Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Lossless vs. Redbook tests? (Read 116540 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Lossless vs. Redbook tests?

Reply #100 –
Quote from: 2Bdecided on
And when precisely did we "eliminate any location for the difference to exist"? I thought we discussed several scenarios where real and audible differences were possible - even likely. It's true all those scenarios occur due to problems with the hardware, software, drivers or OS - but I'm betting the week before each of those specific problems was first discovered, someone was smugly thinking they had "eliminated any location for the difference to exist" - and yet there it was.
I more or less agree with what you're saying (and to add potential insult to injury of my argument, my wife made almost exactly the same objection before she had better things to do than hear me tell some forum story).

I stand by my assertion that, in the context of what is being argued, the merely "corporeal" components (heh) of computer playback have already had their potential flaws excluded from consideration. Nobody's argued (yet) that differing levels of jitter or power supply noise or decoder bugs or whatever cause this. The turks are, specifically, calling into question the storage and decoding components of computer playback.

Quote
Rubbish. The first thing you do is check that they don't. Basic experimental method. Who cares what lunatics think? What matters is following the scientific principle. Within the realms of science it is trivial to prove that the ABX switches are benign, but it would be an idiot who commenced a lengthy ABX test without first checking that the ABX hardware was working correctly. And yes, before you ask, at the end of the day, if there was some genuine concern that the ABX hardware was causing some issue, the logical approach is to perform a manual ABX of the ABX hardware itself. You should also measure it, assuming you know what to measure.
OK, so, again, I more or less agree with everything you're saying, but I think we're speaking past each other. Let me restate my meaning.

We bandy about the terms "lunatic" and "idiot" a lot here, but seriously: very, very few audiophiles have reduced intelligence and/or mental illness. I can only think of a few people whose audiophilia is truly symptomatic of something I would call mental illness (hi Patrick82! George Louis!). It's all fun and games to call somebody a lunatic, but when you argue that one should not take their opinions into consideration because of that, well, that's armchair psychology, and it's unscientific, and honestly, I don't think it is acceptable in polite conversation. But that is a little besides the point.

What these people are is delusional - albeit sometimes with ideas that are not entirely set in stone - and they are deluded by their perceptions, their emotions, their peers, or their lack of education - not necessarily due to a mental illness per se. Most of them seem to be intelligent (and yes also manipulative/insecure). The distinction is important, because audiophiles are still capable of logical thought, tortured though it is; it's just based on faulty assumptions. Those many audiophiles, like JA, who dismiss ABX testing solely because it contradicts their experiences fall into this camp, and they are the loud ones who spout out ad hoc hypotheses whenever some new factoid comes up. The truth of the matter is that JA, Bob Stuart, the CA "pros", and most guys of their ilk seem to be a pretty level-headed on most matters and who simply are often good debators for the pro-subjective camp.

My understanding of the context of this topic was that it was about the testing of lossless==lossless to a wide audience: not just to us skeptics. Recall my point a couple years ago about how I believe subjective audiophile thinking forms a competing paradigm to our skeptical approach. What this means is, every authoritative ABX test we release on the subject can (and will!) be rebuffed by a counterargument from the subjective camp, explaining why such a test result has no meaning. This counterargument will be internally consistent. To somebody who looks at all of this from the outside, judging arguments solely based on their truth or falsehood, both the pro-ABX and anti-ABX positions will have equal merit (just like phlogiston did in fact have equal "merit" to combustion theory under some criteria).

So, in relation to ABX switches: I believe the objection has been levelled in the past by some audiophiles that an ABX test with such a switch is flawed because the switch itself has a characteristic sound that can obscure whatever difference is being evaluated. Such an opinion is entirely internally consistent, and many perfectly sane people have a hard time spotting its flaws. My point is, this situation that has come up in the past with ABX switches matches is going to come up with lossless comparisons in exactly the same way, and the reason I'm so sure about this is that each situation involves questioning fundamental aspects of science and engineering in a manner left completely unresolved.

So if we're doing this purely for our own edification, in the context of our own body of ABX tests, I have no disagreement with you and I have no problem with ABXing this. But I don't think it will be persuasive in the wider world. Maybe I misinterpreted the scope of the topic? I dunno.

Quote
btw, though I'm here debating this with you, I'm sure we're both on the same page. The thing is, if an audiofool can bring a scientifically valid criticism against a test I've done, I'd be upset with myself for not spotting it before hand and avoiding whatever the problem was (even if I knew full well it was a non-problem). However, if they can bring an unscientific / unjustified criticism against a test I've done, I don't care.

If someone thinks there's an audible difference between FLAC and WAV on their PC, let's see the ABX results, and (probably more interestingly) let's have a recording of the output to check for real differences. If there are differences, let's have a complete description of the hardware/software/drivers etc to find out what's going wrong. If there are no differences, that's the end of it. No more silly talk.
Right. I'm not disputing that anybody who wants to actually make a case like that here, or anywhere, needs proof of that level of certainty to present it rationally. The point here isn't to take into account unscientific/unjustified criticism. It's that
  • Some of the guys saying that pseudoscientific stuff genuinely place it above their own hearing, and these people can in a sense be "saved" - no blind testing is even necessary. I'm actually making headway on CA dispelling some myths about the algorithmic nature of computer playback. Swatting away architectural/intrinsic stuff like that is really important when arguing with audiophiles who believe that intrinsics are everything (and whose opinions on that are mutable!).
  • For the others who can't be "saved", no amount of blind testing will convince others that they are wrong.
Quote from: Ron Jones on
If you're going to insult other posters, you may as well be specific rather than being so irritatingly vague and ambiguous. Otherwise, people (like myself) end up wondering if they're included in your insult(s) or excluded. I don't know about you, but I don't come here just to sit around wondering if I'm being insulted or not.

Quote from: krabapple on
Tsk, tsk, such a deplorable tone.  I'm never coming to this shop again.    (FWIW, I scored in the >95th percentile on all my non-maths SATs/ACTs/GREs.  Admittedly that was a long time ago...I've become a lot more deplorable since then)
Y'all are right; I was in a particularly ripe tone last night and I apologize for that. (Heh, I'm pretty sure I scored worse than any of you on SAT verbal.)

I still express my exasperation at how this thread turned out, but truthfully, the signal-to-noise ratio is still far higher than any other forum, even on this thread.

Lossless vs. Redbook tests?

Reply #101 –
Quote from: AV-OCD on
... I've searched this forum as best I could but cannot not find any test results for comparing the sound quality of lossless compression (FLAC, ALAC, etc) files to the uncompressed file (Redbook, WAV, AIFF).

I did find a couple of threads on the subject, and the answers were usually a terse "lossless = no loss."  Which is fine and dandy, but I'd like to see the actual tests that lead these people to this conclusion.

My motivation is this comment made in that other forum:

"I know of some very knowledgeable engineers who have measured differences between Apple Lossless files and AIFF files. They measured the files, once a sonic difference was heard in an A/B comparison, in the first attempt to quantify the difference. The measurements were carried out on several computers with some of the best test equipment available."...


Quote from: AV-OCD on
...Well, while I didn't get the direct answer to my original question, it is clear that most here feel there is no compelling need to ABX test something that that is so unlikely to show a difference...


I did a test and I think I successfully proved that also the playback chain can be "lossless" when a lossless file is played. I.e. the additional processes that are running when the losslessly compressed file version is played do not cause any differences to the output. (Assuming the complete playback chain is correctly configured and also an uncompressed file always produces the same output when it is played repeatedly.)

Here's my test setup:

- Original sample in wave format (30 seconds, cutted from a standard ripped red book audio track. I added two sharp and short "peaks" (to the beginning and to the end of the sample) so that it would be easier to align the samples later. file: WaveSource.wav

- I converted this sample to ALAC. I chose ALAC because you mentioned it, but any lossless format would have been suitable for this test. file: WaveSource_converted_to_alac.m4a

- In addition I created a short silent wave file that I used as a spacer file: 2sSilence.wav

- I placed the test files in a foobar2000 playlist in the following order: 2s silence, wave, 2s silence, ALAC and 2s silence.

- then I played the playlist and simultaneously recorded the rerouted final output on the same PC (as it is just before it reaches any hardware in a standard listening situation). file: recording_(wave_2sSilence_alac).wav

- After that I simply opened the recorded file and accurately cutted & saved the Wave and ALAC parts to separate files. files: playback_result_wave.wav & playback_result_alac.wav


Result: The audio contents of the resulting files are exactly identical. The files pass all possible comparison tests.

I tested only one 30 second audio clip, but I don't think the 2x 1,323,020 samples in the resulting files can be identical by chance. I'd be happy to reproduce the test with any given sample.


The recording procedure was this: start playback in foobar > switch to Wavelab > start recording > stop recording when foobar's playlist ends.

At first the resulting files contained some dithering noise but after doing several tests I found the correct combination of "bit perfect" programs and settings.

The playback and recording chain was this: foobar2000 > Virtual Audio Cable instance 1> Virtual Audio Cable's "Audio Repeater" > Virtual Audio Cable instance 2 > ASIO4ALL driver > Wavelab 5

Virtual Audio Cable (aka VAC) is a software emulator that emulates a hardware audio device, Audio Repeater is tool that can route audio between separate audio devices. foobar's playback settings: kernel streaming to the VAC instance 1. Wavelab was set to record from the VAC instance 2

I used the following programs and versions: foobar2000 0.9.5.6, foo_input_alac 1.0.1, foo_out_ks 1.2.2, Asio4all 2.9, Virtual Audio Cable 4.9.0.1652, Wavelab 5.01b.

I did this test with an old 2.4 GHz Pentium 4 machine (no hyperthreading). To add some stresss I used only a single external USB2 hard drive for the source files and for the recorded output. I had all usual background tasks running in addition to the simultaneous playback and recording processes (tasks like a virus scanner & firewall). The OS was XP Pro.

The test files are available here: http://rapidshare.com/files/218469597/wave...lac_samples.zip (13 MB). I converted the wave files to FLAC to make the package smaller. To reproduce the test they need to be decompressed first. The included ALAC file is naturally the actual file I used in the test.

EDIT: fixed a couple of typos

Lossless vs. Redbook tests?

Reply #102 –
Quote from: krabapple on
It may be that 'very few' on CA are hopeless dark age audio mystics.  Unfortunately, the host and a few of the 'authority figures' on that thread,  appear to be.  In that, CA reflects the 'dominant paradigm' of  nonacademic 'high fidelity audio' discussion since the 80's or so -- the mystics have set themselves up as the authorities.  Perhaps you could lead the charge to 'convert' them, using soft-sell tactics. (Good luck with that.)


Trouble is, when you go up against dogma, you don't get rational responses. Remember that many people from CA are 'Audiophile' first and 'Computer' second. Audiophiles are generally conservative (in outlook and often politics) and are - at least until recently - big spenders on their hobby. That they wrap their hobby up in pseudo-science is almost immaterial; the rationale for their buying decisions is not 'science' - hard, soft or faked. They are the people who buy Tiger Woods golf clubs because they think the association will help their handicap. They try to justify spending money on a big, expensive 'timepiece' by praising the engineering or saying how 'taken over several thousand years, the Daniels co-axial escapement makes my Omega more accurate than any digital watch'. They buy hi-fi because it 'sounds good'.

In other words, audiophiles use their empirical side to choose hi-fi about as often as you use your empirical side to pick out the right shade of tie.

And I'm dipped in the same audiophile ink, I'm afraid. At least I understand that a lot of it is predicated on a bunch of nonsense*. That won't stop me from buying that Krell amplifier when I can afford it; I'm a label junkie, and because of that, I'll probably get as much from it because of what it is as I will from what it does. Does that make me sad - probably. It also makes me a consumer, and without that consumerist drive, there would be no codecs for you to test and no music to test them on.

This is also why I still cling to uncompressed formats. Not for any rational reason - enough tests have demonstrated that I am just wasting disc space. But what rational person would spend Krell money when they know they could get exactly the same performance from something like a HK amp at a fraction of the price?

However, I'm trying, which is why I'm here.


Quote
I left on a note suggesting that  *they*  come *here*.  Of course, the sort of CA goofs who think advocating testing of claims is akin to 'religion',  are advised to 'stay home'.


OK... I'm listening. Or not. I get confused at the right terminology round here.

*if you really want to know how deep the nonsense goes, check this out: My first 'real hi-fi' amplifier was made by a company called Exposure. I bought it back in the before time purely on the basis of a rave review from a journalist called Ian Rankin, who wrote for a UK magazine called Hi-Fi Review. I used that amplifier for more several years, telling everyone 'how good it sounded' and how it was Ian Rankin's 'fave'. Ian Rankin has since gone on to become a well-known crime writer and in a recent interview, he admitted that all his hi-fi reviews were made up. So, my first foray into buying stuff on the basis of subjective reviews was in fact a foray into buying stuff on the basis of creative writing.

I doubt other subjective reviewers have so little 'integrity' - Rankin's tenure in the reviewing game was short-lived - but it invites the question 'how can you tell?'

Lossless vs. Redbook tests?

Reply #103 –
I'd like to know what it is the various members at CA can hear in the difference, since the hypothetical problems we've revealed would probably at most cause some kind of clicks/pops or something like that. If they can hear a difference, they should be able to describe how it sounds.
Can't wait for a HD-AAC encoder :P

Lossless vs. Redbook tests?

Reply #104 –
Quote from: pdq on
"Scientific fact" is a bit of an oxymoron. In science, nothing is ever a "fact", but simply the currently accepted explanation of observed behaviors of the universe. There is always the possibility that the currently accepted explanation will be replaced in the future. In fact, it happens all the time.

Okay, I mean one of these that can be "checked and either confirmed or denied", via a scientific methodology, as per wiki's:

Quote
In the most basic sense, a scientific fact is an objective and verifiable observation; in contrast with a hypothesis or theory, which is intended to explain or interpret facts.[21] Thus, for example, it is a fact that most things stick to the earth and the theory of gravitation explains why this is so.


It seems to me that if I conduct an ABX test, WAV vs. FLAC we can establish a scientific fact, that I (and only I) cannot perceive the difference between them. When someone on CA or HA provides ABX results that say that they can, that "observation" clearly does not generalise.

I'm happy with Wiki's definition: "an objective and verifiable observation".

C.

EDIT: By the way thanks for highlighting the need to define terms. Many disagreements are caused by people using different words for the same thing, or the same words for different things.
PC = TAK + LossyWAV  ::  Portable = Opus (130)

Lossless vs. Redbook tests?

Reply #105 –
Quote from: krabapple on
It may be that 'very few' on CA are hopeless dark age audio mystics.  Unfortunately, the host and a few of the 'authority figures' on that thread,  appear to be.  In that, CA reflects the 'dominant paradigm' of  nonacademic 'high fidelity audio' discussion since the 80's or so -- the mystics have set themselves up as the authorities.  Perhaps you could lead the charge to 'convert' them, using soft-sell tactics. (Good luck with that.)


Trouble is, when you go up against dogma, you don't get rational responses. Remember that many people from CA are 'Audiophile' first and 'Computer' second. Audiophiles are generally conservative (in outlook and often politics) and are - at least until recently - big spenders on their hobby. That they wrap their hobby up in pseudo-science is almost immaterial; the rationale for their buying decisions are not made on science - hard, soft or faked. They are the people who buy Tiger Woods golf clubs because they think the association will help their handicap. They try to justify spending money on a big, expensive 'timepiece' by praising the engineering or saying how 'taken over several thousand years, the Daniels co-axial escapement makes my Omega more accurate than any digital watch'. They buy hi-fi because it 'sounds good'.

In other words, audiophiles use their empirical side to choose hi-fi about as often as you use your empirical side to pick out the right shade of tie.

And I'm dipped in the same audiophile ink, I'm afraid. At least I understand that a lot of it is predicated on a bunch of nonsense*. That won't stop me from buying that Krell amplifier when I can afford it; I'm a label junkie, and because of that, I'll probably get as much from it because of what it is as I will from what it does. Does that make me sad - probably. It also makes me a consumer, and without that consumerist drive, there would be no codecs for you to test and no music to test them on.

This is also why I still cling to uncompressed formats. Not for any rational reason - enough tests have demonstrated that I am just wasting disc space. But what rational person would spend Krell money when they know they could get exactly the same performance from something like a HK amp at a fraction of the price?

However, I'm trying, which is why I'm here.


Quote
I left on a note suggesting that  *they*  come *here*.  Of course, the sort of CA goofs who think advocating testing of claims is akin to 'religion',  are advised to 'stay home'.


OK... I'm listening. Or not. I get confused at the right terminology round here.

*if you really want to know how deep the nonsense goes, check this out: My first 'real hi-fi' amplifier was made by a company called Exposure. I bought it back in the before time purely on the basis of a rave review from a journalist called Ian Rankin, who wrote for a UK magazine called Hi-Fi Review. I used that amplifier for more several years, telling everyone 'how good it sounded' and how it was Ian Rankin's 'fave'. Ian Rankin has since gone on to become a well-known crime writer and in a recent interview, he admitted that all his hi-fi reviews were made up. So, my first foray into buying stuff on the basis of subjective reviews was in fact a foray into buying stuff on the basis of creative writing.

I doubt other subjective reviewers have so little 'integrity' - Rankin's tenure in the reviewing game was short-lived - but it invites the question 'how can you tell?'

Lossless vs. Redbook tests?

Reply #106 –
Quote from: Alex B on
I tested only one 30 second audio clip, but I don't think the 2x 1,323,020 samples in the resulting files can be identical by chance.

If the output chain had 1,323,020 opportunities to err during the ALAC decode and didn't, that's pretty solid as far as I'm concerned. Nicely done.


Lossless vs. Redbook tests?

Reply #107 –
Quote from: carpman on
Quote from: AV-OCD on
I am at a "user-level", while it seems that many here are at the engineering and coding level.  As such, I may need to be "sold" on (explained to) why you believe what you do.

I think part of the problem is language and in particular meaning.
There are a number of terms that need to be defined, some of which are:
  • Opinion
  • Belief
  • Proof
  • Scientific fact.

I've read some of the stuff (not all) on the CA thread and the above terms seem to all be treated the same, i.e. all are taken to mean "opinion". Thus, we end up with "you have your scientific facts and I have mine". But that is not the nature of science, that is the nature of belief. Science is not a belief system.

Thus, the sentence "why you believe what you do", is indicitive. To go back to my breathing under water issue. It's not a belief of mine that humans don't breathe under water, it's a proven scientific fact; one that can be proven both theoretically and practically.

C.


This thread has certainly become more of a philosophical debate than I intended, but here goes.  While I understand that science is not a belief system, you can chose a belief system that is based on science.  You are making what you believe to be an informed decision about the truth based on scientific findings, while others may choose to believe only what they have experienced.

So when I ask "why do you believe what you do?", I know that the base answer within this forum is likely to be "because science has proven it to be true", but what I am really asking is that you show me the scientific results that brought you to your current belief.

But I get what you are saying.  An opinion, by definition is: a belief or judgment that rests on grounds insufficient to produce complete certainty.  While a scientific fact is based on a deeper investigation of the matter and therefore provides a greater degree of certainty.  So no, the two are not equal and the "scientific fact" is probably closer to the "truth" than a loosely based opinion, but I don't think that it is inappropriate to challenge the current scientific findings if there is reason to believe that they are incomplete.

So in this case, it has been proven that an ALAC file will match the original file bit for bit when the ALAC is converted back to the original format, but the nay sayers are claiming that the proper conditions were not met.  Granted, the majority here believes that there is no reason for there to be a difference when the file decompression is done on the fly, but no one has done a bit compare or ABX test of a file being decompressed on the fly to prove that there is no difference.  Well, not until Ron Jones did his recent  experiments.  (Thanks for taking a stab at it Ron and I think your results are a good start.) 

I completely understand why no one has tested ALAC vs. AIFF--because no one that I know of here claims to hear a difference, so why bother.  I do not feel that it is the burden of anyone within HA to prove that there is no difference, that is up to the individual to decide if it is worth pursuing.  I just thought that since I keep reading about claims that there are audible differences, that someone here might have taken up the challenge to test it.  Again, I have no vested interest in the outcome since I do not hear any difference.

Lossless vs. Redbook tests?

Reply #108 –
Quote from: Gag Halfrunt on
And I'm dipped in the same audiophile ink, I'm afraid. At least I understand that a lot of it is predicated on a bunch of nonsense*. That won't stop me from buying that Krell amplifier when I can afford it; I'm a label junkie, and because of that, I'll probably get as much from it because of what it is as I will from what it does. Does that make me sad - probably. It also makes me a consumer, and without that consumerist drive, there would be no codecs for you to test and no music to test them on.
It also makes you brutally honest, and a man of luxury who is more secure than to have to couch his buying decisions on pseudoscience.

Lossless vs. Redbook tests?

Reply #109 –
Quote from: Alex B on
I did a test and I think I successfully proved that also the playback chain can be "lossless" when a lossless file is played. I.e. the additional processes that are running when the losslessly compressed file version is played do not cause any differences to the output. (Assuming the complete playback chain is correctly configured and also an uncompressed file always produces the same output when it is played repeatedly.)

Result: The audio contents of the resulting files are exactly identical. The files pass all possible comparison tests.

I tested only one 30 second audio clip, but I don't think the 2x 1,323,020 samples in the resulting files can be identical by chance. I'd be happy to reproduce the test with any given sample.


Alex -

Thanks for taking the time to test this.  It's another feather in the cap of lossless = lossless and hopefully those that believe that they hear a difference will be interested enough to conduct the same test themselves.

I have a question.  You gave this qualification up front: "Assuming the complete playback chain is correctly configured and also an uncompressed file always produces the same output when it is played repeatedly." 

What sort of things would you have to alter in the system configuration to induce a change?  In other words, what is an incorrect configuration?

Lossless vs. Redbook tests?

Reply #110 –
Quote from: AV-OCD on
It seems that you are making a few assumptions that simply aren’t true,

Please state what they are so that they can be discussed.

Quote from: AV-OCD on
but put yourself in my shoes.

I do when I post on audiophile sites. However, I understand that audiophiles reason based on a different set of beliefs to myself. I recognise that it is my view that is out of step and I post and respond accordingly. You are not doing this in your posts here and want answers based on your view rather than the dominant view on this site.

Lossless vs. Redbook tests?

Reply #111 –
Quote from: Alex B on
I did a test and I think I successfully proved that also the playback chain can be "lossless" when a lossless file is played.

OK, OK... I'm convinced. And I should have an extra half terabyte of NAS freed up in a few hours. Thanks.

Moderation: Reduced the size of your quotation of Alex B's post.  We don't need to read the whole thing twice.

Lossless vs. Redbook tests?

Reply #112 –
Quote from: AV-OCD on
I have a question.  You gave this qualification up front: "Assuming the complete playback chain is correctly configured and also an uncompressed file always produces the same output when it is played repeatedly." 

What sort of things would you have to alter in the system configuration to induce a change?  In other words, what is an incorrect configuration?

Actually, the answer is already in your question. Play a wave (or AIFF) file once and simultaneously record the output. Compare the files and find that the audio contents are identical. Play and record the same file a thousand times and always find that the recorded audio content is  identical with the source file's audio content. Any configuration that cannot pass that test is incorrect.

It is difficult to make a list of things to avoid. For instance, I don't know if a bit perfect setup that can play ALAC and simultaneously record the played output is possible at all on Mac OS X.

In my test, I got only dithered output until I found out that a setting called "Stream fmt" must be set to "Cable Format" in the VAC control panel.

Lossless vs. Redbook tests?

Reply #113 –
Quote from: Gag Halfrunt on
I doubt other subjective reviewers have so little 'integrity'
I'm fairly convinced some subjective reviewers never unpack the equipment - if they ever receive it at all. Some at least bother to read (and re-write) the press release.

I'm sure the more interested / committed ones do it "properly" (not "properly = blind testing" - merely "properly = actually listen to the thing"), simply because they (like me) wouldn't pass up the opportunity to play with some audio equipment, listen to some music, and get paid for it.

Cheers,
David.

Lossless vs. Redbook tests?

Reply #114 –
Quote from: AV-OCD on
It seems that you are making a few assumptions that simply aren’t true,

Quote from: honestguv on
Please state what they are so that they can be discussed.


Well, I thought I made it clear where I stand in my last post to you by describing the events that took place up to my OP, but I'll be more direct.  You said the following:

Quote from: honestguv on
Given the community here want to encourage discussions based on rational arguments and discourage statements of a belief in audiophile magic that dominate almost every other home audio site, how would you suggest handling a post from a person that rejects the logic of "lossless = no loss", believes that an "actual test" is more valid, supports it by claiming "knowledgeable" engineers have measured differences between lossless files and, presumably, the originals but fails to mention what was different, reacts with WTF and DON'T POST, etc... Should such posts be opposed or encouraged?


First, I do not reject the logic that lossless = no loss.  I have stated several times that I hear no difference between ALAC & AIFF, so lossless = lossless holds true in my experience.  In my OP, I was passing along the claims from another thread that some people heard differences between lossless and the original.  Furthermore, the quote I made in the OP was a comment made by the head of CA (Chris), and was not done so to support my own view.  I was simply passing along a comment from someone claiming that they had "proof" and came here to see if there was any opposing test data.  (I've been meaning to get back to CA to ask Chris to provide the test results he says he's seen.)

Somewhere along the way, you also grouped me into the "audiophile" category, which you imply is someone that bases their beliefs on irrationtal, subjective findings.  While I do consider myself an audiophile (a person that seeks high fidelity audio), my definition does not include irrational thinking as a prerequisit. 

My beliefs (as they apply to the subject at hand) are largely rooted in objective and repeatable proof, which is why I cam here.  As such, it is my belief that if someone claims to hear differences between ALAC and AIFF, that this would be revealed as either real and repeatable or imagined with proper objective testing.  ABX testing seemed to fit the bill when I posed the original question, but testing the file size of the real-time output would suffice as well.  I would conduct the tests myself, but I don't know how, and I figured that they would have already been done by now.

Quote from: AV-OCD on
but put yourself in my shoes.

Quote from: honestguv on
I do when I post on audiophile sites. However, I understand that audiophiles reason based on a different set of beliefs to myself. I recognise that it is my view that is out of step and I post and respond accordingly. You are not doing this in your posts here and want answers based on your view rather than the dominant view on this site.


What have I said up to this post that leads you to believe that my view differs from the "predominant view" on this site?

Lossless vs. Redbook tests?

Reply #115 –
Quote from: Alex B on
Quote from: AV-OCD on
I have a question.  You gave this qualification up front: "Assuming the complete playback chain is correctly configured and also an uncompressed file always produces the same output when it is played repeatedly." 

What sort of things would you have to alter in the system configuration to induce a change?  In other words, what is an incorrect configuration?

Actually, the answer is already in your question. Play a wave (or AIFF) file once and simultaneously record the output. Compare the files and find that the audio contents are identical. Play and record the same file a thousand times and always find that the recorded audio content is  identical with the source file's audio content. Any configuration that cannot pass that test is incorrect.

It is difficult to make a list of things to avoid. For instance, I don't know if a bit perfect setup that can play ALAC and simultaneously record the played output is possible at all on Mac OS X.

In my test, I got only dithered output until I found out that a setting called "Stream fmt" must be set to "Cable Format" in the VAC control panel.


Thanks, that makes sense.  Bummer for me, since I use a Mac Mini and it would be interesting to see if this still holds true on OSX. 

Lossless vs. Redbook tests?

Reply #116 –
I think far too much credence is being given to the idea that decoding on the fly gives a different result than decoding prior to playback.  There are people here who have the expertise and authority to give you a definitive answer regarding this point.  But like the first response that was given in this thread ("yawn"), they aren't interested and won't bother. 

Well, Josh gave a response; consider yourselves lucky.

Lossless vs. Redbook tests?

Reply #117 –
Quote from: AV-OCD on
Thanks for taking a stab at it Ron and I think your results are a good start.

Well, you're more than welcome, but I can only wish my tests weren't basically meaningless in comparison to Alex's more well-executed and meaningful tests

Lossless vs. Redbook tests?

Reply #118 –
Quote from: Ron Jones on
Quote from: AV-OCD on
Thanks for taking a stab at it Ron and I think your results are a good start.

Well, you're more than welcome, but I can only wish my tests weren't basically meaningless in comparison to Alex's more well-executed and meaningful tests


I think the fact that all you were left with was a constant level of dithering noise says enough.  Thanks again for even spending even an iota of time on this "pointless" task. 

Lossless vs. Redbook tests?

Reply #119 –
Quote from: MichaelW on
I have to agree with poop. OP was NOT audiophool nonsense; it was a reasonable question from someone who had read something they were very doubtful about (the difference between lossless and PCM), but wanted to know how he could be sure the fringe stuff was wrong. That is NOT trying to start a war, or be rude.



I meant to comment on this earlier.  This is half true for me.  Based on my experience I was very doubtful that there is a difference between lossless (on the fly decoding) and PCM, but was open to the possibility that there may be bugs or other known issues with some decoders.  I just wanted some sort of objective proof one way or the other (though I expected to find that testing has shown that there is no difference).

Quote from: MichaelW on
There's also been impatience with the questions asked here about how we know what we know (epistemology, if long words aren't frightening). Questioning the basis of our knowledge is as important to a scientific approach as performing experiments to attempt to refute conjectures. N00b-bashing is as much evidence of a faith-based approach as is being certain there's an audible difference between silver and copper interconnects.


I think this comment shows great insight into the irony of some of the responses to my original question.

Lossless vs. Redbook tests?

Reply #120 –
Quote from: Axon on
It's all fun and games to call somebody a lunatic, but when you argue that one should not take their opinions into consideration because of that, well, that's armchair psychology, and it's unscientific, and honestly, I don't think it is acceptable in polite conversation. But that is a little besides the point.

What these people are is delusional - albeit sometimes with ideas that are not entirely set in stone - and they are deluded by their perceptions, their emotions, their peers, or their lack of education - not necessarily due to a mental illness per se. Most of them seem to be intelligent (and yes also manipulative/insecure). The distinction is important, because audiophiles are still capable of logical thought, tortured though it is; it's just based on faulty assumptions. Those many audiophiles, like JA, who dismiss ABX testing solely because it contradicts their experiences fall into this camp, and they are the loud ones who spout out ad hoc hypotheses whenever some new factoid comes up. The truth of the matter is that JA, Bob Stuart, the CA "pros", and most guys of their ilk seem to be a pretty level-headed on most matters and who simply are often good debators for the pro-subjective camp.



Hmmm..polite conversation.  I suggest you try telling 'these people' sometime that they aren't idiots, merely *delusional*.  Just don't be shocked when they respond with something less than gratitude for clearing all that up.

Personally, I'm more in favor of seeing the delusional authority figures *ridiculed* and *humiliated*  -- that can really change the paradigm.  Rome/Carthage.



Quote
So, in relation to ABX switches: I believe the objection has been levelled in the past by some audiophiles that an ABX test with such a switch is flawed because the switch itself has a characteristic sound that can obscure whatever difference is being evaluated. Such an opinion is entirely internally consistent, and many perfectly sane people have a hard time spotting its flaws.


This is akin to the denial of evidence of evolution, on the grounds that it can also be God testing us and WE CAN'T PROVE IT ISN'T.

(Though actually an ABX switch can itself be compared to the direct path.  The only key point is to keep the comparison blind.  And it's been done.)

Lossless vs. Redbook tests?

Reply #121 –
Quote from: krabapple on
Quote from: AV-OCD on
So I see that there has been some cross-pollination going on between the Computer Audiophile thread that inspired my OP and this one, with a couple of the guys from HA posting over there.  I just have to say that the attitude of the HA member posts over there (and many of the early ones that I encountered here) is deplorable (save Axon's), and that this attitude certainly isn't going to help HA win anyone over to the objectivist side of the audiophile war. 

It is my experience, that very few of the people that post on CA are the hopeless dark ages mystics that you assume them to be.  Many are just as interested in the "truth" as you here, and they could probably learn something about finding that truth from this group, but not if you dash in, insult their intelligence and berate them.  Frankly, I was surprised by the intellectual beat down I received when I made the OP.  I consider myself a logical, rational, left-brained thinker, yet I was treated like some sort of subjectivist troll.


It may be that 'very few' on CA are hopeless dark age audio mystics.  Unfortunately, the host and a few of the 'authority figures' on that thread,  appear to be.  In that, CA reflects the 'dominant paradigm' of  nonacademic 'high fidelity audio' discussion since the 80's or so -- the mystics have set themselves up as the authorities.  Perhaps you could lead the charge to 'convert' them, using soft-sell tactics. (Good luck with that.) 


Quote
One of the key elements of making a sale, is to know your customer.


And how do you know that no 'sales' were made by my 'deplorable' tactics? The dark age audio mystics, I could care less about, except as target practice.  The hypothetical somewhat skeptical lurker who maybe needs some moral support for going against the audiophile party line, is the 'customer' I typically get 'thank yous' from by PM.


Quote
If you take a moment to put your assumptions about "all subjectivists" to the side, and actually get to know some of these people I think you'll find that there is enough common ground that you could enter into a meaningful conversation about the pursuit of high fidelity audio.



I left on a note suggesting that  *they*  come *here*.  Of course, the sort of CA goofs who think advocating testing of claims is akin to 'religion',  are advised to 'stay home'.


Yes, one person on CA (who was already inclined to accept scientific findings) stated publicly that he appreciated your input over there, despite your brash, take-no-prisoners approach, but take a look at how effective Axon was in opening the minds of the rest of the people in that thread when he stepped in after you and treated them with common decency.

http://www.computeraudiophile.com/content/...Lossless?page=2

If your goal is to alienate most of the people that don't share your belief system, or that you feel are less knowledgeable than you, well then mission accomplished.  If your goal is to convince others of the benefits of scientific methodology, you may want to refine your approach.  It doesn't have to be a "soft sell"; just a rational, logical counter argument conveyed with respect for the recipient.  When you have no respect for the other party, an absence of insult will probably suffice. 

Lossless vs. Redbook tests?

Reply #122 –
Quote from: krabapple on
Hmmm..polite conversation.  I suggest you try telling 'these people' sometime that they aren't idiots, merely *delusional*.  Just don't be shocked when they respond with something less than gratitude for clearing all that up.
OK - you got me. I'm being slightly fast and loose here when I refer to this kind of stuff as delusional. I'm admitting that this (somewhat hypothetical) group of people may be able to have their beliefs changed through facts; real delusions are absolutely unshakable. Like I said, I think audiophiles can be grouped into two camps reasonably cleanly on the matter. But I still think that quite a number of audiophiles have beliefs on the matter than can quite accurately be described as delusions.

Nevertheless, who ever said that delusion was ever a good allegation to bring up? It's an absolutely terrible rhetorical device. These people prove it well enough by themselves. If you're right, and you show how they're wrong, and you be assertive but also respectful about it and don't let them change the subject, they pretty much dig their own hole on their own and jump in. Either you win the argument (by acceptance or by avoiding the question), or they get extremely hot and bothered.

Quote
Personally, I'm more in favor of seeing the delusional authority figures *ridiculed* and *humiliated*  -- that can really change the paradigm.  Rome/Carthage.
Sure. So, how much paradigm have you shifted?

Ridicule runs the distinct risk of radicalizing people who would not otherwise be radicals on an issue. It can convince some people but drive away many others.

Quote
This is akin to the denial of evidence of evolution, on the grounds that it can also be God testing us and WE CAN'T PROVE IT ISN'T.
Exactly. And you're not necessarily going to win that fight with more biology.

Quote
(Though actually an ABX switch can itself be compared to the direct path.  The only key point is to keep the comparison blind.  And it's been done.)
Like what I am predicting for ALAC/AIFF, this does not eliminate some specific tweako objections against ABX switches - especially in the context of testing tweako things to begin with, like cables. That said, yeah, you are right on this.

Lossless vs. Redbook tests?

Reply #123 –
Just checked my X-Fi XtremeMusic and when in "audio creation mode" with the master sampling rate set to 44.1 kHz and bit-matched playback on the "What U Hear" input did in fact record a bit-perfect recording of the output from FB2K (kernel streaming mode.) Was frustrated by Audacity for a few minutes when I discovered that when its sampling rate conversion dither option is enabled, it ALWAYS dithers, even if there isn't any sample rate conversion being done.

Lossless vs. Redbook tests?

Reply #124 –
I guess confirmed tweakers are going to stay that way, whatever. What's more to the point is people on audiophile forums who are simply concerned to get good information, and have no way a priori of knowing who's got the good oil, who's got cloth ears, and who's just willy waving. Like, perhaps, OP.

Such people are more likely to be led to the way of truth and light by a calm and measured approach. Assuming that that matters.