MP3 vs WAV: The Blind Test - Can YOU tell the difference?

[beclemund]

I think you are willfully misunderstanding the real problem which has never been the individual buyer's use of the music.  The issue has always been the large counterfeiters who flood the stores with thousands of fake recordings that are visually and digitally indistiguishable from the real thing.

If you say so. Again, you are redirecting to a similar but still different scenario in the later case. Obviously, we are not going to convince each other which of us is correct here, so there is little reason for me to continue this. I will add a few quotes from selected sites for you, however:

From PBS:

With new technology, people can put song files on compact discs. Hardware like a CD burner can be hooked up to a computer and copy, or "burn," files onto a blank CD. CD burners can also copy regular music CDs, so music fans can make a mix of their favorite songs.

"There are two types of blank CDs," Marino said. "CD-Rs are regular blank CDs that can hold stuff like music files. CD-RWs, on the other hand, can be rewritten, sort of like a cassette."

Senior Mike Jackson uses his burner regularly.

"The burner came with a program that lets me copy computer files or music," he said. "I buy blank CDs, copy a music CD on them, then resell the original to a used record store, like the Record Exchange."

(This is illegal, but more on that later... )

Here's a blog discussing the vagueness of the law with an interesting tidbit from the RIAA:

A spokesperson for the Recording Industry Association of America, the group known for filing lawsuits to stop digital-music swapping, pointed me to a document from the Copyright Office to support the idea that you aren't allowed to keep a digital music file once you sell the physical CD.

And finally, from About.com's internet security:

You can sell or give away the original CD, but only as long as you no longer have any copies of the music in any format (unless of course you have another copy that has been legitimately paid for). You can not copy the CD onto your computer and load MP3's of it onto your portable MP3 player, and then give the original CD to your best friend because you don't need it any more.

I do not dispute that it is largely a gray area and you are not likely to be pursued by this. But the fact is, it *is* wrong. You do not believe it is so, so obviously our moral compasses are not alike. That is fine. For others concerned about the legal implications of such activity, I made my initial comment so that they would be aware that it was a good practice.

It is pointless for us to continue going in circles about this as we are not going to convince each other. I certainly do understand that there are other copyright issues that are of larger concern to the recording industry, but that is not what we are discussing here.

[71 dB]

Well, my clock here in the UK says 00:40hrs, so it's officially Sunday July 1st and time to reveal which clip was which.


Clip A was originally a 192kbps MP3 file

Clip B was the original WAV file

Clip C was originally a 320kbps MP3 file

Clip D was originally a 128kbps MP3 file

I had no clue about the order. I was distracted by the low quality of the original recording. I have avoided comparing lossy sound to original because when you learn to hear the difference you hear them very easily (so I have read). That would spoil 128 kpbs mp3 for me. Why is this critical? Because 128 kpbs MP3 is everywhere! It's good to have wooden ears with lossy sound. People who say even 320 kbps mp3 is not good enough for portable players should see a doctor. Damn, people listen to them in traffic etc.! Who cares if there is some minor artifacts? In 1990 I bought a Sony Walkman that was more expensive than my iPod. It was a state of the art Walkman but still even 64 kbps mp3 sounds superior to those damn C-casettes!

People demand the products with the best numbers on paper while they have lost contact with what's really important or relevant. Sad.

[Greta]

I meant to come back and guess again, but didn't get around to it in time!  I'm guessing Mark meant me that got close - I mixed up the 192 and 320 but got the other two. And I went back and forth on those several times. I suspected it wasn't quite right, but hated to change it and get something else wrong!

The 128 was really obvious right off. And the WAV seemed crisper first listen. I wanted to mention, not just on this recording but others, that with the WAV I usually "hear" the hall much more than in others, the fine details, especially in high frequencies and headspace if you will. Bloom is another good way to describe it.

That was definitely missing from the other two for me, but I am not at all surprised that I got 192 and 320 mixed up. With a recording I am familiar with I can tell, but on the first few listens it's hard. A very good example is with music I have bought online in 192, become familiar with, and then got the physical CD for because I liked it a lot. I really realized on hearing the real thing the extra details that had compressed out.

But a correct rip (ex. EAC, also CDex is good) on a CD in decent sound, it can be hard to tell. Unfortunately, my computer system actually shows the differences more than my stereo, because it has a brighter, sharper, less weighty sound, and also I'm closer to the speakers! I used to rip in 256k, then switched to VBR 320 (LAME VBR - great in 320), but now am using FLAC a lot. I hope fervently that music downloads within at least the next two years completely switch to it, or I'll take Apple or WMA Lossless, as long as its DRM free. No real beef I guess, just less warning messages and hullabaloo.

Does anyone here like 128k AACs? I have an audiophile Apple-crazy friend that loves them, and while iTunes stuff I have gotten doesn't really sound bad, I am just wondering what makes it equivalent to say, 192k MP3.

[M forever]

I had no clue about the order. I was distracted by the low quality of the original recording. I have avoided comparing lossy sound to original because when you learn to hear the difference you hear them very easily (so I have read). That would spoil 128 kpbs mp3 for me. Why is this critical? Because 128 kpbs MP3 is everywhere! It's good to have wooden ears with lossy sound.

And maybe a wooden brain to go between them. What a lame excuse for the fact that you probably can't hear the difference. But then you can judge that the original recording is so "low quality". It's indeed not that great, but pretty decent. Maybe not the "ideal" test material for such a comparison. But certainly quite representative for a lot of what's out there.
Just how do you come up with such lame nonsense? I have to admit there is an element of surreal genius to some of your statements that is quite entertaining in a bizarre way - but only when I ignore the fact that you are allegedly an acoustical engineer. And of course statements like these complete contradict your claims to have such incredibly advanced hearing that can pick out structures "normal" people can't perceive.

You don't even have to "learn" to tell 128kbps apart from uncompressed audio - it's glaringly obvious. 192kbps is still pretty easy to tell with somewhat complex material. With most material it is not very easy at all at 320kbps to tell the difference.

I got the "perfect score", but only when I listened to the clips for the second time on my "premium" setup at home. When I first listened to them on my still quite good travel setup, I couldn't decide confidently which one was 320 and which WAV, and at first I thought C was the wav file because it appeared "cleaner" to me, but then I revisited the clips on my home setup, and then I was pretty sure B is the wav file.

So, the actual *sound quality* delivered by 320 is really very, very close to uncompressed files. Which extensive blind listening tests with experienced listeners have already shown. But there are limitations to that format. Some more complex and "busy" textures can get audibly degraded even by such a modest level of compression. And there can be artifacts, interestingly not just in very complex situations.

I remember ripping a recording of Higdon's "Blue Cathedral" to 320. In some places, there are very fine ringing sounds made by players operating those metal Chinese balls with little bells inside which you can role in your hand to exercise it. These sounds simply did not work in 320 - the result was a strange sound like high frequency electronic noise. It only worked in uncompressed sound.

Or, when you listen to some of the clips in Mystery Orchestra 15 very carefully, best with headphones, surprisingly enough, on some of them you can hear compression artifacts right at the very beginning of Zarathustra when the low C in the organ, contrabassoon, and contrabasses begins. They sound like very high, somehow chirping sounds (tchi-tchi-tchi). You may only hear them when you listen carefully for them - IIRC, they can be most clearly heard right at the beginning of Bonus Clip 1. It seems that the sound of the room atmosphere is simply to complex for the codec to figure out and represent even at 320kbps.

And of course we want the best possible sound. What do we care if some people have wooden ears and brains. At the end of the day, it doesn't matter either that for some recordings, for some listening situations lower qualities are enough. We want sound that will not let us down in those listening situations in which we lean back, concentrate, and want to enjoy high quality music in hopefully good recorded sound on hopefully good playback equipment.

There aren't that many really good *recordings* to begin with, so there is no reason to further degrade that. And with storage capacities getting bigger and cheaper all the time, there is really no reason at all not to use a lossless format like FLAC as first choice. I use 320 for Mystery Orchestra because it is indeed *very* close, but when I heard the artifacts in the MO15 clips, I hesitated, then decided to go ahead and use the clips and see if anybody would complain about that. Nobody did, but that's not a "failure" of the listeners - the actual quality of the sound is still very good, very close to the originals. But I wouldn't want to archive the recordings in a format which introduces artifacts, no matter how weak they are.

[The Mad Hatter]

I've always found that it's possible to hear the difference, but in a more instinctive way. When I used to listen on my mp3 player to CDs that I owned, I always found that the mp3s just didn't 'grab' me in the same way. It wasn't that I was hearing it more or less sound (though when I did the test - and by the way, I was one of the correct answers - I distinguished the 320kbps from the wav by virtue of the fact that the strings felt slightly weaker), or more or less quality, it was just that what on CD was powerful was just loud on mp3.

[71 dB]

And maybe a wooden brain to go between them. What a lame excuse for the fact that you probably can't hear the difference. But then you can judge that the original recording is so "low quality". It's indeed not that great, but pretty decent. Maybe not the "ideal" test material for such a comparison. But certainly quite representative for a lot of what's out there.
Just how do you come up with such lame nonsense? I have to admit there is an element of surreal genius to some of your statements that is quite entertaining in a bizarre way - but only when I ignore the fact that you are allegedly an acoustical engineer. And of course statements like these complete contradict your claims to have such incredibly advanced hearing that can pick out structures "normal" people can't perceive.

Let me explain my nonsense. Human hearing has developped in evolution to perceive physical sounds around us. Lossy formats utilisize this. They try to create "un-physical" distortion to the sound so that human ear would have big difficulties detecting them. This works well as long as people do not train their hearing to detect these "un-physical" artifacts. The idea of lossy sound stops working when you really search for the artifacts.

I'm not an advocate for lossy sound. I prefer CDs and SACDs at home. Outdoors I use iPod and mp3s. There's noise, traffic, wind etc. that masks the music and the artifacts. The listening experience is heavely compromised even with original wav but it's better than no music at all.

Hearing is not an mechanical thing. Hearing differences between lossy and lossless sound is lower level hearing. Identifying complex musical structures is higher level hearing. Hearing can be good in many ways. 

[M forever]

Hearing is not an mechanical thing. Hearing differences between lossy and lossless sound is lower level hearing. Identifying complex musical structures is higher level hearing.

Not true. The ability to hear sounds so precise that one can detect fine differences between sounds (such as comparing different qualitiy playbacks of the same source material or similar situations) is basically the same as being able to hear complex structures clearly. No, it *is* the same. That ability is a *requirement* for "higher level" hearing. One can't separate that. It all has to do with the ability to perceive sonic structures precisely rather than just emotionally. Tonal color is one of the important parameters of music, if one can't hear it very precisely, one can not hear complex structures precisely either.

But one can still *imagine* or at least *pretend* to be able to. 

[71 dB]

Not true. The ability to hear sounds so precise that one can detect fine differences between sounds (such as comparing different qualitiy playbacks of the same source material or similar situations) is basically the same as being able to hear complex structures clearly. No, it *is* the same. That ability is a *requirement* for "higher level" hearing. One can't separate that. It all has to do with the ability to perceive sonic structures precisely rather than just emotionally. Tonal color is one of the important parameters of music, if one can't hear it very precisely, one can not hear complex structures precisely either.

But one can still *imagine* or at least *pretend* to be able to. 

Start here:Hearing to study hearing. Maybe you then realise how little you know.

[M forever]

I started studying hearing decades ago. Unlike you, I am actually currently work in a professional audio environment, so I don't need you of all people to point me to wikipedia articles about hearing. Or have you found a new job in the meantime? I mean, other than MacDonald's or similar?

Do you also have a link to an article in which it is explained that it is better not to be able to hear differences in sound quality? And is that what they taught you at the university? Or is that simply one of your free-thinker ideas with which you "challenge conventional wisdom"?

[71 dB]

I started studying hearing decades ago. Unlike you, I am actually currently work in a professional audio environment, so I don't need you of all people to point me to wikipedia articles about hearing. Or have you found a new job in the meantime? I mean, other than MacDonald's or similar?

Do you also have a link to an article in which it is explained that it is better not to be able to hear differences in sound quality? And is that what they taught you at the university? Or is that simply one of your free-thinker ideas with which you "challenge conventional wisdom"?

Oh, now you are a professor of acoustics?

My job situation has nothing to do with my hearing.

Is it better to hear clearly artifacts in lossy sound and whine about it? How does that make you enjoy life more?

I listened to the sound samples (original & 128 kbps) again now that I know which is which. I don't hear a clear difference but the original "feels" better but that's probably placebo effect. What I DO hear is that the original recording is of awful quality. Especially cymbals and brass instruments do not integrate to other instruments. The stereo image is very unnatural and the sound is colored/distorted.   

[M forever]

Oh, now you are a professor of acoustics?

I didn't say that (reading comprehension problems?), but compared to you, almost *everyone* is a professor of acoustics.

My job situation has nothing to do with my hearing.

Probably not. Unless you work or want to work in a profession in which good and judicious hearing abilites are a requirement. Then you are totally screwed. Well, MacDonald's are always hiring, I hear.

Is it better to hear clearly artifacts in lossy sound and whine about it? How does that make you enjoy life more?

It's not a matter of choice. Either you hear the *extremely audible* degradation of the sound and the compression artifacts at lower bitrates, or you don't. But I have a hard time imagining anyone *not hearing* them, especially someone who claims he has especially great aural perception.

I listened to the sound samples (original & 128 kbps) again now that I know which is which. I don't hear a clear difference but the original "feels" better  

You godda be kidding me (us). Well, at least you are honest...

What I DO hear is that the original recording is of awful quality. Especially cymbals and brass instruments do not integrate to other instruments. The stereo image is very unnatural and the sound is colored/distorted.  

How would you know? If you can not tell the difference between the different clips, you are the last one who should make such statements. Besides that, you have no idea what the Wiener Philharmoniker sound like in real life.
It's not the greatest recording I ever heard, but it's fairly good. Definition could be better at the low end, but it's OK. If you hear a lot of distortion there, then it's either your equipment or your wooden ears. Or both. Probably both.

Hey, but don't worry about these things. Just enjoy life, OK?

[71 dB]

I didn't say that (reading comprehension problems?), but compared to you, almost *everyone* is a professor of acoustics.

I said that on purpose because I knew you would ask about reading problems.

I
How would you know? If you can not tell the difference between the different clips, you are the last one who should make such statements. Besides that, you have no idea what the Wiener Philharmoniker sound like in real life.
It's not the greatest recording I ever heard, but it's fairly good. Definition could be better at the low end, but it's OK. If you hear a lot of distortion there, then it's either your equipment or your wooden ears. Or both. Probably both.

I can hear the difference between a good recording and a bad one. The differences between the clips are insignificant compared to the problems of the recording itself.

[M forever]

I said that on purpose because I knew you would ask about reading problems.

That's a pretty lame comeback. Even you can come up with something better. I know you can.

I can hear the difference between a good recording and a bad one. The differences between the clips are insignificant compared to the problems of the recording itself.

No, you most definitely can't. If you can't tell the huge differences between the heavily compressed clips and the original, there is no way in hell you can judge recording quality in general. That should be obvious even to you.
Not even when hell freezes over, thaws again, all the little devils drown, Satan has to declare bankruptcy, sell the property way below market value, and MacDonald's move in. At least you could apply for a job there then.
And again, you have never heard the WP live, and certainly not in their own hall. Have you ever actually been to an orchestral concert anywhere? Or do you chose freethinkingly not to go because they never play Elgar (or because you are broke, or both)?       

[71 dB]

That's a pretty lame comeback. Even you can come up with something better. I know you can.

Why bother? It's Sunday!

No, you most definitely can't. If you can't tell the huge differences between the heavily compressed clips and the original, there is no way in hell you can judge recording quality in general. That should be obvious even to you.
Not even when hell freezes over, thaws again, all the little devils drown, Satan has to declare bankruptcy, sell the property way below market value, and MacDonald's move in. At least you could apply for a job there then.
And again, you have never heard the WP live, and certainly not in their own hall. Have you ever actually been to an orchestral concert anywhere? Or do you chose freethinkingly not to go because they never play Elgar (or because you are broke, or both)?       

Think whatever you want. I have been to orchestral concerts and I have heard even Elgar live.

[M forever]

Why bother? It's Sunday!

Not bad. Much better. Thanks. But - isn't every day like Sunday for you?

Think whatever you want.

Thanks, I will. You know, I am free-thinker, too!!! Except that I do sometimes check if my ideas have any relationship with reality. You should try that once in a while, too.

[71 dB]

Not bad. Much better. Thanks. But - isn't every day like Sunday for you?

You are welcome.

Unemployment has it's advances and disadvances. Less money, more free time. Btw, yesterday I worked 10 hours measuring loudspeakers for a hifi magazine.

Thanks, I will. You know, I am free-thinker, too!!! Except that I do sometimes check if my ideas have any relationship with reality. You should try that once in a while, too.

People do disagree about things even if the reality is the same for us all...

[M forever]

Unemployment has it's advances and disadvances. Less money, more free time. Btw, yesterday I worked 10 hours measuring loudspeakers for a hifi magazine.

You mean the cabinet dimensions?

[71 dB]

You mean the cabinet dimensions?

No. Frequency/power responces + THD/compression measurements for subwoofers.

[M forever]

You actually know how to do that?

[71 dB]

You actually know how to do that?

That's what happens when you study acoustics in university. You learn how to make acoustic measurements. 

I actually wrote the MATLAB script that calculates to power response approximation.