Audible difference between analog interconnects

[wwaldmanfan]

This subject has been bandied about repeatedly here in various threads, so, rather than search back, I've created a new thread just to make a quick comment.

 

I did a simple test which may or may not be relevant to a high-end audio system, but I can tell you that I certainly heard a major difference between two sets of stereo analog interconnects on my video setup:

 

I have the variable analog audio outputs of my Cablevision digital cable TV converter connected to a NAD integrated amp which is connected to a decent pair of Klipsch booshelf speakers by short lengths of 14AWG OFC speaker wire. While I primarily watch broadcast TV and cable movies on this setup, I also listen to cable music channels.

 

The previous interconnect was a 1/2 meter length of the familar Monster Cable paired black wire with the red/blue gold-plated RCA plugs. In order to get the tone set right for the majority of material that I watch/listen to, I kept the bass knob at 9 o'clock and the treble knob at 3 o'clock on the NAD amp using the Monster interconnect.

 

I swapped it for a handmade "boutique" 1/2 meter interconnect that I had purchased some years ago. I think I paid $100 for it. It is constructed of very thin pure silver wire encased in Teflon insulation and tightly twisted, with good quality gold-plated RCA plugs on the ends.

 

I've always felt that silver conductors sound "hot" compared to copper, even with digital cables, like USB. This is why I was not using said silver interconnect on my audio system, which lacks tone controls (other than EQ AudioUnit plug-ins). But, with this cable on the video rig, I now set the bass and treble controls flat to get the same, balanced tone I was experiencing before with MC at 9/3 o'clock. Whether this is preferable is not the point; I just say that there can be a considerable audible difference between cables of like purpose, based on materials and construction.

[esldude]

I have done nulling comparisons of a cable like you describe. I have of course no idea whether it is the same or not. It has two silver wires (I believe they are 22 gauge) inside teflon tubes and tightly twisted with excellent RCAs.

 

They of course nulled down around the thermal noise floor of my gear at the time like some inexpensive cables, some very expensive AQ copper and some even more expensive AQ silver cables did. Meaning I can send music over the silver boutique cable, record the result with a good ADC. Do the same for the other cables. Invert the signal for one cable versus another and see what is left. Effectively there was nothing left other than a bit of noise way down so far it couldn't matter.

 

Of course I didn't use bass and treble controls. So maybe my calibration and comparison wasn't precise enough.

[sandyk]

Effectively there was nothing left other than a bit of noise way down so far it couldn't matter.

 

Perhaps you had the reason staring you in the face, but refused to accept it as YOU believe that the differences were at too low a level ?

If somebody told you that the artifacts from a USB Regen were also at that low level, you would more than likely seize on that as the reason why some (mainly from another forum) believe they don't work properly ?

[Panelhead]

This subject has been bandied about repeatedly here in various threads, so, rather than search back, I've created a new thread just to make a quick comment.

 

I did a simple test which may or may not be relevant to a high-end audio system, but I can tell you that I certainly heard a major difference between two sets of stereo analog interconnects on my video setup:

 

I have the variable analog audio outputs of my Cablevision digital cable TV converter connected to a NAD integrated amp which is connected to a decent pair of Klipsch booshelf speakers by short lengths of 14AWG OFC speaker wire. While I primarily watch broadcast TV and cable movies on this setup, I also listen to cable music channels.

 

The previous interconnect was a 1/2 meter length of the familar Monster Cable paired black wire with the red/blue gold-plated RCA plugs. In order to get the tone set right for the majority of material that I watch/listen to, I kept the bass knob at 9 o'clock and the treble knob at 3 o'clock on the NAD amp using the Monster interconnect.

 

I swapped it for a handmade "boutique" 1/2 meter interconnect that I had purchased some years ago. I think I paid $100 for it. It is constructed of very thin pure silver wire encased in Teflon insulation and tightly twisted, with good quality gold-plated RCA plugs on the ends.

 

I've always felt that silver conductors sound "hot" compared to copper, even with digital cables, like USB. This is why I was not using said silver interconnect on my audio system, which lacks tone controls (other than EQ AudioUnit plug-ins). But, with this cable on the video rig, I now set the bass and treble controls flat to get the same, balanced tone I was experiencing before with MC at 9/3 o'clock. Whether this is preferable is not the point; I just say that there can be a considerable audible difference between cables of like purpose, based on materials and construction.

The Eichmann or KLE plugs make a big improvement. Wire and construction is less critical.

Using PCOCC or 7 nines 26 - 28 copper with urethane varnish twisted pair construction works well. A cotton sheath for spacing is even better.

The real improvement is the RCA's.

[Speedskater]

When doing interconnect testing, you can't even touch any of the balance & level pots. The smallest change can be audible.

 

With RCA unbalanced analog interconnects, a cable without a shield (screen) invites increased interference & noise.

[Paul R]

Perhaps you had the reason staring you in the face, but refused to accept it as YOU believe that the differences were at too low a level ?

If somebody told you that the artifacts from a USB Regen were also at that low level, you would more than likely seize on that as the reason why some (mainly from another forum) believe they don't work properly ?

 

At what DB level do you consider noise to be inaudible? Remember, Dennis noted the noise was at the thermal noise floor - you cannot get electrically quieter than that.

 

-Paul

[CR250]

I have done nulling comparisons of a cable like you describe. I have of course no idea whether it is the same or not. It has two silver wires (I believe they are 22 gauge) inside teflon tubes and tightly twisted with excellent RCAs.

 

They of course nulled down around the thermal noise floor of my gear at the time like some inexpensive cables, some very expensive AQ copper and some even more expensive AQ silver cables did. Meaning I can send music over the silver boutique cable, record the result with a good ADC. Do the same for the other cables. Invert the signal for one cable versus another and see what is left. Effectively there was nothing left other than a bit of noise way down so far it couldn't matter.

 

Of course I didn't use bass and treble controls. So maybe my calibration and comparison wasn't precise enough.

 

I set up a fairly elaborate listening test with some AQ silver and copper IC's, and was able to tell the difference between the 2 100% of the time. The other people who took the same test scored about 90%. There was no question whatsoever that there's a difference in sound between the 2 metals. (In the test system. I can't say the same difference will be heard in other systems.)

[sandyk]

At what DB level do you consider noise to be inaudible? Remember, Dennis noted the noise was at the thermal noise floor - you cannot get electrically quieter than that.

 

-Paul

 

How could Dennis see the differences he noted in the thermal noise floor ? You need special techniques to do that.We have had similar arguments before, even with Barry D. involved, about Dennis's Diffmaker results and his conclusions about their lack of audibility.

Differences between Amplifier distortion artifacts in the -130 to -140dB area are quite audible BTW, as I showed in posted measurements of a front end balanced amplifier on several occasions.

[esldude]

I set up a fairly elaborate listening test with some AQ silver and copper IC's, and was able to tell the difference between the 2 100% of the time. The other people who took the same test scored about 90%. There was no question whatsoever that there's a difference in sound between the 2 metals. (In the test system. I can't say the same difference will be heard in other systems.)

 

So what were the details of the fairly elaborate listening test?

[esldude]

How could Dennis see the differences he noted in the thermal noise floor ? You need special techniques to do that.We have had similar arguments before, even with Barry D. involved, about Dennis's Diffmaker results and his conclusions about their lack of audibility.

Differences between Amplifier distortion artifacts in the -130 to -140dB area are quite audible BTW, as I showed in posted measurements of a front end balanced amplifier on several occasions.

 

What special techniques are you talking about? The theoretical thermal noise would be for all the resistance in the circuit. Simple to calculate. The EIN for the gear I was using was something like 5-6 db above the bare thermal noise. And that is about the difference I obtained for the various nulling on the cables. A few db from just thermal noise. The tiny additive noise of the active circuitry. People can hear 15 db to perhaps 20 db into the wideband noise level. FFT can look deeper than that into the noise present. There simply isn't much there. Nothing to show an audible effect from those measurements.

[CR250]

So what were the details of the fairly elaborate listening test?

 

To be honest, I kind of regret my last post. I've detailed the story a couple of times and was attacked over and over by people that didn't want there to be a difference. I was a fool for even bringing it up.

[mansr]

To be honest, I kind of regret my last post. I've detailed the story a couple of times and was attacked over and over by people that didn't want there to be a difference. I was a fool for even bringing it up.

 

In other words, it didn't hold up to scrutiny.

[Jud]

In other words, it didn't hold up to scrutiny.

 

You''ve proved a point. I'm just not sure it's the one you intended.

[CR250]

You''ve proved a point. I'm just not sure it's the one you intended.

 

He's right. I shouldn't have brought it up. Cables are a very touchy subject. You usually end up looking like a fool no matter what you say.

[mansr]

You''ve proved a point. I'm just not sure it's the one you intended.

 

I proved whatever point the reader wants to see. It's all subjective.

[sdolezalek]

If cables were a religion, it is clear to me that virtually everyone here has chosen not to risk going to hell, because their buying behavior indicates making choices that are only logical if you believe.

[Paul R]

To be honest, I kind of regret my last post. I've detailed the story a couple of times and was attacked over and over by people that didn't want there to be a difference. I was a fool for even bringing it up.

 

I am sorry you got hit with such disrespect. Usually people here don't get argued with when they report hearing a difference, though the same cannot be said for the reasons they might posit for hearing that difference.

 

Mostly, one has to just grow a thick skin, which isn't a lot of help when one feels they have been dumped on.

 

But in any case, can you drop a pointer to the details, or post them here please? I for one am very interested, though I say up front, I rarely if every hear differences from interconnects, save for USB, Coax, Optical, and speaker cables.

 

-Paul

[Paul R]

How could Dennis see the differences he noted in the thermal noise floor ? You need special techniques to do that.We have had similar arguments before, even with Barry D. involved, about Dennis's Diffmaker results and his conclusions about their lack of audibility.

Differences between Amplifier distortion artifacts in the -130 to -140dB area are quite audible BTW, as I showed in posted measurements of a front end balanced amplifier on several occasions.

 

Hard to reply to this, because I am not really interested in arguing with you over the subject, but it is not difficult at all to calculate the level of (thermal) noise in a system.

 

As for -140db being audible, I suppose it depends upon what you are using for zero DB. Normally, in measurements, zero db is the quietest sound that a human can hear - and that is near total silence. Each +10db means the sound is about twice as loud to our ears. (Obviously, each -10db means the sound is half as loud to our ears.)

 

What did zero DB represent in your measurement?

[Allan F]

Normally, in measurements, zero db is the quietest sound that a human can hear - and that is near total silence. Each +10db means the sound is about twice as loud to our ears. (Obviously, each -10db means the sound is half as loud to our ears.) .

You don't know what you are talking about. 0 dB refers to the maximum or "peak" value of a digital audio sample. Or, in an analog context, it generally refers to the maximum level of a signal before it distorts or clips. References to -10 dB, -20 dB, or -130 dB for that matter, are relative to that peak.

[kumakuma]

You don't know what you are talking about. Not even close. 0 dB refers to the maximum or "peak" value of a digital audio sample.

 

Both of you are partly correct

 

0 db SPL is the threshold of human hearing

 

0 dbFS is the maximum possible digital level

[esldude]

You don't know what you are talking about. Not even close. 0 dB refers to the maximum or "peak" value of a digital audio sample.

 

Come on now. Yes, often people count 0 dbfs as max level in these days of digital because it can be precisely stated what 0 db is. Then you work downward from there.

 

In Paul's case, he was referring to the threshold of hearing with ears which is 0 dbSPL and working upward from there.

 

I had the same thought about those differences at -140 dbfs. If your max digital level of 0 dbfs represents 120 dbSPL in your listening room (not many systems will do this) then -140 dbfs would be 20 dbSPL below the zero threshold for hearing something. That 0 dbSPL threshold is also only going to work in a specially quieted place. Very rarely are normal listening rooms even as quiet as 20 dbSPL above threshold or in this example 40 dbSPL above the -140 dbfs distortion products. Very hard to see how you can possibly hear that even if you were listening only to the distortion with fundamentals filtered out. Throw in masking effects and someone is straining credibility hugely, and then some to think one is going to have any chance of hearing that.

 

It occurs to me another reason to find such an idea lacking believability. Our ears threshold is about what it is partly because if our ears were more sensitive in some frequencies, we would hear the random brownian motion of air molecules bumping about. So we are supposed to believe 20 dbSPL below threshold well into brownian air motion effects we can hear such things picked out separately from other sound and the random motion? Oh boy, that is really spreading it thin.

 

edited: to differentiate SPL and FS db values once kumakuma pointed it out quite correctly.

[Paul R]

You don't know what you are talking about. Not even close. 0 dB refers to the maximum or "peak" value of a digital audio sample.

 

(sigh)

 

Allan, you so often come off like a rude, ignorant ass. You must practice it a lot.

 

Did I not specify I was talking about hearing? As in audibility measurements?

 

And did I not ask specifically what he is using as a zero measurement? Do you not understand that the decibel scale is logarithmic? That means it is a linear representation of non-linear values, and what something in decibels *means* depends almost entirely on what is being measured. Sound pressure, hearing, sound intensity, voltage, power? All are *different*.

 

I know you are a lawyer, so you should be able to recognize when someone is wording something carefully and precisely to get information without starting an argument.

 

By the way, the peak value you refer to is not a "digital audio sample" - it most often refers to a VU meter reading. (Volume Unit) That means zero is a voltage of exactly 1.288 (AC) volts at 600 ohms and 1Khz. I think it is slightly different in Canada, but I can't recall what it is from memory. VU meters are, by design and definition, integrating an analog signal. Digital VU meters work differently.

 

[Edit - And yes, I should have specified more clearly I was talking dbSPL vs dbfs. I was *trying* to be non-argumentative, because I was rather interested in what Alex was talking about.]

[Allan F]

My point was that when one speaks of audibility to -130 dB or whatever, that is relative to the peak signal of 0 dB. That was clearly the context of Alex's post. The meaning of 0 dB obviously has different meanings depending on the context. For example, Paul, you pompous ass, 0 dB does refer to the maximum or peak value of a digital audio sample, i.e. all ones. If you are unable to reply appropriately to Alex's post in context because you don't understand it, that is clearly your problem.

[esldude]

My point was that when one speaks of audibility to -130 dB or whatever, that is relative to the peak signal of 0 dB. That was clearly the context of Alex's post. And, FYI Paul, you pompous ass, 0 dB does refer to the maximum or peak value of a digital audio sample, i.e. all ones. If you can't reply appropriately to Alex's post in context because you don't understand it, that is clearly your problem.

 

Allan, please, stop digging.

 

Paul obviously always understood that. It was inherent in his post for him to bother commenting as he did. And understanding it just as you stated it, one is at a loss to figure out under what conditions the reproduction of such low levels would be possible or audible once everything is a sound in the air.

[tarquineous]

I still have a set of Music Metre silver interconnects which I purchased in 1995-96. They sound pretty good, and are smooth sounding compared to some others. In general, and after trying at least 25 interconnects since then, I do not prefer silver cables over copper for audio use. Silver tends to soften the bass, highlight the midrange, and lessen full harmonic content.

Silver plated is just as bad or worse for interconnects, but is sometimes very good on tweeters and midrange drivers.

The better sounding interconnects do not have heavy insulation on the metal wires, base on what I have actually heard over the last 15 or so years.