Losing Audible Information.
- Thread starter maxflinn
- Start date
Robert
Tapehead
Audible information is the key here.
Information is always lost in the recording/replay process since no parts of the chain achieves 100% technical perfection, ie. there is always noise and distortion of some sort at some level. What we have to establish is the threshold where these things begin to obscure/modify the signal in audible ways.
With digital I'd say the format is largely unimportant (excluding lossy compressed formats for now) since most reasonable implementations will capture and replay with imperceptible noise and distortion in normal use - and I mean normal use, a pro environment can place different demands on things.
The best recordings played via bog standard 16/44 can sound stunning, and the insertion of such processing loops into the chain pass unnoticed. They aren't the problem by a long stretch.
Superb recordings also sound superb when rendered using 16 bit 320kbs lossy systems to drive home the point.
It's interesting to look at analogue too, for here we see the introduction of errors far closer in nature to low bit rate digital combined with the effects of lossy compression, primarily noise, phase, frequency response and inter channel crosstalk all of which are a factor with lossy digital codecs.
Ironically, analogue 'throws away' the most information yet is often preferred. Which just goes to show how tenuous is the link between truthful, transparent audio and what the listener might like or regard as good hi-fi. Don't get too fixated on the preservation of every last scrap of information or the goal of ultimate transparency. Go with what you like.
Information is always lost in the recording/replay process since no parts of the chain achieves 100% technical perfection, ie. there is always noise and distortion of some sort at some level. What we have to establish is the threshold where these things begin to obscure/modify the signal in audible ways.
With digital I'd say the format is largely unimportant (excluding lossy compressed formats for now) since most reasonable implementations will capture and replay with imperceptible noise and distortion in normal use - and I mean normal use, a pro environment can place different demands on things.
The best recordings played via bog standard 16/44 can sound stunning, and the insertion of such processing loops into the chain pass unnoticed. They aren't the problem by a long stretch.
Superb recordings also sound superb when rendered using 16 bit 320kbs lossy systems to drive home the point.
It's interesting to look at analogue too, for here we see the introduction of errors far closer in nature to low bit rate digital combined with the effects of lossy compression, primarily noise, phase, frequency response and inter channel crosstalk all of which are a factor with lossy digital codecs.
Ironically, analogue 'throws away' the most information yet is often preferred. Which just goes to show how tenuous is the link between truthful, transparent audio and what the listener might like or regard as good hi-fi. Don't get too fixated on the preservation of every last scrap of information or the goal of ultimate transparency. Go with what you like.
Max, no offence but any chance of you posting a topic that doesn't involve your belief all formats/digital media/cables sound the same? Perhaps something about all amps sounding the same might be a welcome change 
.
In fact don't worry, I'll do it for you as it's something I'm interested in at the moment.
.In fact don't worry, I'll do it for you as it's something I'm interested in at the moment.
maxflinn
pfm Member
Ponder the question, factor in all the recent talk of high-res and DSD.Max, no offence but any chance of you posting a topic that doesn't involve your belief all formats/digital media/cables sound the same? Perhaps something about all amps sounding the same might be a welcome change
It's relevant and designed to spark interesting and informative debate. Robert's excellent reply a perfect example.
You're clearly a clever chap, let the intelligence supersede the suspicion and get involved in an open way, you'll feel better
.
YNWOAN
100% Analogue
That itself is quite a big assumption and not a very solid premise in my opinion.
The constant assumption is that we listeners are 'passive' and that we just sit there receiving 'information'. In fact, the human brain is known to process/manipulate/interprete that external 'information' in all sorts of ways.
Listening involves the equipment and the listener. No two listeners are alike, how they 'interprete' the sound is unlikely to be identical. The tendency to ignore the active involvement of the listener is what leads to the 'Serge' school of audio design...where the listener is an unresponsive 'given' with no real role other than not to argue.
NB: I know Serge is not the only proponent of this, he's just good at stating it.
Listening involves the equipment and the listener. No two listeners are alike, how they 'interprete' the sound is unlikely to be identical. The tendency to ignore the active involvement of the listener is what leads to the 'Serge' school of audio design...where the listener is an unresponsive 'given' with no real role other than not to argue.
NB: I know Serge is not the only proponent of this, he's just good at stating it.
Julf
Facts are our friends
I don't think that is the assumption at all. I think the assumption is that the listener interacts with the music produced by musicians and studio engineers, and reproduced through a set of loudspeakers in a room. The role of the reproduction system, in this model, is to bring the recorded music to those loudspeakers in a form that is as accurate (or pleasing, depending on the tastes and choices of the listener) as possible within the criteria of human hearing.
Do tell: Just how wide should we make the 'criteria for human hearing'? Two people can respond entirely differently to the same sound, indeed the notion of 'same' might become problematic when talking of consciousness. Suppose you play a note, and the listeners all describe it quite differently (in a range from bright to dull.)
Even if you went for the statistical approach ('most people find that sound bright') it would not prevent other listeners from describing it as 'dull'. The fact that they were in a minority wouldn't make their description invalid.
So you can't ascribe a specific human response to a specific set of sound waves. At this point the 'science' breaks down; in the sense that no set of measurements logically entail a particular human response. There is a conceptual gap, which isn't solved by ignoring it.
Even if you went for the statistical approach ('most people find that sound bright') it would not prevent other listeners from describing it as 'dull'. The fact that they were in a minority wouldn't make their description invalid.
So you can't ascribe a specific human response to a specific set of sound waves. At this point the 'science' breaks down; in the sense that no set of measurements logically entail a particular human response. There is a conceptual gap, which isn't solved by ignoring it.
Julf
Facts are our friends
Two people can respond entirely differently to the same sound, but if none of hundreds of test subjects respond at all, you might possibly conclude that the sound is inaudible. So that is my criteria - the limits of absolute audibility. A frequency range of 20 Hz to 20 kHz, and a dynamic range of 90-100 dB.
formbypc
pfm Member
Darn, beat me to it.
formbypc
pfm Member
Even if a number of the test subjects did respond, Max would automatically say they were imagining things or suffering expectation bias.... and I think you might, too.
Julf
Facts are our friends
I think you might well think that. Seems you have no idea of experiment protocols.
maxflinn
pfm Member
Could be I guess..
maxflinn
pfm Member
I'm not sure how any of this is relevant?
Max,by chance I looked at your home forum a couple of seconds ago and I find exactly the same topic from exactly the same person. What are you trying to achieve here? Pinkfish is rather more liberal than your home site, and less likely to swallow the orthodoxy.
As it stands I think myself and others are more prepared to buy SACD or other hi res formats now than before you started to push the orthodox line. You seem a nice chap, but maybe you need to start questioning (metaphorically) the drum you're banging.
As it stands I think myself and others are more prepared to buy SACD or other hi res formats now than before you started to push the orthodox line. You seem a nice chap, but maybe you need to start questioning (metaphorically) the drum you're banging.
maxflinn
pfm Member
Formby, stop trying to cause trouble for me please.
BTW, have you thought of a non-condescending way that someone might suggest that your observations may not be real, yet?
maxflinn
pfm Member
You're just not happy unless you're targeting someone, are you?
Why not let others decide whether the question I posed is worthy of discussion.
Thanks.
Robert
Tapehead
That doesn't reflect his view at all, nor do I know of anyone else who sees the listener as an unresponsive given. It's pure invention, something against which to argue perhaps.
With respect to Max and his question, it's pretty open for folk to interpret that in many ways and take the discussion down several paths.
flatpopely
Prog Rock/Moderator
Max.
If I did a blind ABX between DACs and then afterwards between different bit depths , codecs etc. would you accept the results even if it meant people heard a difference?
If I did a blind ABX between DACs and then afterwards between different bit depths , codecs etc. would you accept the results even if it meant people heard a difference?
maxflinn
pfm Member
Andrew, it's nice of you to ask but completely off topic.
My rather simple logic tells me that unless a given format is unable to contain all of the available audible information on a given master, then there is no reason to think it possible that music replayed from it should sound any different to the same music replayed from any other format that is also capable of containing all of the available audible information. Once the recording process was competently undertaken in all cases, from the same master.
What do you think about that?