz. I can't even hear at 20 khz anymore,

 

I don't think I've said anywhere that cost=performance. It's an interesting result, nonetheless. It sort of shows that that particular $2/meter cable sounds "better" than the 'audiophile cable'. So if we call the $2/m cable an audiophile cable, would that change things? The point here is whether or not cables can make a system sound different or not (i'm refraining from 'better' here). The $2/m cable seems to say that it does.

Better is subjective. That cable has less distortion, but the amount is insignificant for most cable lengths.

 

So at the magic 3 m, it somehow matters. As I do not see anything intrinsically magical about the number 3,  i assume the same factors whatever they are that matter at 3 m remain important at 10, 19 or 1 meter, correct? May be at a lower magnitude but still there, may be? This kindda goes against the "all cables are the same and if you think they're different you're imagining things" posts I've been reading.

 

You know, I'm sort of with you on this one, though. Between my gungnir and mjolnir I use the schitt cables at 6" and I can't think of anything that will better them cos 6" of decent cabling should at least do less damage (if it does at all, granted) than anything 1 m long, no matter how amazing. So up to a point, if you go extremely short, things probably don't matter. I think there's a wireworld youtube clip where the guy talks about this. But I don't think this discussion has any limits on length however.

Obviously it depends on the length of the cable. Those posts are simplifications designed to prevent clueless people from spending huge amounts on cables without reason. Of course if someone wants the full explanation, nobody here is going to stop them.

 

You seem to have respect for proper engineering of things and so do I. Given that, I'm sure you also know that there is no such thing as "the best" design in engineering. That $2/m cable I'm sure can be made better too. If the task is taken up by someone who works for a company that makes audio cables and he/she ends up improving on it and sells it for $20/m, would that be somehow a scientific impossibility suddenly?

Better cables already exist. The point here is that nobody needs to pay more for a cable that delivers an improvement below the threshold of audibility.

 

Finally, I included the second link to which you refer as an example of how a dbx test may not be so successful at times. Or at least sometimes results are not entirely reliable.

Not all experiments will be well designed, this is real life. Just look in a scientific journal. However, the more effective the blinding is, the better the experiment. This is why we have things like peer review, clear reporting of methodology and data, etc. Sadly some 'research' has none of these.


Edited by higbvuyb - 4/8/14 at 6:08am
post #306 of 1186
The thread is a whole lot of words that doen't mean anything. Sometimes I think being an audiophile has more to do with being able to spew our reams of vague descriptions and logical obfuscation than talking about getting better sound.

For me, anyone who honestly wants to improve the quality of sound in their setup is my pal. X838NWY, you mentioned your system has a significant flaw... harshness. I'd be happy to help you solve that problem, and I'm sure I'd be able to help you, but I'm not interested in talking in circles about wires until you do as much research on the thresholds of human perception as much as you have manufacturer sales sheets.

The differences between properly designed and manufactured audio cables are inaudible. Controlled DBT has shown that.

NEXT!
post #307 of 1186

Well it's not really call "significant" but it does affect my enjoyment of the system. Good news is, I think I've narrowed it down to the actual quality of the mastering itself. Seems a lot to do with clipping actually in the recording and some occasional pops and clicks. I first thought it was my DAC but it showed up on other DACS I own too once I hook them up to a decent amplifier or use a more revealing pair of iem's (borrowed). I tried similar songs at similar levels and played perfectly fine. I think it was just that my amp was capable enough at digging up the details that this became apparent. To be honest, if we're going OT and talking about system improvements, I must say I'm a little disappointed by my PWD (mkII). Not THAT disappointed, but sometimes it does sound a bit artificial sounding for some reason. Compared to something like a Chord Qute, it seems a little dry. I'm not good enough to put it to words, but that's about it. I'm hoping the DirectStream upgrade where PS goes from DAC chips to FPGA's will go some way to help this. I'll have to wait and see.

 

Still, I take your point about cables. To summarize, you say that cables must be properly designed to function correctly. But onec a certain level of functionality is reached (let's call it a functionality threshold),  any change as long as it does not reduce the functionality below a certain point, is inaudible to humans. Do I get this correct? I just happen to think that there is no threshold and that things can be improved. I am still looking for time to look properly into the details of those DBT's and am waiting (foolishly perhaps) for replies from cable manufacturers.

post #308 of 1186
Quote:
Originally Posted by x838nwy View Post
 

To summarize, you say that cables must be properly designed to function correctly. But once a certain level of functionality is reached (let's call it a functionality threshold),  any change as long as it does not reduce the functionality below a certain point, is inaudible to humans. Do I get this correct?

 

Yes... but I've never run across a cable that messed up sound myself. In anything under a 20-50 foot length, it is extremely unlikely. The cables from China that come with your blu-ray player, three dollar cables at mono price, fifty dollar fancy cables... all of them are the same. However very high end boutique cables marketed to audiophiles are much more likely to be deliberately designed to not conduct properly. The screwed up performance is referred to as "house sound".

 

I'd be happy to try to help you figure out the "artificial sound" in your system. I seriously doubt your guess about the cause of it is correct. I suspect something quite different.

post #309 of 1186
Quote:
Originally Posted by x838nwy View Post
 

.... a $5k cable wrapped in angel's pubic hairs.

 

I bet that would sound heavenly!

 

Here are my 2cents on the cable debate (and whether or not we should be having it).

 

Look over at the non-Science forums here at head-fi. There are many active threads which contain discussions about cables as if there are important sonic characteristics that only high-end cables can provide.

 

Now look at Sound Sicence, there is only this one discussion of cables in a scientific setting and it's filled with skeptics who refuse to discuss the topic.

 

Do you guys know how that appears to budding audio enthusiasts from the outside? People don't often do an exhaustive search inside of headfi and read through all the thousands of posts about cables (or whatever they happen to be searching for). Many folks are either googling a topic and clicking google links direct to threads, or they're browsing the active topics at the top of each forum listing. These causal head-fiers, head-fi-lurkers, and google-fiers are getting a very low signal-to-noise regarding high-end cables, and unless they happen to have a heavy training in electrical engineering, they are most likely getting a very skewed idea about the importance of high-end cables and their supposed audible effects.
Most folks read a statement like "silver cables sound bright, copper cables sound warm. This is because if different resistivity of copper vs silver" and think 'ooh! that sounds like a scientific explanation' rather than thinking "hey, the difference in resistivity between a 10 foot silver and copper interconnect is vanishingly small compared to the input impedances of a typical line level device, that supposed audible difference doesn't make a lot of sense, let me find my electrical engineering textbook and refresh my understanding of impedance bridging..."

 

If we don't continue to address the notion that high-end cables sound different than reasonably well built cables, then the myth of cable sound will continue to be perpetuated to the next generation of hi-fi enthusiasts.

 

Also, I'm going to gripe a bit about this: there is no bijection between double blind testing (i.e., DBT) and science. Double blind testing is a method by which one can test the hypothesis that a difference exists between two populations. Science is the generation of new knowledge by testing falsifiable hypotheses using systematic methods and reasoning. Double blind testing is a valuable tool for performing scientific tests, but it is only a subset of the tools available to science.

 

Here's free information on the internet: http://en.wikipedia.org/wiki/High-end_audio_cables

 

Cheers

post #310 of 1186
The Admins should sticky this link in the header for the forum.
http://www.roger-russell.com/wire/wire.htm
post #311 of 1186

there is "too cheap" construction in RCA cables at the lowest end of the retail chain

 

the shield can be pigtailed under the moulding, leaving a open loop for magnetic couping when it should be 360 degree terminated for effective RF shielding

 

very cheap RCA cable may have < a dozen strands in the shield, coverage factor <10% - so allows capacitive coupling of external noise, EMI/RFI, may only be loosely spiraled rather than braid and can open up area on bends

 

and these skimpy "shields" have too high resistance - in SE IC like RCA the only defense against chassis current errors is extremely low shield/gnd resistance

 

Bill Whitlock from Jensen - not just an anyone's opinion:

 

Quote:
Originally Posted by jcx View Post

MOGAMI - The Cable of the PROS

or if you can diy:
"'Exotic cables will not stop noise. Expensive cables, even if double or triple shielded, made of 100% unobtainium, and hand woven by virgins will have no significant effect on hum or buzz. Only the resistance of the grounded conductor can make a difference.' If you have to use coax, Belden 8241F, with a low-resistance copper-braid shield, works well for audio and video."
from:
Washington June AES - SMPTE.org

Edited by jcx - 4/9/14 at 6:14am
post #312 of 1186
Quote:
Originally Posted by x838nwy View Post
 

I just happen to think that there is no threshold and that things can be improved.

This is trivially false - there must be some level where a difference is too small to be heard (unless you can demonstrate the existence of a perfectly accurate sensor)

The concept of a sensory threshold has been around for decades in neuroscience and is not controversial.

 

Yes, you can use sophisticated measuring devices to assemble a set of equipment that is 'improved' beyond the threshold of audibility. However, this forum is about human beings listening to music, not about people seeing how close they can get graph A to graph B.

post #313 of 1186
Researching the thresholds of human perception is just as important as researching specifications and measurements. Without applying perceptual thresholds, all those graphs and charts are meaningless.
post #314 of 1186

Leaving audiophile jargon aside, long as the wire can carry the required amps, has essentially no resistance and has proper shielding to keep out stray emi nothing else matters. Now its totally possible the stock cable has no shielding and is such a low gauge it can't handle the amps so re-cabling might change the sound signature but when it comes right down to it, who cares?

 

If you like pretty cables(which i do) then by all means go nuts. Like the saying goes, if it aint broke mod it!

post #315 of 1186

Before people take my posts the wrong way, what I meant by "threshold" in the context of my post was not the auditory threshold. Well, not exactly.

 

What I meant is this:

 

If there's a specification to which a cable must conform (say have a certain value of LCR at a particular frequency and voltage input and a bunch of other parameters) what most seem to suggest is that if a reasonably made cable can adhere to these specifications at the stated conditions, the cable is deemed to be functional. If the specification calls for a particular tolerance, say X +/- Y microH for example, as long as a cable can keep within these limits, it is functional. Let's call this cable "A".

 

At the same time, what most seem to be saying also is that if another cable (let's call it "B" since I have lots of imagination) can perform to within X +/-G microH, it will sound identical to cable "A" as long as G < or = to Y.

 

In effect, the spec +/- Y is like a threshold beyond which no audible difference will result at the ears of the listener. Effectively that the cable spec tolerances for a threshold below the human auditory threshold. Perhaps I used the same word too many times and I must apologize for any confusion.

 

What I question (foolishly according to most of you guys) and it's really the essence of all that I've been blabbing about is whether or not the difference between G and Y matters in ways that produce audible results. I propose that it does. Perhaps as a result of the design choices of input/output stages or s/pdif implementation and so on and so forth. A lot of people it seems are of the view that the specifications are suitable and the differences between wires cause inconsequential changes to their variations from the spec. Or something like that.

 

*Please note that I am simply clarifying what I'm saying and not dragging up anything in an attempt to troll.

 

It's been a few days now and no response from manufacturers. Some of you might enjoy this fact.


Edited by x838nwy - 4/10/14 at 2:43am
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.