Do speaker cables make any difference?

Status
Not open for further replies.
rdf said:


EMI, RFI; my thinking is frequency agnostic. :magnet:

Sometimes it is confusing. I tend to use EMI, "electromagnetic interference", for signals whos wavelengths are much longer than the system that is being spoken of. This class of signals is not broadcasting through the "ether" like a radio wave, but is just a result of the magnetic field being produced by an object in the vicinity. Because it is a result of local field generation, it is picked up by the system through conductive loops and faraday's law of induction. This class cannot be easily stopped by the use of simple shielding, braids are useless at 60 hz up to 10 or 20 khz, braids cannot stop the magnetic intrusion because the skin depth at these frequencies is more than the braid thickness.

RFI uses different shielding techniques. While RFI can be induced through the line cords, I am not referring to them, just what I call EMI.

Cheers, John
 
soongsc said:
4. With 3 cables the statistical success rate should be 1/3. But in reality, if the exact reverse identification counts in some way, the success rate was at least 50% for both sessions.
Lets throw out the "reverse identification" nonsense; it doesn't count. You are engaging in a practice known as mining the data (not to be confused with "data mining," although the two are closely related).

Now, assuming there was no compromising of the blind, and taking the results at your word, you got 8 out of 20 correct.

Choosing strictly at random gives you a 33% probability of getting at least 8 correct out of 20 trials.

In other words, a p-value of 0.33. Not much to hang your hat on, and certainly wouldn't get you published in any journal.
 
Dumbass said:
Lets throw out the "reverse identification" nonsense; it doesn't count. You are engaging in a practice known as mining the data (not to be confused with "data mining," although the two are closely related).

Now, assuming there was no compromising of the blind, and taking the results at your word, you got 8 out of 20 correct.

Choosing strictly at random gives you a 33% probability of getting at least 8 correct out of 20 trials.

In other words, a p-value of 0.33. Not much to hang your hat on, and certainly wouldn't get you published in any journal.

I think the point is not whether I can hang my hat or not, but exploration of how the results will differ based on different test methods. If you just want to get personal. I don't think there will be any progress but meaningless arguments. If you really are interested, please conduct your own test and share the results. I know I can hear a difference and can generate a preference, but how repeatable it is is what's interesting to find out.

But, if everyone here just want to argue, please just keep on doing so and have fun.

I took the time to find out, and was willing to share the experience.
 
soongsc said:


I think the point is not whether I can hang my hat or not, but exploration of how the results will differ based on different test methods. If you just want to get personal. I don't think there will be any progress but meaningless arguments. If you really are interested, please conduct your own test and share the results. I know I can hear a difference and can generate a preference, but how repeatable it is is what's interesting to find out.

But, if everyone here just want to argue, please just keep on doing so and have fun.

I took the time to find out, and was willing to share the experience.

soongsc, I do not believe his intent was personal.

What he pointed out was the relative merit of the data you have provided to predict correctly the entity being tested. From a statistical point of view, your results can easily be explained by random chance.

I thank you for the effort you have put into this.

Cheers, John
 
soongsc said:
I think the point is not whether I can hang my hat or not, but exploration of how the results will differ based on different test methods. If you just want to get personal. I don't think there will be any progress but meaningless arguments. If you really are interested, please conduct your own test and share the results. I know I can hear a difference and can generate a preference, but how repeatable it is is what's interesting to find out.

But, if everyone here just want to argue, please just keep on doing so and have fun.

I took the time to find out, and was willing to share the experience.
I have done wire swaps before, and honestly did not hear a difference. This was when I was pretty agnostic on the whole wire issue.

Well, it turns out that I was right -- interconnects don't make an audible difference, as your experiment and others have shown.

I still have half a mind to build that ABX unit. Actually, I would build two, so I could verify the transparency of the units. But the real point of building them would be for the evaluation of much more important components than mere wire.

The psychology of belief is fascinating to me, which is why I have enjoyed participating in this thread.
 
Anyone confident enough to do a serious test with either > 50 listeners or > 50 repeats? Lets finally do it, all examine the results as statistically significant or not and lay it to rest (suggest a believer and disbeliever take part in the design so it does not 'aim' for a result as did poor soongsc.

How are you getting on with your Jordan Watts restoration project soongsc? These were great units for their day!
 
keladrin said:
Anyone confident enough to do a serious test with either > 50 listeners or > 50 repeats? Lets finally do it, all examine the results as statistically significant or not and lay it to rest (suggest a believer and disbeliever take part in the design so it does not 'aim' for a result as did poor soongsc.
Beyond the great difficulty and expense of arranging such a test, there will never be results "definitive" enough to satisfy the subjectivists.

Witness how several people in this thread question the validity of even the protocol of ABX (i.e. identify an unknown signal as type A or type B), not to mention the technical specifications surrounding the test (i.e. source, associated equipment, room treatments, the "pressure" of taking part in a scientific test, etc etc etc ad nauseum).

It's the audio version of creationism. Rather than confronting the data that stare you right in the face, wriggle and contort and equivocate to come up with any answer other than the obvious one.
 
P.S. I too would be interested to hear about these Jordan-Watts units.

I am considering putting together a two-way active system based on either the 92mm or 50mm Jordan drivers plus 12" woofers. Although I am veering toward the larger drivers, I am somewhat hemming and hawing having heard very good things about the smaller ones.
 
Dumbass said:

Witness how several people in this thread question the validity of even the protocol of ABX (i.e. identify an unknown signal as type A or type B), not to mention the technical specifications surrounding the test (i.e. source, associated equipment, room treatments, the "pressure" of taking part in a scientific test, etc etc etc ad nauseum).


Where I come from we call that 'science'. This 'good enuf' :bs: we call 'belief system'.
 
Dumbass said:
I have done wire swaps before, and honestly did not hear a difference. This was when I was pretty agnostic on the whole wire issue.

Well, it turns out that I was right -- interconnects don't make an audible difference, as your experiment and others have shown.

I still have half a mind to build that ABX unit. Actually, I would build two, so I could verify the transparency of the units. But the real point of building them would be for the evaluation of much more important components than mere wire.

The psychology of belief is fascinating to me, which is why I have enjoyed participating in this thread.

Many friends here actually hear significant differences that we actually did shootouts. A group would just get together, mostly in stores, bring their own interconnects and start comparing. Not everyone will find their cables are as good as they though. Most of the equipment we used at that time were really expensive. Those were the times when there were not that much else to do, and people just had fun doing this. We have heard significant differences in power cables as well, but I really think it's more caused by poor PSU design. One manufacturer even went so far as to specifying Mission speaker cables with one line in the reverse direction(mae no sense to me).

I've spend the least amount of money on speaker cables, because I really did not mind the difference either because it was not significant, or it was just different but did not improve the system performance in the way I felt was correct. I'm sure as I go to more poweful systems, my perspective might change.

This is what I think is very critical. If the difference is just different, it is more difficult to pin-point which one you listened to. I think this is why most listening tests turn into statistically reasonablew numbers. But if the difference is making things sound more real and correct, it is much more easier to identify. Another thing is the way we listen. If I get myself in the analytical listening mode, then I get mixed up quite easily. If I'm in the "feel the perfomance" mode, the more accurate one is more obvious. This is why during my second listening session, the correct ones were identified just listening to the very first few seconds of selected track. During that session, only two tracks on different CDs were used.
 
keladrin said:
Anyone confident enough to do a serious test with either > 50 listeners or > 50 repeats? Lets finally do it, all examine the results as statistically significant or not and lay it to rest (suggest a believer and disbeliever take part in the design so it does not 'aim' for a result as did poor soongsc.

How are you getting on with your Jordan Watts restoration project soongsc? These were great units for their day!

I have the suff to do it, just had't had the time to get to that. But I do have a pair that's working okay, probably have about 6+ of those drivers. I'm sort of trying to collect Teds drivers. It's interesting to see how they evolved. Most driver I have really are based on sound principles. I'm actually looking into 6"5 range drivers to see if the cone breakup usually existing in the 5~6K range in these types of metal cone drives can be eliminated or not.

Back to cables. I am also interested in designing a test to challenge the 1M if it's for real. Now you know why I ran two sessions. The last time I was Florida was probably 2 years ago.
 
Dumbass said:
P.S. I too would be interested to hear about these Jordan-Watts units.

I am considering putting together a two-way active system based on either the 92mm or 50mm Jordan drivers plus 12" woofers. Although I am veering toward the larger drivers, I am somewhat hemming and hawing having heard very good things about the smaller ones.

For two ways I would recommend the JxR6. We are using a 3" that comes close, but still not quite as good in the higer frequencies.

Jordan Watts are out of production. So is the original 50mm one. I am a proud owner of both.:angel:
 
Dumbass said:
Audiophile creationism, as surely as jerk follows knee, with a cute little smiley icon to boot.

Creationist, too funny. If you have cogent reasoning to back up your smug dismissals bring them forward. Otherwise these continuous ad hominem rebuttals are everything you believe yourself to be fighting against.
 
although this talk mostly concentrates on speaker cables,I will share findings of a simple comparative test on two Intrrconnect cables we did last night with two friens.NO ABX, NO 16VALVE etc...,just simple change as we surely trust our ears and willingness to reject what we concider as worst even if it is our cale or other's cable.Both cables were from van den hul,namely Jubilee silver hybrid and First Ultimate carbon.Straight to the result:the silver one has a clear sound.ok ''sound'',very good detail,control and dynamics.Also,slight mid prominence and hf hardness.Going to the carbon one differences were huge and not very small as many say in this forum.No mid prominence,no hf hardness,and enviroment acoustics of four OPUS3 LP's were astonishingly reproduced in contrast to the dryer less ambient ''sound'' of the silver one.Theese for now.
 
Well, if the differences were indeed huge, you're in line to win a million dollars! I mean, heck, how hard could it be to distinguish HUGE differences in a simple blind test?

Just remember your old friends here while you're out guzzlin' Dom Perignon and cruisin' for babes in your Ferrarri.
 
Status
Not open for further replies.