[OJNeg]

Why are you so adamant on defending lossy formats bigshot? Just because you've ripped your whole library to AAC?
 
Again, I'm not going to make any claims that I or anyone else can hear the difference. Whether the difference is audible or not is irrelevant; there is a difference. The fact of the matter is, when you apply lossy compression, you take away information and no longer have an original copy. That's not up for debate, it's very much a technical reality. I know you're going to say that an end-user has no use for a lossless copy, but that is very much a blanket statement.

 

[bigshot]

I do tests on everything I use in my system, comparing it and finding out if there is any effect on sound quality. It's a lot of work, but that's the kind of guy I am. I want to know. I'm not guessing, repeating the general consensus, or advising that people do something I'm not sure about "just to be safe". I know.

If I've carefully tested and determined that a particular codec and bitrate is audibly transparent, and that it maintains that sound quality for ten generations, why should I equivocate? How would that benefit anyone?

I don't listen to bits. I llisten to sound. The conversion from electrical signal to mechanical sound is a process that changes audio in many orders of magnitude beyond lossless vs high bitrate lossy. People should be aware of exactly how much of a difference things make so they can choose their battles. As long as they understand that, they can assuage their OCD and save to WAV files backed up to a hundred different drives running a bitcheck algorithm before each play... As long as they know that all that doesn't make a lick of difference.

I'm operating on a wild and crazy theory here. I know it's revolutionary. I'm wearing a hard hat and keeping my head down. Here is my theory...

People want great sound, and they don't want to pay more than they have to, spend more time and energy than they have to, or complicate the process more than they have to.

I know that is an unusual and unique theory. I've been told recently by one fella that he actually *wants* to spend thousands more for a high end amp that sounds exactly the same as a midrange one, because he considers the cabinet to be "art". Other people tell me they encode their music lossless, even though they can't hear any audible difference "just to be safe". Other people buy SACD players based on frequencies they can't hear, and dynamic ranges that are only audible at ear damaging volume levels. Still others buy cables that cost hundreds of dollars because they think "silver sounds sharp and clear" and copper sound "warm".

I'm not speaking for the benefit of those folks. They can cheerfully ignore my posts and I won't care. I'm speaking to the hifi nuts who want to get great sound and not be silly or wasteful about it. I think these folks deserve being told what matters and what doesn't. I know it's a crazy idea....

 

[xnor]

I think everyone (ok, every sensible person) agrees with you on acoustic transparency. But I don't see how making backups using FLAC is silly or wasteful.

 

[Clarkmc2]

I seem to remember that Bigshot's backups are ten thousand CDs in his garage. I use the same system. When I download original files I get the FLACs. No transcoding up from lossy to burn a CD and I archive the FLACs on DVDs. Media is cheap. The downloads are not. Hard drives all die eventually.

Lossy makes no sense for me because I don't play from files. Old school, and I don't multitask listening to music.

 

[OJNeg]

It sounds like you're trying to group people who keep lossless libraries with cable believers and other audio nonsense. That's silly.

 

[bigshot]

I'm saying that there really is no real practical need to keep a lossless version. You can keep one if you want to, but I transcoded AAC 256 VBR ten generations and it still sounded fine. I can't imagine in my lifetime needing to transcode more than that. I let record companies keep masters. All I need is players.

Backups are imperative though. It just doesn't matter if the backup files are lossless or high bitrate lossy. I back up my AAC files because I've invested a lot of time preening the tags since I've ripped them. Quite frankly, the tagging I've done is more valuable to me than the CDs in the garage.

 

[nanaholic]

Quote:

No disrespect, but the fact that a forum poster convinced YOU that he could tell the difference between high bit rate lossy and lossless hardly constitutes "scientific fact" that high bit rate lossy is audibly distinguishable from lossless, or that there is a select group of individuals who can make the distinction.  The incident may be interesting.  It may even be worth investigating further.  What it's not, though, is proof.

 
You've misinterpreted - I never said that it is a scientific fact that lossy and lossless audibly distinguishable, however it IS a scientific fact that lossy and lossless is not completely transparent on the waveform level - otherwise it would not be called a "lossy" codec and a "lossless" codec.  The only point up  for debate is whether that difference could be heard, and data showed that most can't yet some can.  Again here please refer to my 100m in 10s example, so I think my stance is perfectly logical and rational - my default position is that unless you are super-special (ie you trained really hard or you have a special biological advantage) you won't be able to tell the difference, but I don't completely rule out the possibility that such people can and do exists because it is biologically possible for someone just to have better hearing, this isn't something like believing in a Lizard-man or something like that.

 

[bigshot]

What you can't hear is never the problem.

 

[HiFi1972]

There is a HUGE difference in sound quality from something originally recorded at 24bit and a 320kbps MP3. I know this because I master audio for a living, and I get to hear what a 24bit master sounds like over a full-range speaker system (B&W) and once in a while, I'll listen to the 320 MP3 (LAME encoded, usually using CBR).
 
Whenever I hear people say "I can't tell the difference between the CD and the MP3" I then question the playback system (not to mention that they're unaware of the material before it was dithered down to 16 bit, from 24). If it's a pair of headphones (no matter how great the amp and pair of cans are, they're likely to NOT hear a difference because headphones don't resonate through your body the way audio does coming out of speakers). The more refined the playback system is (high end DAC, speakers and a system that reproduces from 20Hz to 20kHz as neutral as possible) the more obvious it will be; MP3s sound "narrow" and even though most of us can't hear 16kHz, there will be a sense that something is missing or not completely there when you cut those frequencies out.
 
One thing that is usually not in this discussion is the difference between "hypercompressed" audio that is typical these days on most modern releases and material that has optimal dynamics. When you maximize the levels of a song by use of dynamics processing (upward compression/limiting) you lose low end frequencies and there is little dynamic crest between peak and average levels. Most of the time, you're listening to predominantly middle frequencies and combined with loudness, you start becoming subjected to the Fletcher Munson effect, where we're hearing more mid frequencies (which is natural, as we perceive those than lows and highs). If you take a track that is so loud that you mostly hear mid frequencies, but can't "feel" the low end, you are less likely to be able to hear any difference, especially if you're playing back material over headphones.
 
I'm not trying to argue with anyone over the above, I *know* this is true because I listen to approximately 40 different songs, every week, that come in as 24bit mixes for me to work with and there is ALWAYS a trade off in sound quality when you 1. apply dynamics processing to make the material louder (therefore reducing the dynamic range) and 2. removing "inaudible" frequencies which are likely more "felt" than "heard".
 
EDIT: One (non-scientifical) observation that I've made for myself is the fact that MP3s sound better when made from the 24bit masters and not from dithered, 16bit files ripped from a CD. It's gotten better recently, but at one point, most digital file distributors only created MP3 versions of albums to sell online from 16-bit files, and I found this out when calling a distributor for a client who simply could not take my 24bit masters as a source for making digital (metadata-tagged) MP3 files. The reason was that their system only accepted 16bit files for the source. I'm not too familiar with the kind of system that is used by most distributors, but this seemed a bit odd to me, since CDs are not perfect, and there's always a degree of lost data that is acceptable for burning CDs (known as C1 or "recoverable" errors). Whenever possible, I provide 16bit (dithered) files and not a CD as a source for distributors, to avoid them ripping files that have some degree of data loss, so when you hear that CD is "lossless", it's not exactly so, unless you're talking about a straight transfer from DDP files, which is 1:1 and even so, the number of CD releases that are done from DDP is quite small, most CD replicators use Red Book CD-Rs, which have a degree of data loss, even the best burns have approximately 0.6 average/sec digits lost.

 

[bigshot]

There is a HUGE difference in sound quality from something originally recorded at 24bit and a 320kbps MP3. I know this because I master audio for a living, and I get to hear what a 24bit master sounds like over a full-range speaker system (B&W) and once in a while, I'll listen to the 320 MP3 (LAME encoded, usually using CBR)..

24 bit to MP3 involves two conversions... 24 bit to 16 and uncompressed to compressed. Which of those steps is introducing the huge difference? Is there a huge difference between 24 and 16 bit uncompressed, or do you hear a bigger difference between 16 bit compressed and uncompressed?

By the way, VBR is better sound quality than CBR with compressed audio.

 

[xnor]

Quote:

There is a HUGE difference in sound quality from something originally recorded at 24bit and a 320kbps MP3. I know this because I master audio for a living, and I get to hear what a 24bit master sounds like over a full-range speaker system (B&W) and once in a while, I'll listen to the 320 MP3 (LAME encoded, usually using CBR).

What track was that? Did you match the levels, and do a ABX test? Anything to back up that claim?
 
 
Quote:

Whenever I hear people say "I can't tell the difference between the CD and the MP3" I then question the playback system (not to mention that they're unaware of the material before it was dithered down to 16 bit, from 24). If it's a pair of headphones (no matter how great the amp and pair of cans are, they're likely to NOT hear a difference because headphones don't resonate through your body the way audio does coming out of speakers).

There's a couple of things wrong with this. First of all there have been tests done comparing 16 and 24 bits and no, there's no huge difference as you claim. Secondly, several listening tests over at HA have shown that headphone reveal compression artifacts more easily than speakers (which is only logical since there is no room acoustics, lower distortion ...). Thirdly, resonating bodies have nothing to do with MP3 compression. Before artifacts appear in the bass, they usually show up quite audibly in the higher frequencies.
 
 
Quote:

The more refined the playback system is (high end DAC, speakers and a system that reproduces from 20Hz to 20kHz as neutral as possible) the more obvious it will be; MP3s sound "narrow" and even though most of us can't hear 16kHz, there will be a sense that something is missing or not completely there when you cut those frequencies out.

MP3 320 CBR doesn't filter out stuff above 16 kHz. As someone who masters audio you should know that.
 
 
Quote:

One thing that is usually not in this discussion is the difference between "hypercompressed" audio that is typical these days on most modern releases and material that has optimal dynamics. When you maximize the levels of a song by use of dynamics processing (upward compression/limiting) you lose low end frequencies and there is little dynamic crest between peak and average levels. Most of the time, you're listening to predominantly middle frequencies and combined with loudness, you start becoming subjected to the Fletcher Munson effect, where we're hearing more mid frequencies (which is natural, as we perceive those than lows and highs). If you take a track that is so loud that you mostly hear mid frequencies, but can't "feel" the low end, you are less likely to be able to hear any difference, especially if you're playing back material over headphones.

Really? Isn't it typically easier to make out artifacts in heavily DR compressed tracks..
 
 
Quote:

I'm not trying to argue with anyone over the above, I *know* this is true because I listen to approximately 40 different songs, every week, that come in as 24bit mixes for me to work with and there is ALWAYS a trade off in sound quality when you 1. apply dynamics processing to make the material louder (therefore reducing the dynamic range) and 2. removing "inaudible" frequencies which are likely more "felt" than "heard".

People *knew* the earth was flat. So what?
 
I agree with 1) but not with 2). The MP3 psychoacoustic model doesn't remove frequencies which are more likely to be "felt" than "heard". Overly simplified: it looks at strong tones and removes much weaker signals that are masked by the stronger tone. Only when you reduce the bitrate the encoder has to remove more and more tones eventually introducing artifacts.
 
 
Quote:

EDIT: One (non-scientifical) observation that I've made for myself is the fact that MP3s sound better when made from the 24bit masters and not from dithered, 16bit files ripped from a CD. It's gotten better recently, but at one point, most digital file distributors only created MP3 versions of albums to sell online from 16-bit files, and I found this out when calling a distributor for a client who simply could not take my 24bit masters as a source for making digital (metadata-tagged) MP3 files. The reason was that their system only accepted 16bit files for the source. I'm not too familiar with the kind of system that is used by most distributors, but this seemed a bit odd to me, since CDs are not perfect, and there's always a degree of lost data that is acceptable for burning CDs (known as C1 or "recoverable" errors). Whenever possible, I provide 16bit (dithered) files and not a CD as a source for distributors, to avoid them ripping files that have some degree of data loss, so when you hear that CD is "lossless", it's not exactly so, unless you're talking about a straight transfer from DDP files, which is 1:1 and even so, the number of CD releases that are done from DDP is quite small, most CD replicators use Red Book CD-Rs, which have a degree of data loss, even the best burns have approximately 0.6 average/sec digits lost.

What MP3 encoder are you using?

 

[bigshot]

I'm suspecting his problem is when he's bouncing from 24 to 16. The wrong dither can indeed make a huge difference. If he's exporting from a 24 bit ProTools mix directly to MP3, ProTools does the downconvert in the background using a default dither. I'm betting that default is set wrong.

Also, xnor, I've met perfectly good recording engineers who knew jack diddly about the MP3 format. If it isn't something they use in their everyday work, they really don't need to know. I think he is mistaken because he misdiagnosed his problem. As a side note, he doesn't seem to know much about the CD replication process either. The only people who manufacture disks from CD-Rs are consumer services like Diskmakers. In industrial applications a glass master would be made from a disk image or raw audio files with a TOC delivered on hard drive.

 

[HiFi1972]

Okay, lot's of comments! Here's what I know (and as an audio engineer, look, we're not scientists - some of these arguments reach scientific levels to the point where it's no longer about what's important, which is perception). All I can comment on is what I've known for a while, and what other engineers concur with (we leave the scientific arguments to "audiophiles"). My responses to the key points:
 
"24 bit to MP3 involves two conversions... 24 bit to 16 and uncompressed to compressed." - No, it depends on what bit resolution the encoder can handle. LAME 3.99 can encode an MP3 directly from a 24bit file (the LAME encoder is one of the most, if not the most popular encoder in the mastering community, likely what most of the MP3s out there have been encoded with).
 
"By the way, VBR is better sound quality than CBR with compressed audio." - Perhaps, the idea of keeping things CBR is (from other engineers who share this POV) to keep the encoder consistent; VBR will reduce the size of the file even more and when creating MP3s, my idea is to keep as much of the data as possible (a bit counterintuitive, I suppose). Pondering whether VBR sounds better than CBR is right up there with saying one fake Sugar brand is better than the next; they're still NOT real sugar
 
"Secondly, several listening tests over at HA have shown that headphone reveal compression artifacts more easily than speakers" - This comment is a bit argumentative, I will avoid that aspect of this conversation, all I have for this comment is this: Not all headphone drivers are the same, and the same thing goes with Speakers. I don't know what was used for these listening tests, the room (who designed it) and how neutral the environment was, all I know is the equipment I use, whose purpose is there for wide-ranging translation of the material I work on to whatever my clients listen to, and besides, who is "HA"!? (LOL, don't answer)
 
"Anything to back up that claim?" - Just a little over 15 years of experience working with loads of recording/mixing/mastering issues from various sources (no science degree here, sorry). Trust me, the only ones obsessing over things like ABX testing aren't typically the people responsible for delivering this content to the masses (tends to be really analytical audiophiles that love to argue more than just chilling out and listening to the tunes!)
 
"MP3 320 CBR doesn't filter out stuff above 16 kHz. As someone who masters audio you should know that." - I mentioned the LAME encoder, other encoders might not do it, but I mentioned that LAME is one of *the* most popular encoders, so likely some commercial tracks that people download have everything past 16kHz filtered out. Versions prior to 3.99 didn't, but I see a lot of MP3 files with missing frequencies past 16kHz; it's not that uncommon.
 
"The MP3 psychoacoustic model doesn't remove frequencies which are more likely to be "felt" than "heard". - On paper, this is the truth. In the real world, I'll tell you what I did one time. I grabbed my wife's favorite CD and made a 320kbps MP3 version of it and I popped it in the car stereo on our way to run errands. After about 10 minutes, she said "Something's wrong with this CD, it sounds like it's going bad or something." My wife completely hates anything to do with analyzing audio and is simply someone who loves music without having any affection to anything technical (my favorite kind of person). Once you start talking science, it's possible to start believing things because someone can present a very articulate reason to the cause; I don't care for written theories of how a person perceives sound.
 
"I'm suspecting his problem is when he's bouncing from 24 to 16. The wrong dither can indeed make a huge difference. If he's exporting from a 24 bit ProTools mix directly to MP3, ProTools does the downconvert in the background using a default dither." - The Pro Tools dither not the best, that's for sure. Again, LAME allows direct encoding from 24 bit files and I personally batch process MP3 files (or simply have the clients send whatever the digital distribution house needs to create their metadata-encoded MP3s for iTunes, etc.) 24bit masters printed into PT are exported without any other processing from my end. I often don't create MP3s for clients, this is often done by distributors and once in a while, I load the MP3s into my system and can tell that there's a huge difference, it's almost like fat free vs. the real whatever it is you want to compare fake stuff to.

 

[El_Doug]

oh boy, the "even my wife could hear it" anecdote... especially ironic, given the paragraph concludes with suggesting scientific evidence can lead one astray from the truth, so we should all trust our own perceptions

 

[nick_charles]

Quote:

Okay, lot's of comments! Here's what I know (and as an audio engineer, look, we're not scientists
 
Actually some of us here are scientists
 
- some of these arguments reach scientific levels to the point where it's no longer about what's important, which is perception)
 
But we know that uncontrolled listening tests using only perception can be influenced by all sorts of factors unrelated to the sound stimulae themselves, our perceptions are so easily misled it is just not funny, for instance you know 24 bit uncompressed is better so you hear it as better, this is a slam dunk psychophysics wise
 
. All I can comment on is what I've known for a while, and what other engineers concur with (we leave the scientific arguments to "audiophiles"). My responses to the key points:
 
This is like the Stereophile response to Masters and Clarks 1980s amp tests, the Stereophile writers know that amps sound different so the test must be wrong !
 
 
"Anything to back up that claim?" - Just a little over 15 years of experience working with loads of recording/mixing/mastering issues from various sources (no science degree here, sorry). Trust me, the only ones obsessing over things like ABX testing aren't typically the people responsible for delivering this content to the masses (tends to be really analytical audiophiles that love to argue more than just chilling out and listening to the tunes!)
 
It would be trivial for you to set up a set of DB tests in Foobar, run 16 or 20 trials and prove that you can reliably tell uncompressed from good MP3, lots of members here do it, this subforum is one of the few where "I know I hear a difference" is just not considered strong evidence.