How To: Process High Resolution Music Files

[upstateguy]

 

Aye, but you are claiming there is no difference.

se

 
The burden of proof is on those who claim to hear a difference. Those who do not hear a difference don't have to prove anything.

 
You're skirting the issue "MA". 
 
Your claim was not that you didn't hear a difference. 
 
If you didn't hear a difference and said "I don't hear a difference between 24 bit and 16 bit"  it would be accepted as just your subjective observation.
 
But that wasn't the case, was it?
 
Your claim, to refresh your memory, was "Take a 24-bit file. Convert it to lossless 16-bit / 44.1 kHz. Sounds exactly the same. Always."
 
Then you go on to say, "Those who do not hear a difference don't have to prove anything"  and that would be true until they post it as a claim, like you did.
 
It's funny that all of a sudden, the burden of proof falls elsewhere, when you have to prove what you're claiming.
 
RE: magical dragon analogy
 
My analogy is just as ridiculous:  "a 45rmp record "sounds exactly the same" as a compact disc. Always."
 
The willingness to accept either statement without proof is the hallmark of a Subjectivist.

 

[Music Alchemist]

Unless my recollection is bad, I believe you said no one would hear a difference. Not just that you didn't hear a difference. And sure, if you say you personally didn't hear a difference, that's fine and there's really no onus on you to prove that claim unless someone want to call you a liar. But if you said no one would be able to hear a difference, that's another matter entirely.

se

  You're skirting the issue "MA". 
 
Your claim was not that you didn't hear a difference. 
 
If you didn't hear a difference and said "I don't hear a difference between 24 bit and 16 bit"  it would be accepted as just your subjective observation.
 
But that wasn't the case, was it?
 
Your claim, to refresh your memory, was "Take a 24-bit file. Convert it to lossless 16-bit / 44.1 kHz. Sounds exactly the same. Always."
 
Then you go on to say, "Those who do not hear a difference don't have to prove anything"  and that would be true until they post it as a claim, like you did.
 
It's funny that all of a sudden, the burden of proof falls elsewhere, when you have to prove what you're claiming.
 
RE: magical dragon analogy
 
My analogy is just as ridiculous:  "a 45rmp record "sounds exactly the same" as a compact disc. Always."
 
The willingness to accept either statement without proof is the hallmark of a Subjectivist.

 
It has already been proven by science. There cannot possibly be a difference unless the system used does not properly play the files.
 
https://xiph.org/~xiphmont/demo/neil-young.html

 

[Steve Eddy]

But that wasn't the case, was it?

Your claim, to refresh your memory, was [COLOR=800080]"Take a 24-bit file. Convert it to lossless 16-bit / 44.1 kHz. Sounds exactly the same. Always."[/COLOR]

Then you go on to say,"Those who do not hear a difference don't have to prove anything"  and that would be true until they post it as a claim, like you did.

It's funny that all of a sudden, the burden of proof falls elsewhere, when you have to prove what you're claiming.

Sorry, MA, but I'm afraid my hair stylist has you on this one. You have indeed put yourself in the position of having to prove a negative.

se

 

[Music Alchemist]

Sorry, MA, but I'm afraid my hair stylist has you on this one. You have indeed put yourself in the position of having to prove a negative.

se

 
I don't need to prove something that has already been proven. I already explained why hi-res has no benefit and linked to an article going into detail about it.

 

[Steve Eddy]

It has already been proven by science. There cannot possibly be a difference unless the system used does not properly play the files.

I'm sorry, but that is evidence, not proof.

se

 

[Music Alchemist]

I'm sorry, but that is evidence, not proof.

se

 
What would satisfy your requirements for proof?
 
Humans cannot hear above roughly 20 kHz, so frequencies above that are inaudible. 16-bit can handle all the dynamic range in any recording, so the added dynamic range of 24-bit does nothing for playback. It is physically impossible for there to be an audible difference, unless there is a problem in the system used.
 
There is already no room for doubt, so I don't see how it's not proof.

 

[nick_charles]

Sorry, MA, but I'm afraid my hair stylist has you on this one. You have indeed put yourself in the position of having to prove a negative.

se

 
 

I'm sorry, but that is evidence, not proof.

se

 
 
As it happens there have been several tests of downsampling/resampling and/or bit depth changes on the interweb. To the best of my knowledge at least 3 different groups of people including one AES publication (McGill University Masters students) and one HydrogenAudio contributor and a chap called Amir on the "What'sBest" forum Linky
 
have posted evidence that contradicts the general tenor of MA's claim that going from anything above 16/44.1 to 16/44.1 is undetectable. As is almost always the case all these claims have been challenged to some extent but they are all based on double-blind listening tests so they constitute prima facie evidence at least. While the stats on the McGill study are possibly dodgy the other two at least bear up under initial scrutiny. So MA really does have to do some work to support his claim or gracefully back down.
 
This does not mean that I personally claim to hear any differences between the uber formats and ye olde red booke but I'm not making that claim...

 

[Steve Eddy]

What would satisfy your requirements for proof?

Humans cannot hear above roughly 20 kHz, so frequencies above that are inaudible. 16-bit can handle all the dynamic range in any recording, so the added dynamic range of 24-bit does nothing for playback. It is physically impossible for there to be an audible difference, unless there is a problem in the system used.

There is already no room for doubt, so I don't see how it's not proof.

There are actually very few real proofs in science.

But I think I'm just going to exit this discussion.

se

 

[Music Alchemist]

  As it happens there have been several tests of downsampling/resampling and/or bit depth changes on the interweb. To the best of my knowledge at least 3 different groups of people including one AES publication (McGill University Masters students) and one HydrogenAudio contributor and a chap called Amir on the "What'sBest" forum have posted evidence that contradicts the general tenor of MA's claim that going from anything above 16/44.1 to 16/44.1 is undetectable. As is almost always the case all these claims have been challenged to some extent but they are all based on double-blind listening tests so they constitute prima facie evidence at least. While the stats on the McGill study are possibly dodgy the other two at least bear up under initial scrutiny. So MA really does have to do some work to support his claim or gracefully back down.
 
This does not mean that I personally claim to hear any differences between the uber formats and ye olde red booke but I'm not making that claim...

 
Would you be able to link me to this evidence? (Via PM if it's a link that is not allowed here.)

 

[nick_charles]

   
Would you be able to link me to this evidence? (Via PM if it's a link that is not allowed here.)

 
 
The Amir forum thread is now linked in my post as linky 
 
The McGill study is discussed here linky2
 
The other is a listening test also at HA linky3

 

[Music Alchemist]

  The Amir forum thread is now linked in my post as linky 
 
The McGill study is discussed here linky2
 
The other is a listening test also at HA linky3

 
Thanks for these. I honestly think it would be interesting if I could tell them apart, but in all the tests I've done, they sounded exactly the same.

 

[Steve Eddy]

As it happens there have been several tests of downsampling/resampling and/or bit depth changes on the interweb. To the best of my knowledge at least 3 different groups of people including one AES publication (McGill University Masters students) and one HydrogenAudio contributor and a chap called Amir on the "What'sBest" forum Linky

Don't know if it's the same person, but a friend was telling me some time back about a person at HA was known to cheat.

have posted evidence that contradicts the general tenor of MA's claim that going from anything above 16/44.1 to 16/44.1 is undetectable. As is almost always the case all these claims have been challenged to some extent but they are all based on double-blind listening tests so they constitute prima facie evidence at least. While the stats on the McGill study are possibly dodgy the other two at least bear up under initial scrutiny. So MA really does have to do some work to support his claim or gracefully back down.

That's why I say it's better to simply say no one has yet demonstrated it to be true instead of absolutes. Many a scientist has been tripped up by that.

se

 

[RRod]

Can anyone who can access the McGill paper confirm the findings as laid out here (post 9)?
http://www.sa-cd.net/showthread/58757/58908

 

[Music Alchemist]

  Can anyone who can access the McGill paper confirm the findings as laid out here (post 9)?
http://www.sa-cd.net/showthread/58757/58908

 
I won't pay for it, but anyone who is interested can try reproducing a similar experiment using the standard procedures.

 

[RRod]

   
I won't pay for it, but anyone who is interested can try reproducing a similar experiment using the standard procedures.

 
The ones who need to do it won't do it. It's really best for all this to be hashed out in the literature, because at the end of the day none of us internet schmucks really trust any of the other internet schmucks