Do you remember the great fun we had 11 years ago when I ran audio signals through fruits and vegetables, beer, coke, steel wool and other stuff? No? Not to worry, here is the banana mud redux!
The idea of this blind test is that the same audio signals were run through copper wire, then through mud and then through bananas and recorded them. You get to listen to the files and figure out what differences you hear. It's good fun for all and you might be surprised by how each different material sounds. Have a listen, make some notes and let me know what you hear. Do Not post your results here (because spoilers) just send me a direct message with your findings.
In this these Google folder you will find the test files.
FLAC format. Lossless and faster downloads.
https://drive.google.com/drive/folders/1hxRL2L_M5O0fD8Ll9Y2dphqQgaeCKW_I?usp=sharing
And in WAV format if you can't deal with FLAC.
https://drive.google.com/drive/folders/1fe8-0kYWexOEveq5DPJ1zqPFP3RQZQ4S?usp=sharing
Pick a song or three that you like, and compare the versions. In those folders you will find many tracks arranged by musical genre. All tracks are 30 secs long. You should easily find something with which you can judge sound qualities. There is rock, jazz, classical and others - pick what you like and have a listen.
Remember, DO NOT post your results here, you don't want to bias the other listeners. Send me a private message. I will respond to your message and answer questions.
In the next post I will explain more about the test.
The idea of this blind test is that the same audio signals were run through copper wire, then through mud and then through bananas and recorded them. You get to listen to the files and figure out what differences you hear. It's good fun for all and you might be surprised by how each different material sounds. Have a listen, make some notes and let me know what you hear. Do Not post your results here (because spoilers) just send me a direct message with your findings.
In this these Google folder you will find the test files.
FLAC format. Lossless and faster downloads.
https://drive.google.com/drive/folders/1hxRL2L_M5O0fD8Ll9Y2dphqQgaeCKW_I?usp=sharing
And in WAV format if you can't deal with FLAC.
https://drive.google.com/drive/folders/1fe8-0kYWexOEveq5DPJ1zqPFP3RQZQ4S?usp=sharing
Pick a song or three that you like, and compare the versions. In those folders you will find many tracks arranged by musical genre. All tracks are 30 secs long. You should easily find something with which you can judge sound qualities. There is rock, jazz, classical and others - pick what you like and have a listen.
Remember, DO NOT post your results here, you don't want to bias the other listeners. Send me a private message. I will respond to your message and answer questions.
In the next post I will explain more about the test.
How the test works:
Each 30 second musical excerpt has four versions.
You have the original file for comparison, but you will not know which is the original. All files have a 4 digit code at the end of file name that allows me to identify them, you will not know which is which. You will know that each clip has four versions, as listed above.
While you might not be able to tell a banana from mud, you should at least be able to tell those two conductors from copper wire or from the original file.
Nothing fancy was used, just ordinary consumer products. This is within reach of any audio hobbyist.
Test probes are crude, made from old microphone cable and a US penny soldered on as the test probe. They worked well!
The green wire is to tie the cable shield to the aluminum foil. Without the foil, noise is too high for a good test. Putting the mud in plastic trays wrapped in foil worked to keep the noise floor very low. It worked well for the bananas, too.
Each 30 second musical excerpt has four versions.
- Original File taken from CD
- Loop recording via 180cm of pro audio copper wire
- Loop recording via 20cm of wet mud (and 120cm of copper wire)
- Loop recording via 13cm of banana (and 120cm of copper wire)
- Pick a song or a few songs that you like and feel that you can easily judge.
- Listen to all four versions of that recording and determine which is which.
- Report to me via private message what you hear. DO NOT post your results here.
- I will reveal the key to which is which at a later date.
You have the original file for comparison, but you will not know which is the original. All files have a 4 digit code at the end of file name that allows me to identify them, you will not know which is which. You will know that each clip has four versions, as listed above.
While you might not be able to tell a banana from mud, you should at least be able to tell those two conductors from copper wire or from the original file.
Nothing fancy was used, just ordinary consumer products. This is within reach of any audio hobbyist.
- Dell Laptop - Windows 10
- M-Audio Fast Track Pro USB audio interface for playback and record.
- Goldwave 7.0 for file trimming and level correction of recorded files.
- Reaper DAW for simultaneous playback and record of files to be looped.
- REW RTA function for finding optimal input and output levels of the M-Audio interface.
Test probes are crude, made from old microphone cable and a US penny soldered on as the test probe. They worked well!
The green wire is to tie the cable shield to the aluminum foil. Without the foil, noise is too high for a good test. Putting the mud in plastic trays wrapped in foil worked to keep the noise floor very low. It worked well for the bananas, too.
Please DO NOT post your results in this thread. Send them to me via direct message.
Feel free to post general comments, such as "They all sound so different!" or "I hear differences, but can't tell what they are" or "I don't hear anything at all." Just don't tell us in public which is which. Send that to me via message.
This test is easy, you can go as fast or slow as you like, you can use whatever method you prefer and you can use any audio system that you can get your hands on, it doesn't matter. Just listen, take notes and guess which version of the song is which.
Software jockeys; remember that this is a LISTENING test. The idea is to find differences by ear. If you use software to help you guess, that's OK, just let us know what you did. Maybe hold back on details in public, until the test is done. Have fun!
Feel free to post general comments, such as "They all sound so different!" or "I hear differences, but can't tell what they are" or "I don't hear anything at all." Just don't tell us in public which is which. Send that to me via message.
This test is easy, you can go as fast or slow as you like, you can use whatever method you prefer and you can use any audio system that you can get your hands on, it doesn't matter. Just listen, take notes and guess which version of the song is which.
Software jockeys; remember that this is a LISTENING test. The idea is to find differences by ear. If you use software to help you guess, that's OK, just let us know what you did. Maybe hold back on details in public, until the test is done. Have fun!
Piano solo 8B9D is only 3 seconds long and ECC5 is 45 seconds with 12 seconds of lead in silence.
That makes sense, but post them somewhere I hope? I'm going to try this but no longer have wired headphones. I'll see what happens FLAC --> Apple Lossless --> Apple Music --> AirPods Pro 2. My prediction? One excerpt will sound muddy, one will sound mushy, one will sound very metallic...
Thanks. I did the piano solo separately and seemed to have made a mess of it. I will fix that and upload the correct ones.
Once again the unwritten rule seems to be confirmed that in this forum we are asked to measure even that which has no reason to be measured. 😉🙂
Do you mean we really should establish the type of soil and the degree of humidity of the mud before proceeding to listen to this intriguing test?
Then someone will come and ask us how ripe the banana has to be... 🙄
If you give me time I'll get there too and listen. 😉
You were not asked to measure. I said that if you DO use software to analyze the files, please let us know.
No, that was not stated. You are simply asked to listen to the files and report what you hear. The mud is saturated, Central American and Volcanic.
This should be obvious from the photos. 😉
Very logical, Mr. Spock. Don't forget also that one will sound brilliantly original. 👍
Pano, I simply answered those who asked about the type of terrain and how wet the mud should be, just to prevent other questions of the kind that imho have no real reason. 😉
Indeed it is and it is very clear 🙂, at least for those who interpret the test as it should...
To be fair, I have used both ripe and green bananas in the test and could tell no difference. The green bananas measure at 5.1K ohms at DC. They probably have some stray capacitance.
The sheer imaging, openness, and inky-black, almost tangible space between instruments has to be heard to be believed with a matched pair of organic Ghanese plantain...
Since the ground is common to the inputs and outputs which are in the same device, just a few inches apart, no “return” is needed, it’s built in.
I did connect the ground to the aluminum foil for shielding, but it carried no signal current.