But you told us that cables don't matter

[olivier salad]

So I've run the RMAA loopback test of my Asus STX card with 2 RCA to 3.5 mm cables and got quite different results. One cable is 1 m and the other is 5 m and obviously the long one measured worse. I didn't expect that much of a difference given the amount of jokes about cables that exist here.

 

[Blumlein 88]

A little more detail as to which is which in the graph would be nice.

I see what looks like one cable picking up a bit more noise. Nothing in the measurements look like they'll be audible at all. So effectively they will sound the same to you. Or cables don't matter.

 

[solderdude]

It looks like your 5m cable is picking up more noise. This is an unbalanced connection after all.
Add to that, besides the measurements being suspect, that the difference falls far below audible differences.

Of course 1m and 5 m will measure somewhat differently.

(cross-posting with Blumlein)

 

[olivier salad]

A little more detail as to which is which in the graph would be nice.

Noise level, Intermodulation distortion and Stereo crosstalk. Crosstalk is especially high. Do you think it will be still inaudible?

 

[solderdude]

Crosstalk measurement is incorrect. This cannot possibly be the way it shows. It shows L and R are shorted over the entire audible range.
You do realize RMAA isn't the greatest testtool...

 

[Beave]

^Exactly.

The first two are inaudibly different. The crosstalk measurement results don't add up - basically excellent channel separation for one cable and impossibly bad separation for the other cable.

 

[Frank Dernie]

Noise level, Intermodulation distortion and Stereo crosstalk. Crosstalk is especially high. Do you think it will be still inaudible?

In experiments to try to understand why LPs don't sound dire given their poor performance we tried different levels of introduced crosstalk.
Already at 30dB any better was not audible, which shows accurately set up pickup cartridges are pretty well good enough for crosstalk too. I was amazed.

 

[Blumlein 88]

Unbalanced connection near a pc. If you move the longer cable mostly away from the pc except for the ends noise will probably go down.

 

[olivier salad]

It shows L and R are shorted over the entire audible range.

This can't be RMAA's fault so there is something wrong with the cable.

 

[pma]

So I've run RMAA loopback test of my Asus STX card with 2 RCA to 3.5 mm cables and got quite different results. One cable is 1 m and the other is 5 m and obviously the long one measured worse. I didn't expect that much of a difference given the amount of jokes about cables that exist here.

Depending on cable length and shield impedance, there is always a difference in measured mains frequency related noise. 50Hz + multiples are inevitable in SE link, especially if there is a loop or two class I components. I have dozens of such measurements. The problem of most ASR contributors is that they are oversimplifying.

I could continue to infinity. Another fact is that the cables have very different immunity to EMC RFI pick-up.

However, it is much easier to say that "cables make no difference"

My advice is - take the best coax cable with lowest shield impedance. See the difference between the RG-59 and the cable that was supplied with the DVD player. All the measurements are made with one USB card as a loopback.

 

[thewas]

However, it is much easier to say that "cables make no difference"

Most people here don't say that but that "cables make no audible difference" which is still true when seeing that the noise components you measured are under -100 dB.

 

[olivier salad]

Most people here don't say that but that "cables make no audible difference" which is still true when seeing that the noise components you measured are under -100 dB.

But crosstalk is surely audible

 

[thewas]

But crosstalk is surely audible

Depends also on the level and which music listened, upmost music doesn't have total silence on one channel when there is music on the other.

 

[pma]

Most people here don't say that but that "cables make no audible difference" which is still true when seeing that the noise components you measured are under -100 dB.

Probably. The question might be with some classical music with silent passages about -50dBFS.

Another example, and now the one that is audible.

balanced but pin 1 connected to PCB board gnd

balanced and pin 1 connected to chassis

 

[Wombat]

Probably. The question might be with some classical music with silent passages about -50dBFS.

Another example, and now the one that is audible.

View attachment 101591
View attachment 101592


View attachment 101593
balanced but pin 1 connected to PCB board gnd

View attachment 101594
balanced and pin 1 connected to chassis

That is a connection configuration issue. Choose the correct cable(interconnect).

https://www.ranecommercial.com/legacy/library.html

 

[thewas]

Probably. The question might be with some classical music with silent passages about -50dBFS.

Another example, and now the one that is audible.

As so many Hifi people have their sources 10 meters away from their amps.
Of course there can be noise and hum problems with non optimal long distance non-symmetric connections, that's why you won't find such in the audio pro world where long cable distances are the standard, but I wouldn't call first as cable sound but flawed implementations.

 

[ZolaIII]

There are good & bad cable's nothing in between, no magical ones & cetera. Read what AWG specification is and how it scales whit length. However most important thing isn't cable it self but that it has a proper shielding. You can buy perfectly good cheap cables. For example I use Procab CLA 711 5m which I paid around 9€ which is perfectly fine all do wire is thin but shielding is perfectly OK.

 

[Rock Rabbit]

Many problems in the test setup to make any conclusions
Crosstalk at 0 dB flat on 5 mts?
At least don't make a pickup coil with the cable (non inductive coiling is easy)

 

[solderdude]

Probably. The question might be with some classical music with silent passages about -50dBFS.

You would be hearing hum (continuous) and some of its harmonics at best, it won't 'tighten bass' or give deeper or more extended lows, 'smoothen' or increase treble or clarity, provide better imaging etc. This is what is commonly associated with the 'better sounding cables'.
As soon as there is hum or continuous 'noises' one starts to fiddle with cables first... but to get rid of the continuous 'noise' not to improve SQ.

I have encountered quite a few 'generic' RCA cables that came with equipment that didn't even have a screen, just merely a wire for screen.
In many cases this did not lead to any induced hum. This is also dependent on the output resistance of the source b.t.w.
So sure there is quite a difference between interlinks when it comes to shielding, groundloops etc... but not in sound quality as claimed by cable believers.

This can't be RMAA's fault so there is something wrong with the cable.

The crosstalk here is typical measurement error or a short which is easily confirmed with a multimeter or by using the cable in a stereo setup.
I don't see RCA cables shorting the 2 channels happening though. That leaves measurement error.

 

[Lorenzo74]

So I've run the RMAA loopback test of my Asus STX card with 2 RCA to 3.5 mm cables and got quite different results. One cable is 1 m and the other is 5 m and obviously the long one measured worse. I didn't expect that much of a difference given the amount of jokes about cables that exist here.

Yes, they don't.

Pls. go for balanced XLR cables for signal if DAC and Amp are far away as all pro industry does.
Forget about snake oil & fairy dust is not there where you get Nirvana.
Treat you room instead, you'll be much more rewarded.
Best
L.