Jump to content
IGNORED

OK, I tried hydrogen audio. It didn't go too well.


Recommended Posts

I would consider myself objectivist, but instead of trying to disprove my own subjective observations I try to find ways to figure out objective methods that correlate with those subjective observations.

 

If I find that flowers smell better than shit, I don't immediately go and say that there cannot be a difference if both release same amount of molecules. And that the topic cannot be discussed because it has not been scientifically shown that flower molecules would be technically better than shit molecules.

 

If we would always stick to what has already been proven and is status quo, nothing would ever progress.

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
I would consider myself objectivist, but instead of trying to disprove my own subjective observations I try to find ways to figure out objective methods that correlate with those subjective observations.

 

If I find that flowers smell better than shit, I don't immediately go and say that there cannot be a difference if both release same amount of molecules.

 

To ask for 'objective' support in the form of results from double blind listening tests seems pretty sensible to me, certainly for extraordinary claims like bit-identical files sounding different from each other. Personally, I know from experience that I can't trust my hearing in a sighted setting when I have knowledge about what I'm listening to and/or what I should be hearing. I'd be more than a little surprised if that wasn't true for others as well.

 

Link to comment
To ask for 'objective' support in the form of results from double blind listening tests seems pretty sensible to me, certainly for extraordinary claims like bit-identical files sounding different from each other. Personally, I know from experience that I can't trust my hearing in a sighted setting when I have knowledge about what I'm listening to and what I should be hearing. I'd be more than a little surprised if that wasn't true for others as well.

 

6 separate Blind A/B/A 3 minute sessions were performed in this area by Chartered E.E. and technical journalist Martin Colloms, and results published in HiFi Critic Vol.6 No.1.

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment
I would consider myself objectivist, but instead of trying to disprove my own subjective observations I try to find ways to figure out objective methods that correlate with those subjective observations.

 

If I find that flowers smell better than shit, I don't immediately go and say that there cannot be a difference if both release same amount of molecules. And that the topic cannot be discussed because it has not been scientifically shown that flower molecules would be technically better than shit molecules.

 

If we would always stick to what has already been proven and is status quo, nothing would ever progress.

 

A fly will enjoy more the smell of shit than flower and bee the opposite,who is right?

 


Link to comment
To ask for 'objective' support in the form of results from double blind listening tests seems pretty sensible to me, certainly for extraordinary claims like bit-identical files sounding different from each other.

 

One problem is that HA is fairly religious about A/B/X testing (and rarely "double blind" - I've seen people using that phrase who evidently don't know what it means), to the exclusion of other objective testing protocols that are perfectly acceptable in both the "hard" and "soft" sciences. There's no scientific reason to limit what one considers evidence exclusively to two testing protocols (single or double blind A/B/X).

 

(This is totally leaving aside what Miska mentioned with regard to subjective experience making him curious about whether there might be something objective behind it. I see nothing wrong with that. Nor do I see any scientific reason to think that what we know now about audio reproduction and human hearing is all we will know, forever and ever amen, but now we're getting even further afield.)

 

Personally, I know from experience that I can't trust my hearing in a sighted setting when I have knowledge about what I'm listening to and/or what I should be hearing. I'd be more than a little surprised if that wasn't true for others as well.

 

Absolutely, I think we've all got to allow that our senses are fallible. That shouldn't absolutely foreclose our inquiring into what we believe we are hearing and why, it's just that we've always got to be ready for the possibility that "I was fooled" or "I imagined it" may be the correct answer.

One never knows, do one? - Fats Waller

The fairest thing we can experience is the mysterious. It is the fundamental emotion which stands at the cradle of true art and true science. - Einstein

Computer, Audirvana -> optical Ethernet to Fitlet3 -> Fibbr Alpha Optical USB -> iFi NEO iDSD DAC -> Apollon Audio 1ET400A Mini (Purifi based) -> Vandersteen 3A Signature.

Link to comment
Some time ago I came to the conclusion that any DAC that is audibly affected by upstream or external perturbations is defective. All the evidence I have seen, such as that which you just gave, supports this. I have a choice between buying a DAC which is unaffected, or trying to fix all the external influences. I enjoy tinkering and tweaking as much as anyone, but my primary goal is to listen to music.

Well said.

Jim Hillegass / JRiver Media Center / jriver.com

Link to comment
Well said.

 

Eagerly awaiting the list of non-defective DACs.

One never knows, do one? - Fats Waller

The fairest thing we can experience is the mysterious. It is the fundamental emotion which stands at the cradle of true art and true science. - Einstein

Computer, Audirvana -> optical Ethernet to Fitlet3 -> Fibbr Alpha Optical USB -> iFi NEO iDSD DAC -> Apollon Audio 1ET400A Mini (Purifi based) -> Vandersteen 3A Signature.

Link to comment
To ask for 'objective' support in the form of results from double blind listening tests seems pretty sensible to me, certainly for extraordinary claims like bit-identical files sounding different from each other.

 

It is not efficient way, I rather use MOS/MUSHRA/PEAQ -type methods (yeah, I have strong background in telecomms).

 

There are multiple good documents about how to properly conduct listening tests if you want to go that way, but these are good starting point:

https://tech.ebu.ch/docs/tech/tech3286.pdf

https://tech.ebu.ch/docs/techreview/trev_274-hoeg.pdf

 

But without much effort I can think of number of reasons why bit-identical files could sound different. It doesn't even need much effort to measure that same file played twice through same hardware results in different analog output every time. There are multiple time-varying interference and noise factors in analog and mixed-signal environments always, and sample clocks belong to that category too. Since we don't feed the bits directly to our brains in digital format, keeping inspection purely in digital domain is completely useless.

 

I'm mostly interested about correlation between these factors and listening experiences, instead of questioning existence of those factors.

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

OK, but in Bystander's case we are talking about "extraordinary claims like bit-identical files sounding different from each other." Extraordinary claims require extraordinary evidence (to paraphrase Carl Sagan).

 

Normally, if you wanted to check to see if two files were bit-identical, you would check the md5sum of each, and if they matched, you would be done. No need for a double-blind test or anything else. (The clinically paranoid might opt for a sha1 hash of each as well, but it is basically the same idea.)

 

Now we have the claim that two bit-identical files, with all else being equal, can sound different. Personally, I think this suggestion is implausible to the point of being insane, but if we want to test the claim, how can we do it? Implicitly, we are rejecting things like checksum hashes, so all we are left with is people's impressions of how they sound. A properly-conducted double-blind test is the only way you can evaluate such a claim.

Link to comment
OK, but in Bystander's case we are talking about "extraordinary claims like bit-identical files sounding different from each other." Extraordinary claims require extraordinary evidence (to paraphrase Carl Sagan).

 

Normally, if you wanted to check to see if two files were bit-identical, you would check the md5sum of each, and if they matched, you would be done. No need for a double-blind test or anything else. (The clinically paranoid might opt for a sha1 hash of each as well, but it is basically the same idea.)

 

This is going a bit OT from the players sound different -subject.

 

Problem with the hash test method in this case is that it tests only the files, not the entire playback. So it tells nothing about playback.

 

Now we have the claim that two bit-identical files, with all else being equal, can sound different. Personally, I think this suggestion is implausible to the point of being insane, but if we want to test the claim, how can we do it? Implicitly, we are rejecting things like checksum hashes, so all we are left with is people's impressions of how they sound. A properly-conducted double-blind test is the only way you can evaluate such a claim.

 

As a starting point you'll at least need to measure the playback from analog domain. For example playback from RAM disk vs HDD vs SSD may have vastly different EMI/RFI signature. Even two files from the same HDD may result in different signature if the other one is more fragmented than the other, and the reading head servos are quite powerful.

 

No, I'm not saying I'd be hearing differences with my hardware and software, but that probably doesn't mean anything to anybody else.

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
This is going a bit OT from the players sound different -subject.

 

Problem with the hash test method in this case is that it tests only the files, not the entire playback. So it tells nothing about playback. ...

 

As others have said, "identical" files can produce different analogue outputs for all sorts of reasons. My original point was that if the (sum of the) differences are large enough to be audible, the DAC can be considered to be defective / poorly designed and implemented.

"People hear what they see." - Doris Day

The forum would be a much better place if everyone were less convinced of how right they were.

Link to comment
Feel free to suggest one. I didn't even suggest or imply that I'd provide a list.

 

With all due respect

 

 

Some time ago I came to the conclusion that any DAC that is audibly affected by upstream or external perturbations is defective. All the evidence I have seen, such as that which you just gave, supports this. I have a choice between buying a DAC which is unaffected, or trying to fix all the external influences. I enjoy tinkering and tweaking as much as anyone, but my primary goal is to listen to music.

 

Strongly implies that you know of one or more such DACs in which case it's disingenuous not to identify them.

 

A

Link to comment
As others have said, "identical" files can produce different analogue outputs for all sorts of reasons. My original point was that if the (sum of the) differences are large enough to be audible, the DAC can be considered to be defective / poorly designed and implemented.

 

Now you can turn this from subjective to objective by obtaining bunch of DACs and measuring their output in different configurations. I've been doing that, while my primary focus is still on inspecting effectiveness of my DSP algorithms in real world situations.

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
... I'm mostly interested about correlation between these factors and listening experiences, instead of questioning existence of those factors.

 

Yes. Hear a difference, find out why. Fix it. For example, bit identical files might sound different. If the system were perfect, you know they wouldn't. Where we differ is in the fix. For example, a DAC might be susceptible to jitter in the incoming bitstream. Some might work to fix the causes of the jitter. Others might work to fix the DAC's susceptibility to jitter. I happen to be in the latter group.

"People hear what they see." - Doris Day

The forum would be a much better place if everyone were less convinced of how right they were.

Link to comment
... Strongly implies that you know of one or more such DACs in which case it's disingenuous not to identify them. ...

 

I already said, I did not intend to even imply such a thing. I do have a personal list, but it's my opinion and would likely be vociferously debated if I did post it. We're far enough OT already.

"People hear what they see." - Doris Day

The forum would be a much better place if everyone were less convinced of how right they were.

Link to comment
I already said, I did not intend to even imply such a thing. I do have a personal list, but it's my opinion and would likely be vociferously debated if I did post it. We're far enough OT already.

Don

Feel free to start a new thread. (grin)

 

4ktLH2.jpg

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment

Miska

Please check your PMs

Alex

 

How a Digital Audio file sounds, or a Digital Video file looks, is governed to a large extent by the Power Supply area. All that Identical Checksums gives is the possibility of REGENERATING the file to close to that of the original file.

PROFILE UPDATED 13-11-2020

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...