Win 7 Optimisations

[chill]

Cognitive process passes through a kind of mental representation that may need some more time.

Except that differences due to improved ethernet cables and server priority tweaks are always 'instantly' apparent.

[superbonham]

No difference. But claiming these statements are false without evidence of it is no genuine wisdom.

For unicorn, I believe like you probably that there are enough scientific evidence they don't exist so we are not in any of above categories.

So what makes you so confident in this case and so undecided in the "cable and server tweak voodoo" case? What kind of "scientific evidence that there are no unicorns" are you referring to? I am not aware of any, still I don't believe in unicorns.

It's not "our" task to prove that there is no such cable/tweak voodoo, because common sense at least suggests it does not exist. This thread still contains some counter arguments as well as a simple test set up ("pull the damn cable") for further assessment.

Everyone claiming "facts" that contradict common sense is - in contrary - deemed to provide verifiable evidence for his claim. The more outrageous and quixotic the claim is, the more supporting evidence it needs.

If not, we just end up with the simple

... there are unicorns, because I've seen one ...

aka

... there is cable voodoo because I can hear sound improvement ...

Cheers

superbonham

[vett93]

Except that differences due to improved ethernet cables and server priority tweaks are always 'instantly' apparent.

I fully agree with the server priority tweak. The improvement is instantly apparent. The improvement is that the menu display is much more responsive to the remote control.

[darrell]

What is it with some people who think they can second-guess qualified engineers who obviously know what they are talking about (and often provide references to reputable sources to prove it)? It's a disease, this "everyone's opinion is equally valid" mentality. No, Some people know what they are talking about, others don't. The post-modern hell which thinks otherwise is unfortunately made much worse by the internet, which allows any crackpot to reinforce their prejudices before giving them a platform to broadcast their "ideas" to the world.

The problem is that the more of this stuff that is posted, the more reinforcement there is for the next wave of crackpots to feed on, which is why it is necessary for those with real knowledge to challenge this stuff whenever it appears. Just think of the kind of use links to this thread would be put to if the crazy posts were left unchallenged!

I should say that although I am scientifically trained, and work in IT, the kind of stuff relevant to this discussion is not my area, so I trust those whose area it is. And before someone says that this is a recipe for being stuck for ever at current levels of understanding, remember it is the scientists, researchers and engineers in the field who will make the progress which leads to refined and new theories in the future. That is how science works - there are no dogmas, no tablets of stone. The more complete understanding of phenomena in the future will be built on top of the current state of the art, not on foundations of subjectivism and ignorance.

Let's save subjective discussion for questions of music and performance, and give proper respect to the scientists and engineers whose discoveries and work allows us to experience frighteningly high quality sound reproduction in our own homes, sometimes separated by many decades from the original performances.

[Phil Leigh]

The fact that the buffer is too short to judge might be precisely the consequence that we have bad aural memory.

Cognitive process passes through a kind of mental representation that may need some more time.

No we are very good at detecting immediate changes and processing them... All human senses are optimised for delta detection! - thats what has kep the species alive :-)

It would be impossible to mix an album otherwise! What DOES happen is that you can spend many hours making minor cumulative changes and lose all sense of macro perspective as you get fatigued, so now you think it sounds great but actually it doesn't... And this only becomes apparent when you listen to playback the next day.

This tells me that the longer we listen to something, the less perspective we retain. essentially the brain gets tired and bored and actually starts to fill in the bits it remembers from memory and stops listening to what it really sound like... And we have no idea when this is happening and when it isn't...
This probably explains why certain types of blind listening test are also fatally flawed...

So I would argue that the longer we listen to something being repeated, the less we actually hear and any judgements we make are flawed. On the others hand, quick a/b tests are ruthlessly revealing.

[adamdea]

You don't get the point here: I said people who claim they don't know if unicorn exist are wiser than people who claim they don't exist without any scientific evidence of it.

For unicorn, I believe like you probably that there are enough scientific evidence they don't exist so we are not in any of above categories.

But I was not talking about unicorn here and You did not answer the questions I raised.

You started this off by saying you knew who was wiser. Now tell me
a. who was saying they don't know if server tweaks affect the sound of a dacf connected to a squeezebox connected via ethernet to the server?
I have heard some people asserting not that it might make a difference but that it DOES What do your questions have to do with the rationlity of that position?

b. what conclusion should be drawn by a person who says that they don't know if server tweaks.....etc
the bottom line is that there are many categories of doubt and evidence between conclusive proof and conclusive disproof. SO the bottom line is this

c. what GROUNDS are there for suspecting that .......etc. The simple answer is that there are no grounds for belief in these tweaks.

[evdplancke]

You started this off by saying you knew who was wiser.

I did not say that. I said I have my opinion about it. Opinion are not facts. Therefore I let anyone make its own opinion.

Now tell me
a. who was saying they don't know if server tweaks affect the sound of a dacf connected to a squeezebox connected via ethernet to the server?
I have heard some people asserting not that it might make a difference but that it DOES What do your questions have to do with the rationlity of that position?

What is the rationality of a perception? that's the question! I believe they have perceived a difference. I don't know if this difference can be objectivated or not but I respect the fact that they are doing experiments, because I suspect a lot of those who pretend to know the truth didn't even experiment anything.

b. what conclusion should be drawn by a person who says that they don't know if server tweaks.....etc
the bottom line is that there are many categories of doubt and evidence between conclusive proof and conclusive disproof. SO the bottom line is this

c. what GROUNDS are there for suspecting that .......etc. The simple answer is that there are no grounds for belief in these tweaks.

You have the right to believe what you want but without answering to my questions, I don't see enough reason to believe that your opinion is the truth.

[evdplancke]

The fact that the buffer is too short to judge might be precisely the consequence that we have bad aural memory.

Cognitive process passes through a kind of mental representation that may need some more time.

No we are very good at detecting immediate changes and processing them... All human senses are optimised for delta detection! - thats what has kep the species alive :-)

It would be impossible to mix an album otherwise! What DOES happen is that you can spend many hours making minor cumulative changes and lose all sense of macro perspective as you get fatigued, so now you think it sounds great but actually it doesn't... And this only becomes apparent when you listen to playback the next day.

This tells me that the longer we listen to something, the less perspective we retain. essentially the brain gets tired and bored and actually starts to fill in the bits it remembers from memory and stops listening to what it really sound like... And we have no idea when this is happening and when it isn't...
This probably explains why certain types of blind listening test are also fatally flawed...

So I would argue that the longer we listen to something being repeated, the less we actually hear and any judgements we make are flawed. On the others hand, quick a/b tests are ruthlessly revealing.

My opinion, founded on my own listening experience, is that it is very difficult to judge of sound quality on short extracts because of the very changing nature of the music making one second very different from the previous one. For instance, how to compare the quality of an attack without replaying it twice, one time on each system we want to compare? And in my opinion, attacks are very relevant to judge of the sound quality, especially with strings. The most efficient methodology for me is to replay short extracts in turn on the systems we compare, but unfortunately this is not easily applicable to the experiment of unplugging the ethernet cable.

[SBGK]

what is the pull the cable test, can someone explain it together with the expected results ?

[Phil Leigh]

Another test would be: Put a SD card with the same track as on the server into the Touch and play the music from that card. You bypass the entire netwerk transmission so it should produce a much better sound.
But personally I believe one won't hear any difference.

My opinion, founded on my own listening experience, is that it is very difficult to judge of sound quality on short extracts because of the very changing nature of the music making one second very different from the previous one. For instance, how to compare the quality of an attack without replaying it twice, one time on each system we want to compare? And in my opinion, attacks are very relevant to judge of the sound quality, especially with strings. The most efficient methodology for me is to replay short extracts in turn on the systems we compare, but unfortunately this is not easily applicable to the experiment of unplugging the ethernet cable.

I think there is a way... Find a segment of a piece of music with a repeating motif or pattern and pull the plug during the repetitions... If there is a difference it should be apparent straight away. After all, if you had tone controls and adjusted the bass or treble or even volume you would hear it straight away... You don't need to replay the whole track to hear the difference.

Now I know that some people are going to jump up and down and say that is a very crude difference compared to a subtle change in noise induced jitter... Well I say "nonsense". trained professional engineers with many years of experience can JUST hear a 0.5dB change in EQ in the context of a complete mix. A typical listener needs 1.5-2dB change to be sure. In either case that would be a "night and day" change. On the other hand you can wind up the jitter to levels that Stereophile would call dreadful and neither trained professionals or average listeners can detect the change.

If you really want to try this yourself and are prepared to sacrifice a cheap TOSLINK cable, just cut one in half and hold the ends together with your hands... That will produce large amounts of jitter...

Just try it and be amazed...