1. Home
  2. tech-talk
  3. 197614-214267

What do you all make of this psuedo-science

https://www.tomshardware.com/speakers/in-a-blind-test-audiophiles-couldnt-tell-the-difference-between-audio-signals-sent-through-copper-wire-a-banana-or-wet-mud-the-mud-should-sound-perfectly-awful-but-it-doesnt-notes-the-experiment-creator

The Experiment

The goal was to see if listeners could distinguish between a high-quality original audio file and versions of that file that had been "re-recorded" after being passed through different conductive mediums:

  1. Original CD File (The control).

  2. 180cm of Professional Copper Wire.

  3. 13cm of Banana (connected via 120cm of copper).

  4. 20cm of Wet Mud (connected via 120cm of copper).

The Results

After a month of testing with 43 participants (many of whom consider themselves audiophiles with high-end equipment), the results were:

  • Identification Rate: Only about 14% (6 out of 43) of the guesses were correct.

  • Statistical Significance: This is actually lower than random guessing (which would be 25% for 4 options). Tom’s Hardware noted that there was a 6.12% chance of getting these results by accident, meaning the data is consistent with people simply guessing.

  • The "Mud" Surprise: The creator of the test noted, "The mud should sound perfectly awful, but it doesn't." Most listeners couldn't reliably tell it apart from the professional copper wire or the original digital file.

Why does "Mud Audio" work?

The physics behind this is simpler than it sounds:

  • Resistance: Both mud (due to water and salts) and bananas (due to potassium and moisture) are conductive. While they are worse conductors than copper, at the "line level" (the low-power signal between a player and an amp), they mostly just act as a resistor.

    +1

  • Signal Integrity: A resistor reduces the volume (gain) slightly but doesn't necessarily distort the frequency response or add noise in a way that the human ear can easily detect over short distances.

  • The Conclusion: The experiment reinforces a long-standing skeptical view in the audio world: while shielding and build quality matter for cables, the "exotic materials" used in expensive high-end cables often provide no audible benefit over basic conductors—even if that conductor is a piece of fruit or dirt from the backyard.

I realise the being the "best of the average" is not the same as the comparing the mud, banana, "hi-grade copper" may not be the same as a "high-end" interconnect or speaker cable. But I do wonder if there is a price point when we get negative return in regards to sound quality. When I do the AB testing at home or in a shop (blind AB testing's) I rarely hear a significant difference and usually not one worth an extra $1000+.  I know it is horses for courses with this hobby and appreciate that. 

jspnz
kofibaffour
718 posts
 

it is not pseudo science.  it is very valid and coming to this conclusion about cables frees your mind for actual useful changes that bring tangible and useful benefits to your system.

Your room is such an imposing element that any supposed cable benefit get drowned out by the room issues. so unless you fix the room, you ain’t hearing any supposed cable changes

I did the correct thing of approaching this from a physics standpoint and it saved me years of whack-a-mole

sold my cables, bought two Captivator RS1s and dug into DIY bass traps and ceiling clouds and all the issues the cables didn't solve, vanished

logistics
64 posts
 

All recordings are re-recorded.  And a master is just some engineers idea of what it should sound like or in partnership with the producer.  But if you decide that is what should sound best to you, as in the baseline, you've already failed yourself in audio.  We dont hear alike, and you should do whatever you want to make a recording sound better to you, even if that means EQ'ing, because your ears decide, not a guy from the past, sitting at a mixing board. 

richardbrand's avatar
richardbrand
1,395 posts
 

Richard Feynman, the great US physicist, described science as an acceptable philosophy of ignorance.  We have to be comfortable knowing what it is that we don't know.

What we do know of this experiment is that the starting point is digital data stored on a CD, being transferred to a digital file.  Most likely, the data transfer has end-to-end error detection and recovery.  CDs can have 4,000 consecutive bits wrong, yet the original data can be reconstructed perfectly, thanks to Cross-Interleaved Reed-Solomon coding.   In reality this means completely error-free - something relied on every time software is distributed on CD.

The experimenter has not told us precisely what input and outputs were used, or what protocols were used for the digital file transfer.  But it is reasonable to assume computer-grade error detection and recovery at every stage of the transfer.

A computer can be easily programmed to determine if there is any bit variance between two files.  It is reasonable to deduce that the listeners were exposed to bit-perfect copies, no matter whether bananas or wet mud formed part of the transmission chain.  No wonder they could not guess a difference.

It is worth mentioning that the input impedance is likely more that 10,000-Ohms which swamps the likely impedance of a wet banana!

Caveat

Now this result cannot possibly be extrapolated to wires carrying analogue signals, or to wires carrying digital signals which are not error corrected (I am thinking I2S here, but USB streaming and Ethernet without error correcting protocols are also guilty).

More to discover