Can you hear a difference between audio cables? ABX test

[SIY]

Random is exactly 50/50. 70/30 is success

That's not even vaguely correct.

I did, it was almost 50/50.

Res ipsa loquitur.

 

[ctrl]

The RCA cable with S/N 71dB (post #61) file cable3 is audibly distinguishable from both cable1 (XLR) and cable2 (RCA) with 16/16 ABX result, due to error voltage across the signal ground wire due to error loop current. I understand that you and several other members do not want to admit this fact, but you should be able to do so. It is easily explainable by basic laws of electrical engineering. You guys debate too much and do a little of real measuring and experiment job.

A photo of the setup would be helpful.
Also, the information on whether the cables were near an EMI source during the test or if realistic conditions were present.

The cheap cable, at 1.5m (4.4 ft) for an RCA cable, is already very long, and nobody disputes (as with speaker cables) that this will eventually have negative effects. Therefore, it can be read everywhere that RCA cables should always be as short as possible. That's why purchased audio devices usually come with only 0.5m (2ft) long cheap cables for testing.

Archimago conducted a similar experiment with RCA cables in 2013. However, without an "error-loop" and under normal conditions. His results tend to show similar trends (shielded better than unshielded) but differ significantly in THD values and noise from your measurements.

Perhaps you could measure the 1.5m cheap cable without an "error-loop" and with distance from potential EMI sources. Then the results should actually be similar to those of Archimago.


Analog RCA interconnects test by Achimago 2013:

 

[Ken Tajalli]

A photo of the setup would be helpful.
Also, the information on whether the cables were near an EMI source during the test or if realistic conditions were present.

The cheap cable, at 1.5m (4.4 ft) for an RCA cable, is already very long, and nobody disputes (as with speaker cables) that this will eventually have negative effects. Therefore, it can be read everywhere that RCA cables should always be as short as possible. That's why purchased audio devices usually come with only 0.5m (2ft) long cheap cables for testing.

Archimago conducted a similar experiment with RCA cables in 2013. However, without an "error-loop" and under normal conditions. His results tend to show similar trends (shielded better than unshielded) but differ significantly in THD values and noise from your measurements.

Perhaps you could measure the 1.5m cheap cable without an "error-loop" and with distance from potential EMI sources. Then the results should actually be similar to those of Archimago.


Analog RCA interconnects test by Achimago 2013:

View attachment 357179

genuine question:
Why are you suggesting he should?
I think we can gather that two pieces of coat-hanger wire in a proper Faraday's cage, would perform possibly better that any cable. So what would an error loop tell us?
I was under the impression that the point was in two folds:
- at what level noise or distortion becomes audible.
- how well a certain cable behave in an average environment subjected to possible EMI. Include the DAC or amp in there too.
Did I get it wrong?

 

[ctrl]

Why are you suggesting he should?

To see if the experiments are repeatable.
Therefore, one should always provide all information about the test setup.

I think we can gather that two pieces of coat-hanger wire in a proper Faraday's cage, would perform possibly better that any cable. So what would an error loop tell us?

The normal case when using RCA cables is without a "ground loop" (I've never had problems with it) and usually, you don't wrap your RCA cable around a power supply or similar, but rather lay them out so that EMI radiation is minimized.

I was under the impression that the point was in two folds:
- at what level noise or distortion becomes audible.
- how well a certain cable behave in an average environment subjected to possible EMI. Include the DAC or amp in there too.
Did I get it wrong?

The thread started with "Can you hear a difference between audio cables?", then it turned out to be a comparison of two different techniques: RCA vs XLR.

Then @pma provided measurements and an audio sample of an RCA cable with extremely high noise and THD values, which better fits the thread title. However, the measurement values of the cheap cable are so drastically different from another source (Archimago) that one should indeed question how this can be.

It makes little sense to construct extreme test scenarios that are irrelevant for 99% of users.
With a cheap 0.5m RCA cable, THD+noise is very likely to be inaudible. If someone needs a 1.5m long RCA cable, the forum will likely recommend a "reasonable" cable for $10, which will probably be hardly distinguishable from XLR - as @pma's first audio samples had shown.

What @pma's measurements clearly show is that, from a measurement standpoint, XLR is always superior to an RCA connection - but that's probably not something anyone would doubt.

 

[solderdude]

You probably missed the post #67 and #61.

Can you hear a difference between audio cables? ABX test

There is definitely noise background error and non-harmonic background modulation with the single ended link / USB ground loop (cable2 file). System limits DAC DX5 + ADC (still source limited) Cable1 system (DacMagicPlus - XLR - ADC) Amir's AP measurement of DM+

www.audiosciencereview.com

The RCA cable with S/N 71dB (post #61) file cable3 is audibly distinguishable from both cable1 (XLR) and cable2 (RCA) with 16/16 ABX result, due to error voltage across the signal ground wire due to error loop current. I understand that you and several other members do not want to admit this fact, but you should be able to do so. It is easily explainable by basic laws of electrical engineering. You guys debate too much and do a little of real measuring and experiment job.
Cables are not any forbidden topic. They behave according to laws of physics, they may become an issue in certain system configurations and thus their exchange may bring audible difference. This is the fact and is easily proven.

See the post #61 for differences in error voltages with RCA cables and these differences between the worst and best cables re error voltage are audible even with music files at low level passages.

The zip I downloaded only contained 2 files, RCA and balanced.
I did not know there was a 3rd (2nd RCA) cable added later on.

Of course the electrical resistance of the screen (or ground wire) can make a difference as well as the ground loop current and even the PCB design can make a difference.
Screening itself can also be an issue. I have seen OEM interlink cables that did not even have a screen and just 2 or 3 thin wires.

Then again, which is my point, the sound does not change by changing cables. Only an unwanted signal is added which of course can create aliasing. The audibility of this depends on the music, listening SPL, and above mentioned points.

Even cables with poor contacts do not have a sound. They can introduce noise, distortion and whatnot but this is because it is (nearly) broken. When that is not the case and there is no extra signal introduced it will sound the same as any other cable.
A fault condition is not a 'sound'.

Of course, to the majority of users it will be a 'poor sounding' cable.

All 3 cables sound exactly the same when there is no unwanted signal added. Of course there can still be a difference between balanced and unbalanced connections because of differences in the method (different distortion profiles etc.)
Hardly the cable's fault.

 

[KSTR]

FWIW, I failed to find *any* measurable difference between a short balanced patch cable (TRS-to-TRS) and a very long unbalanced cable (6m of TS-to-TS), of course factoring out any noise/buzz/hum, just looking at the pure signal differences. There weren't any... and I've developed and used probably the most elaborate and highest sensitivity test procedure that we know of (way beyond of what even the best Audio Precision machine can measure directly).

I did of course try to compare many different cables, using even one XLR with solid silver conductors but to no avail. All cables behaved exactly as expected from basic theory.

Making very small distortions/errors audible with music signals. Some examples.

Hi, Over the course of the last weeks I managed to set up and stabilize a procedure that allows to expose the error residual of Null-Tests á la DeltaWave. Actually I'm using DeltaWave for the final stage of display and analysis but it does only level matching fine-tining here, the rest is done...

www.audiosciencereview.com

(note there is a download of the music snippet and the residual signals).

Together with the current test (thanks @pma) and previous similar ones my conclusion can only be that it is differences in noise susceptibility what makes cables "sound different" in real systems, sometimes. Cables don't have a sound on their own but they may change a system's sound due to a different noise profile.
This applies to all cables in a system, including mains cables, USB and SPDIF cables. Speaker cables as well, but those do also actually change the signal a bit.

 

[pma]

@KSTR - agreed, Klaus. At this point, it would be wise to close the thread.

 

[Doodski]

Together with the current test (thanks @pma) and previous similar ones my conclusion can only be that it is differences in noise susceptibility what makes cables "sound different" in real systems, sometimes. Cables don't have a sound on their own but they may change a system's sound due to a different noise profile.
This applies to all cables in a system, including mains cables, USB and SPDIF cables. Speaker cables as well, but those do also actually change the signal a bit.

I respect and admire your excellent commentary and the deep knowledge that you bring with your A-game whenever you come to ASR. For that reason this latest take on cables has tossed me on my ear and forces me to challenge everything I have ever thought correct about cables and digital fundamentals. Not sure how to do this. I need more information to digest this and then I might be able to adjust everything I have ever known to good effect.