Meridian Unplugged - Powered by MrTechGuy.co.uk Forums Discussion HiFi & HT Chat Can sound quality be measured?

And these are the people that bring us the telecoms & computers that function : the planes that (safely) take us on holiday, the car that (repeatably) takes us to work & to the shops etc etc etc....
Yes, as you say, what credible & ethical scientists / engineers say is true : they work with & take advantage of proven fact. When these things do not work, they usually make all efforts to understand & fix the problems. Not always, but usually.
Sorry - and this is not a personal attack - but that's nonsense. And this is exactly the kind of statement that has to be addressed.
The digital data arrives, is checked, and then it's processed. If the processing is sub-standard, because of system design (whether hardware or software, or both), then it's not the fault of the data. If the data is so corrupted that it exceeds the system parameters / capability, then there is an error and you hear noise, not music.
Note : the enjoyment of music is an emotion, strongly & potentially overwhelmingly affected by physiology and environment. What happens after the creation of the physical sound-wave at the boundary of the transducer (speaker cone / element) before it is ultimately translated into an emotion can be - in the majority of cases - far more damaging to "fidelity" than anything that happens during the signal-chain before the pressure-wave is created.
Oh dear. Why would this situation be any different at all ? Generally speaking, the digital video environment is considered to be horrible in terms of RFI / EMI !
In the quoted test, no errors were found from the sum of 1.1 billion transmitted pixels - where each pixel itself also consists of a packet of data. This test also showed that a more expensive cable actually transmitted a "faulty" pixel - which was identified and displayed as an error. You can repeat this experiment. Anyone can. It doesn't even really require (relatively speaking) complicated or expensive equipment. And we know with relative certainty - because of the way the test is designed - that the result will always be the same.... Or ?

There is no doubt that RFI / EMI exists. And there is no doubt that this can affect a digital signal that's being transmitted electrically down a wire. You can very simply see this in a live / visual representation on an oscilloscope.
But part of the reason for the creation / existence of digital data transmission is its fundamental robustness and ability to employ error-correction when compared to analogue.

Anyway, and returning to the original question again, can we measure Fidelity ? Yes. And those who avoid or obfuscate this only have something to lose - usually financially - or even emotionally.

Finally, the point (often) raised - alluding to blind dogma when related to the (worth of) continuing the understanding of / research into such subjects - is disingenuous.
The pursuit of deeper understanding should not & does not stop, but there are things that we really do know about. Whether or not you want to accept this is a personal choice. If you are so sure that something isn't true, then prove it.
Extraordinary claims require extraordinary evidence !

A too-infrequently repeated mantra (which we're privileged to be able to repeat in a free & democratic society) : everyone's entitled to their own opinions, but they're not entitled to their own facts !

Please note : all posted in good humour, not intended or addressed to anyone personally, and meant in the positive spirit of open & rational discussion.
At the end of the day - much repeated by many - it's all about the enjoyment of music. The mutual goal, which has been clearly displayed so often on this site in particular, is to help & encourage our fellow-enthusiasts, and where possible ensure none are "led astray"....

I have been itching to get back to this discussion all weekend and then Rolski goes and answers all of the points I was intending to make, thanks also for finding the link to that test!

T

Unhindered?

The only way that would make sense was if someone believed that the receiver absorbs 100% of the data, leaving behind everything it didn't understand.

Reality: 100% of the incoming Ethernet signal is unwanted RFI to the analog side.

This will be everything from 0Hz (i.e. DC-POE), to as near to infinite Hz as the system will allow (squarewave rise time) added to any problems created by cable shielding defeating intercomponent isolation.

Persisting in refering to distorted sine wave components doesn't make sense - the analog side doesn't care how pretty the square waves are (*), if the analog side sees any of the incoming signal, the designer has FAILED, you will not fix that with a prettier cable (unless attenuating the signal lowers the noise floor - although that then means you are depending on the signal being degraded).

(*) square waves are better at coupling unwanted noise than sine waves.

This is like complaining that the noise of pins dropping is spoiling the view - with a pneumatic road drill a meter away - the volume of noise from the actual signal will swamp out the noise from the 'noise'.

Telling people to connect up a scope to see how bad things are because of the 'cable' is a good way to fake up a fail - overlooking that the signal has been through at least one transformer (the ethernet interface) and high frequency square waves are not going to look good on anything other than a very high bandwidth scope and appropriate probe.

Cable arguments tend to be 'won' by people repeatedly pasting cable sellers literature full of technospeak that at first glance seems applicable, but takes more time for the people that actually understand the subject to explain why it's not appropriate, eventually the 'copy/paste' wins over the thinking leaving behind a forum with lots of advice based on guesswork/sellers literature/individual perception rather than actual reality.

Matt.

All right guys, I give in. Sorry for mentioning it.

DaveyC,

No worries, it made for a good discussion.

I have been away and so not able to respond to some of the issues raised. But this thread is about more than cables, albeit that the point of departure was inspired by a cable-discussion thread.

My suggestion was that sound quality can be affected by factors that we cannot initially measure as per Meridian, the 808 and jitter. It can also be affected by factors that had not previously thought significant such as timing information only available in hi-res recordings and now brought to us by MQA after Meridian's brave change of mind. And the Manchester research paper that I cited found that actual music rather than tone bursts had to be measured in order to arrive at worthwhile conclusions.

I am not aware that anyone has been quoting cable manufacturers!