Is cable sound real? A more holistic approach trying to track it down.

[KSTR]

Phase shifts only or slightly different propagation delay. So the samples are in ns/10ns shifted.

Exactly. Plus an even more microscopic amplitude drop at HF for the higher capacitance cable.
Similar to this (not a measurement but a synthesized test case) : https://www.audiosciencereview.com/...-comparison-software.6633/page-23#post-649391

 

[UliBru]

Yeah, that averaged spectrum looks suspect in the lower frequencies.

Yes, I've found the reason. The test tone is flawed. I'll further check.

 

[pma]

Contrary to popular belief, music signal is the least revealing of small differences in signal distortion and frequency response irregularities. For distortion, most revealing are pure sine tones and for changes in frequency response it is the noise signal, white noise. I made a test with white noise here

https://www.audiosciencereview.com/...mplifier-abx-listening-test.19352/post-652046

and it should be quite easily discernable in the ABX, though almost impossible with music.

 

[UliBru]

My multitone test signal has been long but a single shot only. Whereas other similar signals are made of repeated shorter blocks.
So now I have used my signal including repetitions.
I can now get such a picture with two different cables with different construction and length:

The top amplitudes show some small difference with a little more damping at high frequencies for the long cable = RLC parameters (for visualization the colors are switched):

So it seems possible to measure cable differences with quite different cables. For cables of same construction and length but e.g. different conductors (copper, coppy 5N, silver ...) I expect much less differences.

 

[pma]

My multitone test signal has been long but a single shot only. Whereas other similar signals are made of repeated shorter blocks.
So now I have used my signal including repetitions.
I can now get such a picture with two different cables with different construction and length:


The top amplitudes show some small difference with a little more damping at high frequencies for the long cable = RLC parameters (for visualization the colors are switched):


So it seems possible to measure cable differences with quite different cables. For cables of same construction and length but e.g. different conductors (copper, coppy 5N, silver ...) I expect much less differences.

To me, the result of your test is suspicious. I get the same plots with 1m and 10m cable lengths.

Someone should independently repeat your test with the fully described test conditions.

 

[KSTR]

@UliBru , I think you are just measuring ill effects of you Fireface UC. The 10dB noise floor change definitely is coming from there, not the cables.
The only thing that looks real is the increased hum and idle tone pickup with the unbalanced cable (bad shielding quality).

 

[pma]

I can see absolutely no reason for rising of the noise floor with the longer cable. The only difference I would admit would be that based on ground loop induction, that would reflect in changed mains spuriae.

Another multitones of 1m x 10m cables:

 

[KSTR]

And more from me:

Some Mains hum but no other spuriae.

 

[UliBru]

If so, I stand corrected.

Ok, now some enlightenment: The measurements have been carried out with symmetric and asymmetric cables. So I picked up first a short symmetric connection and then quickly an asymmetric cable flying around.
Now the Fireface has symmetric inputs and outputs which require jack plugs. The symmetric connections require stereo jack plugs and the asymmetric connection uses mono jack plugs.
So the noise difference in my previous comparison is caused by connecting the cold pin to the ground pin.

Here is a comparison of a short and long symmetric cable:

And here is a short and long asymmetric cable:

The noise floor for the asymmetric connection is about 10 dB higher. But the difference between short and long cable is much less and possibly neglegible.

And the world is back in order

 

[KSTR]

 

[pma]

This is the advantage of ASR, there are some qualified participants here that can give a feedback if something looks improbable. The reason why is found then. Unfortunately this is not the case of some popular magazines or forums based on subjective opinions only.

 

[Pio2001]

The top amplitudes show some small difference with a little more damping at high frequencies for the long cable = RLC parameters (for visualization the colors are switched):
View attachment 108973

So it seems possible to measure cable differences with quite different cables. For cables of same construction and length but e.g. different conductors (copper, coppy 5N, silver ...) I expect much less differences.

At 1000 Hz we can see a 0.0155 dB difference between both cables.
At 20 Hz, the difference looks more like 0.016 dB
At 20 kHz, it looks like 0.017 dB

Which means that the long asymmetric cable has a frequency response of -0.005 dB @ 20 Hz and -0.015 dB @ 20 kHz.
I got the same results measuring the addition of a 5 meters standard asymmetric extension to a standard asymmetric interconnect : 0.01 dB of loss at 20 kHz.

 

[ahofer]

Cable believers will argue that an ADC loop will not record the differences (for a bunch of made-up reasons) and invalidate the test for them.

+1

 

[Gregm]

Full ack. We must distinguish between the personal perceptual reality which, by definition, is real no matter what caused it, and hearing facts, which is what is left when any biasing is removed to best effort. Once we have established some hearing facts (within the limits of statistical significance), it starts to make sense to look hard for the reasons for it, which should be real differences in the signal as other causes have hopefully been ruled out.

I'm with you on this.

 

[MC_RME]

The noise floor for the asymmetric connection is about 10 dB higher.

Let me guess: you used the rear outputs into the front Mic/Line inputs 1/2. No good idea when going unbalanced. Use front 3/4 or rear inputs, then there will be no 10 dB noise difference between balanced and unbalanced cabling.

And the world is back in order

Not yet.

 

[UliBru]

Let me guess: you used the rear outputs into the front Mic/Line inputs 1/2. No good idea when going unbalanced. Use front 3/4 or rear inputs, then there will be no 10 dB noise difference between balanced and unbalanced cabling.

No need to guess, I have told about in https://www.audiosciencereview.com/...ach-trying-to-track-it-down.19807/post-652325. The underlying idea: use same IO when switching cables.
And yes, you are right. Thanks for the hint. I have now tested input 3, the noise is 10 dB down.
A question: is that the reason why the manual (which I have studied now again based on your comment) does not list noise parameters for input 1+2, only for 3-8?

So I have learnt again. It never stops.

 

[MC_RME]

The underlying idea: use same IO when switching cables. And yes, you are right. Thanks for the hint. I have now tested input 3, the noise is 10 dB down. A question: is that the reason why the manual (which I have studied now again based on your comment) does not list noise parameters for input 1+2, only for 3-8?

Use the same I/O is the correct approach, but in this case these are not the 'same' I/Os. One is a servo-balanced output that handles TS and TRS jacks perfectly, the other is a balanced input with not perfect performance if used as unbalanced Line input via TS. The servo-balanced inputs 3 to 8 are the correct 'same'.

The noise values of 3-8 are also valid for 1/2 (as stated in the Tech Specs). Unbalanced operation of input 1/2 is not specified, though.

 

[ESRA]

Hi

I did comparison with foobar , ABX, in learning mode:
1) Original versus Solidcore
2) Original versus Thomann
3) Solidcore versus Thomann

I could not note any difference in all 3 cases, so I did not do the real ABX test as the result would be pure guess, 50% 50%.

 

[Cbdb2]

It would be a good idea to include a ‘sanity check’. Obtain at least one cable that’s objectively, and audibly, inferior: highly inductive, or something like that. You could either fabricate this yourself or stare at one of Audioquest’s more expensive models for a couple of minutes so it breaks.

This will mean that the test contains at least one presentation that should elicit a positive discrimination and provide a baseline.

Good luck. Over at Diyaudio, Pano did exactly this. He included the original file ( no loop thru) for reference, a good quality copper cable ( Mogami), a potatoe and a banana, and the last file thru mud. These were blind tested and no one could reliably tell the difference.

https://www.diyaudio.com/forums/everything-else/236248-fancy-interconnects-potato-mud-36.html

Results are halfway down the page.

 

[Cbdb2]

An the conclusion from Sy.

Originally Posted by SY
"I'm sure you can get a difference in DiffMaker (the flac files should convert easily), the more interesting question is whether or not you can actually hear a difference. After all, if there's night and day audible differences between copper and silver or PVC and Teflon, you'd think that banana/potato vs copper or mud ought to be blatantly evident."