Almost ridiculous difference

Jan, perhaps your not into Techno then 

I am trying to follow your point of directionality.. Yes you have a source and destination address.. and the protocols flow both directions, and each directional flow is as important to each other.. So true if PCM is flowing from upnp server source to the renderer destination.. There is a direction of the transfer of the PCM data.. But the Ethernet cable will need to efficiently handle both directions if one is not to affect the other. So if there is some un yet quantified betterment of having directional ethernet cables  it will need to be bidirectional     Or at least the internal send and receive pairs would need seperate directional treatments.

Anyway I suspect this path is a deadend.. My mind is trying to be open but it is tired this evening.

Simon 

The things that I am trying to understand is how using two cables is supposed to improve things (let alone one!).

Let me explain:

Some people are saying that they have one of these cables going from NAS to switch, and then another one from switch to renderer (technical uPnP term for your streaming music player).

So, lets be a 'bit' of information travelling from NAS to switch: The computer in the NAS sends this bit from its circuit board (it is a timed electronic voltage change running through that circuit board) out on to the ethernet cable. In order to do that the bit is not actually sent, instead an encoding of what that bit represents is created and sent down the wire towards the switch.

Now, at this point the encode signal may pick up some EFI/RFI, i.e. this encoded signal may get a slight distortion imposed onto it. However, when we get to the switch, this slightly distorted signal is reanalysed to create a copy of the original bit from the server, that can then flow through the circuit board of the switch as a timed electronic pulse. All traces of the EFI/RFI distortion introduced by the cable will have been removed. Indeed, if the EFI/RFI had had a significant effect and changed the bit in anyway then the system would know it and re-request the bit (not a strictly accurate description but good enough for this). The bit in the switch is not the bit from the server, but a copy of the bit in the server generated in the switch.

This bit will then flow through the switch until it gets sent to the renderer by exactly the same process that it was sent from the server to the switch.

So finally a perfect copy of the bit sent from the server is generated in the renderer where it is stored in some sort of buffer until it is decoded by a DAC. All RFI/EFI added during cable transit will have been removed. The resulting analogue signal from the DAC is only dependent on the bit(s) that are in the buffer of the renderer and so EFI/RFI introduced into the signal as it travels downs those cables can have no effect on the resultant analogue signal.

Please note that analogue signals in analogue circuitry do not have this advantage and that all introduced distortion is cumulative and generally degrading.

Now, this leaves us with the only other possibility: That the EFI/RF introduced into the cables somehow electrically interferes with the outgoing signal of the DAC directly. I have to say that your system must be of a very poor quality and design for this to be an issue,and if this were the case then there is no point buying these expensive cables in the first place.

Computers and microchips create a MASSIVE amount of electrical noise (I remember putting one of my early computers near a radio so that the radio interference could be used for sound (it was a TRS-80)), and this would be way in excess of any EFI/RFI picked up during transmission through ethernet cable. If computer equipment was susceptible to trace levels of EFI/RFI interfering with their functioning, none of this stuff would work.

So, having the cable from the NAS to the switch is even more pointless than having the cable from the switch to the renderer, since any distortion added there will have been removed and it can have no influence on the rendered sound!

To conclude - the information in the buffer of the renderer is totally electronically isolated from any 'memory' (added distortion) of how it got there; as long as it got there it is a perfect copy. If your system is susceptible to the EFI/RFI introduced into the analogue encoding of those bit in the ethernet cables and computer system that got it there then it is faulty and you should sell it.

Conrad

Sorry but that's a pseudo-logical story full of assumptions and unsubstantiated conclusions, or to abbreviate it BS.

Cheers

Aleg

I have long thought that a fairly high percentage of audio "tweaks" would fail double blind testing - I am not saying this subject is one of them, because I have not experienced the high end ethernet cable personally. 

When I worked at the audio store where I was introduced to Naim is where I encountered this phenomenon.  Obviously some tweaks work, and some are snake oil.  (Often very expensive snake oil.)

I just would love to see any number of audio tweaks subjected to this type of blinded comparison.  I feel fairly certain the manufacturers of said tweaks would not be quite as excited about the outcome of such an unbiased test.

This thread not dead yet?

Hi Aleg,

please enlighten me - what are the assumptions that I have made, and what are my unsubstantiated conclusions? I really would like to know.

Conrad 

Originally Posted by MangoMonkey:

No, I tried several times but they keep coming up with these funny stories 

A friend was showing off her new shoes that she was in love with (well, more of a guilty pleasure actually). I thought they looked quite swanky and since she was making such a play of them I figured maybe £150. Heh, boy was I wrong, by about the same degree...

So I reckon it's horses for courses on the pricing front.

Regards,
Frank.
All opinions are my own and do not reflect the opinion of any organisations I work for, except where this is stated explicitly.

The AQ marketing blurb describes in glowing detail its materials and construction, but makes no claims as to its sound quality. This is because they don't know why it sounds better and would be in trouble with Advertising Standards if they speculated On the reasons.

Keith

It's like the dark ages never happened - "Galileo we're really sorry but your pseudo-logical assumptions are never going to cut it with The Inqusition chaps they tend to adopt a faith based approach I'm afraid.... "

If it wasn't so funny it would be depressing 

Tognatius of Seville

There are sound scientific reasons for arguing that hi res files are not better than red book. All they do is extend the frequency and dynamic range beyond the reach of human hearing and actually have the potential to damage sound because the inaudible information can cause distortion in the audible range. This is backed up by double blind testing which shows that people can't hear the difference between hi res and 16 bit. 

Simon has given us technical information which suggests that ethernet cables can make a difference, but perhaps not a significant one. 

Based on above, I could conclude that ethernet cables are more likely to improve sound than 24/192 files. 

Blimey!

Ironically, the ethernet test is easier because you just have to swap the leads. Apparent differences between red book and hi res can be caused by many factors from the recording studio to the speakers. 

On the other hand, I don't really know what I'm talking about - I'm just regurgitating stuff I've read on the internet so it's probably best to ignore me.

Keith

Chaps,

I think we should have stopped at Pure Readers excellent and cogent post way back (the one with the quiz), as in my opinion, it managed to  "really tie the room (thread) together" whilst maintaining the status quoe of 'difference', shame this debate was not concluded then.

And indeed, an excellent summary by Aleg further up.

Not sure if the quiz has been answered yet, but always happy to oblige, is it 'thought'?

Jason.

I thought Garyi summed it up perfectly in post 2.

Amen.

Lol.

p.s. Hallelujah!

Indeed 

"May your god go with you"  Dave Allen

This seems an appropriate time to roll the closing credits:

A production of Audioquest, based on an original idea by Jonas.

Research: Simon

Research assistant (demoted): Conrad

Research assistant (understudy): Wat

Editing: Aleg

Action scenes: Jason, Tog

Esoterism and quantum physics: Pure reader

Naive questioning: Jan

Sociological questioning: Tog

Viewing from a respectable distance: James

Cats: Tog

Bunnies: Jan

Asparagus: Mark

Director: Garyi

and to Keith, for believing.

Sir Thomas Beecham made a rare visit to the recording studio in the guise of film music director for the Powel and Pressberger film:

The Red Shoes

And during the final very long credits the orchestra were playing the extroit, but an irritated Beecham stopped them and shouted,

"Who are all these non-entities?!"

Another take was made ...

Sorry, but Jan's last post reminded me of the story!

ATB from George

Jan, very good, what would the rolling credit music be ?

Research by Jonas, Simon, Hungry Haiibut, James, Keith and Harry. It should be noted that this research, while incomplete, has failed peer review dismally. 

I would suggest A Few Good Men :-)

Yep, nicely done Jan.

Jason.