Buying and Playing Lossless formats
Sat, Sep 11 2010, 11:27PM
#2
RE: Buying and Playing Lossless formats
A Gramophone technician tells us FM radio has CD bitrate quality??
And anyway, difference between 192 & 320 on DAB is going to be more like the difference between 160 & 256 mp3 because the DAB compression is outdated. You certainly can hear the difference without recourse to scientific waffle. Just try a piece with a spread of instruments and frequencies. The busier it is, the less well the compression can follow the higher frequencies. But mainly, just listen to it.
Sun, Sep 12 2010, 2:31AM
#3
RE: Buying and Playing etc
Very interesting, this thread. A lot to think about here. Particularly for someone like me - software engineer that I am, I have never really gotten into the whole thing about sound file formats and qualities vs physical products and the different types of them even .... I don't even really know what SACD is.
But let me throw this one idea out there, for a different angle.
Is the whole topic even that important ? We are talking about the medium here, not the product. We are not talking about music here. We are talking about sounds. And I don't know about you, but sounds aren't really my thing. I like music.
For example. What if we were movie buff video people. To go around and around about what kind of screen, and if this one is better than that one, and on and on.....when its really all about the movie, and your reaction to it - not the physical equipment it's rendered on, and the brightness of the lights on the screen.
But somehow, when its music, people seem to get so caught up in the sound quality because soundf SEEMS like its music, but it's not.
So if you have the best video equipment in the world, and the movie was recorded in high-def IMAX and on and on and on.... unless you like the movie, its just crummy. The equipment and quality of delivery is obviously pretty irrelevant.
Similarly, if you have a music performance you like, and you have a fairly adequate recording of it, delivered through decent equipment...... you will not enjoy it significantly more by listening to it in the most scientifically advanced sound format and the most expensive equipment available. You just won't. Because the MUSIC is exactly the same. Even thought the sound may be some tiny bit, or even lets say possibly marginally perceptibly better.
And so to me, all this quibling about this and that minor detail about formats and mediums and so on just seems peevish and kind of petulant. Its just not that important..... unless we are focused on something other than music, like that our perspective should prevail....or whatever.... It really all has very little to do with music.
And so Gramophone, being the best music magazine on earth, is not a scientific periodical, but one about art. So to me, all this discussion about sound formats and so on, its misplaced here. I could see some certain small amount of it, being that it is relevent in a tangential way, but that's it. The rest of it is just people confusing sound for music. Its like someone thinking that the best car in the world will make his drive from point a to point b better or more interesting or more fulfilling in some way.....well, if its about the car, I guess it might be.....but if its about points a and b and the space between...... this is a delusional idea.
The magic of Mozart is not to be found in little bits and bytes and sounds signals and all this. No one will understand what Mr Chopin is saying one whit better by listening to it in some certian format better than another one.
Sun, Sep 12 2010, 8:34AM
#4
RE: Buying and Playing Lossless formats
Notwithstanding my post above, I applaud DJC's approach here.
"It's the music, stupid" to paraphrase a successful (if not so recent) mantra.
I would sneak in two comments however:
1) These posts are in the audio section of the forum.
2) When I first became interested in hi-fi a very long time ago my listening was mostly pop. The closer I got to music, the broader and deeper my interest became. Now I listen almost exclusively to classical music. There isn't a genre that I don't enjoy now. It could be coincidental but I don't think so.
Sun, Sep 12 2010, 9:19AM
#5
RE: Buying and Playing Lossless formats
A Gramophone technician tells us FM radio has CD bitrate quality?
Just to clarify, JD isn't 'a Gramophone technician': he's responsible for the technical side of the delivery of websites across a number of Haymarket titles.
Sun, Sep 12 2010, 10:32AM
#6
RE: Buying and Playing Lossless formats
...who happens to take a keen interest in audio reproduction.
But sorry, you're right - BBC's FM transmissions are only about 728kbps on FM.
Sun, Sep 12 2010, 3:09PM
#7
RE: Buying and Playing Lossless formats
...who happens to take a keen interest in audio reproduction.
Indeed, JD...
Sun, Sep 12 2010, 8:02PM
#8
RE: Buying and Playing etc
"In real life, however, who tests new equipment purchases with double blind tests? At what point is double blind testing necessary to make a statement of preference and below that point not so? "
Since we're talking about a publicly-funded body making decisions about how to invest in mass-market deployment, then yes, double-blind testing is a very sensible idea.
Mon, Sep 13 2010, 8:00AM
#9
RE: Buying and Playing Lossless formats
I would love to live in world where Radio 3 could be described as "mass market".
Is there no issue that can't accommodate a bit of BBC-bashing?
Fri, Sep 17 2010, 10:59AM
#10
RE: Buying and Playing Lossless formats
A couple of further comments on some previous comments:
I wouldn't think any scientist would say the differences people claim to hear between different digital formats are not "real" to the listener. Scientists wouldn't say the listener is lying or not really experiencing the difference. But what science is trying to do is to establish whether and why we perceive (sometimes huge) differences when there are none. It's exactly the same as optical illusions, or the placebo effect with drugs etc. Most people would have no problem with accepting the nature of these, but many people seem to have a problem with the audio equivalents (perhaps because of a lack of early familiarity with simple demonstrations of how our hearing can be tricked).
I disagree that science is a trump card simply being played in this thread to put down opposing views, or that scientists have their heads in the sand. The sciences of sound, electricity and magnetism, etc, are of course well established over more than 100 years and are being used as a basis to make the scientific claims in this thread. However what is being discussed here goes beyond simply stating some long established theory and leaving it there. One group of people claim 2 sound files sound different. One group claim that they can't. Science is about doing experiments in the present day to determine who is right. These have been done exhaustively around the world in the last few years. The conclusion is that humans think they hear differences between digital formats when there are none.
Finally as I've touched upon before with mp3 you cannot make statements that there will never be audible differences to the original. This is true even at high bit rates. In simple terms the mp3 format does not prescribe exactly how the audio compression is to be achieved. This is up to the people who write the mp3 encoder. Mp3s from different encoders will be slightly different, and how similar they are to the original depends on how well the developers have done tests and tweaked their code. But as a generalisation (again backed up by proper scientific tests) at higher bit rates (~192-320kbs) most of the differences people claim at first they can hear are not actually detectable by them in blind tests. Many people initially claim that such mp3s lack bass, or are tinny, or lack dynamic range etc. In almost all these cases these are a placebo effect (or the result of only listening on cheap ear plugs etc). The defects in high bit rate mp3s will be much more subtle than this, effecting only strange percussion transients and the like, and most people would never detect any difference. But it cannot be ruled out, especially if they have a particularly unusual interest: for instance I seem to remember the LAME encoder only tweaked the reproduction of casternets in the last few years!
Ted
Sat, Sep 18 2010, 5:15PM
#11
RE: Buying and Playing Lossless formats
This thread has been most interesting. Different viewpoints have been expressed, and various aspects explored most thoroughly. Thanks to everyone for sharing their knowledge. I think sometimes a writer expresses his/her enthusiasm for something openly (e.g. a new technology), and then someone else, who has a different or sceptical opinion, puts their argument across quite strongly. Perhaps a bit of competitiveness creeps in, making for a lively exchange. When someone quotes or paraphrases what another has said before, they may offend the first writer, who considers that his/her words have been distorted. I shall certainly be seriously considering some of the new possibilities, such as FLAC and high bit resolution streaming of music, that some of you have been enthusiastic about (otherwise I will have to put up more shelves in my house for an ever-increasing CD collection). At the same time I shall try to bear in mind the psychological factors, e.g. if I think some new system or format, sounds better, is the difference “real” or just the result of my brain keenly anticipating a particular result? I have tried out a few blind listening tests on myself - comparing different versions of the same recording in .wav format and MP3 formats (without knowing which format was playing). But the results are inconclusive so far. I ran out of time on the experiment, when I put on Mahler’s 7th Symphony, because after a few bars I decided to listen to the whole CD, instead of being disciplined and comparing short extracts. I have to admit that my sheer enjoyment of the music got in the way of being scientific.
Regards,
Geoff
Sat, Sep 18 2010, 6:48PM
#12
RE: Buying and Playing Lossless formats
Hello - I have not been able to understand some of this forum discussion. I have been buying classical music from Passionato (available at least here in US) and have been paying extra money for FLAC as opposed to MP3, then converting them to WAV files with xACT and importing to Apple's iMusic. Am I a fool for paying that extra money?
As someone said, this is about music not sound. I am not a good judge of sound per se and prefer Maria Callas in flat old mono, for instance, to some more recent sopranos recorded in better sound. I don't know whether the WAV conversion destroys what is good about FLAC. Again not sure about any of this really. I must not be alone in that confusion. This topic is surely worth a Gramophone article someday.
Sat, Sep 18 2010, 7:40PM
#13
RE: Buying and Playing Lossless formats
I have been buying classical music from Passionato (available at least here in US) and have been paying extra money for FLAC as opposed to MP3, then converting them to WAV files with xACT and importing to Apple's iMusic. Am I a fool for paying that extra money?
No, assuming what you want is a lossless file you can play in iMusic.
I don't know whether the WAV conversion destroys what is good about FLAC.
It doesn't destroy anything about the audio content, which is the main thing. Both FLAC and WAV are lossless - you can convert between them with no loss of information.
However, FLAC is a compressed lossless format, so a FLAC file will be smaller than a WAV file with the same content. Given ever-increasing storage capacities/decreasing cost of storage, that may or may not be an issue for you. (Yes, WAV files can contain compressed content, since WAV is a container format, not a encoding scheme. But they typically contain uncompressed content.)
And FLAC files include tags, which WAV files don't. (And yes, WAV files can contain metadata - but most music playing software won't read it - so for all practical intents and purposes, they don't.)
Again not sure about any of this really. I must not be alone in that confusion.
You're not, don't worry!
This topic is surely worth a Gramophone article someday.
Andrew? :-)
Sat, Sep 18 2010, 10:20PM
#14
RE: Buying and Playing Lossless formats
As someone said, this is about music not sound.
True, for us.
Is that also the case, however, for audio equipment sellers or music distributors (whether downloaded audio files or broadcast on radio)?
If claims are made about a superior audio experience, that raises expectations that risk being translated into perceived "reality", even in those cases when there is no actual measurable improvement in sound quality within the range of human hearing ... thereby, not coincidentally, helping to sell (more or higher margin) product or to attract (more) listeners.
While the fact of commercial interests does not necessarily mean audio claims are wrong or exaggerated, as in any consumer situation (or where public money is involved) a healthy degree of scepticism is never a bad idea, and the more disinterested, objective information consumers (and taxpayers) can access to help make informed decisions, the better.
Thus the references in this thread to acoustics and testing protocols.
In a nutshell, the old journalistic adage about "following the money" with respect to audio quality claims is a good strategy.
Mon, Sep 20 2010, 1:38PM
#15
RE: Buying and Playing Lossless formats
I take it as given (perhaps foolishly in this thread) that what we pursue through our audio equipment is the closest we can get to a live performance of music in an ideal acoustic environment.
I tread warily, but posit that when we listen to live music the sound waves travel in an analogue signal from instrument(s) to ears and brain. Vinyl's claims to audio superiority lie in this analogue region, I believe, even if it requires “compromises” beyond what happens with groove and stylus.
All equipment creates an illusion but unattainable, “perfect” audio equipment would, if it could, recreate the live listening experience, and “disappear”. Research, development, innovation, seek to get as close as possible to this ideal and have made enormous gains in our purchasing lifetime. It's not uniform across all equipment, of course, as someone stated earlier. Many think the recording quality of early stereo remains unsurpassed.
I understand that digital breaks down the analogue signal into “bits” and the more the bits, the closer it gets to the perfect analogue wave, and it would take an infinite number to perfectly mirror that wave.
We are in dispute about where the increase in bit rate is no longer detectable by any human ear, but not, I assume, that the number makes no difference, or thatany form of compression moves us further from that analogue ideal. I genuinely stand to be corrected here, but is it not true that digital sampling is a percentage of the (hundred percent) analogue wave? If so, why would we want less when we can get more, especially when we concede that live analogue sound is the ideal?
I might be wrong to make an analogy with television picture resolution, but could it, should it, ever be argued that human eyesight cannot detect an improvement and that research should cease short of recreating the illusion of being there in reality? No one has the eyesight of a cheetah but who wants to be told that the picture is good enough for them already? Like hearing, sight is on a continuum across populations and the aim anyway is the recreation of reality: being there on the savannah or in the concert hall.
Who knows what route the next technological breakthrough will take but at the moment, it lies with bit-rate sampling. The road towards the goal in my third paragraph is work in progress. Should the arguments used in this thread be used to halt that pursuit?
I am at the limit of my understanding here and would genuinely welcome correction if in factual error concerning these processes.
It is a parody of my position to state, “But you've been arguing in this thread against the results of a scientific approach to discovering perceivable differences in sound quality, in favour of the "evidence" supplied by popular wisdom”. (Seal4us)
I am not claiming popularity as proof of anything.
I am an enthusiast, excited about a recent shared audio experience and want to suggest, in a section about audio, that people try it for themselves. I am bewildered by the opprobrium this has attracted.
I hope I have as much respect for science as anyone here, but just claiming the high ground by playing the science card is no more valid than the creationists doing so. In technical details I defer to others who know more than I. But it is the wholesale denigration of people's experience that I challenge.
Is it or is it not being argued that beyond a certain bit rate no improvement in sound quality can be discerned by anybody? None at all. By 100% of any sample. Under any conditions.
If it is not, then the only point of discussion is the relative value of that difference, however small, to each individual hearer. But the whole tenor of these criticisms is, however unstated, that it is indeed the case that there is no difference to be perceived. Why is it not explicitly stated? Because if there is a perceived difference, it takes breathtaking presumption to question the value of that difference for each individual. If there is no difference to be perceived, everyone who claims to hear it is either a fool or a liar. There's no wriggle room on this.
Of course double blind testing will remove beyond any reasonable doubt whether the individual taking part in that particular test is or is not detecting improvement between two bit rates. Indisputable.
In real life, however, who tests new equipment purchases with double blind tests? At what point is double blind testing necessary to make a statement of preference and below that point not so? Do Gramophone audio reviewers use double blind testing to justify their evaluations of every piece under review? I doubt it.
There are claims being made for the good sound quality of the new Gramophone Player. On what scientific basis is that claim for 256kbps being made? The same basis that some are claiming for 320kbps. Where are the calls for double blind tests to prove Gramophone's claim?
I do not, sir, underestimate the capacity for self-delusion (any more than I underestimate the capacity of “experts” to resent challenges to their defence of audio orthodoxy.) Self-delusion might play a huge part in the enthusiasm for any innovation, and must indeed be doing so in this one to some extent.
But that is beside the point on the question of whether or not anyone can perceive differences at high bit rates.
So I seem to be countered by the argument that beyond an unspecified bit rate there is not and cannot be any improvement detectable by anyone under any conditions.
To sustain this argument you have to believe that everyone – every single person involved, from the technicians who developed the technology, the BBC engineers supporting the current experiment, the developers who market the products in a competitive market place, every convinced listener and purchaser, and so on and so on – every single one of them is not detecting any improvement in quality and is deluded or lying. That's a big claim to make from any point of view, let alone a scientific one!
In the posts above are quotes including, “in blind tests most listeners...”, “In virtually all cases...”, “...in general too small to be detectable...” “Most people cannot hear any major difference between...” and so on. John Duncan writes, “Won't get into 192 vs 320 here (other than to say I don't find that much difference)”. I agree. I don't find that much difference either. One person might be indifferent to the difference, another excited about it, but that's not the point at issue.
In double blind tests at key bit rates between say, 28kbps and 320kbps with the same listeners to the same music at the same volume on the same equipment, would my detractors like to say beyond what point no improvement would or could be proved? I'm guessing not.
I acknowledge that we are talking, for most listeners, of small improvements, perhaps on relatively expensive equipment, with some just unable to hear. But that still leaves a large number who are neither fools nor liars. They might not be as expert as you; they might be lucky enough to be able to afford relatively expensive equipment; they might have exceptional hearing. There might be all sorts of things about them that provokes your ire.
But to deny their experience with generalisations, innuendo and nit-picking and to call it scientific is graceless at best.
In any post in this thread I have made no more claims than that: I am excited about and enjoying the high bit resolution streaming of music; there is enthusiastic take-up of its development; and that music lovers should try to hear it for themselves.
Whatever is there about this that provokes such sneering condescension?
(EDITED BY MODS to remove 'blank verse' formatting)