debunking myths -- objective and subjective observations from a Sunday afternoon

[Anonymous]

storsvante:

So how do you store the data and what are the challenges?

Formats

There are many audio file formats, generally in 3 categories: uncompressed (eg WAV or raw PCM), compressed (eg FLAC or Apple Lossless), and formats with lossy compression (eg MP3, AAC, Vorbis). The format describes how the audio is organised in the audio file.

The most 'vanilla' format is raw PCM or WAV, which is not much more than the PCM data, sample by sample, organised in a long sequence.

FLAC is getting a lot of attention right now. It uses linear prediction to convert the audio samples to a series of small, uncorrelated numbers which are stored efficiently, and usually reduces the overall size by about 50% (compared to raw PCM data). The encoding is completely reversible, ie the data can be decompressed into an identical copy to the original, hence the term lossless.

changing emphasis slightly I think this is worth reiterating. I've seen people say that there are listenable differences between WAV, Apple lossless and FLAC. I believe this to be impossible and any differences down to placebo. uncompressed to PCM these formats are identical in terms of the data they present. A application can be compressed as a zip file, this is also lossless compression. A play of a CD and a play of a FLAC might produce differences due to variations in how the CDP interprets the disk each time.

 

[professorhat]

zzgavin:Interconnects did the job well enough and they fitted into the world view of how hifi worked.

This is a very good point in so much as we're really only talking about streaming digital data - something that is just generally not done in many hi-fi setups, even today (though obviously times are a changing now with computer based solutions as we know). But in the past, in most hi-fi setups, a CD player was connected to a stereo amp - all the digital to analogue conversion was done internally within the player itself. With this being the case and thus since hardly anyone is going to be using them, why bother putting in ethernet ports and the necessary processing power and software to transport digital music perfectly into a CD player when virtually no one wants to do this and all it's going to do is raise the price of the component.

I'm sold

 

[idc]

storsvante:idc:

What is the advantage of ethernet and what is going to stop an ethernet cable debate springing up?!

Ethernet is inherently two-way and when used with standards like TCP/IP (which is what drives the Internet and most computer networks also what Linn uses) there is robust error handling built in which makes the cable pretty uninteresting. This would be a tough case for the high-end cable manufacturers to argue, because if we couldn't rely on TCP/IP over Ethernet over standard Cat5e cables to guarantee 100% free error transmissions every time then the world would stop tomorrow. Corporations would cease to function, the internet would grind to a halt, and if you walk into a bank branch to withdraw money you'd better check your balance afterwards, as the Ethernet cable from the counter to the servers may have sneaked an extra zero into your withdrawal amount. You get the idea... ;--)

But of course this wouldn't necessarily stop another cable debate...

Totally understand, that, I didn't pick it from your first post. I am now far more dubious about digital cable transmission that I was before. Your above post is like a previous one, where someone said 'how can cables be directional when the current is AC?' That made sense and I am now dubious of directionality, which is something I have not heard with cables. Now I see your point that with error correction, unless there is something inheritly wrong with cable it will transmit data and error correction is the key.

My own experience is that, after using three USB cables (a printer one to get me going, a 'designer' one at £27 which as the correct length and kept its shape and now the one that came with the new DAC) I have noticed no difference. I have also used two different laptops and two different hard drives and again I cannot hear any difference.

However, if Andrew Everard reports hearing a difference then to my mind he is a very credible source and there may still be something that we are missing.

 

[PJPro]

DLNA is effectively the ethernet standard as used on my humble PS3.

 

[Anonymous]

PJPro:DLNA is effectively the ethernet standard as used on my humble PS3.

Just looked at the www.dlna.org site, I had not seen this before. Looks very promising, but I assume it will take some time before everything is implemented. Did not see Apple in the contributors though.

 

[Anonymous]

This is the best post I've read in a while. Very strong sensible facts clearly explained.

I don't think we are far off ethernet being the regularly used format with Hi-Fi. I know that the Yamaha DSP Z11 and the new RX-V2065 has an ethernet port for streaming internet radio.

So how far away are we from an ethernet output to a DAC?

Regarding the S/PDIF not having any error detection/correction, I read somewhere although I cannot find the source, that someone used a straightened coat hanger with no errors and debunked expensive cables. I assume they were able to monitor the data from either side to prove this.

 

[idc]

adamshaw:

This is the best post I've read in a while. Very strong sensible facts clearly explained......... I read somewhere although I cannot find the source, that someone used a straightened coat hanger with no errors and debunked expensive cables.

Adamshaw, in light of the quality of this post can you find that link or withdraw that comment as an internet myth

 

[Anonymous]

no withdrawl required http://consumerist.com/362926/do-coat-hangers-sound-as-good-monster-cables

 

[idc]

Nice one one off, coat hangers as speaker cables. From the theme of this thread I did think that adamshaw was meaning coat hangers had been used to cable a PC to a DAC, which I would not have thought was possible...or is it?

 

[PJPro]

I think we are getting off topic here. The OP has been voicing option on all things digital not analogue. That's a whole different argument.

 

[Tony_R]

PJPro:.... Even within the various DAC chips themselves there appears to be general agreement that they are not all equal. Clearly, as a unit, these differences can be even more marked.

Obviously the DAC chips themselves (amongst other things) influence the sound - but the receiver chip is the most important part where connectivity is concerned.

Whether the connectivity medium is SPDIF, COAX or USB the receiver chip ultimately takes the digitally signal and turns it into something the DAC chip can make use of.

Modern receiver chips can reclock (amongst other things) the incoming signal thereby eliminating any jitter issues.

For the more technically minded this site offers a good explanation SPDIF and it's vagaries.

Ultimately, for SPDIF the signal is a square wave - and the better preserved that square wave is, the better chance the DAC has of retrieving all the bits off the disc.

 

[Anonymous]

Thanks Tony very interesting link and seems to disprove the coat hanger. To think I was actually tempted to try it this weekend!

 

[The_Lhc]

adamshaw:Thanks Tony very interesting link and seems to disprove the coat hanger. To think I was actually tempted to try it this weekend!

You were actually going to try replacing an OPTICAL cable, with a coat hangar? How exactly would that have worked then?

 

[Anonymous]

you can get transparent plastic coat hangers now just glue them into you inputs and outputs

 

[Anonymous]

the_lhc:
adamshaw:Thanks Tony very interesting link and seems to disprove the coat hanger. To think I was actually tempted to try it this weekend!

You were actually going to try replacing an OPTICAL cable, with a coat hangar? How exactly would that have worked then?

The COAX not OPTICAL. I might be stupid but not that stupid.

 

[PJPro]

I'm still not sure how it would work with a coaxial connection. You'd need two coat hangers, wouldn't you?

 

[Tony_R]

adamshaw:Thanks Tony very interesting link and seems to disprove the coat hanger. To think I was actually tempted to try it this weekend!

All that's really happening by experimenting with different cables, is finding the cable that provides the best match between source and DAC.

If the source equipment was designed correctly in the first place, this sort of tomfoolery would not be necessary, but unfortunately, no hard and fast standard prevail for the SPDIF connection (there are some 'standards') but they're rather vague.

Take any RF equipment for example, connecting an aerial on a radio transmitter requires that both ends are carefully matched. This is usually achieved by tuning the transmitting antenna (often by cutting it to length - but some offer more sophisticated tuning methods).

This is to tune the "standing wave ratio" (SWR) in the transmitter / cable / antenna interface.

The theory applied above is quite similar to the SPDIF connection - the better the match between 'ends' the better the quality of the signal transfer. Instead of tuning your CD player output, or the input on your DAC, you tune it by selecting a cable that provides optimum transfer.

However - if the source and DAC were built around a proper standard, this tuning would not be necessary, as a particular length of cable with the correct impedance and termination (plugs) would provide identical results time after time, with any combination of equipment.

 

[Anonymous]

Tony_R:
All that's really happening by experimenting with different cables, is finding the cable that provides the best match between source and DAC.

If the source equipment was designed correctly in the first place, this sort of tomfoolery would not be necessary, but unfortunately, no hard and fast standard prevail for the SPDIF connection (there are some 'standards') but they're rather vague.

[...]

The theory applied above is quite similar to the SPDIF connection - the better the match between 'ends' the better the quality of the signal transfer. Instead of tuning your CD player output, or the input on your DAC, you tune it by selecting a cable that provides optimum transfer.
However - if the source and DAC were built around a proper standard, this tuning would not be necessary, as a particular length of cable with the correct impedance and termination (plugs) would provide identical results time after time, with any combination of equipment.

Tony, can you clarify this a little bit. In which way do you feel the the S/PDIF standard 'vague' and how do you think the different cables differ from one another in effect of this? The example you give is impedance - surely there is no room for ambiguity on that (ie 75 ohms)?

 

[Tony_R]

storsvante:Tony_R:
All that's really happening by experimenting with different cables, is finding the cable that provides the best match between source and DAC.

If the source equipment was designed correctly in the first place, this sort of tomfoolery would not be necessary, but unfortunately, no hard and fast standard prevail for the SPDIF connection (there are some 'standards') but they're rather vague.

[...]

The theory applied above is quite similar to the SPDIF connection - the better the match between 'ends' the better the quality of the signal transfer. Instead of tuning your CD player output, or the input on your DAC, you tune it by selecting a cable that provides optimum transfer.
However - if the source and DAC were built around a proper standard, this tuning would not be necessary, as a particular length of cable with the correct impedance and termination (plugs) would provide identical results time after time, with any combination of equipment.

Tony, can you clarify this a little bit. In which way do you feel the the S/PDIF standard 'vague' and how do you think the different cables differ from one another in effect of this? The example you give is impedance - surely there is no room for ambiguity on that (ie 75 ohms)?

Poor choice of words on my part to have said that the standards are vague - rather that the implementation of these standards appears to be vague. Although signal levels and impedance are specified, it would appear that various manufacturers choose different ways to implement it, i.e. some use pulse transformers to decouple the output, others use capacitors, some use buffers (or a combination of all the above). The same applies to the receiving end - even DAC manufacturers choose varying ways of terminating the incoming (SPDIF) signal.

This is why when measured, some CD players (and DVD players) exhibit better waveforms than others, into a given load. And it is my belief that this has a strong influence on the resulting sound. How could this affect the sound? Well my theory is that with non optimal signal transfer, there could be ringing on the signal, causing some kind of interaction with the reciever in the DAC, which is influencing the sound to some degree or another.

So this is where my cable matching theory applies - that the various cables on the market present different values of impedance, inductance and capacitance to the coaxial output stage in the CD player, so we the public at large select one that "sounds the best" when in reality this cable is probably providing the most optimum signal transfer.

So... if both the transmitting and receiving end were optimised (standardised?), a standard type of cable could be adopted which would remove this variable from the equation, as every connection should then yield the optimum signal transfer.

I hope this makes sense as I'm not very adept at "putting to paper" what's in my head!

Tony.

 

[Fred_Barker]

I thought this was a very interesting vid about what differences in audio can be perceived:-

https://www.youtube.com/watch?v=BYTlN6wjcvQ&list=FL-5vFTshVaksmQj11TvZ-UA

Polite comments cordially invited

 

[Petherick]

I have no comment to make on the outcome of any test performed using coat-hangers (at the moment at least), but purely from a mechanical engineering standpoint, how can you make a pair of 2 metre cables from FOUR coat-hangers??

 

[cheeseboy]

To the op - :clap: nice bit of reading and very well written - thank you

However, if Andrew Everard reports hearing a difference then to my mind he is a very credible source and there may still be something that we are missing.

If he reports a difference than it's placebo effect. It's has been proven many times about digital interconnects, yet what hi fi still manages to find differences in every single digital cable they review, and unfortunately, the more they do it, the less repsect they have about their reviews, and even more so, the more foolish they look. There's a good piece on cnet proving that a selection of HDMI cables all produce exactly the same results, yet if you cross reference those with the what hi fi reviews, what hi fi somehow manges to find myraids of differences with them!

It's a hangover from an analogue age and people are still trying to force analogue ways of thinking on to digital systems. It wasn't that long ago that people would have laughed you out of a room for using a computer as a credible hifi source...

 

[MajorFubar]

Holy thread resurrection, Batman (not me this time though!)

But seeing as we have revived this thread...here's one myth which raises its head now again:
Lossless (ALAC/FLAC) and uncompressed (WAV/AIFF) versions of the exact same file sound different.

(I can hear the groans from here.)

Because some people would just not be told, some time ago I compared lossless vs uncompressed in Audacity and published the results on this very forum. The results showed beyond any doubt that the two files were so indisputably identical they precisely and absolutely cancelled-out each other mathematically when I inverted the phase of one and played them together. All that was left was absolute silence.

Yet now and again, someone turns up like a bad penny and 'puts it about' that lossless and uncompressed files sound different on his player/DAC/computer, usually backed-up with a comment along the lines of "well I know what I can hear".

Can we agree once and for all that any player which genuinely makes lossless and uncompressed files sound different is either broken or badly designed?
It just is.

 

[Fred_Barker]

Holy thread resurrection, Batman (not me this time though!)

As with my jokes and my body - The old ones are the best!!

I know it's an hour of your life you'll never get back - but i'ts worth the effort - some good theories in that YouTube vid eh??

In case you missed "Audio Myths Workshop" - here it is again:

https://www.youtube.com/watch?v=BYTlN6wjcvQ

Fred