Can You Really Hear the Difference Between Lossless, Lossy Audio?

CWmike writes "Lossless audio formats that retain the sound quality of original recordings while also offering some compression for data storage are being championed by musicians like Neil Young and Dave Grohl, who say compressed formats like the MP3s being sold on iTunes rob listeners of the artist's intent. By Young's estimation, CDs can only offer about 15% of the data that was in a master sound track, and when you compress that CD into a lossy MP3 or AAC file format, you lose even more of the depth and quality of a recording. Audiophiles, who have long remained loyal to vinyl albums, are also adopting the lossless formats, some of the most popular of which are FLAC and AIFF, and in some cases can build up terabyte-sized album collections as the formats are still about five times the size of compressed audio files. Even so, digital music sites like HDtracks claim about three hundred thousand people visit each month to purchase hi-def music. And for music purists, some of whom are convinced there's a significant difference in sound quality, listening to lossy file formats in place of lossless is like settling for a Volkswagen instead of a Ferrari."

Comments

Depends on the bitrate

[Anonymous Coward]

Usually if the bitrate is above 256kb/s, i dont notice any difference.
Ofcourse it still effects some songs (especially the percussion parts).

Re:Depends on the bitrate

[jlfose]

It could be dependent on the gear that playback occurs on and the quality of the listener's ears. In watching Stan Lee's new show about "superhumans" it becomes clear that some people have, by training or genetics, better reflexes then the bulk of humanity. On my home gear I can't tell the difference above 160Kbs, but I'm more then willing to believe that some people can, either because they have much better gear to listen to, and/or they have superior hearing.

Re:Depends on the bitrate

[Anonymous Coward]

I'd say it depends on what you're listening to.

Most people, including most slashdot armchair pundits, who listen to Lady Gaga or some similar shit will never notice the difference. However, if you listen to things like Tchaikovsky's "1812 Overture", you will notice just how crappy lossy codecs really are. Especially towards the end.

Re:Depends on the bitrate

[Anonymous Coward]

I'd say it depends on the person. Much like there are people who can see more colours, I have no doubt that there also exists people who can perceive subtle differences in sound far better than normal people. In fact, I assume there's probably a term for it, but I don't feel like looking it up.

So while a lot of audiophiles (or perhaps most) are just saying they can hear differences between lossless and virtually lossless... I assume to look "cool", or whatever the appeal is of self-identifying as an audiophile, there's probably a handful that actually CAN. Not to say those $10000 audio wires aren't a complete scam, but it would be foolish to say that there aren't people who have no problem telling the difference between 256kb/s and lossless.

I may be wrong of course, but for a while people didn't think tetrachromacy existed either. And like tetrachromacy, synesthesia, or hyperthymesia, I imagine there's a number of people who possess these traits, but simply aren't aware that they do, assuming that everyone does and that it's normal. Although for the last, I imagine that would be a lot easier to determine.

Re:Depends on the bitrate

[Anonymous Coward]

That's a very apt description. Genetic factor, age, absence of damage, training to understand the difference/subtleties of overtones, and of course the equipment to playback sounds truly. I found the wired article about Peter Lyngdorf and Steinway building speakers good enough to detect the difference between an American and German manufactured pianos a fascinating read. http://www.wired.com/reviews/2012/10/steinway-lyngdorf-model-ls-concert/

Re:Depends on the bitrate

[cayenne8]

Well, the reproduction environment and the equipment makes a lot of difference too.

I mean, if you're only listening to ear buds (even $$$ ones are limited in bass response, etc), or in a car (one of the worst listening environments conceived)....then sure it won't make a difference, and portability makes a lot of sense too.

However, in a nice listening environment, with good equipment...it is worth the effort IMHO.

For instance, I have a pair of Klipschorns ...paired with a couple of the much older models of the Decware SET amps , running mono to each channel..plus an older 15" 800W Klipsch sub, etc......

Even with my older ears, I can hear differences in recordings and formats. Not as well as I used to be able to, but I figure, WHY would I want anything less than the best I can get for the given time/situation? When listening at home, I rip my music to flac, and have it play on my living room stereo.

And hey....kinda fun to watch the Flintstones in concert volume on tv too from time to time, or hell, once hooked the MAME machine to it....Robotron 2084 is fun with the room shaking around you.

God, my neighbors used to hate me when I live in a place where I had to share walls...

Re:Depends on the bitrate

[Joce640k]

You can actually practice listening to music, it's something you learn.

Sometimes the difference between two sets of speakers can be as little as one clarinet in the middle of an orchestral piece. On one set it sounds good, on another it doesn't (or it's hardly there at all).

It's not something you can pick out just by putting on a rap CD for ten seconds and turning the bass up to maximum in a store (which is how most "HiFi" systems are chosen these days and why the manufacturers produce so much garbage).

Re:Depends on the bitrate

[Bengie]

For me, MP3 knocks out a lot of highs no matter the bitrate. Listening to most Jazz really brings out the flaws of MP3.

Re:Depends on the bitrate

[Joce640k]

It's not mutually exclusive. Some of us manage to listen to more than one type of music..._including_ classical.

Re:Depends on the bitrate

[Bengie]

No amount of equalizer tuning will fix a bad lossy compression. When I listen to any music with real horn, string, or cymbals, MP3 literally hurts my ears. It will give me an ear ache and a headache after only a minute or two of listening. Other better compression algorithms like ogg will not do this, even at higher volumes. Pop music does not have this issue for me.

Re:Depends on the bitrate

[The Mighty Buzzard]

Hey, it still counts if you're listening to it on an episode of Tom & Jerry.

Re:Depends on the bitrate

[Panaflex]

10 years ago, MP3 encoders couldn't encode decent cymbals and saxophones below 384kbps... it was just a stream of high pitched garbage.

These days they're both really good encoders. I still prefer AAC over MP3 just because the high freq nuances are better captured, but at AAC@256 and MP3@320, the differences are practically imperceptible to my ears.

The only time I'd look at lossless music is for Orchestral pieces. Compressed pieces still sound flattened and don't have the wideness because there's a lot more overtones, harmonics and variety of tones in live recordings. Microphones, recordings and engineering have adjusted in the past 5 years to compensate - so recent pieces are not too bad however.

Like anything, it's best to just try a few different methods and see what sounds best to you.

Re:Depends on the bitrate

[asliarun]

In my humble opinion, this old hoary debate will always remain a debate for several reasons. As you right mentioned, the reproduction environment in most cases is woeful at best. Most speakers are not even full-range to begin with, their cabinets resonate, their drivers cannot often keep up in complex multi-layered music, their passive crossovers do a half-assed job in distributing the sound to the various drivers, and so on. Then, the amps are weak so they start bottoming out and start clipping when the speaker impedance and phase dips sharply in certain frequency bands. Then the electronics, especially the capacitors and power supply cannot keep up. Then the cables are not fat enough or are not shielded enough so they load up the power amp even more. Then the pre-amp adds its own coloration to the already feeble signal coming from the source. Then the DAC does its own thing and further colors or degrades the source signal even more. Then the source adds its own share of noise and jitter to the audio signal that screws up not just the signal quality (bad enough) but even the timing of the music.

On top of it, the room comes into play. The room adds its own coloration and effect that is often a far bigger factor that the audio system itself - boosting certain frequencies while muddying and deadening others, and even adding echoes, reflections, etc.

Then there is the human being at the end of the chain. I personally can't even listen above 16KHz, and I have average ears. I suspect many people are like me too, at either end of our audible spectrum. On top of it, we humans hear music very differently - while our audio range may be fairly similar (20hz to 20khz by popular definition), our sensitivity to *variations* in tone and timing varies drastically - many often have off the charts sensitivity to even slightly off-key music (I do) or slightly off-beat music (I do not at all).

All in all, a decent headphone setup is far far more revealing than a decent audio setup. At a thousand dollars, you can probably assemble a decent headphone, but an audio system will sound atrocious, unless you are willing to spend a whole lot more effort and research in second hand discrete gear OR are willing to do serious DIY.

Anyway - I also wanted to say one thing - the thing that gets neglected the most in all this is actually the quality of the source recording - or what people call "mastering".

Most people who say something like "SACDs sound far better than redbook CD" or "vinyl sounds far better than CD" are most likely saying this because a whole lot more care went into recording the SACD or vinyl compared to the cheaper mass market CD or mp3.

If I look back at all the albums I have purchased or listened to (in whatever format), the one thing that stands out to me personally is that I have found less than 10% of them to be "recorded with care". And I'm not even being picky! Across the board, I can say that recording quality sucks when it comes to rock (which is what I listen to most often) - and I mean all kinds of rock.

If Neil Young's initiative (and even his Pono device) and Dave Grohl's initiatives are successful in improving the audio quality of music in general, I strongly suspect it will be because recording quality will be done with greater care, not because they decided to use a fancier digital format or use higher number of bits and samples to store their music. While everything becomes a factor by the time the music reaches your ears (heck, by the time it is processed by your brain, you even have to factor in psychoacoustics and gear bias and the "burn-in" syndrome) - the recording quality in general needs to improve (except for the jazz and classical pieces that audiophiles love to love, and are hence recorded with care), and this improvement will arguably make the biggest difference in audio quality.

Re:Depends on the bitrate

[Lagmo]

Quite so, AAC also is perceptibly better on my stereo once you crank up the volume. Digital PWM amplifier (sometimes referred to as Class-D) seems to be especially vulnerable to this harsh distortion in MP3 encodings.
Extra bitrate does mitigate the issue, but with lossless it lessens considerably. But it can never get better than the source material at any rate.

Stuff that usually sounds like crap in MP3 at any bitrate:
Live recordings with lots of audience participation
Treble heavy songs(hihats, cymbals, lots of complex chords)
Heavily stereo panned recordings.

Don't need to go bonkers on HD audio though, a straight FLAC CD rip is perfectly adequate for most listening and can be trivially stored and later downconverted if space is needed, but there might just be that occasional gem that does sound even better in 24bit/48Khz if you have the right equipment(Ears, Hardware and proper training/experience).
Anyone claiming >48Khz samplerate is needed probably just wants to sell something or they have a very large tinfoil hat collection.

Re:Depends on the bitrate

[jonadab]

It also depends on the music you're listening to.

No, seriously. If you're listening to Yousei Teikoku, you can't hear the difference. If you're listening to a string ensemble playing BWV 1080 The Art of Fugue, you can't *help* hearing the difference.

Re:Depends on the bitrate

[catchblue22]

I suspect that if you are willing to spend $10 000 on a sound system, properly installed in the right space, you will hear the difference. But if you are just using your iPod to play your music, there will be no difference at all.

Re:Depends on the bitrate

[AliasMarlowe]

However, if you listen to things like Tchaikovsky's "1812 Overture", you will notice just how crappy lossy codecs really are.

Exactly. Wagner's "Kill the Wabbit" was ruined in precisely this way...

Re:Depends on the bitrate

[c 0xFF]

The problem is the question. "Can you hear the difference?" can never be answered, exactly for the reasons you suggest. "Can you hear the difference under normal listening conditions?" is a much better question.

I can already answer that question, too! The answer is: "Maybe, but only if you really try." Throw in a bit of poor acoustics and other real-world situations, and the 5% difference is lost.

Re:Depends on the bitrate

[Anonymous Coward]

I'm not an audiophile, but I listen pretty much exlusively to Classical music (mostly full Orchestra / Symphonies,) and can tell you that the recording & performance quality is still HIGHLY variable. Even from the same orchestra. I've heard stuff from the London philharmonic that is downright terrible, but they're seen as one of the best. I've listened to Wagner's ring cycle recorded by the Berlin philharmoniker, and is without a doubt the best I've ever heard, even compared to the parts I've heard live.

Re:Depends on the bitrate

[Anonymous Coward]

I'd say it depends on what you're listening to.

The people who care about the difference aren't even listening to the music. Totally different goals.

Normal people use their stereo to listen to music.
Audiophiles use music to listen to their stereo.

Re:Depends on the bitrate

[chmod a x mojo]

It is a mixed bag of all the above plus what you listen to as well. If you listen to stuff that is all in the midranges you won't care if the lows and highs get clipped off since there really are no lows and highs... If you have decent ears + decent equipment + decent surroundings + listen to something with a full range you probably won't want bitrates much lower than 256-392kbps, depending solely on what you personally hear.

Once you start removing parts of the equation the needed bitrates can drop since you won't be able to hear / reproduce the sounds. Ideally when buying ( for audiophiles anyways ) you would get a very high bitrate lossy file or flac file for your main listening area and a lower ( smaller file sizes ), but still decent, bitrate file for your mobile setup.

Re:Depends on the bitrate

[gTsiros]

ignore the DAC the amp the source and everything... ...except the speaker drivers themselves. even the best in the world are wildly non-linear.

and then there's the air between your ears and the speakers

another non-linearity

Best source? .0001% THD. best amp? .0001% THD. Speakers? 1% THD haha good luck.

Re:Depends on the bitrate

[Cillian]

I bought Tchaikovsky's Violin Concerto on Amazon. Great quality MP3s and all, but apparently they were ripped from an ancient piece of vinyl with a frankly ridiculous level of crackle and several loud pops which made me fear for my speakers. (I wouldn't have minded except the 30s preview carefully avoided any obvious noise). Happily I got a refund.

Re:Depends on the bitrate

[Cillian]

We don't need to argue about whether they exist or not, or say they might or probably do exist. Get the people who claim to have magic ears, apply double blind testing, and now we know.

Re:Depends on the bitrate

[AaronLS]

The quality of the composition and style of music seems completely orthogonal to a discussion of audio quality.

Re:Depends on the bitrate

[CrimsonAvenger]

"Spear and Magic Helmet?"

Re:Depends on the bitrate

[Anonymous Coward]

No amount of equalizer tuning will fix a bad lossy compression. When I listen to any music with real horn, string, or cymbals, MP3 literally hurts my ears. It will give me an ear ache and a headache after only a minute or two of listening. Other better compression algorithms like ogg will not do this, even at higher volumes. Pop music does not have this issue for me.

FYI, "ogg" and "mp3" (and AAC too) aren't algorithms. They're audio file formats designed around the concept of permitting the encoder to throw away sonic information that your ears can't perceive. But the file formats don't dictate exactly what's thrown away -- at best they constrain the mathematical set of possible ways to throw information away. Each is flexible enough that there are essentially infinite ways of writing encoding algorithms which emit valid MP3, or valid Ogg, or valid AAC.

Which is a roundabout way of saying: get a better encoder. Whatever one you're using for mp3 isn't doing a good job. Don't get me wrong, there are improvements over MP3 in the newer formats, but they're incremental, not revolutionary. Essentially, MP3 was pretty good, but hindsight identified some ways to increase the possible set of ways to encode things. Assuming state-of-the-art encoders all around, what it really buys you is a higher compression ratio while holding quality constant. For example, AAC at 128 Kbps might sound equivalent to MP3 at 160 Kbps on a particular reference file.

Also, once you have a good encoder and sufficient bit rate, MP3 is going to be just as indistinguishable from the original source material as any other lossy format. The open source MP3 encoder "LAME" has a rep for being capable of extremely high quality at 192 Kbps or more.

Re:Depends on the bitrate

[donaldm]

Of course bit rate is important but there are other important factors to be considered as well especially since mp3 files are normally encoded for 2 or 2.1 stereo and are lossy although it is possible to get 5.1 channel mp3's (see here) which again are lossy. The problem you have with mp3's or any encoded format is the source of the sound and how it was originally recorded as well as the playback sound system including the ears and personal temperament (ie. likes) of the person who is listening.

You are always going to get people who claim that they can detect the difference between lossy audio files to lossless ones and in many cases this will be true especially if you consider 7.1, 5.1 audio compared to 2.1 audio which the majority of mp3's and mp3 playback devices support. Assuming the original sound source was of high quality then an mp3 rip will normally be acceptable especially when the play back device is of good quality and listened to via head-phones or ear buds.

Re:Depends on the bitrate

[PhunkySchtuff]

If I look back at all the albums I have purchased or listened to (in whatever format), the one thing that stands out to me personally is that I have found less than 10% of them to be "recorded with care". And I'm not even being picky! Across the board, I can say that recording quality sucks when it comes to rock (which is what I listen to most often) - and I mean all kinds of rock.

If Neil Young's initiative (and even his Pono device) and Dave Grohl's initiatives are successful in improving the audio quality of music in general, I strongly suspect it will be because recording quality will be done with greater care, not because they decided to use a fancier digital format or use higher number of bits and samples to store their music. While everything becomes a factor by the time the music reaches your ears (heck, by the time it is processed by your brain, you even have to factor in psychoacoustics and gear bias and the "burn-in" syndrome) - the recording quality in general needs to improve (except for the jazz and classical pieces that audiophiles love to love, and are hence recorded with care), and this improvement will arguably make the biggest difference in audio quality.

Yes, this is it in a nutshell. What goes into mixing and mastering an album has far more effect on the final result than whether it's played back as a 96 kHz 24 bit file, or compressed down to a 256 kbs AAC (or, around 4 Mbs compared with 0.25 Mbs).

Whilst the data rate is sixteen times as much for the high resolution audio, there is nowhere near 15/16ths of the sound lost - and even on good quality hifi equipment, I'd challenge anyone to successfully pick the difference in a proper blind test.

What's more, there are now things like Mastered for iTunes which gives a lossy AAC the potential to sound better than redbook CD audio as the AAC files are created directly from the high res masters with, among other things, better floating point conversions and a very high quality sample rate conversion (and, yes, I have verified the quality of apple's "bats" sample rate converter in afconvert)

Re:Depends on the bitrate

[jenningsthecat]

ignore the DAC the amp the source and everything... ...except the speaker drivers themselves. even the best in the world are wildly non-linear.

and then there's the air between your ears and the speakers

another non-linearity

Best source? .0001% THD. best amp? .0001% THD. Speakers? 1% THD haha good luck.

There is a fundamental problem with your argument, and that is the failure to take into account the nature and type of the distortion. It's not your fault - you share the misconception with most audio engineers, (who ought to know better), that THD figures correlate well with listening tests.

Quoting from my own comment in an earlier Slashdot story:

"THD measurements are taken as the ratio of the total power of all harmonics to the power of the fundamental, with no weighting of any kind applied. The trouble is, human hearing doesn't respond to harmonic distortions in this linear fashion - our ears find higher order harmonic distortions much more apparent and objectionable. This deficiency was noted by prominent BBC engineers D.E.L. Shorter and Norman Crowhurst in the 40's and 50's, when they proposed weighting harmonics by the square or the cube of the order; but their voices were drowned out by market forces that wanted a simple, flattering figure of merit that made the newer, more powerful pentode-based amps, (with lots of negative feedback), look better on paper than their lower-powered triode predecessors. The market won out over scientific and technical accuracy, (it usually does), and today engineers the world over, ignorant of this history, mistakenly believe that low THD is the gold standard for measuring and defining audio amplifier quality. (For a good technical analysis of distortion and the sound of an amplifier, see Lynn Olson's excellent investigation)."

Yes, speakers are hugely non-linear - but their non-linearity doesn't make distortions earlier in the reproduction chain inaudible, even though those distortions can be several orders of magnitude smaller. And that applies to all earlier distortions, whether they originate in an amplifier, a DAC, the digital encoding, or the recording equipment itself. Also, an amplifier with 1% THD can sound much better than one with 0.001% THD, not because the distortion in the 'poorer' amp sounds good, but because the distortion in the supposedly 'blameless' amplifier sounds bad. Then there are Intermodulation Distortion and Transient Intermodulation Distortion, which are difficult to measure thoroughly and seldom appear in amplifier specs, yet are often audible.

Audio quality isn't nearly as simple as THD figures imply, nor as simplistic as most manufacturers of audio equipment would have you believe.

Re:Depends on the bitrate

[jenningsthecat]

You can actually practice listening to music, it's something you learn.

Yes, I've noticed that as I've gotten older. Until I was in my late 40's I never cared much about bass, (the instrument, not the frequency range), in most songs. I heard it, but I felt the song would have been pretty much the same without it. Now, I delight in a good bass line - there's a lot going on there that I simply never heard before. I'm also much, much better at picking out the words a vocalist is singing - lyrics are more meaningful now, because I can hear more of what's being said.

I've always loved listening to music; had lots of records and CD's, built and modded my own equipment, did listening comparisons between the same CD pressed in different countries, etc. But in mid-life, the depth of my appreciation for music has grown considerably, and I hear so much more detail in it than I used to.

Re:Depends on the bitrate

[chrismcb]

However, if you listen to things like Tchaikovsky's "1812 Overture", you will notice just how crappy lossy codecs really are.

Right, because only good codes will encode such a precise and perfect instruments as a cannon?
Just because a bunch of noise was created recently doesn't mean it is any less perfect than a bunch of noise that was made 200 years ago. Some people like classical music, and some people don't. And just because you don't, doesn't mean the music you don't like is "shit"

Re:Depends on the bitrate

[Tablizer]

who listen to Lady Gaga or some similar shit will never notice the difference...

Lady Gaga may be considered the female Beethoven 300 years from now. You never really know. Many of the "greats" were considered ho-hum in their day. Bach flunked the Brandenburg "audition" and Mozart couldn't get the top conductor/composer position and title he struggled for. Salieri, whose only claim to fame is (flimsily) alleged sabotage, had a better career money-wise compared to Mozart.

Music critics and fans change their tastes and opinions over time.

P.S. I like some of Gaga's works.

Re:Depends on the bitrate

[DedTV]

There's several sites (Ex. MP3 or Not and Noise Addicts) that have tests where you can find out if you can hear the difference between two different quality sound files on your equipment.

Generally, Lossless audio is like expensive wine or water. Ego drives people to find a difference much more than them actually being able to tell a difference.
If you take boxed wine or tap water and put it in a fancy, expensive looking bottle then have people compare them; there's a few experts and connoisseurs who can reliably tell they're the same but most people will say the liquid from the expensive bottle is superior.
Lossless audio has the same placebo effect. Tell people it's lossless and put it in a 100MB file and they'll be certain it sounds better than a 12MB compressed file. But there's really only a relative few people who can actually hear any difference.

Re:Depends on the bitrate

[Gen_Music]

Not entirely true. The best sounding in the world are not reference, yes, but there are some fairly flat speakers out there. I would be far more worried about transient response times anyway as that has a far bigger impact on the sound.

Re:Depends on the bitrate

[ormondotvos]

So listening to classical music is just striving for audio cred? I thought it was because the music was so good that it's classical.

Re:Depends on the bitrate

[K10W]

actually classical is more forgiving, now technical metal on the other hand has a lot more sound going on and stresses the drivers a lot more. I listen to both metal and classical an have experimented with lossy/lossless boundaries and it is easier to here degradation in metal at reasonably higher bitrate compared to even multi voice classical. Artifacts, bass muddying and bleeding into everything, stereo compression and so on really take it's toll quickly under 200kbs. I listen on sansa clip+, HP amp, sunrise xcape v1 out and about but at home on my PC with headphone amped soundcard and FA03 clone cans I can hear it and especially on denon ah p372 which don't tolerate lossy artifacts very well at all and show it up. As above it depends on mastering of source to begin with as well as ears and equipment.

Re:Depends on the bitrate

[conufsed]

Hey, I listen to both types. Country and western

Re:Depends on the bitrate

[MurukeshM]

I'd agree about the '1812 Overture'. It's difficult to get it to sound anything but crappy in common hardware/software combos.

Re:Depends on the bitrate

[gravis777]

I am going to say that it also depends on what you are listening to them on. A 192kbps MP3 on $2 earbuds is going to sound pretty darn close to the Flac version.

Even for me, a lot of times it depends on just how well it was compressed, and then listening to the original and the lossless side by side. I have had 256kbps lossless files that sound pretty darn good, and only if I listen to the original and am looking for the differences do I notice.

The same can be said for DVD audio versus lossless audio. Like 5.1 Dolby Digital at 256k really doesn't hold up that well on some giant action movie versus a Dolby TrueHD track, although if you are watching a romantic comedy, you probably won't notice much difference. The difference between a DTS soundtrack on a DVD, though, and the DTS MA track is less noticable.

And when you look at a Dolby Digital Plus track versus Dolby TrueHD, you are going to find even less people who notice the difference.

That said, I always perfer the highest quality I can get. Its not necessarially whether or not I can hear it, but that I am not left wondering if there is something going on.

Re:Depends on the bitrate

[cayenne8]

But if you have a choice between the two, why would you choose the lessor quality of the two?

That's kind of my deal...I can buy mp3's...or I can buy a CD and rip it to flac, or buy a flac......with the higher quality one, I keep for the liviing room. I can then rip to a lessor quality one for the iPod for the gym or the car.

But given a choice between two versions, and you KNOW one version is of lessor quality, why would you choose the lessor one?

Re:Depends on the bitrate

[phorm]

Just out of curiosity, how do you find 320 compares VBR? Does a variable rate manage to capture the highs properly at appropriate points?

Re:Depends on the bitrate

[Bengie]

I don't, but my ears feel like they want to "pop", like from a pressure differential, when listening to classical/jazz encoded in MP3. Popular music does not do this, but I assume that's because pop music is highly modified compared to a more "live" style music. My ears will actually start to ring after a bit from MP3 in these cases, but it does take a few minutes. It is very uncomfortable for me.

Fortunately for me, I listen mainly to pop music, so my issues with MP3 doesn't matter much.

I also find it interesting that I get a similar affect when I have only one ear covered with a headphone. This applies to all music for me. Even with the music quite turned down, if I only have one ear covered, that ear will feel like it wants to "pop", but within seconds of covering the other ear or stopping the sound, the sensation goes away. This sensation is almost identical to the above MP3 issue, but quite a bit stronger. This issue takes under 10 seconds for me to notice, while the MP3 issue takes more like 30 seconds. Really depends on the music.

Re:Depends on the bitrate

[Bengie]

Definition of "highs" may be different or my hearing may be different, so it is subjective to definitions and perceptions. Unless someone gives a frequency range with certain amplitude that is.

Re:Depends on the bitrate

[gTsiros]

the main idea is that in the chain source-amp-speakers-air-ear

the worst offenders are the speakers

one could raise a point saying that ears (and the air, perhaps, haven't done any measurements or even *seen* any) are even worse (psychoacoustics etcetcetc) but since we can't do anything about it we ignore them

Re:Depends on the bitrate

[DedTV]

If the lossless file was the same size, price and obtained with the same convenience then sure, I'd take the lossless file. But lossless files tend to cost more, are usually slightly harder to obtain, always take up far more space per file and I can't tell the difference between them and the 256k files I can very conveniently get for $0.99-1.29 on Amazon or iTunes and don't have to spend time to convert it for for the iPod or the car.

Considering all that, why would I choose lossless? The only benefit it gives is that it would sate my ego knowing I had "the best" while having several quantifiable disadvantages.

A lengthy, thorough, and well-explained discussion

[EmagGeek]

There is a long discussion among very qualified individuals on this subject. You can read it here

Depends on the source

[Stentapp]

I am quite sure I prefer a lossy compressed version of a 24 bit, 96 kHz track than a lossless compressed version of a 16 bit, 44.1 kHz track.

Re:Depends on the source

[Hatta]

44.1hkz 16bit audio is completely transparent to the human ear. No one has ever been able to detect when a 16bit DAC ADC pair has been placed in a 24/96 audio path.

Your preference for 24/96 audio as a listener is entirely due to the placebo effect. There are good reasons to master audio in high res, but for listening 16 bit 44.1khz audio is as good as anything.

Re:Depends on the source

[fa2k]

Depends on how good the sound engineers are. A lot can be gained by higher resolution and sample rate in the mastering stage, but by using a good low pass filter and dithering (and dithering is not really necessary, http://developers.slashdot.org/story/13/02/27/1547244/xiph-episode-2-digital-show-tell ) basically all audible information is captured in 44.1kHz / 16. Your speakers probably don't go much above 20 kHz anyway, so anything beyond 44.1kHz will only cause distortion (aliasing), see post by MetalliQaZ "Debunked" below.

Re:Depends on the source

[Anonymous Coward]

44.1hkz 16bit audio is completely transparent to the human ear. No one has ever been able to detect when a 16bit DAC ADC pair has been placed in a 24/96 audio path.

Your preference for 24/96 audio as a listener is entirely due to the placebo effect. There are good reasons to master audio in high res, but for listening 16 bit 44.1khz audio is as good as anything.

This man says the truth.

Re:Depends on the source

[fatphil]

> I am quite sure ...

In other words, you've never done an ABX test and are just spouting ill-informed supposition. The ABX is the gold standard, get back to us once you can distinguish those sources that way with a 95% confidence level.

Re:Depends on the source

[Rougement]

Mine are flat up to 50kHz. The problem with 44.1kHz is that the highest frequency possible is 22.5kHz, dangerously close to the upper range of human hearing of around 20kHz. Add to that possible DAC high pass filtering artifacts, etc and there's a good argument for moving to a sample rate of 48kHz or higher.

Re:Depends on the source

[QRDeNameland]

Your preference for 24/96 audio as a listener is entirely due to the placebo effect.

Well, in all fairness, listeners may actually hear perceptible differences between 24/96 and 16/44.1 audio sources due to different mastering, but of course that says nothing about whether they can actually tell the difference between the two bitrates when everything else is equal.

This article is a pretty good explanation of why 16/44.1 is as good as anyone needs for playback.

Re:Depends on the source

[femtobyte]

You sure can hear the difference if you stick a 44.1kHz DAQ in a 96kHz signal chain before filtering out ultrasonic high frequency components (if there are enough to make a difference). The advantage of 96kHz recording isn't that it can capture any more human-audible frequencies than 44kHz can, but that you have a lot more leeway to prevent aliasing of signals above the Nyquist limit down into the audible range (a 25kHz tone sampled at 44kHz results in a spurious, highly audible (25-44/2)=3kHz aliasing signal).

It's pretty much impossible to build analog frequency filters with a sharp cutoff (e.g. everything below 20kHz and below gets through, everything above 22kHz is -60dB attenuated), so recording at 44.1kHz sampling requires either being absolutely certain the original sound source has minimal high-frequency harmonics, or heavy analog filtering that cuts well into the audible high frequency range. With 96kHz sampling, it's much easier to build an analog filter that gradually rolls off high frequencies between 20kHz and 40kHz (...producing a >40kHz sound is tricky in the first place), preventing aliasing without the filter cutting into the audible range. Once digitized, it's trivial to make a *digital* filter with a perfect frequency cutoff to downsample the 96kHz to aliasing-free 44.1kHz.

Re:Depends on the source

[hairyfeet]

You are 100% correct, I have sat in a $100k studio with $5k reference monitors and heard my tracks played back at both 192k and at 44.1k and honestly? Couldn't tell the difference, i really couldn't. And while my midrange hearing may not be the greatest I'm picky as hell when it comes to low end and that is usually the first thing that goes when you compress but standard 44.1k? Couldn't tell the difference which if there was gonna be a difference i would have heard it on that system, it was top notch. I'm sure many here can bring citations showing double blind tests which i have no doubt show its all placebo, because if I can't hear it in a nice studio with the actual live instrument right beside it i doubt seriously anybody is gonna hear a difference with home gear, even high end home gear.

Re:Depends on the source

[chipschap]

44.1hkz 16bit audio is completely transparent to the human ear. No one has ever been able to detect when a 16bit DAC ADC pair has been placed in a 24/96 audio path.

Your preference for 24/96 audio as a listener is entirely due to the placebo effect. There are good reasons to master audio in high res, but for listening 16 bit 44.1khz audio is as good as anything.

As a former audio professional (specialized in location recording of choirs and orchestras) I must agree. But even my aging ears can hear the difference between 44.1 (or 48)kHz 16 bit uncompressed and a typical MP3. Side note: 24-bit has a few audible advantages for music with extremely wide dynamic range (from ppp to fff, say) where 16 bit will struggle a little at the very soft end.

Re:Depends on the source

[amicusNYCL]

FM radio is a touch above phone quality, *everything* sounds compressed. Not to mention the static mixed in.

Re:Depends on the source

[dgatwood]

Speaking as someone who frequently does recording, your comment suggests that no one has done that test with classical music in a properly controlled listening environment using quality gear while giving the test subject the ability to control the volume arbitrarily. When you crank up the volume, the noise floor difference in soft passages alone should make the difference between 16-bit and 24-bit signal paths a dead giveaway, even for someone with moderate to severe hearing loss. It isn't even subtle. Of course, if the person doesn't turn it down for the loud passages, he/she will likely suffer hearing damage, but perhaps that's why he/she has moderate to severe hearing loss in the first place. :-D

The 44.1 vs. 96 kHz difference is more subtle, requiring someone with top-notch hearing (very rare), headphones that can accurately reproduce frequencies above 20 kHz, and 96 kHz DAC hardware that does not have a bandpass filter starting at 16 kHz. If you fail to verify even one of those requirements, you would expect no one to be able to hear the difference, because there won't be any difference.

Re:Depends on the source

[Dahamma]

No, not at all like 640K.

Re:Depends on the source

[QRDeNameland]

kinda like 640K?

Unless you want to argue that human hearing is improving similarly to Moore's law, then no.

Re:Depends on the source

[Dorianny]

Agreed but this is largely due to the power of suggestion rather than the placebo effect. If I suggest to you that the music will sound better because it is encoded in a better format, than you will expect it to sound better and likely conclude that it does so even if your brain really can't tell the difference.

Re:Depends on the source

[hedwards]

The point of the equipment is that you have quality in reserve as you go through the process of mastering the tracks. The more quality you have in reserve the more you're able to do before you start having to deal with artifacts and other nastiness. As with all such things, you have to think about the order in which you do things and the order in which you throw out data to get the best results.

The point of buying lossless music isn't so much that it's better for listening to, it's that you can compress it however you like later on without having to worry as much about the sound quality you get. Since you have more data to work with, you can get a better quality at a lower bitrate than if you were starting with an already compressed track.

Re:Depends on the source

[hairyfeet]

Well to be really REALLY fair I have noticed it also matter if the original music was recorded in analog or digital, as I've taken some tracks we've cut in a classic studio with the analog 8-track and its really fricking hard to get those to sound really..."right" for want of a better term as its really hard to describe, when it is compared to digital.

The closest I can get to describing it is this and sorry if you aren't a musician but they'll know of which I speak...you know how you have that great old tube amp for the guitar and it has that nice warm fat feel to it? Notice how the same amp when modeled digitally doesn't doesn't quite have the warmth? Its kinda...artificial sounding? That was the trouble we had, the tapes sounded nice and warm but trying to get that to switch over to digital was fucking hard, frankly it was easier to just cut the tracks again in a digital studio than it was to get the analog tapes to really convert well.

Sorry if I'm not describing it correctly but music is one of those things where my terminology often fails me, its so hard to describe feelings and emotions and music for me is an emotional expression so I end up having to try to describe how I felt as I listened or played and my vocabulary fails me, the analog was a little fuzzy but it was warm and lived in feeling while trying to convert that to digital something was lost in translation, no other way I know how to say it. the same tracks recorded natively in digital sounded great, analog sounded great, but putting the two together was just something we never could get to work.

Re:Depends on the source

[Dahamma]

You don't have to do a personal ABX test when there are many others who have done them and confirmed his statement. In fact, it's a much more powerful statement citing many others than just yourself. One is a statistic and the other is an anecdote.

And for a MUCH more exhaustive and scientific discussion than any post on this article will ever make (anther post in this thread already linked it, but you must have missed it, and it's a great article): http://people.xiph.org/~xiphmont/demo/neil-young.html

Re:Depends on the source

[Jane Q. Public]

"There are good reasons to master audio in high res, but for listening 16 bit 44.1khz audio is as good as anything."

The reasons for having "extra" fidelity in master recordings is the same reason for having high-resolution photos in "raw" format: there is lots more wiggle room for editing while still maintaining good enough fidelity that the end user can't tell the difference.

For example: take a large (say 16M pixel) 8 x 10 photo, and reduce it to 4 x 5 at 600 dpi. Then take the same photo, edit it (for example, change some colors, remove a cloud from the sky, etc.) and reduce that to the same size and resolution. Even though the resulting photos are higher resolution (at arm's length) than the eye can perceive, they look different.

Re:Depends on the source

[juniorkindergarten]

I will tell you now that the average person cannot hear to 20khz. Young children can. Anybody who has listened to loud music for any length of time have blown away the top couple of khz of their audio range.
If you have ever gone to a rock concert and been near the front or gone to most dance clubs and you will have sustained hearing damage. If you have ever left one of these venues with ringing ears, or been around loud machinery and noticed the same, then you have sustained hearing loss. Your hearing will recover mostly after the trauma and that will be indicated by the subsiding of the ringing of your ears.
If you want to find out how your good/bad hearing is, spend the money and see an audiologist. You will be surprised on to find out what your hearing is really like.

Re:Depends on the source

[Damnshock]

Humans can usually listen to sounds from 20Hz to 20kHz but that doesn't mean than there might be some people that are able to listen to higher frequencies, say 22,5kHz for example

Now, as per the Nyquist-Shannon theorem we need to sample at double the bandwidth ( f>2B ) in order to get *all* the information from the source. 22,5*=45 which is higher than 44,1... therefore, there are few people that *do* actually notice the difference between 44.1kHz sampling and 96kHz ( although that is very very rare).

Re:Depends on the source

[fa2k]

Yeah, 100 % agreed. It's not like we can't afford at worst 9 % extra storage space, just to have a little more headroom

Re:Depends on the source

[femtobyte]

Right, which shows why 96kHz digital sampling *is* critical, even if you immediately digitally downsample on-chip before passing it along to the next device in the processing chain.

Re:Depends on the source

[Jane Q. Public]

When I wrote "take the same photo", I meant the original, not the reduced copy.

Re:Depends on the source

[Anonymous Coward]

According to Wikipedia the audible range for human hearing is around 130dB. 16 bits can in best case offer a dynamic range of 96 dB, whereas 24 bits offer 144 dB.

So it should be pretty obvious that you can't fit the entire audible range into 16 bits. This might not be relevant to modern day music. But if you want to record what the ear is actually capable of hearing (not including sound levels above the pain threshold) you will need those 24 bits.

Re:Depends on the source

[ozydingo]

Two nits to pick:
1) You can get arbitrarily close but you can't get "perfect" frequency cutoff.
2) A 25 kHz tone sampled at 44 kHz gives you a 19 kHz tone. Remember the [-pi:0] (or [pi:2*pi]) frequency range comes first.A 41 kHz tone would get you a 3 kHz tone after sampling.

Otherwise all true, which is why most recording devices do exactly that, sample at a high rate and digitally filter before downsampling to 44.1. But none of that has much to do with whether or not, once you've gotten past the aliasing problem as you say, you can tell the difference between a 44.1 kHz playback and a 96 kHz playback.

Re:Depends on the source

[hondo77]

The 44.1 vs. 96 kHz difference is more subtle, requiring someone with top-notch hearing (very rare), headphones that can accurately reproduce frequencies above 20 kHz, and 96 kHz DAC hardware that does not have a bandpass filter starting at 16 kHz. If you fail to verify even one of those requirements, you would expect no one to be able to hear the difference, because there won't be any difference.

And that's really the heart of the matter, right there. I can hear the difference between uncompressed and compressed. No, really, I can. However, I do all my listening, these days, in conditions where 256 mp3 is just fine. Headphones at work, through my iMac at home, through the nice home stereo but while I'm doing stuff. For that, the tradeoff between space and quality makes uncompressed not worth the disk space.

Re:Depends on the source

[Dahamma]

If you consider 15Khz stereo a touch above 3500Hz mono with lower S/N, sure. Though I agree it's a joke to claim you can hear MP3 compression artifacts though that 15Khz and relatively poor FM S/N.

Re:Depends on the source

[Jane Q. Public]

This.

Re:Depends on the source

[hairyfeet]

I think the problem is people listen to crap once, think "oh well if X is crap then X-Z is crap" and there ya go. I mean sure you listen to a 128k MP3 and its gonna suck on decent phones, but just because 128k sucks on decent phones doesn't mean 256K and up is gonna suck, its really apples to oranges.

Personally I think they all have their place, i use low bitrate in my truck because all the background noise means 128k will do just fine there whereas in my apt I prefer 320k, its all about the right format for the right place.

Re:Depends on the source

[loneDreamer]

22.5kHz, dangerously close to the upper range of human hearing of around 20kHz

Man, we clearly have very different definitions of "danger" :-)

Re:Depends on the source

[femtobyte]

1) Digitally, yes you can. Take the DFT of the data; zero out all components above your frequency cutoff; reconstruct the signal as the sum of below-cutoff frequencies. Voila, a perfect sharp cutoff. The only subtlety is that you can only choose an exact cutoff corresponding to some integral number of cycles in your sampling window, so you can't cutoff at exactly sqrt(e*pi)kHz --- but you do have plenty of wave numbers from which to select a perfect cutoff (increasing with the size of your DFT window).

2) Untrue: a 44kHz *sampling rate* has a 44/2=22kHz Nyquist cutoff. Frequencies f>22kHz Nyquist limit "wrap around" to f-22kHz difference frequencies.

But yes, I agree, on the playback side there's no audible difference between a (sufficiently well made) 44.1kHz and 96kHz DAC.

Re:Depends on the source

[Technician]

For most source material, this is true. Take a Denon Technical Audio CD from way back when, recorded from a digital source (the original DDD mastering) and play it on any current system. Use the 20Hz to 20 KHZ sweep.. I have not found a system yet that does not have ailising at the upper end of the sweep that is audiable. This is UNCOMPRESSED 16 bit 44.1 KHZ stereo encoding.

You may be able to find the track online, but avoid any compressed versions of it. Compression makes the matters much worse.

I have the original CD.
http://avaxhome.ws/music/denon_tech_cd.html

Re:Depends on the source

[fatphil]

I've read that Xiph article several times. It brought back happy memories of the work that I did decades ago in the field. That article fully supports my point of view, and negates the parent poster's - "I am quite sure I prefer" is *meaningless*, scientific testing is all that matters - just see the "Listening tests" section, for example.

Re:Depends on the source

[locopuyo]

I am a dog you insensitive clod!

Re:Depends on the source

[jonsmirl]

When the music gets soft in 16b you have a lot of zeros in front of the number. So you effectively only have a three or four bit signal being fed into the DAC. This is fixed point math, not floating. With 24b you can put all of those zeros in the front and still have eight or more bits to feed into the DAC. This is even more beneficial when the amp implements power supply volume control. PSVC raises the effective noise floor the DAC has to deal with.

Re:Depends on the source

[router]

I second this, I couldn't hear above 16kHz by my early 30s (using a frequency generator and oscope to verify). I think folks get caught up in the theoretical, not the reality.

andy

Re:Depends on the source

[Rougement]

Check out the low pass filter slope on a DAC, it's not like 20kHz is played back accurately and 20,001Hz is filtered completely. Filtering causes distortion. I'm just saying that it's better to take all of that and move it up to 24kHz by using a 48kHz sample rate. Ideally it would be even higher at 60kHz. 96kHz is overkill once the material has been mixed and mastered though.

Re:Depends on the source

[Entropius]

Yes! Someday, instead of having real dog whistles, we'll just play back mp3's of dog whistles for our dogs, and those will only work if recorded in 24bit/96kHz!

Also Monster Cable.

Re:Depends on the source

[Entropius]

Doesn't this perfectly sharp cutoff also require an infinite temporal extent (i.e. each sample in the output is a linear combination of an infinite number of samples in the input) to achieve?

Re:Depends on the source

[Entropius]

OT, as a choral performer:

Classical music has a stupid wide dynamic range, more than any other genre I know of, and (in particular) soprano sections have a nasty talent for pegging meters that were supposed to be set with plenty of headroom.

Re:Depends on the source

[ozydingo]

(This is fun; I know we agree on all substantive points but I'm still going to take you on here :-)>

1) This is a misconception. The DTFT only represents samples of the DFT, and you can only work with the DTFT with any real machine with finite computing resources. If you zero out the DTFT samples, you are *not* zeroing out all DFT samples in between them.
Example MATLAB code:
x = [1 0 0 0 0 0 0 0];
X = fft(x);
Y = X; Y(5) = 0; %zero out the highest frequency component
y = ifft(Y);
stem(abs(Y,8)) %Look at the pretty DTFT with zero amplitude at the pi frequency component!
stem(abs(fft(y,256))) %Plot a finer sampling of the DFT. What happened to your perfect cutoff??

2) True, despite the 22 kHz cutoff. f>22 kHz wraps around to the negative frequency region first. That is, w>pi wraps around to the [-pi:0] region before getting back into the [0:pi] region; remember, we have 2*pi periodicity in the DFT, and 0:pi here represents 0:22 kHz. 22:44 kHz is pi:2*pi, which by periodicity is the same as -pi:0. An aliased, rising tone falls continuously from fs/2 to 0 before rising again.

Your move, good sir

Re:Depends on the source

[arth1]

But yes, I agree, on the playback side there's no audible difference between a (sufficiently well made) 44.1kHz and 96kHz DAC.

No, but what makes a big difference is when you have a 48 kHz sound card that resamples everything to 48 kHz for an internal DSP stage that cannot be bypassed, and then back again. Yes, Soundblaster Audigy, I'm looking at you.
44.1 -> 48 kHz gives a lot more audible artifacts precisely because they're so close. Think of it as audible moire.

Also, for newer computer audio cards, if you have a choice, use 88.2 kHz for the internal rate instead of 96 kHz. The reason is that most high quality sound is in 44.1 which converts perfectly to 88.2. For 48 kHz, it's less of a problem in the first place, and likely also worse quality sound to start with.
Of course, unless the rest of the audio path is good, it doesn't matter much, but if you like to listen to FLACs with high end headphones, it sure won't hurt to use 88.2 instead of 96 kHz.

Re:Depends on the source

[nabsltd]

The closest I can get to describing it is this and sorry if you aren't a musician but they'll know of which I speak...you know how you have that great old tube amp for the guitar and it has that nice warm fat feel to it? Notice how the same amp when modeled digitally doesn't doesn't quite have the warmth?

The reason for this is that it's hard to capture distortion accurately.

That "warm sound" is a result of the inacurracies of the tube amp. You may like it better (and that's just fine), but it is does not accurately reproduce the original signal. For me, it's really no different than the current "loudness war" where re-mastered releases are much louder. Many of today's listeners like that sound beter, but it isn't accurate.

Re:Depends on the source

[Omestes]

The point of the equipment is that you have quality in reserve as you go through the process of mastering the tracks.

This is how I've seen it as well. Its like the difference between a RAW file, and high quality jpeg. The jpeg is good enough for normal use, but you pretty much kill all the information you need for further editing during the compression process. The RAW is your master, but is pretty pointless for for anything else due to its size.

Re:Depends on the source

[femtobyte]

In a finite window, *any* signal can be represented as a sum of elements with frequencies corresponding to n=0 (DC offset), 1, 2, 3, ...., infinity integral cycles in the window. A signal corresponding to a non-integral number of cycles, e.g. 100.5, is indistinguishable over the window from some (infinite) combination of integral cycle waves. If you measured in a window twice as long, the 100.5-cycle signal would now be a unique, identifiable 201-cycle component. So, in an important sense, in a finite window the "intermediate" frequencies "don't exist" --- they can't do anything different from the (infinite series) of integral frequencies. Thus, you can create a cutoff that is as "perfect" as is meaningful in a finite window.

Re:Depends on the source

[Goaway]

Try listening to them and then tell us the difference.

Just because you can tell the difference between a 2 kHz sine and sawtooth wave does not mean you can do the same at 20 kHz.

Re:Depends on the source

[Rougement]

My hearing tails off at 16 khz too. Even so, why design a system that works for most people when it's just as easy to design one that works for everybody?

Re:Depends on the source

[Joce640k]

This is why they do a "radio edit" of popular songs.

It's really a "radio mix" - zero dynamic range, designed to be broadcast over FM and listened to on crappy speakers over a backdrop of engine/tire noise.

Re:Depends on the source

[crgrace]

It's pretty much impossible to build analog frequency filters with a sharp cutoff (e.g. everything below 20kHz and below gets through, everything above 22kHz is -60dB attenuated), so recording at 44.1kHz sampling requires either being absolutely certain the original sound source has minimal high-frequency harmonics, or heavy analog filtering that cuts well into the audible high frequency range. With 96kHz sampling, it's much easier to build an analog filter that gradually rolls off high frequencies between 20kHz and 40kHz (...producing a >40kHz sound is tricky in the first place), preventing aliasing without the filter cutting into the audible range. Once digitized, it's trivial to make a *digital* filter with a perfect frequency cutoff to downsample the 96kHz to aliasing-free 44.1kHz.

But the fast majority of analog-to-digital converters used for audio use delta-sigma modulation. They are already sampling far above 96 kHz (delta-sigma modulation is a combination of oversampling and quantization noise shaping).

Your argument is specious. If audio converters used Nyquist-rate ADCs I would agree with you, but they don't. The absolute vast majority of audio ADCs are of delta-sigma type so they are already doing your "trivial digital filter with a perfect frequency cutoff to downsample". It's an inherent part of the modulation.

Re:Depends on the source

[ChrisMaple]

16 bits only provide a 98 dB dynamic range. If the system is set up not to clip at 120 dBA, a 10 dBA signal will disappear entirely in a 16 bit signal path.

Re:Depends on the source

[Overzeetop]

Actually, you've proven the GP's point. You can't tell the difference if you are listening to the program. Turning a program up in the "soft sections" is exactly what you should never, ever do when listening to a program. You may as well put on the IR headset with compression that came with your TV so you can watch late night TV without disturbing your wife.

Mastering is an entirely different ball of wax and, yes, you want all the headroom you can get. It's no different than photographers using RAW formats instead of JPGs (even lossless JPGs) out of the camera. You want all the bits you can get. But after your done mastering, dropping to 16bits isn't going to affect the outcome. That's the whole point of mastering - if we didn't want to be that soft, we would have engineered it to be louder.

Re:Depends on the source

[Goaway]

No, really, I can.

In ABX tests?

Re:Depends on the source

[guantamanera]

I can't hear the difference, but is weird I get different feelings. Maybe just because you cannot hear does not mean exactly you will not feel it. Bethoven was deaf and he used to lie on top of his piano so he could feel it. Get a high quality sample of an explosion and encode them at different levels and I bet you will feel the difference.

Re:Depends on the source

[ChrisMaple]

Modern (last 20 years) audio ADCs are of the delta-sigma type that effectively sample at a multiple (8 or higher) of the output sample rate. Filtering is applied in the digital domain using FIR filters with very sharp corners and no phase distortion. An analog filter of similar quality is quite simply impossible.

Re:Depends on the source

[femtobyte]

Same response as I gave to the AC post above making the same point: this indicates a non-specious importance for the initial >>44.1kHz sampling, even if the on-chip digital circuitry immediately downsamples to properly-filtered 44.1kHz. It's a semantic distinction whether you should really call an internally-higher-than-44.1kHz-sampling DAC a 44.1kHz DAC, given that it is relying on information outside the 44.1kHz sampling range to produce the output rendered down to 44.1kHz. But yes, I agree that in practice dropping a good "44.1kHz" DAC (which is "really" a 96kHz DAC + internal downsampling) into your 24/96 signal chain won't result in audible problems.

Re:Depends on the source

[fatphil]

Agree. The people who are making the statements are not the ones who are informed enough to be able to make such statements. I didn't ABX, but I did do the closest blind test that my g/f could manage via the medium of just having 2 xmms's open, and I came up with a 160k-bad-192k-OK guideline. When VBR became widely supported, it was clear that 320kVBR, averaging at between 200k and 220k, was all that I would ever practically need, and was also small enough to be convenient in any context. Enough headroom, but not too much waste.

I still buy mostly CDs, but they typically only get played once to rip them. (In particular since my Nakamichi CD player got fucked up when I moved flat, so I have to play them via my PC's soundcard anyway now.)

Re:Depends on the source

[turkeyfeathers]

When the music gets soft in 16b you have a lot of zeros in front of the number.

I just listened to Smokey Robinson and The Miracles's "Oooh Baby Baby" and as it fades out at the end I heard him sing "Oooooooooooh Oooooooooh, Baby Baby, Ooooooooooohoooooooooooooo...". But the end was so quiet that I didn't hear the number after all the zeros.

Re:Depends on the source

[chuckugly]

It can faithfully record all the signal below the Nyquist threshold in both examples. As long as that threshold is high enough, the ear can't tell the difference. TL;DR; - they both sound like a 22KHz sine wave to your ear anyway.

Re:Depends on the source

[Atzanteol]

This article is a pretty good explanation of why 16/44.1 is as good as anyone needs for playback.

kinda like 640K?

Not even a little.

Re:Depends on the source

[toadlife]

You just described the placebo effect.

Re:Depends on the source

[ozydingo]

Your 44.1khz sampler can't distinguish them.

And I'll bet neither can you. The sawtooth is the sine wave plus a lot of higher-frequency harmonics. But your ear can't detect any of the higher-frequency harmonics of a signal with 22kHz periodicity, so you probably can't tel them apart.

Re:Depends on the source

[dgatwood]

Turning a program up in the "soft sections" is exactly what you should never, ever do when listening to a program.

That's not necessarily the case. Consider a classical piece with three movements. The second movement is soft and slow. The entire CD is almost certainly mastered so that the relative volume from one track to the next is preserved—that is, the soft movement will be significantly quieter than the other two movements.

If you are listening only to the soft movement, however, it is perfectly reasonable to crank up the volume so that you can hear more of the details—details that could easily be buried in the digital noise when listening to a 16-bit recording.

Re:Depends on the source

[femtobyte]

The trick you're playing on yourself here is:

x = [1 0 0 0 0 0 0 0]; % x is only defined on 8 samples over the interval. There are an infinite number of continuous signals that could be sampled this way.

Following your procedure through to y:
octave:5] y = ifft(Y);
octave:6] y
y =
      0.87500 0.12500 -0.12500 0.12500 -0.12500 0.12500 -0.12500 0.12500

so y is also defined at 8 sample points; as for x, there are an infinite number of curves that could fit these. One of these curves is the sum of frequencies indicated by Y. But what does fft(y,256); mean? From the Matlab documentation,

"Y = fft(X,n) returns the n-point DFT. fft(X) is equivalent to fft(X, n) where n is the size of X in the first nonsingleton dimension. If the length of X is less than n, X is padded with trailing zeros to length n."

So, now you have y defined in a larger window (y = 0.87500 0.12500 -0.12500 0.12500 -0.12500 0.12500 -0.12500 0.12500 0 0 0 0 0 .... 0). See my response above to another poster's question: when you enlarge the sampling window, you "create" a lot of possible "intermediate" frequencies that "don't exist" (i.e. are indistinguishable from sums of integral frequencies in the shorter window). By padding y with zeros to a larger window, you're looking at a *different* signal from the un-padded y alone; consequently, you need the "extra frequencies" that you ascribe to the "non-sharp-cutoff" to correctly describe the different "y+0,0,0,0,...,0" signal (which is distinct from y). But that doesn't mean the cutoff isn't perfect as defined on the original signal x->y. In fact, if you periodically *repeat* y (y->y,y,y...,y instead of y->y,0,0,0...) you'll see the "sharp cutoff" still applies since the periodic signal is still the sum of the original frequencies in y.

Re:Depends on the source

[djdanlib]

As a live sound engineer dealing with vocalists who do that regularly (sing at normal program levels and then BELT A PHRASE OUT)... let me say... ARGH.

I put a steep compressor on someone who's prone to doing that, and let me tell you, it makes my life much easier. I can't fix the clipping, but I can make sure they don't cause the audience to cover their ears.

Re:Depends on the source

[EvanED]

Thanks for the link. I've seen it before, but I didn't notice the footnote which leads to this post which gives the first actually convincing explanation I've seen (not that I've looked very hard) for the adequacy of 16/44.1. It's easy to find people who go "Nyquist's theorem! Your limit of hearing is 20Khz!" but that ignores the effects of the bit rate, and that poster did not.

Re:Depends on the source

[QRDeNameland]

The article sounds convincing, unfortunately it is completely false.

linked PDF

I gave this a quick skim, and couldn't help but notice that it says "There is no convincing argument for using 24-bit data in a distribution format" and under the heading "Do we need more than 44.1KHZ?", he pretty much hedges both ways and does not present a convincing case that any actual human being can actually perceive a difference between 44.1 and higher sampling rates.

I won't say it doesn't present a valid alternative view to the Xiph article, but to say it renders the article I linked "completely false" is a fair stretch.

Re:Depends on the source

[slinches]

I agree, I just wish there more good sound engineers. It's inexcusable, but way too common, to see horribly clipped signals with no dynamic range on 44.1kHz / 16bit CDs. The only reason I prefer 24bit files is that they haven't been as badly affected by the loudness war.

Re:Depends on the source

[t4ng*]

The vast majority of tube amps are Class A amplifiers that cause non-linear distortion of the input signal. This distortion, especially when over-driven, adds even harmonics to the signal which is what you, and other musicians that got their start with tube amps, describe as "warm" and "fat." It has absolutely nothing to do with the sound being more accurate; in fact, it is less accurate.

Class B amplifiers became the norm during the switch to transistor circuitry. There is no reason you can't build a Class A transistor amp, or a Class B tube amp, but Class B amps are more efficient and cheaper to build. Class B amps are actually two amplifiers, one amplifying the positive part of the signal and the other amplifying the negative part of the signal, with each half being mixed together in the output. This causes what is called cross-over distortion. Early designs with more cross-over distortion would add odd harmonics to the audio, which people perceived as a "cold," "harsh," or "metallic" sound. Modern Class B amps, even cheaply made ones, have very little of this distortion now.

Bottom line is that the warm, fat feeling you fell in love with was distortion, not reality.

Re:Depends on the source

[w0mprat]

Your preference for 24/96 audio as a listener is entirely due to the placebo effect.

Well, in all fairness, listeners may actually hear perceptible differences between 24/96 and 16/44.1 audio sources due to different mastering, but of course that says nothing about whether they can actually tell the difference between the two bitrates when everything else is equal.

This article is a pretty good explanation of why 16/44.1 is as good as anyone needs for playback.

There's plenty of articles from "experts" on why 16/44.1 is all you need, however these kind of opinions risk being being wrong: What about actual data? I see little solid data where the hypothesis has been put to the test.

I think the point is, some say 24/96 survives lossy compression better, it also produces less artifacts as the higher frequencies have less data points to describe their waveform. Perhaps some won't hear the difference in uncompressed audio, but I bet some can hear the difference in compressed.

Re:Depends on the source

[ultrasawblade]

I won't believe that support for sample rates that can record frequency ranges above 20KHz are for any other reason than to embed watermarking data streams in the inaudible ranges - think something like Cinava. Same thing with any color depth over 24 bit.

Disclaimer: The above is a joke.

But maybe I really want to record a song with parts listenable only by my dog or pet bats.

Re:Depends on the source

[Lagmo]

Actually what people may be hearing is atrocious sample rate conversion, they might benefit from improved up-sampling algorithms from 44.1Khz -> 48/96Khz. Some soundcards and drivers are notorious for this(looking at you Creative Labs) but that's probably only after you've spent an hour just finding out how to turn off all the default EQ booster/plug-in/effect crap as well.

Re:Depends on the source

[femtobyte]

TL;DR version of my verbose answer above:

Tacking on 248 samples of "0" to an 8-sample signal isn't the same as "doing nothing"; from a mathematical standpoint, it's little different from tacking on 248 samples of "airplane full of nails crashing into a chalkboard factory". It's true, there's no digital filter I can apply to my 8-sample-long window to control the frequency spectrum if you expect to listen to my 8 samples, then tack on 248 samples of your own noises (including silence) and FFT the whole stretch. But, I can filter the 8-sample segment so that, measured within that 8-sample window I was responsible for filtering, the frequency spectrum has a "perfect" cutoff (not my problem what you do before or after).

Re:Depends on the source

[ChrisSlicks]

This.

Also the steppings at the low end of the sampling range are pretty course when you consider the energy level on the dB scale. The audio is sampled linearly, however it's energy can be interpreted as 20 * log10(iSample/iMaxVal), which you will see the stepping gets increasingly course as you get into the lower amplitudes, especially as you approach -60dB for 16-bit (relative to the 0dB maximum). The last 36dB from -60 to -96 is represented by only 32 values in 16 bit (just 5 bits). The same thing with 24 bit sampling would have 256 as many steps thanks to that extra 8 bits, so 8192 values instead of 32.

So yes, while 16-bits might theoretically have 96dB of range, in reality not all that range is really useful in terms of quality.

Re:Depends on the source

[ozydingo]

I'm about to head out but I will respond quickly. I can get back at it tomorrow.

There are an infinite number of continuous signals that could be sampled this way.

But only one for an ideal band-limited sampling / reconstruction scheme such as pulse-amplitude modulation.

My main point is that those "intermediate" frequencies do in fact exist and are important for the analog reconstruction of the signal, even if they do not contribute any extra information to your digital signal (and they don't). You are not creating them by using a larger window, that's just a mathematically-equivalent operation to computing more samples of the DTFT. Were you to put y through a DAC, those frequencies will be there in the FT of the analog signal. The samples of the DFT of y produced by zero-padding the 8-sample signal (you read the documentation correctly) before taking the DTFT tell you exactly what the value of the DFT are at those "intermediate" frequencies.

The bottom line argument is that you cannot represent a finite-bandwidth signal with a finite-time signal (using fourier transforms)

However, for the periodic series created by repeating y (which is infinite-time), then your arguments hold. The fourier series represented by the DFT of y is a sum of delta functions in continuous frequency.

So our difference is just that you're considering the periodic signal of y repeated (which is of course how you derive the DFT in the first place), while I'm considering the signal y which is assumed to be zero outside of its support. I'll argue that mine is more relevant for analog signal reconstruction. If I reconstruct my y signal in the analog domain, I don't want that click repeating every 8/(sample freq) seconds out to infinite time (which is truly band-limited and can by further band limited in the way you describe), I want just the one click (and this signal cannot be band limited, analog or digitally, because it is time-limited).

Sorry, no time to proof read. Post may, as always, contain errors.

Re:Depends on the source

[oh_my_080980980]

That's not the placebo effect. You are describing an "auditory illusion."

Re:Depends on the source

[jonsmirl]

So if I had to chose, I'd always take 24b audio. 96K sampling makes some difference but not nearly as much as 24b does. Not sure if there is anything gained from 96K to 192K. Of course compression totally makes a mess of the few bits you have to work with in soft 16b passages.

This problem impacts classical music the most. Other genres don't utilize as much dynamic range.

Re:Depends on the source

[MessageApprovalMan]

"However, dither doesn't change the fact that once a signal sinks below the noise floor, it should effectively disappear. How is the -105dB tone still clearly audible above a -96dB noise floor?

"The answer: Our -96dB noise floor figure is effectively wrong; we're using an inappropriate definition of dynamic range. (6*bits)dB gives us the RMS noise of the entire broadband signal, but each hair cell in the ear is sensitive to only a narrow fraction of the total bandwidth. As each hair cell hears only a fraction of the total noise floor energy, the noise floor at that hair cell will be much lower than the broadband figure of -96dB.

"Thus, 16 bit audio can go considerably deeper than 96dB. With use of shaped dither, which moves quantization noise energy into frequencies where it's harder to hear, the effective dynamic range of 16 bit audio reaches 120dB in practice [13], more than fifteen times deeper than the 96dB claim.

"120dB is greater than the difference between a mosquito somewhere in the same room and a jackhammer a foot away.... or the difference between a deserted 'soundproof' room and a sound loud enough to cause hearing damage in seconds.

"16 bits is enough to store all we can hear, and will be enough forever."

http://people.xiph.org/~xiphmont/demo/neil-young.html

Re:Depends on the source

[muridae]

In my 30s, my hearing is still fine above 20kHz. Those 'mosquito' type ring tones that float around the internet are still easily heard by my ears. Except for the ones right around 23kHz, those drop out before coming back at 24 and then fading down from there.

So, maybe I've preserved my hearing; I doubt it since I've been to concerts with no ear plugs, listened to death metal through ear buds, and hung around server rooms and heavy machinery. Maybe I just started with good high range hearing, I dunno. I did have an audiologist test my hearing years ago, they were surprised by the high range. So while your 'average person' may not be able to hear it, I can hear a ton of rattling in badly compressed 16/44.1 audio. The cymbals and snare drums are horribly distorted compared to lossless. Flutes and some woodwinds suffer the same problem, hell even the harmonics of a soprano can get into the range of 'wow, that's a terrible mix/badly compressed'.

Re:Depends on the source

[femtobyte]

I agree that our only actual point of disagreement is interpreting what time window the filter designer is "responsible" for --- indeed, one can't control the frequency spectrum over an infinite (or at least lifetime-of-the-universe) window when producing a soundwave with finite support.

On the other hand, maybe you should upgrade your stereo to one that does "true" sinc interpolation (starting infinitely before you hit the "play" button, and stopping infinitely after) --- then I can give you a sampled signal which, when properly interpolated to an infinite time window, has a frequency spectrum with finite support :)

Re:Depends on the source

[Belial6]

Back to the original point, I think that your point shows why we should be keeping and distributing music in lossless formats. RAW is gigantic compared to jpg. Pictures also can be huge. Music on the other hand is tiny when compared to our storage methods now. TFS talks about some lossless music collectors having "Terabyte" sized collections. That means that their storage costs are a tiny fraction of the cost involved in collecting their music. A 3TB drive costs less than $150 at Costco. Portable TB drives are well under $100.

I have a relatively large collection of music with greater than 20k tracks. In flac, this is only about 400GB. The entire collection would easily fit on a 500GB portable drive. Those with huge collections would have no trouble putting the entire thing on a cheap 3TB drive.

While available services and software push us towards lossy audio, the only time we really should be using it is in last step streaming, and that is only in the rare instances that the bandwidth is too low to handle lossless streaming.

Re:Depends on the source

[timq]

a 25kHz tone sampled at 44kHz results in a spurious, highly audible (25-44/2)=3kHz aliasing signal

As another poster already said, no, it would result in a 19 kHz tone.

And this tone would not be audible, let alone highly audible, because the A/D conversion filter blocks frequency content above half the sample rate very effectively. (Unless you're using broken converters.)

Re:Depends on the source

[micromoog]

That's correct, there is no audible difference to a human between a 22kHz sine wave and a 22kHz any-other-shape periodic wave. Not to mention, no adult human can hear 22kHz anyway. I hear 16kHz. My 9-year-old can hear 19kHz. Get a frequency generator app and test yourself -- it's fascinating.

Re:Depends on the source

[Vireo]

Ok so you can't fit the human hearing's 130 dB dynamic range into the 96 dB dynamic range offered by 16 bits. Now take a $1500 Emotiva XPR-1 Mono Block amplifier. It only amplifies a single channel for all that money. It's not necessarily the best amp out there, but it sure is a very nice one. At 1 W, its SNR is 93 dB. So you won't fit the 16-bit dynamic range into it. Of course the SNR gets better at higher volume, and eventually you'll be able to fit 96 dB. But with 24 bits... You'll have 30 dB of dynamic range buried in amp noise. Then consider the average consumer. My $400 AVR has a 81 dB SNR at 1W, and could barely fit the 96 dB at max volume. At max volume, I don't want to be in the same room. It would be waaay to loud anyway. Also consider than an average quiet room registers at about 30 dB SPL. If you want 96 dB of dynamic range over that noise level, that would bring the total to 126 dB SPL, which is around thunder clap levels.

Re:Depends on the source

[c 0xFF]

Everybody needs to read the following very carefully:

http://people.xiph.org/~xiphmont/demo/neil-young.html

Some of it agrees with you, some of it doesn't. Let me summarize some key highlights.

16 bits encodes more than the entire range of the human ear. In practice, you can get 120 dB of range, "greater than the difference between a mosquito somewhere in the same room and a jackhammer a foot away.... or the difference between a deserted 'soundproof' room and a sound loud enough to cause hearing damage in seconds." A 24-bit (or even 32-bit float, for that matter) signal path isn't strictly necessary, but reduces mistakes in recording and processing: "The primary reason to use 24 bits when recording is to prevent mistakes; rather than being careful to center 16 bit recording-- risking clipping if you guess too high and adding noise if you guess too low-- 24 bits allows an operator to set an approximate level and not worry too much about it."

So, a perfect 16-bit signal path will technically be just fine, but 24 works better in practice. For final playback, however, using more than 16 bits is generally a waste.

Now, on to 44.1 vs. 96 kHz sample rate. Oversampling at 96 kHz used to have a purpose back in the days of analog -- more wiggle room for filters and whatnot. But now, digital processing has all but eliminated the benefits of a 96 kHz sample rate.

And then you get into the drawbacks of oversampling. If there's any weak link in the path from recording to headphones, recording ultrasonic frequencies (the ones over 20 kHz, basically) will only produce distortions. So, if you have a 96 kHz recording, you'd better have "headphones that can accurately reproduce frequencies above 20 kHz" and whatnot, otherwise you're going to get distortion.

But, here's the kicker: if the original recorder were done at 48 kHz instead, the exact same sound quality can be produced at the back end, without requiring special playback hardware to work around the distortion produced by unnecessarily recording sounds that can't be heard anyway. With modern digital processing, that is.

Here's the lesson: record at 48 kHz, 24 bit, encode for playback at 48 kHz, 16 bit. Higher sampling rates will only produce distortions unless everybody has great hardware (unlikely) and doesn't improve the sound anyway. The extra bits per sample are unnecessary for playback.

Re:Depends on the source

[Pentium100]

Also, transistor amps clip hard when overdriven resulting in a very distorted sound.
Tubes distort more, but the distortions are lower frequency (lower harmonics). Transistor amps distort less, but the distortion is higher order and in some cases can be easier to notice (7th harmonic sounds like a separate note, while 2nd harmonic just alters the sound a bit but you don't hear two distinct notes). Human ears also distorts similarly to tube amps so the brain can compensate that distortion easier.

But the most important thing is - if you like the sound, enjoy it and stop caring whether or not it is entirely accurate. I sometimes play tapes on my tube tape decks - they are low quality, but I like the sound and enjoy using the device even though I have a higher quality tape deck. Of course, I have a high quality transistor amp too (and some day will build a "maybe not so high quality, but more fun to use" tube amp).

Re:Depends on the source

[Anonymous Coward]

Have you ever listened to classical music? The dynamic range can be enormous within a single piece. You might go from a single quiet instrument to an earth-shaking rumble. To replicate the feeling of listening to a real orchestra you need to crank up the volume *and keep it there*. 16-bit will have audible artifacts during the quiet sections because they are essentially only being sampled at 8-10 bits. Human hearing is non-linear so the perceived difference in loudness is less than you'd expect when you think of amplitudes varying by a factor of a hundred or more. Current pop music of course does not have any dynamic range let alone 40+dB swings, so there's no issue.

Re:Depends on the source

[fa2k]

I've been using 48kHz with pulseaudio because I watch some TV and movies as well as listen to music. It is possible to pick a high quality resampler, and I have no complaints, but this post is interesting and I will probably have to change things... Is it better to resample 44.1kHz to 96kHz vs. 48kHz? I just downloaded some free music from the store in the summary (now *that's* targeted advertisement) and it's at 96kHz, so I'd prefer to use that.

Re:Depends on the source

[hairyfeet]

Uhhh...what EXACTLY does that have to do whether you can hear a difference or not? If the discussion would have been over the merits of selling lossless tracks? i would have agreed 100% that its better to get the lossless track and then compress it yourself for whatever you need but that wasn't the topic, the topic was whether these audiophiles can actually tell a difference sound wise.

As I said I sat in a very nice studio with reference monitors that cost more than most will spend on their entire system and i honestly could not tell a difference so purely on basis of sound there is no difference between 44.1k and 192k, what format you want as far as for re-encoding and whatnot is an entirely different matter.

Re:Depends on the source

[Omestes]

I have a relatively large collection of music with greater than 20k tracks. In flac, this is only about 400GB. The entire collection would easily fit on a 500GB portable drive. Those with huge collections would have no trouble putting the entire thing on a cheap 3TB drive.

Though the photo analogy still holds here; I keep all my RAW files on an external drive, and use JPEG for normal use, and sharing. Music is similar, my portable player can't handle lossless formats, and if it could I couldn't fit my music on it. Sadly I can't stream to my phone (thanks to Verizon having caps), but even then it would be worse than a decent mp3.

Lossy has a place. Lossless is the best archival format, or for good setups, but for mp3 is good enough for most applications.

Re:Depends on the source

[Asmodae]

Not to mention the speakers and amplifiers can't reproduce those frequency components anyway. A sawtooth is awfully sine-like when you hear it. Your ear being a physical system with a limited frequency response can't detect them either so why reproduce them?

Re:Depends on the source

[arth1]

Oh, yes, it's way better to convert 44.1 kHz to 96 kHz than to 48 kHz.

To get a visual representation, take a screenshot of some mixed stuff, text and images both. Crop it to 441x441 pixels and save it.
Now scale it to 480x480. It's going to be very ugly, especially high contrast high detail contents like text.
Scale it to 960x960 and it's quite acceptable.
Scale it to 882x882 and it's perfect.

Similar with sounds - the more detailed, the worse results when scaling to a nearby sample rate. If you start with the musical equivalent of a somewhat blurry and overexposed photo, it won't matter much.
So if you listen to Nickelback, don't worry about it. Go with 48 kHz. But if you listen to something more detailed, avoid running your DSP at 48 kHz if the source is 44.1.

Re:Depends on the source

[Asmodae]

I'll take two of those such stereos!

Re:Depends on the source

[Gr8Apes]

I can honestly say at this point, I couldn't tell the difference between either of those either at 20kHz and a flat wave, because I can't hear 20kHz, and neither can more than 80% of the human race.

Re:Depends on the source

[Gr8Apes]

My entire collection easily fits on a single disk. You'd have to have a very large collection to need a second disk with today's disk sizes.

Re:Depends on the source

[Belial6]

This is why what what we buy should be lossless.

Re:Depends on the source

[Fast Thick Pants]

But I like to slow things down sometimes.

Re:Depends on the source

[BlueCoder]

I think this is the mastering issue. Sound is recorded but what they are producing is the playback they hear in the studio. The playback audio in a recording studio is high end. To be fair you have to playback the recording on the original mastering equipment with the same setting to get the intended music. Analog music inherently uses analog distortion to it's benefit.

When you convert such music to digital the playback equipment is fundamentally different which is why you get a different sound. Analog systems are full of distortion which is an intentional part of the music. Digital system seek to minimize distortion. You would need to setup a playback and rerecord the music to figure out how to properly digitally distort the music.

As far as the 96khz controversy no you can't hear the difference unless your a mutant with super hearing. But where 96khz could possible play a factor is in personal digital distortion of the music, i.e. post processing on your personal audio equipment. It's the same reason that digital masters are higher fidelity, because they need to post modulate the sound. You ideally need a master to be something like 5 times greater in order to properly deal with detectable digital distortion. Think of digital distortion as a type of floating point error that is cumulative. You need more bits to do the calculation in order to get an accurate result with fewer bits. High end equipment will often try to automatically extrapolate a higher fidelity waveform before it digitally applies effects to sometimes good or bad effects. To do it properly you need a human being to do digital remastering. It's all around easier and you get a better result if you just start from a source your don't have to upsample or at least not as much.

Re:Depends on the source

[SkimTony]

Just make sure you have appropriate speakers/headphones. No one can hear the frequencies that their output devices can't reproduce.

I find the discussion half-silly anyway. I find the difference in quality between recorded versus live to be much larger than the difference between playback on nice equipment vs. stupidly-expensive equipment. No audio system** can accurately reproduce the sound of 400 voices singing Carmina Burana in a good concert hall, so I'd rather save the money on audio equipment and spend it on actually experiencing music.

**Someone tried to argue that some combinations of McIntosh drivers and amps in a perfect room could come close. When I asked, no, he'd never been to a choral concert with 400 voices and a full orchestra, so he had no real means of comparison.

Re:Depends on the source

[Carewolf]

44.1 -> 48 kHz gives a lot more audible artifacts precisely because they're so close. Think of it as audible moire.

It really shouldn't unless you are doing it wrong.

Re:Depends on the source

[dgatwood]

Unfortunately, that discussion ignores time-domain processing, such as pitch shifting. The more samples you have, the greater the precision of pitch detection (more FFT buckets for a given sampling period), and the more natural the results are likely to be without doing overlapping windows or upsampling or other tricks that potentially introduce additional artifacts. At least this has been my experience.

Also, although hardware-based digital filters are better than analog filters, as I understand it (and correct me if I'm wrong), the steeper the roll-off in an FIR filter, the more latency you introduce into the filter, which still means you have certain limits that wouldn't necessarily exist in a software-based filter. I'm not sure how important this is in practice, of course.

So, if you have a 96 kHz recording, you'd better have "headphones that can accurately reproduce frequencies above 20 kHz" and whatnot, otherwise you're going to get distortion.

When you're talking about an analog component like a speaker, it acts more like a slow slew rate than distortion. The high frequency components basically just aren't reproduced if the diaphragm can't respond quickly enough. Instead of moving out by one unit and back by one unit, it might move out by half a unit and back by half a unit in that time, or out by a tenth of a unit and back by a tenth. And the farther outside the designed frequency range, the more the motion caused by that signal falls off until it isn't actually having an appreciable effect on the sound at all.

This is not to say that it has no effect whatsoever. It is entirely possible to cause damage to speakers by injecting a strong high-frequency component. All the energy that does not go out as physical motion gets turned into heat. But by distortion, I'm assuming you were referring to the sound quality rather than physical distortion of the voice coils. :-D

Re:Depends on the source

[dywolf]

you're confusing sound level and sound frequency.
they are not the same thing.

Re:Depends on the source

[Dr. Spork]

Yes, this is why the music on my hard drive is in a lossless format. Well, this, and the fact that hard drive space is quickly becoming trivially cheap. I figured out the optimal encoding for my portable player, which is ogg vorbis at q=5.25 and a noisefloor of -15. If my preference should ever change, or I move on to a portable player that can't do vorbis, or if a later and greater encoder algo gets released, it's trivially easy to re-encode my music. I do it with the dBpoweramp Music Converter, a great piece of software. It gives me the perfect interface to the vorbis command line encoder, and any other command line encoder out there.

Re:Depends on the source

[Rozzin]

Your preference for 24/96 audio as a listener is entirely due to the placebo effect. There are good reasons to master audio in high res, but for listening 16 bit 44.1khz audio is as good as anything.

There's a pretty good explanation of this (and other factors) on xiph.org: "24/192 Music Downloads are Very Silly Indeed"

Re:Depends on the source

[hairyfeet]

Maybe when the prices come down to pre-flood levels i'll agree with ya, but so far the only drives I'm seeing close to pre-flood prices are the Seagates which frankly are a game of hardware roulette thanks to bad ARM controllers and shitty firmware.

I mean if ALL you do is music? Then sure, why not. but I have music AND movies AND games AND software and all that adds up pretty damned quickly. You seen how much space a single game takes lately? Jesus Tap Dancing Christ these downloads are getting HUGE, its gotten to the point i fully expect to shell out a good $20-$30 more for my Internet every time that Steam has a sale thanks to my going over the cap.

So until I can snatch a 1TB for $37 or a 2TB for $65 like I was doing pre-flood I'll rip my CDs into 320k MP3 and then just put the CD up in case i need to rip to lossless later.

Re:Depends on the source

[hairyfeet]

Frankly I've been playing music on my PC since the late 90s and I have to say i don't get why anybody would WANT to play a CD or LP more than the once it takes to rip 'em. I mean with all my tunes on my hard drive I can have a kicking back mix, a "wake my dead ass up" mix, hell I can set it to random and hear something I may have completely forgotten I have, its just so much nicer than playing a single disc at a time or even using a disc changer.

Now I'm starting to do the same to my movie collection, found some killer freeware that loads all the info into WMC, artwork, synopsis, all that jazz, and its just so nice to be able to just spin through my movie collection with just a flick of the wrist and have it all ready to go, just so much more convenient.

Re:Depends on the source

[PhunkySchtuff]

You might be onto something here - and this is why Apple have their Mastered for iTunes program - where instead of getting your masters, decimating them and dithering them down to 44.1/16, you can either supply Apple with high-res masters (and they will re-convert them for you if their mastering process changes, or if they up the quality again on the iTMS) or you can use the same tools that Apple uses to directly convert your high-res audio to AAC 256.
When you use the tools Apple provides, they take the high res audio and convert it to 32-bit floating point, apply a "mastering quality" sample rate conversion (and, yes, it is a very high quality SRC - refer to the afconvert examples at the SRC Comparisons page) and then make the AAC from this. They also have a plugin for the workflow where you can get a preview of how the audio sounds when converted to AAC, so you can preview the tracks and adjust it to get the best out of the AAC after it's converted.

Re:Depends on the source

[PhunkySchtuff]

But yes, I agree, on the playback side there's no audible difference between a (sufficiently well made) 44.1kHz and 96kHz DAC.

No, but what makes a big difference is when you have a 48 kHz sound card that resamples everything to 48 kHz for an internal DSP stage that cannot be bypassed, and then back again. Yes, Soundblaster Audigy, I'm looking at you.
44.1 -> 48 kHz gives a lot more audible artifacts precisely because they're so close. Think of it as audible moire.

Also, for newer computer audio cards, if you have a choice, use 88.2 kHz for the internal rate instead of 96 kHz. The reason is that most high quality sound is in 44.1 which converts perfectly to 88.2. For 48 kHz, it's less of a problem in the first place, and likely also worse quality sound to start with.
Of course, unless the rest of the audio path is good, it doesn't matter much, but if you like to listen to FLACs with high end headphones, it sure won't hurt to use 88.2 instead of 96 kHz.

There are also good and not so good ways to do sample rate conversions. High-quality sample rate conversions take quite literally one or two orders of magnitude more processing power to do than a quick one, and the effects of a poor quality SRC can have a dramatic outcome on the sound.
Refer to some of the graphs on the SRC Comparisons page for some good converters (eg, Apple's afconvert in bats mode, iZotope's converters) versus some really bad ones (FL Studio 10 6-point, ffmpeg 1.1.1 swr etc)

Re:Depends on the source

[tbird81]

This article is a pretty good explanation of why 16/44.1 is as good as anyone needs for playback.

kinda like 640K?

I gave some advice to a glove factory once. I told them, you should start making all their gloves with 6 fingers on them - just to be sure. They told me that most people had just 5 fingers, and that five fingers should be enough for anyone. I rudely quipped "yeah, you sound exactly like Bill Gates about that 640kB of RAM!"

They called the police on me, and I was trespassed from the building.

Re:Depends on the source

[sahonen]

No, there is no audible difference, because the harmonics of that 22khz sawtooth which *make* it a sawtooth instead of a sine wave are above the limits of human hearing. Filter out those harmonics and what you have left is a sine wave. Not to mention that 22khz is itself arguably above the limits of human hearing itself.

Re:Depends on the source

[sahonen]

Look up how delta-sigma ADCs work some time.

Re:Depends on the source

[Jane Q. Public]

The point is that someone mentioned it. Granted, GP was apparently replying to the wrong person. But it was the correct answer to what he/she was actually answering.

Re:Depends on the source

[timq]

2) Untrue: a 44kHz *sampling rate* has a 44/2=22kHz Nyquist cutoff. Frequencies f>22kHz Nyquist limit "wrap around" to f-22kHz difference frequencies.

No matter how much you repeat this, it's not true. Look it up in any textbook.

Also, your allegations about the DFT allowing a perfectly sharp cutoff aren't realistic. A perfectly sharp cutoff implies an infinitely steep transition between passband and stopband, and this can only be achieved with an infinite number of points in the DFT.

Re:Depends on the source

[bemymonkey]

I've read that 24bit is essential for maintaining SNR during recording, mixing and mastering, even if the end product will be mixed down to 16 bit... but even then I haven't been able to tell the difference (I have "enthusiast" recording equipment though - nothing high end).

Long story short, I like CDs and FLAC in 16bit/44.1kHz. :)

Re:Depends on the source

[bemymonkey]

That's a completely separate issue though - your old tube amp (I've owned a few, so I can understand where you're coming from) and the tape recorders are anything but transparent. Technically, they're horribly bad at audio reproduction, because they change the sound significantly. In the audio chain, you could think of a tube amp or tape recorder as the equivalent of an effect pedal. There's no reason you couldn't record the single tracks on tape and then transfer to digital....

Re:Depends on the source

[Dahamma]

I suppose I could have let this thread die, but I have to add... looks like I was careless and misread the OP (I thought you were responding to a post above it) when I replied to yours. The *actual* OP you were responding to was in fact the OPPOSITE of my point (which completely agrees with yours). My bad :)

Re:Depends on the source

[fatphil]

Most DACs, and sound-cards generally, were shitty in the 90s. I trust that the DAC I currently have in my PC is better than the first ISA Soundblaster Pro I got in the 90s, as I'm currently using high end Apple hardware, and the build-quality does seem very good. I've not noticed any noise or crosstalk between any components. I get more noise from my monitor refresh than is detectible on line-out (and it plugs into a pretty nifty hi-fi, so there's no discernable distortion once it's bridged that gap).

Re:Depends on the source

[fatphil]

No harm done. I'm glad I didn't post more aggressively ;-)

Re:Depends on the source

[hairyfeet]

Well they should really try to post to the right person because something like that seriously derails a discussion. To use the /. car analogy it would be like talking the merits of various trucks and somebody chimes in "My boat gets great gas mileage!"...uhhh...okay? Now everybody is wondering WTF that has to do with anything and the entire chain of thought is derailed.

What is worse is they were actually arguing with me when I AGREED with their position, if you are strictly talking about what is better to OWN, not what is better to listen to? I keep all my band's multitracks in FLAC so that we lose nothing and can re-edit or compress at any time from source, just makes sense. but that is a completely different ball of wax than whether you can hear a difference between those FLAC files and the same tracks in MP3 320K.

Re:Depends on the source

[hairyfeet]

Well I live in an apt so I consider the DAC in my system "good enough" for the volume levels I'm playing at. Hell I have a brand new Audigy soundcard sitting in the closet a customer gave me but the DAC built onto my board sounds fine as far as I'm concerned, and while I'll be the first to admit the DACs of the late 90s weren't great (which is why I always had soundblaster back then, those AC97 chips were just shitty) the ability to have everything at my fingertips 24/7 just outweighs the possible sound losses in my case. Hell the last few years I've gotten rid of all my consumer electronics like TVs and stereos and just do everything through the PC, its just easier to have one unit that takes care of everything.

Re:Depends on the source

[hairyfeet]

So you say but I primarily play 5 string bass and I'm picky as hell about the sound, it can be damned hard to get the nice fat round tone that comes from that 20 year old swamp ash P5 playing through my Trace Elliot and get it reproduced perfectly in digital. Once you do get it recorded well though I honestly couldn't hear a difference between 192k and 44.1k, both on individual and band tracks and that was with a pretty damned sweet setup, nicer than most audiophiles will ever have.

But to me the perfect example of how its mostly if not all placebo is the picture in TFA, you had the audiophile playing back his music...on a fricking tube headphone amp. Now anybody who knows shit about amps will tell you that while tubes have a nice warm tone accurate reproduction is NOT their strong suite yet this guy is saying how he can tell the difference in all these little variations while playing it through what is frankly probably the most inaccurate system he could possibly get. Hell tube amps sound different depending on the room temp and even the amount of humidity so how does he know what he is hearing isn't just the way the tubes are acting that day?

Re:Depends on the source

[ozydingo]

Heh, I'd also bet good money that you could write up a patent for such a non-causal stereo and get that patent approved.

I thought of one more reason why I still prefer to consider those intermediate frequencies to exist. If you don't--that is, if you just consider the fourier series represented by the samples and nothing else--then it's very easy to run into time-aliasing problems. Without the intermediate frequencies, you're really representing the signal that is made periodic by repeating your N samples indefinitely ([... y y y y y ...]). When you filter this signal, the (n>0) response of the filter to y(8) will wrap around to y([1,2,...]). So you could try to use y as above as a filter to chop off the highest frequency component, but then either suffer time-aliasing or explicitly have to zero-pad the signal to avoid it. Of course, your preferred flavor of the math is your own choice. This is mine, but I'd say it's really 6 of one, integral(sin(x)^2, x, 0, 2*pi) of a dozon of the other.

There's probably also an argument to be made about what happens if you frequency-shift the signal. I suspect you'll find those intermediate frequencies causing some headaches in the math if you do it your way, but it's too early in the morning for me to really think that one through. I'll get back to you once I can get some sound out of this true-sinc stereo.

Re:Depends on the source

[Freultwah]

Well, this guy who came up with Ogg Vorbis seems to disagree – sampling rates that high are a liability and introduce all kinds of unwanted side effects both at DAC and playback level. Taking that into account, a losslessy compressed 16/44.1 track makes way more sense than the other one.

Re:Depends on the source

[Overzeetop]

You don't actually go to concerts of classical music, do you? There is no volume at your seat for the "soft" movements.

Re:Depends on the source

[swalve]

Just because YOU can't hear it doesn't mean it isn't there. Low end is not the first to go when digitally sampling (if done right) because it is the easiest to "hear". A nice fat 30 hz tone will have 88,200 samples per wavelength. A 22 khz tone will have 1.

Can you hear a CRT when it is on simply by the high pitched noise it makes? I can. And I can tell the difference between 44 and 192. I can't describe it, but I can hear it.

Re:Depends on the source

[femtobyte]

The downside of a true-sinc stereo: it takes a heck of a long time to "warm up" before it can play.

Anyways, yes, the Fourier basis representation for the signal (designed for periodic functions) is indeed problematic when you try and shoehorn it into representing a non-periodic signal. Of the infinite number of curves that one could draw through your samples "x" and "y", the sum-of-sines-and-cosines one is awkward if you want to suddenly start/stop the signal without a "click" from spurious frequency components introduced by zero-padding (instead of periodic repetition). I have a personal quirk for liking to think about the intermediate frequencies "not existing" in a short time window, because it's a cute analogue to the Heisenberg Uncertainty Principle --- with a finite measurement window, you're fundamentally unable to uniquely distinguish tones more finely than one wave number (inversely proportional to the window width) apart.

While the "true" sinc stereo is a joke, the basic concept is not. Of the infinite number of curves that match your "x" samples, you can interpret them with sinc interpolation (truncated/approximated when you finally need to calculate final results). To filter x->y to remove high frequencies for downsampling, convolve this signal with a wider sinc (or other bandpass window, possibly with slightly less steep edges in Fourier space to cut down on real-space ringing). The efficiency/accuracy of this bandpass indeed won't be perfectly sharp (as you originally asserted) without infinite computer resources to calculate an infinite-length filter kernel, but it can converge pretty quickly, past the limits of quantization noise in a 16 or 24-bit signal, with finite resources --- with cleaner results than the large spurious frequency tails created by zero-padding a sample sequence interpreted as a Fourier sum. And again, on the playback side, your stereo's DAC may already be using some truncated-sinc-like interpolation filter to generate continuous interpolated output.

Re:Depends on the source

[swalve]

It's not just about the width of the dynamic range, but the granularity. Fewer bits means a more stair-step like waveform. In the visual domain, it's like a picture of a blue sky. You are using almost none of the dynamic range available, but because you only have a few bits of resolution available in what you are trying to capture, you are going to get a banding effect.

Re:Depends on the source

[Samizdata]

Well, and edit out anything that would offend the FCC (and FCC fanfolk).

Re:Depends on the source

[Jane Q. Public]

And this really helps anything?

That's an interesting idea.

Re:Depends on the source

[AK Marc]

I thought Radio Mix was to get the timings right for radio.

Re:Depends on the source

[Mikalek]

Words from Professionals in the music recording industry, is that 24 bit do make a difference in the dept field, dynamics. differnce between 16 sound levels and 24 is a 50% better. But 96 hz is no use as only dogs, cats, rats and mouse's ca hear there. It is just a waste of space.

Re:Depends on the source

[djdanlib]

You conflated my post with the parent post.

I stick compressors on certain vocalists' mics on a weekly or more frequent basis. I do not typically mix choral performances, nor have I claimed to do so - that is the parent post. My events usually only have up to 5 vocalists although I have mixed a larger group of about 8 before. (It was a hassle.)

There are occasions when you would amplify a large vocal ensemble, such as in a large venue, or one with particularly deficient acoustics, or some other genre of music featuring a chorus like gospel or some big rock/R&B/hip-hop band with a choral part. I have attended performances where this was done. You use a different mic setup in that case, however, since mic'ing everyone individually is way overdoing it. Check out the outboard rigs and/or digital mixers at an event like that sometime.

Re:Depends on the source

[djdanlib]

I probably shouldn't discuss who I've worked with. I'm not even sure who you're talking about, actually, but the person in question is a male tenor.

One word: YES.

[Anonymous Coward]

Caveat: You have to have decent headphones (not Apple earbud BS), and/or good speakers, but that's about it. The difference is negligible once you hit ~320Kbps MP3, in my opinion, but anything under 256Kbps, regardless of lossy format, you can *clearly* hear cymbal hits turning to an underwater splooshy mess.

Re:One word: YES.

[gabereiser]

I would tend to agree, it's the highs that get muddy with bitrates below 256kbps Lossy audio. However, it really depends on the clarity of the original recording. Some are not as good as others on defining the highs and lows with clean tone.

Re:One word: YES.

[arth1]

Caveat: You have to have decent headphones (not Apple earbud BS), and/or good speakers, but that's about it. The difference is negligible once you hit ~320Kbps MP3, in my opinion, but anything under 256Kbps, regardless of lossy format, you can *clearly* hear cymbal hits turning to an underwater splooshy mess.

Highhats are even worse than cymbals. Even at 256 kbps, highhats tend to sound like they're being hit with a bag of broken glass, and is the easiest way to identify lossy compression I can think of. Except, perhaps, some of Mike Oldfield's earlier works.

Re:One word: YES.

[devhen]

Good point. I was scrolling through the comments looking for this to be brought up because I think its a very important starting point. Essentially, if you are using the earbuds that came with your phone or mp3 player, or something like Beats, you're likely not going to experience a difference between any of the formats. If you want to do ABX testing to find out if you can tell a difference you're going to need to use equipment that can actually reveal those differences. Those people who use stock earbuds or cheap (though not in price) celebrity endorsed headphones plugged directly into mobile devices and laptops etc. have no need for lossless files. However, if you are using a high quality DAC and proper headphones (High-end Sennheisers, Beyerdynamics, orthodynamics like Hifiman's or Audeze, electrostatics like Stax etc) you're more likely to be concerned with using the best quality music you can find and more likely to notice those differences. I've done ABX tests between 24/96 and 320k mp3 and have been able to consistently spot the difference although not 100% of the time.

Re:One word: YES.

[muridae]

THIS, THIS, THIS!

Sorry, it needed repeating. The natural harmonics of something like a brass percussion disc are way up there even when the main tone is well within human hearing. It's just the nature of the waveform, and why so many drum machine cymbals sound like crap. Even a snare drum has some high frequencies in it (or I think it does since I hear the same mushy sound to a tight snare that got badly compressed under 44.1kHz, can't say I've looked on a scope). I most of my listening cases, it doesn't bother me; who cares if it's compressed a bit, I can store 300 hours of music on my phone and listen over bluetooth headset or in the car where those highs would be lost anyways. But when I want to sit down and listen to something special on my larger speakers (1x8 and 1x10 former guitar/bass amps, need a high in the mix soon) or through the studio headset I found for pennies (someone died and their family sold it for a quarter) then I want to hear the actual sound of the crash, ride, high hat, snares, wind section, violin, and flute. I don't want to hear compressed sounds that are vaguely reminiscent of those instruments.

Re:One word: YES.

[yusing]

Decades of A/B tests suggest this isn't true for anyone who's ever been tested.

Re:One word: YES.

[jmv]

Modern codecs are better at 256 kb/s than MP3 is at 320 kb/s. Also, at those rates, it depends a lot on the actual encoder. The newer formats (be it Opus, Vorbis or AAC) all have the potential of giving you perfect quality at 256 kb/s VBR. In the (very) few cases where you can hear an artefact, it's due to the encoder making the wrong decision (e.g. not detecting a transient).

Re:One word: YES.

[evilviper]

Even at 256 kbps, highhats tend to sound like they're being hit with a bag of broken glass, and is the easiest way to identify lossy compression I can think of.

You're not identifying "lossy compression", instead you're actually identifying frequency-domain lossy compression. If you try the same test with a time domain lossy codec like Musepack or even MPEG-1 Layer II (MP2) at high bitrates you won't get the distortion, and you'll be unable to tell the original from the lossy compressed sample.

http://en.wikipedia.org/wiki/MPEG-1#Quality

Modern lossy codec research (HE-AACv2 SBR, Opus, etc) is focused on very low bitrate encoding, NOT transparency. We had transparency with the first international standard audio codec (MP2) and Musepack improved greatly upon that. Strangely, somehow we ended up with those low bitrate codecs (AAC, MP3, Vorbis) becoming the standard for high bitrate (192k+) encoding, where they do a very poor job, and are easily beaten by the earlier, simpler codecs.

Re:One word: YES.

[peawormsworth]

I agree. It is more obvious with good headphones, since everything is more clear. I will hear a consistent repeating defect, which is more disturbing then using crappy ear buds which make everything worse, and the defect blend together. I would guess that many people who dont notice this simply dont use very good headphones.

I can hear a slight difference

[jgtg32a]

I can't tell which one is better though.

Re:I can hear a slight difference

[Twinbee]

That's not entirely the point. Lossless subsumes both of them. In other words, a lossless signal can emulate the other 'worse' one (as well as itself obviously). However, the lossy one can't emulate the better quality lossless version.

Reminds me of the argument with vinyl records - yes the sound might be subjectively 'warmer', but that's through distortion or frequency bias. Perfect replication of a signal can simply emulate that if need be (perhaps just by a simple turn of the bass/midrange knob).

Re:I can hear a slight difference

[X0563511]

However, if you were to then encode to a lossy format from it... you would most definitely get better quality output from the one that was originally lossless.

That's why I keep my stuff in flac - not necessarily because I want to listen to it in flac, but if i need to format shift for a particular device or whatnot, it just works better.

Re:I can hear a slight difference

[fermion]

This reminds me of the argument that sometimes flares up in european classical music. Some say that the music can only be appreciated if played on original instruments. Others say the music is versatile enough to be appreciated in many different forms, even recorded and amplified, something that was not possible when the form was in it's heyday.

Music changes based on the equipment available. When clarinets and pianos came out everything changed. When we were able to overdrive amplifiers everything changed.Now that people tend to consume music through little headphones using tiny playback devices, everything is changing. It is not better or worse. It is doing what artists do. When technology allowed pigments to be packaged and carried out of the studio, we go the plein air paintings, It is nothing more than this.

Re:I can hear a slight difference

[Pope]

The "warmer" argument is also pretty stupid. Who said we're supposed to hear things through the analog distortion of a record player?

As for Neil's argument up there in the summary, it's also pretty dumb for the same reason. You've never been able to buy music that matches the fidelity and sound of the master tapes, so why bring it up?

I grew up listening to music on the radio

[BenSchuarmer]

... and scratchy/poppy vinyl records. MP3s on my cheap ear buds are good enough most of the time.

Re:I grew up listening to music on the radio

[dugjohnson]

I grew up with the same thing (AM radio, no less) and I've lost most of my highs in both ears and a lot of everything in my right ear at this point, so mono works fine for me...in fact, listening to some OLD recordings from the sixties and seventies when they really thought that separating the voices into different tracks was cool makes listening on headphones nearly impossible...I get left track only. Although a great take on the backup singers sometimes, depending on the mix. Frankly, if you stand behind me with a drum and a bass, I'm pretty much set for rest of my life.

Re:I grew up listening to music on the radio

[Zemran]

I listen in the truck with a blown exhaust and whilst getting high on the fumes, lossy or lossless? I have trouble noticing if the car radio is even turned on.

Re:I grew up listening to music on the radio

[nitehawk214]

I listen in the truck with a blown exhaust and whilst getting high on the fumes, lossy or lossless? I have trouble noticing if the car radio is even turned on.

Lossy was referring to the audio quality, not brain cells.

Re:I grew up listening to music on the radio

[muridae]

Some of The Beatles' stuff was originally mixed for mono, and the stereo mixes were done by others after the band had finished mixing their mono release. In cases like that, the argument is there that the mono version is "what it's supposed to sound like"

No

[Hatta]

No you can't. Not with any reasonably modern encoder and bitrates above 256. Anyone who tells you otherwise is experiencing the placbo effect. BTW, you can't tell the difference between 16bit/44.1khz audio and 24/96 audio either. And vinyl might sound "better" than digital to you, but digital is objectively more accurate.

Audiophilia is saturated with woo. This is the same market that brought us $500 ethernet cables.

I usually can, but I rarely care.

[Clueless Moron]

I'm listening to a performance, not some audio benchmark. If a bit of loss bothers you, it must be some pretty damned uninspiring music you're listening to.

And if you're listening on some random mp3 player with bud headphones while walking around doing stuff, compression loss is the least of your worries.

Re:I usually can, but I rarely care.

[polyp2000]

If i had mod points i would totally mod you up!

Of course its about the music - if you are listening really for encoding artifacts how can you be listening to the song?

Re:I usually can, but I rarely care.

[Anubis IV]

So much this.

While Grohl may be right about the listener not receiving what was intended, at the end of the day it's not about which one is better or even if we can tell the differences between them. The question we need to be asking ourselves is:

At what point does the difference in quality cease affecting my enjoyment?

Note that the question is quite a bit different from asking at what point can I stop telling the differences apart. In some cases, I have trouble tolerating slight imperfections, so they may impact my ability to enjoy a form of media. When that happens, I go higher-end. But for music, not so much.

I have an above-average ear after working sound systems for several venues over the course of almost two decades, but even with that going for me, I have trouble telling the difference between lossless and well-encoded lossy files, assuming I can notice any differences at all. And those differences that I do notice are not sufficient enough to impact my enjoyment of the music, so I see no reason to pointlessly spend thousands or tens of thousands of dollars on audiophile-grade equipment, as well as storing the larger files for lossless playback, when equipment that costs far less and files that take up far less space will be more than sufficient to take my enjoyment to the max. That's true for me, but it may not be for others, and I can respect that.

Similarly, that's why I went for a nice 1080p HDTV now, rather than waiting for 4K. I did the math for the size of TV I was interested in and realized that from the distance I'm sitting, I'm already past the threshold of being able to see the individual pixels on the screen, making any further pixel density improvements superfluous for my purposes. As such, there was no point in even checking to see if they would improve my enjoyment, since I can empirically say that they are incapable of doing so for reasons other than the placebo effect. That's thousands of dollars saved and no enjoyment lost, despite the equipment being inferior.

TL;DR version: get the stuff that maximizes your enjoyment at a decent price, but don't waste money on improvements that don't enhance your enjoyment, even if you're capable of noticing them.

In traffic, a VW will get me someplace

[wiredog]

as fast as a Ferrari.

Since I do most of my listening in a car, and am almost 48, I can't hear the difference between an mp3 and a vinyl album, or a cd, most of the time. Well, except for the lack of skipping. Ever try to listen to an LP in a moving car? But I digress. Sure, people who are younger and $pend lot$ of dollar$ on the Finest Audiophile equipment areound can tell. Me in my Chevy? Not so much.

Re:In traffic, a VW will get me someplace

[serviscope_minor]

I can't hear the difference between an mp3 and a vinyl album, or a cd, most of the time.

Really? I can hear the difference between vinyl and everything else: some of the things I have have been recorded off the original vinyls from the 60's and let me tell you, those things were butchered. With the hiss, crackle and utter mangling of the deep base it is impossible to miss.

Re:In traffic, a VW will get me someplace

[TubeSteak]

I think it's important to note that the audiophile's "4TB 67,000-song music library" is what we now call "one hard drive that costs $200"

Over ten years ago, I remember RAIDing up 4 x 75 GB hard drives and thinking "holy shit I have a lot of space"
A DVD-R could back up a significant chunk of your hard drive.
Now... a lossless library (at home) is within everyone's reach.

Re:In traffic, a VW will get me someplace

[Belial6]

Exactly. Any discussions of file size is pointless. Compressed lossless audio files are are tiny compared to our storage capabilities. We might as well be asking whether 7z, Zip, or RAR is the best choice for compressing our text files.

The only time lossy compression makes sense anymore is when streaming over the internet to low bandwidth devices. Even then it doesn't always make sense.

Re:In traffic, a VW will get me someplace

[dywolf]

plus theres the investment vs return factor.
why spend a lot of time or money for little added benefit?
rather spend that time/money/energy on something else.
such as singing along

Audiophiles might.

[SuricouRaven]

Everyone else listening on the little earphones that came with their cellphone can't.

Now, in grand slashdot tradition, could we please have a debate about the use of 192KHz sample rates between those people who know what they are talking about and those who belive 'fourier' is just a word you say to sound smart?

Re:Audiophiles might.

[msauve]

Mine goes to fiveier.

Re:Audiophiles might.

[rudy_wayne]

Everyone else listening on the little earphones that came with their cellphone can't.

Now, in grand slashdot tradition, could we please have a debate about the use of 192KHz sample rates between those people who know what they are talking about

Saying that A sounds better than B is purely subjective and has absolutely nothing to do with "knowing what you are talking about".

Re:Audiophiles might.

[SuricouRaven]

It does when it is easily testable that human hearing has an upper frequency limit, and mathematically provable that a 44.1KHz sample rate is more than sufficient to reconstruct all possible signals which have no components above this limit. Unless you have superhuman hearing, there is no benefit in >44.1KHz sampling. 48KHz at the very most, and that's only to allow for imperfect filtering in the playback devices and would only be noticed by those of the best hearing. We're not talking about subjective things like the asthetics of different distortions: These are fundamental limitations of physiology and mathematics.

Re:Audiophiles might.

[LordLimecat]

Not if you can mathematically prove that the two sound reproductions are identical; at that point, it isnt the SOUND that is better but the user's perception of it.

Re:Audiophiles might.

[msauve]

You're quoting stats for typical hearing. When I was (much) younger, I could easily hear the "ultrasonic" motion detectors used for entry alarms at some mall stores. It was quite annoying. No one else I knew could, which I would expect if I was hearing a sub-harmonic. Those are typically over 30 KHz. I can't hear much over 16 KHz, now.

But, I don't doubt that there are a small minority of people who can hear the difference between 44.1, 48 and 96K sampling rates.

Re:Audiophiles might.

[Belial6]

I am a bigger fan of convenience than quality with my media, so I certainly can't claim to be an audio snob. My hearing also isn't perfect. What I find is that the low quality speakers/headphones and low quality hearing makes the difference in bigger, not smaller. When I rip a CD using LAME to 192k mp3, there are plenty of albums where the vocals go from understandable to just noise.

Perhaps that has to do with my hearing seeming to be better in the high and low ranges than it is in the middle ranges. I assume this is the case because when I watch TV, I will frequently have to turn up the volume, or adjust the frequencies to be able to hear the actors talking over background music, or effects.

Re:Audiophiles might.

[AK Marc]

Fourier is the year after year 3.

44.1khz ought to be enough for anyone...

[scorp1us]

We recently discovered that human hearing beats the linear response assumptions used in lossy codecs. So yes, their criticisms are scientifically founded.

Re:44.1khz ought to be enough for anyone...

[Hatta]

Unless you have people that can ABX the difference, no their criticisms are not scientifically founded. An actual blind test beats any theoretical reasoning any day.

Re:44.1khz ought to be enough for anyone...

[Trepidity]

In particular, nobody claims that lossy codecs use a perfectly accurate model of human hearing; they don't need to. The goal is to have a psychoacoustic model that captures enough of the general mechanics of hearing, to enable a bunch of constants to be tuned empirically. If the model doesn't come anywhere near to capturing anything important, that would be a problem, because you'd never be able to tune the constants. But once it captures the general outlines, much of the real work on lossy encoders over the past 10-15 years is on tuning a billion constants with listening tests. The goal is empirical transparency (people cannot distinguish the compressed version), not a scientifically valid model of human hearing. Pointing out that there are all sorts of slightly wrong things about the internal model isn't really important if you can't show that they produce audible differences in the end result.

Re:44.1khz ought to be enough for anyone...

[ImprovOmega]

Subject:

44.1khz ought to be enough for anyone...

Body:

human hearing beats the linear response assumptions used in lossy codecs. So yes, their criticisms are scientifically founded.

These have nothing to do with each other.

Re:44.1khz ought to be enough for anyone...

[Kjella]

Unless you have people that can ABX the difference, no their criticisms are not scientifically founded. An actual blind test beats any theoretical reasoning any day.

This. This whole lossy vs lossless was dispelled in the 90s for everyone but audiophiles, most people can't hear the difference and neither can most audiophiles either in an ABX setup. There's always some "magic" about their own system that lets them hear the difference, but not in a controlled environment. There's no doubt that we're throwing away lots of information in pretty much every form of sound recording, a bat would probably be very displeased with the sound reproduction. The question is does it matter, you don't need infinite bits of audio anymore than you need infinite pixels on a screen or a photograph. I think the last person I read about who won such a contest had a hearing disability so he heard things that would be masked to normal ears, we're on that level.

Re:44.1khz ought to be enough for anyone...

[fa2k]

Very cool resource. I didn't know that lossy codecs relied on an "uncertainty principle" like relation, but it seems likely that hearing is not linear. I don't think that anyone seriously claimed that it was *impossible* to tell the difference between lossless and high bitrate lossy compression, and this is the reason that some prefer FLACs etc

Re:44.1khz ought to be enough for anyone...

[scorp1us]

Didja read the article? Some people can tell the difference down to one oscillation per second. That's not theoretical.

Re:44.1khz ought to be enough for anyone...

[Hatta]

There is always loss.

That loss is both theoretically and empirically beyond human perception.

Debunked

[MetalliQaZ]

The concept of improving consumer listening experience using studio quality recording has been thoroughly debunked, right here on Slashdot...
Why Distributing Music As 24-bit/192kHz Downloads Is Pointless

Re:Debunked

[fa2k]

I really wanted to mod this up because it's a great resource.

But: it's actually off topic. The question here is about using psychoacoustic compression like MP3 and AAC, not about sample rate or bit depth.

Re:Debunked

[QuietLagoon]

has been thoroughly debunked

Not really. Lots of pseudo-science and opinions expressed as facts in that thread. But debunking? ummmm.... no.

Re:Debunked

[ImprovOmega]

I really wanted to mod this up because it's a great resource.

But: it's actually off topic. The question here is about using psychoacoustic compression like MP3 and AAC, not about sample rate or bit depth.

The logic and impassioned response is similar for both. Honestly the deal with psychoacoustic compression is a matter of prioritization. It just reorganizes the data so that the most important bits for hearing and recognizing it as a faithful representation of the original are front loaded and saved first. Beyond a certain bitrate you have drastically diminishing returns. 128kbps gathers the most significant and important bits and is probably more than sufficient for your standard earbuds and regular consumer. 256kbit will make a noticeable difference to audiophiles on good equipment. Anything above that and you're getting very close to saving plain old noise.

Re:Debunked

[Psyborgue]

Read the xiph.org article. Yes. Debunked.

yes and i don't care

[alen]

years ago i had music ripped in lossless and yes you can hear the difference
  these days its all MP3 or AAC. and i don't care. only time i listen to my music is while running or sometimes on the train to and from work. most times my iphone doesn't have any music on it and i listen to spotify or pandora for 20 minutes while driving home

some people probably care about the best sound quality, most dont
i like blu rays. my wife will watch TV on the non-HD channels most times and she can't tell the difference in quality

Re:yes and i don't care

[mjr167]

Have you seen Vanna White in HD? There are advantages to non-HD...

It doesn't matter

[Anonymous Coward]

The reason people use lossless compression for audio (i.e. FLAC or SHN) is not because they can tell the difference. Maybe you think you can, maybe you think you can't, but that's irrelevant anyway. The reason people choose lossless is that lossless is the only suitable solution for archiving. If you want to preserve your CD audio exactly as it appears on the CD, the only possible solution is lossless compression. If you choose lossy, you aren't making an archive or the original, but rather an approximation of the original.

That's all there is to it.

Re:It doesn't matter

[Tamran]

EXACTLY!

Re:It doesn't matter

[xorsyst]

Oh, for mod points.

While I can't (mostly) tell the difference between the original CD and a ~140Kbs VBR MP3, I _can_ tell the difference between a 140Kbs VBR MB3 made from the CD source, and a 140Kbs VBR MP3 made from a 256Kbs VBR MP3.

Lossless isn't for listening to, it's for archiving. And make sure you get the cuesheet, pregaps, etc. right when you're archiving too :)

Re:It doesn't matter

[tuffy]

And you never have to re-rip physical discs. 128kb/s CBR MP3 used to be the standard. Then 192 VBR. Then AAC. And so on and so forth. So by keeping a lossless archive, one will always be able to transcode to the latest-and-greatest lossy codec without a lot of hassle.

Re:It doesn't matter

[EvanED]

That's my reasoning too. I've tried to ABX things, and at least with my equipment (I have decent headphones but not everything on the audio path is good) and I've sometimes been unable to discern a 128 kbps OGG from lossless. I'm under no illusion that it provides audible benefits over a

And yet, even though I already had by CD collection ripped to higher-bitrate MP3s, I re-ripped to FLAC a few years ago. Why?

(1) I wanted to re-rip anyway, as some CDs had ripping artifacts (clicks and pops) and I wanted to do a secure rip. (I bought dbpoweramp's ripper for this, and it worked quite well. No relation to the company except a pretty satisfied customer.)

(2) Archiving. Suppose in 10 years some other compression format comes along and supplants MP3. What then? Well, I'd either have to transcode MP3->FutureFormat and suffer bigger losses (perhaps the codecs interact particularly poorly and even though 256 kbps sounds good with both, but the transcode is bad) or do another re-rip anyway. And in that time span, my CDs would be more likely to be damaged, destroyed, etc.

(3) I want to add good metadata. I'm unhappy with the metadata that comes with most digital music purchases, for example; while in some sense it's more accurate than what I want, it behaves far worse with most players. For instance, I bought a collection of symphonies, and the Album Title of each is "Karajan Symphony Edition", the Artist is always "Herbert von Karajan" or something like that, and the track titles are always like "I. Moderato" -- but the actual composer of the symphony is nowhere present. That's a particularly bad case, but I would much rather have the Album Title be something like "Symphony No. 6", the Artist be "Bruckner, Anton", and then the actual album title and performers be in other metadata fields. So I am very slowly replacing all of the metadata in the tracks I've ripped. This interacts with the previous entry, because if I wanted to re-rip my collection in a different format I'd have to either re-do this work or automatically find correspondences between tracks. (Probably not too hard, but you never know.)

(4) Storage is cheap. My CD collection takes up less than 100 GB in FLAC. That's way too much for my portable player, but in another few years it probably won't be. In the mean time, doing transcoding from lossless to MP3 or OGG is easy, and doesn't suffer from any additional problems of problem (2), and metadata is preserved well. If all you have is a laptop and no external drive that could be a problem even there, but I'm a space hog anyway so I barely notice that 100 GB.

(5) There are no real drawbacks -- it's just a very small amount of extra storage cost (but not much; the collection of photos I've taken is three times the size of my music) and a token effort and a bit of CPU time to start an additional transcoding step to OGG for the portable player.

(6) Given (5), why not? Who knows, maybe I'll have biotic ears implanted in 30 years that will make even lossless CD quality sound like crap. :-)

Re:It doesn't matter

[PRMan]

I never used less than 192 anyway. And as soon as LAME came out with the alt-preset stuff, everything I've ripped sounds identical to the CD. I never went back and re-ripped the 192 stuff, but in a quiet room with a very good system I can hear very few artifacts (like an almost perfect JPEG).

Ironically, the same audiophiles will often have cable with absolutely horrible HDTV artifacts and not even care...

Re:It doesn't matter

[AudioEfex]

Yeah an archive that may never be playable. The point of archiving is preservation, but a lot of good that FLAC archive would do someone who found it in 1000 years while sifting through the remnants of Earth - they will have a lot easier time finding a device that still exists that plays MP3 than they would FLAC or what have you.

The quality threshold has been met long ago by all but the most obsessive types who are more concerned about data theory than actual sound. So yeah, it's not a bit for bit copy of the CD, but in truth the original physical CD Is already a better storage medium for archival purposes than anything digitally stored on more volatile media (magnetic or even SSD). Sure it may take up more space (but not as much as one would think, in archival sleeves), but it's a heck of a lot easier to destroy a magnetic hard drive or wipe flash memory than to destroy an actual CD, plus all your eggs are not in one basket (if physical damage occurred, the likelihood of the entire CD archive being destroyed at once is more difficult than it would be to render a drive unusable).

And truthfully - who ever is going to really care more about it being a bit for bit digital copy - the sound is the point, and what we want to preserve - I'd rather leave a recording of Sinatra singing "New York" that can be audibly heard than a bunch of bits and data that are somehow mathematically more sound.

The biggest reason mp3 is and will remain standard is the same as why the red book CD has not been replaced for physical music media, even though better quality choices have been available. Virtually every device that accepts a disc plays red book CDs, and virtually every device out there that plays digital music plays MP3. The world could cease making new CD players and MP3 Players (notice most don't call them "digital music players") and we would all live for a very long time and still be able to enjoy our music. Just about every device in the world which has a speaker and you can get a digital file to plays MP3.

128k was only the "standard" for people that didn't know any better, or those that only pirated music off of Napster. I was ripping at 256 and 320 VBR since before the turn of the century. And my MP3's made back them still sound great, in spite of the supposedly antique methods used to record them.

Any FLAC I come across I generally convert and dump it, to me the importance of my musical archive is that the entire thing is in one format (MP3 of varying bit rates) and I can feed it in its entirety to any device and consume it. It's all at my finger tips no matter what playback device I may put it through, as long as I can get the files to it.

I think you will find most people feel similarly, if they even think about the concept at all. The average person thought cassette sounded just fine - it was the disc based media that sold the deal (much like the adoption DVD, which on image quality alone would not have made it as big nearly as quickly until HDTV came along (ironically, even though DVD is obviously SD). Now you or I may feel differently (for example, I almost exclusively buy Blu-ray now, I only buy a DVD when a Blu is not available) but to most people, they don't care or even know there is a difference that some people can supposedly hear. A least side by side unless you have cataracts you can see the difference between DVD and Blu - very few people are going to say the same thing about a 256k MP3 vs FLAC or what have you.

I think it also boils down to what someone said above - if people are listening that closely for technical details that very arguably human hearing can tell little difference of, then it must be some really poor music. I say the same thing about people who watch Blu-rays with microscopes to figure out if they like the quality or not - if you focus so much on the technical delivery, you are missing the entire point of enjoying media to begin with.

Re:It doesn't matter

[tuffy]

Yeah an archive that may never be playable. The point of archiving is preservation, but a lot of good that FLAC archive would do someone who found it in 1000 years while sifting through the remnants of Earth - they will have a lot easier time finding a device that still exists that plays MP3 than they would FLAC or what have you.

FLAC is about an order of magnitude simpler than MP3. I once implemented a decoder in about an hour over lunch just because I could. And because many lossless codecs feature error detection, they're much more likely to survive as a long-term archive than something like MP3 which doesn't even have a container or any reliable way to verify that the file's contents are correct.

Re:Better question

[Anonymous Coward]

No?

Any good studies?

[Experiment 626]

Anyone know of any good double-blind studies comparing people's ability to tell FLAC from 320kbps MP3? Googling just turns up people debating in forums whether you would be able to tell the difference rather than any serious academic research.

Re:Any good studies?

[FrankSchwab]

I don't know if it's a good study, but I did exactly this test. Ten or fifteen years ago.

I took four musical selections (from the latest Rolling Stones album at the time, a solo piano performance, a classical orchestra, a female vocal), and encoded them at 128, 192, and 256 Kbps with the Fraunhofer codec of the day (remember that?). I re-expanded them to 44.1 KHz CD tracks, and put them on a burned audio CD (remember those?). Each selection on the CD had five versions - the first was always the original bit-for-bit copy from the source CD, then followed (in random order) the 128, 192, 256, and the original again.

I made ten copies, and handed them out to the audiophiles in the office to play on their home stereos, and gave them a test sheet - I asked them to identify for each selection which version was 128, 192, 256, or the original. Nobody came close to having a "golden ear" that could reliably tell the 128Kbps versions from the others, much less the higher bitrates. Overall, there was a slight ability to detect the 128 kbps versions - it got selected as the lowest quality one more times than random chance would suggest, but even it was still well below 50% (I don't remember the exact numbers any more).

And this was with ancient MP3 encoders.

Frankly, if you think you've got the golden ear, first of all I pity you - I'm sorry that you have to put up with all the crap you're going to hear. Second of all, I really recommend running the same test - prepare the tracks, have a friend randomly order them (but keep track), and then see if you can identify them. Don't simply say "Of course I can" - Actually do it and prove it.

And, if I can be an old man with a bit of advice for a minute: if you can't tell the difference, don't go out of your way to train yourself to tell the difference. It'll just be an annoyance to you for the rest of your life. Kinda like the person who taught me about the reel-change indicators on film at the movie theatres - I see it, and my whole body tenses up waiting for the change. I wish I had never known about it. I really appreciate the change to digital projection so I don't have to deal with those anymore. /frank

Re:Any good studies?

[fa2k]

Here's a very un-scientific study in Norwegian. http://www.diskusjon.no/index.php?showtopic=1490576&st=140 The goal was to find the compressed version among 8 different wav files (I know, strange setup). The problem is that the files were numbered, which seems to have affected the result.The right one is #4. [I got the right one, but it may have been luck]

Re:Any good studies?

[Lussarn]

This is what I did, not on MP3 but DD vs TrueHD.

I'm a heavy movie watcher. When I got a receiver capable of lossless sound I just wanted to know if I could hear the difference. If I can't it's really no use for me. I took a movie with a great soundtrack, namely Transformers (shity movie, but one of the best soundtracks). I ripped the blueray and made a little program to play small parts of the movie 2 times, 1 using TrueHD and one using Compressed Dolby Digital (640kbps), randomly switching which soundtrack it played first. I calibrated the volumes to match until I couldn't hear any difference on most parts of the movie (dialogs etc, where the DD bitrate is clearly enough for my ears/equipment/room at least).

Then I sat down with pen and paper in front of the movie and guessed if I was hearing Lossless or compressed sound. My conclusion and my numbers from the guesswork is that Lossless vs DD is very much hearable when the action gets going. The sound is more spread out. sorry I don't know the terms for this (I'm not an audiophile) but, lower low volumes, higher high volumes, better separation beetween different simultaneously playing sound, and possibly better channel separation. My conclusion also is that the difference isn't earth-chattering on my living-room-setup and I can still appreciate DD movies . But the difference is there and it's nice.

Re:Any good studies?

[ninjackn]

I was concerned with scientific studies as absolute proof but then I realized that when it comes down to it, it only matters with what *I* can hear. So I used the ABX comparator plugin in foobar2000 to test my ability to discern FLAC vs 320kbps. You select two tracks you want to compare and then it presents them as blind tracks labeled A,B,X,Y and you need to pair up tracks A and B to X and Y. So I found out I can hear a difference. On select things. I'd have to listen to almost the entirety of the song and only at certain points does the difference become apparent to me. Things like ending a song with a rattling of a drums hi-hat sounds more distorted with 320kbps MP3 than FLAC. Music like the Beatles ends up sounding the same from even at 256kbps to me. So I believe that a person can hear the difference between FLAC and MP3. It might not apply to all music across all equipment for all people but there is a difference. Personally, the majority of my music is encoded in 256kbps MP3.

Re:Any good studies?

[evilviper]

Anyone know of any good double-blind studies comparing people's ability to tell FLAC from 320kbps MP3?

Would you like some similar tests to determine if water is wet, or the sky is blue?

MP3 has some fundamental limitations that will prevent it from ever being indistinguishable from the original. The only thing a study will tell you is how good people's audio equipment is.

Audio perception is a whole field. There are a constant stream of listening tests, mostly funded by the semi-famous companies and organizations that develop lossy audio codecs, based on international testing standards. There don't need to be "studies" of lossy audio, because the psycho-acoustic concepts have been well established and are unchanged since the preeminent studies performed in the 80s. Just go look up "Perceptual Entropy".

If you want a lossy codec that can potentially be indistinguishable from the uncompressed original, you need to look at temporal-domain codecs, such as Musepack/MPC, MPEG-1 Layer2 (MP2), or possibly even AC3 (Dolby Digital).

Re:Any good studies?

[coldsalmon]

You can do a scientific study of your own abilities using an ABX testing program. Some googling should reveal one for whatever OS you're running. When I tried it, I could tell the difference between CD-quality and a 128kbps mp3, but anything above 128kbps was indistinguishable to me.

Re:Any good studies?

[nabsltd]

I ripped the blueray and made a little program to play small parts of the movie 2 times, 1 using TrueHD and one using Compressed Dolby Digital (640kbps), randomly switching which soundtrack it played first.

640Kbps AC-3 does strain a bit on "busy" soundtracks, so it's not surprising that you noticed a difference.

I've done a similar comparison with lossles vs. 1.5Mbps DTS, and the results are much less definitive. I couldn't tell which was which enough extra times to say it wasn't just random luck.

Somewhat unrelated, I also will sometimes re-encode DTS-HD Master Audio using a lower core bitrate to see if I can get the overall file size smaller. Basically, the audio on the disc uses 1.5Mbps as the core so that you get the best quality if you can't decode the lossless. But, if you are decoding the lossless, a smaller core doesn't change the quality, but can save you quite a bit of disk space.

Re:Any good studies?

[Tyler Durden]

Kinda like the person who taught me about the reel-change indicators on film at the movie theatres - I see it, and my whole body tenses up waiting for the change. I wish I had never known about it.

You're welcome.

Re:Any good studies?

[Maow]

Anyone know of any good double-blind studies comparing people's ability to tell FLAC from 320kbps MP3? Googling just turns up people debating in forums whether you would be able to tell the difference rather than any serious academic research.

This may not be exactly what you're looking for: Xiph on 24/192 music, but it likely has links to something that will answer your question.

And it's extremely interesting in its own right.

Re:Any good studies?

[residents_parking]

I have downloaded tracks *meant* to be FLAC, listened to them and gone "NAH", then run them in Audition only to see tell-tale dynamic bandwidth limiting endemic in MP3s.

Now .. if you think it is a subtle effect, you might disbelieve me. But if you have even entry-level studio monitors, it's as clear as day. The staging is more focussed, high frequencies better represented, and reverb tails last as long as they should. I can just about tell the difference in my car (staging), but I don't care so much.

Yes and No.

[nospam007]

People with normal, standard hearing cannot detect a difference, that' s the point of the compression.

If you are hearing a difference, it's because you have a hearing defect. If you can hear something that you don't hear after compression, it's because you're deaf to the sounds that's overlaying it (and killed it in the compression)
You were hearing the original differently in the first place, than anybody else with normal hearing.

Re:Yes and No.

[Belial6]

"If you are hearing a differnce, it's because you ahve a hearing defect."
"Peace is ware."
"Slavery is Freedom"

Better yes, but how much better?

[deathsquirrel]

I listened to the sample tracks hdtracks.com offers for some albums I own & have ripped to 256kbps MP3s and without question the lossless tracks did sound better. The question that I then had to ask was did they sound $20/album better and nope, not even close for me.

Re:Better yes, but how much better?

[fatphil]

Have they been remastered? If they have, they're not the same track.
Did you do an ABX test? If not, then your anecdote is pretty meaningless.

Oblig

[jxander]

We still like XKCD around here, right?

mp3 vs wav

[JonathanP.Bennett]

Yes, I can hear the difference. When working in a small sound recording studio, I trained my ears to pick up on fine details. There was one day in particular I remember listening to a track, and wondering what the strange noise in the background of it was. I realized that I was hearing the audio artifacts from the mp3 compression. Not sure how Mr. Young figures that a CD is only 15% of the master, though. A CD is pure uncompressed audio. If you recorded and mixed in 44.1k audio, then your cd is an exact copy of your master.

Re:mp3 vs wav

[wonkey_monkey]

Yes, I can hear the difference.

At 128kbps? 256kbps? 320kbps?

Re:mp3 vs wav

[JonathanP.Bennett]

I don't remember what the encoding rate was. It wasn't the on the low end, but I can't be sure it was full 320, either.

Re:mp3 vs wav

[bobbied]

Yes, I can hear the difference. When working in a small sound recording studio, I trained my ears to pick up on fine details. There was one day in particular I remember listening to a track, and wondering what the strange noise in the background of it was. I realized that I was hearing the audio artifacts from the mp3 compression. Not sure how Mr. Young figures that a CD is only 15% of the master, though. A CD is pure uncompressed audio. If you recorded and mixed in 44.1k audio, then your cd is an exact copy of your master.

Congrats, you have excellent hearing... Or the MP3 compression was *REALLY* high.. :)

I'd wager that less than 1% of people will be able to accurately identify MP3 compressed material over the so called "high def" version of the same in all but the most ideal situations or really high compression rates. Further, I'd stipulate that nobody would be able to tell for some of the pop music I hear today which is mixed for radio play in the first place and highly compressed heavy metal would likely sound better....

Re:mp3 vs wav

[PRMan]

I have bad hearing and I can hear artifacts in 192 or lower. But once you get up to 220/240 or so, forget it. (This is on the latest LAME.dll, on Fraunhoffer you can definitely hear stuff at 256).

Re:mp3 vs wav

[Hamsterdan]

"A CD is pure uncompressed audio"

Which won't make a big difference when the Dynamic Range is shot to hell

http://en.wikipedia.org/wiki/Loudness_war

And Yes I can hear the difference.

Re:mp3 vs wav

[PhunkySchtuff]

Yes, but no-one is arguing you should record and mix at 44.1/16 - on the contrary, there are well established and accepted reasons for recording and mixing at, say, 96/24 and only mixing down to 44.1/16 at the very end of the process.

If you compare the raw bitrate, and you're mixing at something like 192/24 and the final mixdown gives you 44.1/16 then the size of the output will be on the order of 15% - and that's a good thing, as it means that CDs aren't the size of LPs and MP3 players can store more than three songs.

Where he's wrong (Neil Young) is assuming that he can hear every fine nuance of the high-res audio (he can't, no argument there, he simply can't) and that as the redbook CD audio is 15% of the size, then somewhere along the way you're throwing away 85% of the sound. Now, that's simply crazy talk (and, let's face it, Neil Young has done more than his fair share of substances that may induce a hint of crazy)

Difference is not in the listening.

[Anonymous Coward]

The difference is the ability to transcode to different bitrates and formats without losing anything from the original source.

Re:Difference is not in the listening.

[bobbied]

MOD PARENT UP!

This is about the ONLY reason to even consider higher bit rates. Transcoding and mixing are good reasons to have higher bit rates. Because most folks simply cannot physically hear the differances.

Some lossy sound better than other lossy.

[jslarve]

Not saying that lossy is ever better than non-lossy, but with good "dithering", it can really make a big difference. Probably better for artists to provide their own mp3 (or whatever file type) than leaving it up to a vendor to do it. This Izotope video is pretty informative, despite being kind of an ad. http://www.youtube.com/watch?v=vVNzylf9sGo

Re:Some lossy sound better than other lossy.

[Achra]

You can't recreate signal once it is lost. All interpolation would do in this example is chew up more disk space.

Never been able to tell the difference

[Anonymous Coward]

I've been listening to digital audio since the '90s and have switched back and forth between lossy and lossless a few times. I've even tried to compare formats to see if I could tell the difference. Personally I can't, but maybe that's because I blew out my ears at loud concerts. I certainly can't hear like I used to.

Maybe some people can tell the difference and if they want to devote the time, money and space to lossless audio formats power to them, but it means little to me.

Also depends on age ...

[Anonymous Coward]

I'm nearing 50 and I already feel I'm reaching for the volume control more. 'True' 'audiophiles' will say anything to justify their beliefs. Same goes for 'true' 's elsewhere

I may, of course, be very very very wrong

Oh, it all goes to shit when you have kids too :-)

Re:No

[router]

I can't tell the difference between 128kbit mp3s and CD most of the time. Fiona Apple or other strong vocal artists are degraded at 128, but at 256kbit I can't tell the difference. I would still rip at FLAC now, since storage is cheap, but I haven't gone back and re-ripped my 256kbit rips....

andy

the answer is obvious, isn't it?

[v1]

No you can't. Not with any reasonably modern encoder and bitrates above 256.

And there's the rub of course. That general of a question can't be answered yes/no. It depends on a variety of factors, most notably the content, the codec, the bitrate, and the playback.

I don't even know why this article submission got accepted. It's like asking "can you win a race against a Toyoda?" where do you even start with that....?

Re:the answer is obvious, isn't it?

[rudy_wayne]

It's like asking "can you win a race against a Toyoda?" where do you even start with that....?

Since Akio Toyoda is 30 years older than me, I'm pretty sure I could beat him in a race.

Re:the answer is obvious, isn't it?

[Etherwalk]

It's like asking "can you win a race against a Toyoda?" where do you even start with that....?

Since Akio Toyoda is 30 years older than me, I'm pretty sure I could beat him in a race.

Boy do you need a new car!

Re:the answer is obvious, isn't it?

[Psyborgue]

And the encoder. There are good mp3 encoders out there and there are some very very bad ones.

Re:the answer is obvious, isn't it?

[evilviper]

It depends on a variety of factors, most notably the content, the codec, the bitrate, and the playback.

The constraints of the question are pretty well defined. We're talking about popular music, with common codecs at bitrates used by online music stores.

Re:the answer is obvious, isn't it?

[v1]

The constraints of the question are pretty well defined. We're talking about popular music, with common codecs at bitrates used by online music stores.

isn't it difficult to talk with such a large mouthful of assumptions like that?

Re:the answer is obvious, isn't it?

[evilviper]

isn't it difficult to talk with such a large mouthful of assumptions like that?

What assumptions? Sounds like you didn't bother reading past the TITLE of TFA. This is spelled-out in the FIRST SENTENCE of the /. summary:

"compressed formats like the MP3s being sold on iTunes"

Yes ... but not always

[Anonymous Coward]

1) It depends on the quality of the encoder
2) bps
3) The music itself
4) Good environment (no external noises)

This is what bugs me for a few online streaming services...

Re:No

[Reverand Dave]

Agreed, if anyone is really such a purist that they think a file format is superior over another, they should either skip the argument and go see the band live, or kill themselves to save everyone else the hassle.

Sure, you can tell.

[jtownatpunk.net]

If you've got decent equipment and a quiet environment. With cheapo earbuds, I don't notice the difference. With my good headphones, the difference is obvious. When I'm driving down the highway, I can't tell. In my living room, I can tell.

With storage so cheap and bandwidth so plentiful, there's really no reason not to use lossless audio. My $40 Clip+ with a $25 miscrosd card can hold 40 gigs of content and can play FLAC. There's no reason to use a lossy format.

Re:Sure, you can tell.

[SolitaryMan]

That is exactly why I'm using FLAC. I mean, nobody arguing that mp3 gives you better quality than FLAC, the whole point is to save some space on hard drive. Which is dirt cheap this days, so why bother.

Re:No

[Spy Handler]

Doesn't matter, the audiophile market is not rational (kind of like the wine market). After a certain quality threshold, say 256kbps mp3 or $100 bottle of wine, nobody can tell the difference in a blind test. Yet suckers keep paying money for $500 speaker cables and $1000 bottles of wine. Just stoking ego at that point.

By my estimation...

[fahrbot-bot]

By Young's estimation, CDs can only offer about 15% of the data that was in a master sound track...

... Neil Young is neither a Mathematician or Audio Engineer.

[ -- insert appropriate Neil Young lyric for satirical effect here -- ]

Re:By my estimation...

[gl4ss]

well. he probably did it this way: you got some 6 tracks at 44khz/16bit going into the cd -> the new 44khz/16bit stream has that much less information.

can you hear it though?

Re:By my estimation...

[BenSchuarmer]

Southern man don't need him round anyhow.

Re:By my estimation...

[JBMcB]

He probably meant that most music these days is recorded at 192KHz at 24-bit. Dropping down to 44.1KHz/16-bit equates to around 15% (it really doesn't but who cares)

It ultimately doesn't matter. The nyquist of 44KHz is 22KHz, and few people can hear anything near 20KHz. No unamplified instrument, except maybe a pipe organ or the cannon in the 1812 Overture, have more than 16 bits of dynamic range. So 24-bit playback is a waste, unless you're listening to some really weird techno or something. And, as others have pointed out, not many people have a stereo capable of accurately playing back sounds higher than 20KHz, or anything that dynamic (you'll need a ton of speaker drivers or horns to get that kind of dynamic range)

Re:By my estimation...

[nabsltd]

No unamplified instrument, except maybe a pipe organ or the cannon in the 1812 Overture, have more than 16 bits of dynamic range.

Maybe not a single instrument, but in a whole composition it's pretty easy to go over 16 accurate bits of dynamic range, since the LSBs are fairly inaccurate, which is especially noticable when the overall volume of the signal is low.

Nope, normally.

[BLToday]

Nope. Not if the quality is high enough, I can't tell the difference 99% of the times. There are some musical instruments (harpsichord) and singers (Tori Amos) where compression is very obvious. The lossy version becomes almost unlistenable once you've heard the lossless version.

On "normal" speakers I can rarely tell the difference, but on reference monitors the difference is noticeable on many tracks. Not terrible distracting but still noticeable.

Re:Nope, normally.

[Ichijo]

When you listen to music on electrostatic speakers, you can hear things you couldn't hear before. It makes normal speakers sound muffled as if you're listening through a pillow. So the speakers can mean the difference between hearing the mp3 compression and not hearing it.

Re:Nope, normally.

[triffid_98]

If by normal you mean crappy 'big-box store' bookshelf speakers then yes. We've had efficient 'bright' speakers for quite a long time now, it's just that they aren't the tiny little boxes + a sub woofer that seems to be all the rage today.

Re:Nope, normally.

[sessamoid]

Do you find any recording "unlistenable" after you've heard them live? Because even your lossless recordings are not a perfect reproduction of having the artist playing it in front of you in person.

Re:Nope, normally.

[BLToday]

I go to a lot of live concerts and yes it does sound better than lossless but the "unlistenable" of some lossy track is something different. Lossy compression of some songs produces a lot of missing tonal range. Majority of songs are fine with lossy compression, but some voices and musical instruments sound hollow and lose a lot of depth to them.

Can only speak for myself

[Anonymous Coward]

The problem is not lossy compression per se. In badly encoded mp3 files (plenty of them out there!) drum cymbals sound "watery"- there's somewhat of a flanger-like effect to them as an artefact of the compression. In ogg vorbis files encoded at the same rate I don't notice the same issue.

Also... there's indeed a difference in whether you're listening to the audio in your car over shitty speakers over the noise of a roaring engine, at home in your average living room with decent speakers or in a studio setting with box-in-a-box isolated walls and on the high end studio equipment under the conditions on which the audio was actually mixed.

Otherwise, it's a bit of a mundane audiophile discussion, really. Most audiophiles fail to account for the fact that their mere presence in the room and what they wear makes far more of an impact to the audio quality than the extra money they throw at the problem (for the best listening experience, they should probably leave the room in which the audio is being played).

There are other reasons to choose lossless compression over lossy though- in audio material that underwent lossy compression, some frequency bands are simply no longer present. It is conceivable that this has some impact over how much control you have over the frequency response of the material on playback. You may find that that fancy graphic equalizer of yours won't work as well on lossy audio as it would on lossless audio.

Not that it matters. Stepping away from the intentions of the original audio engineer is blasphemy, anyway.

They're kidding themselves.

[Anonymous Coward]

I've tested the ability of audio professionals to discern differences between high quality MP3's and WAV files in a sound booth. They can't consistently identify which file is playing when going back and forth between the files, even though they often convince themselves that they are hearing distinguishing characteristics. Certainly with lower bit rates people can hear differences, but not with high quality compression settings.

You do much better to spend your money on high quality headphones or speakers than on "hi def" audio recordings or the disk space to store them.

The real question is

[zrbyte]

Can it go up to eleven!?
At eleven who cares, right?

Re:No

[rudy_wayne]

In medical tests, people are given a placebo and yet claim to feel better or feel the same effects as people who are given the real medication. These must be the same people who rail against mp3s.

Just because Neil young and Dave Grohl are famous musicians, it doesn't mean that they actually know what they are talking about. 40 years of exposure to loud music has probably damaged their hearing enough that they really don't know what they are hearing.

Saying that A sounds better than B is completely subjective and affected by many things. Not just how the music was encoded, but the quality of the DAC used for playback and the quality of the speakers/headphones used.

It's the soul, stupid

[deanklear]

For most people, the connection they have is about the lyrics and the memories associated with the song anyway. There's magic in lo-fi recording like this rehearsal, and zero magic in some perfectly recorded crud step (though I do enjoy many electronic artists.) I think the main thing is the yearning in the recording itself.

I'd check out Grohl's keynote at SxSW in any event. He makes some very good points about artists who wouldn't survive American Idol that are a hell of a lot better than anyone who wins it. Music is about expression, about documenting loss and love and joy, and the "karaoke dictatorship" of talent shows that rob people of their voice and replace it with meaningless corporate pablum designed to sell products... well, it's horrible and empty and it doesn't matter what the bit rate is because it's not worth wasting time for.

Wow

[roc97007]

...this really takes me back. I haven't heard a good CD vs LP thread since... well, yesterday.

Re:Wow

[Bengie]

You sound warm.

Trends

[mpol]

There have been more posts on Slashdot in the last 14 years on Slashdot about this topic. What I recall of them, is that people have been tested with blind and double-blind tests. And about ten years ago you could hear a difference between lossless audio and low-bitrate mp3's. The latter has less high and low, and mostly a certain "Hiss" sound through it. The preference was with the lossless audio then.
What struck me in later tests, was that people seemed to favour mp3's above lossless audio. I reckon it has to do with getting used to the Hiss-sound in mp3's, and therefore having it as a preference. A big factor in music taste is how much you are used to hearing similar music and sounds, and the hiss-sound does make a usual sound.

To be fair, I do think that mp3's in a high bitrate like 320 kbit are almost as good as lossless audio. Even though I prefer the lossless audio, just to be sure.

Depends on listener and device

[PseudoCoder]

I remember having to make the excruciating decision of which format to rip my entire CD collection when I was building my HTPC back in 2007. I listened carefully through high-quality studio headphones at the difference and concluding that lossless was going to be the better format for my setup. If I could tell the difference through the headphones, then I figured there would be even more of a difference through my Pioneer Elite receiver and Mirage DefTech speakers.

When I hooked it up, it paid off big time. Sounds heavenly. When I sync from my HTPC to my library and play it through my Samsung Galaxy S3, I convert down to 160 or 192kbps and it sounds as good as I can expect it in a mobile format.

Point is it depends on the setup as a whole. Like any performance chain, your worst component will determine the overall system performance. Furthermore, it depends on the listener. My wife couldn't care if it's coming from my system or from her Coby boom box (WTF?), and I'm the one who's hard of hearing. Big whoop to her.

I can't here the difference, but still want it

[Omnifarious]

The reason I because I want audio I can recompress to the format I like without progressive degradation. Better lossy formats might be created in the future, and I want to be able to re-encode in those formats without suffering the losses due to lossy compression twice.

Yes, I can.

[Nyder]

I don't have the greatest hearing anymore, because most my life I had headphones plugged in my ear. But I can tell the difference between MP3 files and Flac files. Not only that, I can hear the difference been CD quality Flac files and 24bit/96khz Flac files.

My music collection demands I download at least CD quality (16bit/44khz), and prefers I got up a step. At worse, I will try to find 320kbps MP3's, but I like a bunch of older 80's music that I can only find in lower rates.

Sure, I could survive on 320kbps MP3's, after all, that what I have to listen to in my mp3 player. Shit, i survived on Cassette tapes for a couple of decades, and most that music was copied from friends.

There is another part to this story though. Not everyone know how to rip MP3's decent. So when I have a flac of the CD, then I can rip it how I like, the best quality possible.

Does my opinion matter? Fuck no. I'm the same about Video. I can see the imperfections in various codecs that others can't see. And I'm not down with that shit. for example, HDTV via Cable (Comcast) is crap, and I notice it. My dad? He won't notice it, shit, he doesn't even noticed with normal cable gets a bit overworked and gets a little blocky.

Also, I tend to like older music. Led Zeppelin, The Doors, Jimi Hendrix, etc. And you'll find all sort of bad copies of their music. You don't know what the source is, LP, Cassette, 8-Track (joke), that the MP3 was pulled from, so getting flacs of the various CD's is best.

Then worse, you have the volume levels, or compression levels, or whatever they started doing in the 2000+ are a lot higher then previous CD releases. So while you might have a MP3 of Black Dog that is sort of quiet, the latest CD rip would be a lot louder. So now there are a few different sounding MP3 releases around.

Am I a normal consumer? Hell no. I don't buy music anymore, fuck that. I spent enough money on music in the 1980's and 1990's. I'm not making the Record Companies any more money on purpose, they do NOT deserve it.

I'm assuming to most people MP3's are enough, but then I have never followed what "most" people do, I like being myself.

Re:There is a difference

[xorsyst]

To listen to on a player with limited storage. Sure, that's not your only copy, you keep lossless too.

Neil Young and the Placebo effect.

[guidryp]

Neil Young is an aging rocker with hearing damage. There is no way he can tell the difference except in his own mind.

Even among people good hearing, only a minority can detect a difference between lossless and properly encoded higher bit rate (~200K+) lossy.

The hubbub over this is almost all placebo effect and snobbery.

Re:Neil Young and the Placebo effect.

[Rougement]

"Almost all ..." I tend to agree. Most people don't have speakers that can get higher than 20kHz or ears that would be able to hear the benefit anyway. Some people do though.

Remember the context

[zuki]

While I agree that for most consumers it's really a bit of a moot point, the following may need to be kept in mind:

The difference in audio quality may not really be apparent when something is played back on earbuds or tiny computer speakers, rather than on a concert hall-sized system. These differences are very hard to pick out - even in an audiophile home situation - but become far more obvious once these same recordings are played on a 25,000-watt sound rig in a large auditorium.

Like taking a jpg logo you just lifted from a web site and blowing it up to a large billboard on the side of the road. Pixelation will occur, but won't be noticeable until you scale up to those large sizes. And yes, before someone dismisses this as irrelevant, do not forget the thousands of professionals who play recorded music for millions across the planet every week on those large sound installations. (granted, most of whom do not care one bit about audio quality)

But the difference is there, it's just a shame that no one wants to take the time to actually do these listening tests in large-scale environments with proper acoustics (clubs, concert halls, auditoriums). It should be added that if the venue in question has horrendous acoustics and tons of reflections, none of this will obviously matter.

These perceptual compression algorithms do in fact strip out the very essence of what bind the sounds together, the inner dynamics (so to speak) and it's truly a shame that by now it's become the new 'normal'. Even though vinyl is far more imperfect, on large-scale installation it has a much smoother presentation and the bass really comes out in ways that the castrated digital files do not seem capable of generating. The human ear is extremely sensitive to a lot of this once these details become noticeable due to the size of the room.

Re:Remember the context

[Junta]

*Maybe* I could hypothetically buy the argument that the psychoacoustic models of lossy compression fail to preserve data that could be perceived in a concert stadium setting.

However, to broadly say that vinyl is better at scaling up than digital period is silly. It captures all waveforms that the human ear can distinguish. Whether it's 10 watt or 25,000 watt system, the frequencies preserved and available are the same.

Re:Remember the context

[FrankSchwab]

So if I always listened to music in a concert hall with a 25000 watt speaker system then I might care. But instead I always listen to my music through earbuds or my car stereo or even my home stereo - where, as you say, "The difference in audio quality may not really be apparent".

So, to answer the question of the thread, NO, I can't really hear the difference between Lossless, Lossy Audio.

Re:Remember the context

[zuki]

So if I always listened to music in a concert hall with a 25000 watt speaker system then I might care. But instead I always listen to my music through earbuds or my car stereo or even my home stereo - where, as you say, "The difference in audio quality may not really be apparent".

So, to answer the question of the thread, NO, I can't really hear the difference between Lossless, Lossy Audio.

The people doing this are called DJs and their audiences. In case you haven't noticed, this whole EDM trend has sort of really blown up in recent years. It's not a fringe phenomenon anymore.

A few of these DJs (not many) have endeavored to try and keep providing their audiences with the best-possible audio experience. There's no question that lossy audio has less 'depth', punch, 'inner dynamics' whatever you may want to call it which has nothing to do with frequency response, but certainly relates to psychoacoustics and perceptually how it makes people feel.

The very imperfect vinyl still sounds much smoother and deeper (especially the bass) on these very large-scale sound systems. Many UK dubstep DJs still insist on playing vinyl. It's immediately apparent during gigs when the next digital DJ comes on. Their digital files may sound cleaner, but lack that punch, what sound system culture enthusiasts refers to as 'weight'.

Re:Better question

[noh8rz10]

it doesn't matter how lossy or lossless the file is if you're listening with shitty white earbuds.

Oh no....

[QuietLagoon]

... not this discussion again....

Some probably can...

[luckymutt]

...but for me, after years of attending concerts my hearing is shot just enough to not really tell a difference. This includes shows by both David Grohl and Neil Young. I don't mind the hearing loss too much, but don't lecture me now on the best file formats, thanks.

Audio processing is better for lossless

[Dwedit]

If you're trying to cancel the left and right channels by subtracting them, you will get significantly different results depending on whether the files are lossless or not.
When they are lossless, it will work properly. Otherwise, it will have artifacts.

Re:Common misconception

[EvanED]

Thus, the *correct* way to appraise say mp3 is with very good speakers in a treated listening room

No it isn't. At least most of the time it isn't, though that result would be interesting.

If I'm trying to decide whether to archive my CDs to MP3 or FLAC, I don't give a rat's ass what it sounds like with great monitors in a treated listening room, because that's not where I listen to music. If my speakers give a non-linear result that amplifies the distortions from compression, that's what matters; not what it sounds like in an ideal situation.

How does that work, exactly?

[Gordonjcp]

By Young's estimation, CDs can only offer about 15% of the data that was in a master sound track

Where does that figure come from? A CD is a perfect reproduction of the analogue master.

Re:How does that work, exactly?

[gl4ss]

By Young's estimation, CDs can only offer about 15% of the data that was in a master sound track

Where does that figure come from? A CD is a perfect reproduction of the analogue master.

I posted already a guess about this..
It's my belief that neil refers to ALL of the tracks on the master. so if you have 6 master tracks that get mixed into the final product, you'll end up roughly with just x amount of info. not that it matters.

it's a sham though, for selling one product.

Re:How does that work, exactly?

[Junta]

There is no such thing as a perfect reproduction of any analog master. There is, however, a such thing as close enough to be impossible to tell. Mathematically, you can play all sorts of funny games.

Let's say that the master was digital and was 48-bit and 96 khz (to pull almost reasonable sounding numbers out of my ass). CD at 16-bit and 44.1 khz would be about 15% of the data because 85% of the data is impossible for a human to recognize.

This isn't to say the master is captured at a meaninglessly high sampling rate, considering the post-processing involved the overhead may be important to preserve things until the final product is produced. However to take those numbers and imply the listener is deprived of anything is silly. To use those numbers to imply remixers are deprived might be defensible.

Re:How does that work, exactly?

[Gordonjcp]

No, that's the multitrack. The master is the final mix to a stereo pair once it's gone off to the mastering suite and had various EQ and compression applied, ready to stick onto the final print.

If it's a CD, it's printed pretty much flat. If it's going to vinyl then the cutting engineer will tweak the compression and EQ further to make it more suitable for vinyl (mostly, removing all the low and high frequencies so the lathe doesn't freak out).

Re:How does that work, exactly?

[Gordonjcp]

But the analogue master has a lower upper frequency response than the CD, and a higher noise floor...

Sampling rate is irrelevant; what matters is the corner frequency and slope of the shelving filter before the ADC and the reconstruction filter after the DAC. I guarantee you that even the cheapest shittiest CD player will be wider than your analogue master.

Re:How does that work, exactly?

[Junta]

One, are there really 'analog masters' anymore? I assume the recording studios nowadays go straight to digital as a matter of course and any products like vinyl are analog capture of the digital content.

I'm not saying analog is inherently 'better', I'm just saying that any transfer from analog to anything else (including another analog) is not going to be a perfect reproduction. Therefore, you can play numbers with some rational basis to say 'CD loses data from the master', even if the statement is completely irrelevant to the intended audience.

Re:How does that work, exactly?

[ozydingo]

Where does that figure come from? A CD is a perfect reproduction of the analogue master.

Not quite true; even assuming your master is completely band limited to 22.05 kHz (or also at 44.1 kHz sample rate), you still have quantization error.

Re:No

[fa2k]

Depends of how you define "hear the difference". If you gave me two files encoded at 320 with the Fraunhofer codec, and one lossless file, and unlimited time, I could tell the lossless one from the others. Granted Frau isn't quite modern, but its certainly possible

Will hi-def be mastered properly?

[steveha]

I would pay more for audio tracks that are mastered properly.

Far too much of the music released these days is mastered to sound "loud". A sound-level compressor removes the dynamic range, and then the music is gained up about as high as possible, or sometimes higher than that (gained so high there is hard-clipping).

In the best case, the dynamic range is gone and the music loses some of the drama and impact it should have had. In the worst case, the sine waves are hard-clipped into square waves, which sounds terrible. Hard-clipping adds unpleasant harmonics and distortion and you definitely can hear this.

I promise you that a properly mastered track at 16-bit/44.1 kHz will sound dramatically better than a poorly mastered one at 24-bit/96 kHz. Mastering trumps format.

So if they are going to the trouble to make 24-bit/96 kHz tracks, I'm hoping that they will let the mastering engineers do their jobs properly! If they do, I would pay the extra money and bandwidth to buy the music in the higher-quality format.

The music industry is convinced that most of their customers are idiots, unconcerned about sound quality, who can be distracted by shiny things or loud noises; so they try to make every album as loud as possible. But maybe, just maybe, they will be willing to try something different with the high-quality downloads.

http://en.wikipedia.org/wiki/Loudness_war

Re:Will hi-def be mastered properly?

[steveha]

Considering the lyrical content of what tops the charts, I think the music industry's assumption that their customers are idiots is quite correct.

It's hard to disagree with your point. But the problem is that the music industry is remastering old music to be loud, as well as mastering the new music to be loud. I am now deeply suspicious when I see "newly remastered!" on a CD label. Once upon a time that was a promise of improved quality; these days it might mean a "loud" master that is actually worse than the original. And they are doing this, not just for death metal bands but for everything. For example, a Billy Joel pop album is not improved by being overcompressed, but:

http://www.youtube.com/watch?v=3TlQo9k827c

Re:Will hi-def be mastered properly?

[hondo77]

The music industry is convinced that most of their customers are idiots...

In all fairness to the music industry, have you seen/heard the Billboard Top Ten?

Re:Will hi-def be mastered properly?

[bjdevil66]

This is worthy of a "+6, Insightful" moderation.

Assuming we're talking only about audio recorded or remastered by the music industry anytime after around 1990, any debate about lossy vs. lossless file formats for saving music is practically meaningless. This is because the the music industry's sound engineers overdrive the music's levels to the point of distortion and massive sound wave clipping. This is done solely because louder music sells more copies, and if any sound engineer complains about quality, they don't get any future work, so the problem isn't going to go away anytime soon.

In the end, no file format (lossy or lossless) can polish a sonic turd...

Re:Will hi-def be mastered properly?

[TAG13]

I was going to say something similar to this. "Limiting" music is a much harsher treatment of music than encoding it as an mp3. For those unfamiliar with the audio jargon, limiting refers to squashing the dynamics of the sound. The loudest peaks of the song are brought down, and then the whole song is brought up in volume. The net result is that the quiet parts become louder. The current trend in music, especially pop music, is to severely limit the tracks. The loudness of a sound comes from the average volume of the sound, not the peak volume. So, limiting the track makes the song consistently as loud as possible, and thus the perceived loudness of the song is unnaturally loud. Loud things get people's attention, and thus you have the "loudness wars."

Besides the distortion that tends to happen in the limiting process, you lose the dynamics of the song. In some songs I think that's fine and can be exciting, but severe limiting is used far too often. Take Dave Grohl's opinions on sound fidelity with a pinch of salt. Perhaps it's not his decision, but the music he puts out has severe limiting. He just made a documentary about a famous music studio, "Sound City." The audio in the film is nice and dynamic, but the soundtrack for the film (which includes original music that was performed in the film) has been limited hard. There's a video that discusses this case that might be interesting to some: http://youtu.be/O3aCNalLojQ

When audio engineers are mixing and mastering songs for "hi-fi" formats like vinyl and SACD, they are much more delicate with their limiting, if they limit it. I think the hi-fi formats themselves don't offer much value to the listener*, but for songs are treated much better for those hi-fi releases. In the video I posted above, the guy compares the vinyl release of a Foo Fighters album to the CD version. He shows the waveform, so you can visually see the affects of limiting on audio (about 2:45 into the video).

As for lossy vs. lossless: We've gotten really good with our lossy formats. Sure, it's getting rid of information, but it is carefully chosen information that humans ears don't easily pick up. I rip all of my CDs as 320kbps mp3, and I don't hear any difference. Even at 128kbps, only people really focusing in on sound quality will notice a difference. People listening to stuff in their car, listening to cheap ear buds, or just playing it as background music don't care.

(*I think hi-fi formats can be great for archiving history. The human ear doesn't pick up on extra information of higher sample rate or analog playback, but there is still extra information there. For people dramatically manipulating the sound, it's often good to have the extra information. Maybe historians will have reason to comb through some of our recordings in the future and analyze the minutiae. I'm sure they'll appreciate the extra information.)

For what it's worth, I'm study audio engineering in college and will be graduating soon. I've definitely got a lot to learn, but I think my studies of audio give a little weight to my opinions. Limiting and lossy audio is always being discussed in my circles. Hope I offered something useful.

Re:Will hi-def be mastered properly?

[PhunkySchtuff]

Yes, this 1000 times. I'd happily rather have a 256kbs AAC that's mastered properly than have a 24/96 lossless track that's mastered badly with all the dials turned up to 11.

Nope

[Mathness]

I know I can not hear the difference. But for me, the few times I go for lossless, it is simply to have something as close to the original as possible, as I find it worth it to have.

Re:No

[noh8rz10]

dude, my approach is, so what? somebody worked hard to get a little pot of money, and wants to use the money on something that makes him happy. audiophile stuff makes him feel happy. it wouldn't make me feel happy for the price, but who am i to tell him otherwise? Life got a lot easier once i let people be their own people.

Re:Speakers or headphones?

[1u3hr]

What I see almost every day are DJs with headphones plugged into their laptop and it sounds fine to them. The same track out there on the dancefloor sounds like a horrific wall of distortion.

Pumping out enough sound for a disco at high quality is much, much more expensive than a couple of tiny speakers next to your ears. Disco speakers are designed to be loud and have lots of bass. Fidelity? No one comes to a disco to hear a recital. I often listen to videos in my living room using headphones, partly because it doesn't bother anyone else in the house, but also the quality of the (cheap) headphone sound is about the same or better than the "hifi" (again, cheap, so not audiophile "hi"), and much higher than that of the built-in speakers in the TV.

Some scientific mumbo-jumbo

[toxygen01]

This might shine a lot of light into the topic: http://people.xiph.org/~xiphmont/demo/neil-young.html

That is not really the point

[cgimusic]

Most of the time I can't tell the difference. That is not why I use lossless. I use lossless audio because it means I can convert it between hundreds of different lossless formats and it is the exact same quality as it was when I started. It doesn't matter that every audio player I use requires a different audio format. 20 years down the line I can have changed audio formats as many times as I need to to take advantage of better compression or to achieve compatibility with a new player and I will still have high quality audio.

Re:Better question

[MarkGriz]

Or not using Monster Cable

Re:Better question

[Joce640k]

This is the real point: People are so used to listening to music with no dynamic range, on ear buds, in crappy acoustic environments that they wouldn't know where to start listening for a difference.

One question...

[Junta]

I know in imaging that having better than the human eye can see is important in intermediate products as visual manipulation on low fidelity content could produce visible artifacts. Is it the case for audio as well? If someone is going to resample audio for a remix, is there risk of the decreased fidelity ultimately manifesting in the final product?

Re:One question...

[hedwards]

This is always the case when going from analog down to digital. Digital requires a certain amount of rounding. And neither source is likely to be maximum. Ideally, you go with a substantially higher bitrate so that the rounding errors are minimum, then after you've finished wroking on whatever it is, you convert down to something that's much closer to what humans are capable of hearing.

It applies to pretty much anything where you're converting from analog into digital. Now if we had unlimited storage space and unlimited processing power, you could get very close to identical to the original analog source, but even if that were possible, it would be a waste to do that.

Re:One question...

[ozydingo]

Yes, for both bit depth and sampling frequency. Here are two possible reasons why:

1. Bit depth. Remix wants to amplify a sound in the original mix. At 16 bit depth, you have 2^16 possible values to cover everything from silent to max loudness. If you take a soft sound that uses only some of those values and amplify it, the result suffers from possibly noticeable quantization artifacts. This is like magnifying a small picture to produce a pixelated one.
2. Sample frequency. Remix wants to frequency-shift / pitch-shift a sound in the original mix. Your sampling rate determine the max frequency you can encode, so any audio in a 44.a kHz file has a max frequency range of 22.05 kHz. Say you shift something down by an octave (factor or 1/2); the shifted sound will be cut off at 11.025 kHz.

How much these effects are noticeable in typical mixes is up to the listener...

Re:One question...

[Dennis Flynn]

It's the case with everything really. As you make a copy of a copy of a copy of a copy, ad infinitum........entropy becomes the dominant factor. You eventually have nearly random noise. Images, sound, you name it.

Re:One question...

[PhunkySchtuff]

In a word, yes.

You'd be crazy to record, mix and master all at 44.1/16 as you need the headroom to work with, to adjust the volume, to mix things together.
The final result that people end up listening to though is just fine as redbook CD audio.

Monster Audio File

[Stormy Dragon]

It seems to me this lossy vs. lossless compression debate is the information theory version of the $20,000 speaker cable. I'm willing to bet that in any blind trial, 99.99% of the population can't detect any difference. Pretending they can is just a way to conspicously signal that they care way more about music than you do with your $5 HDMI cable.

Could it just be suggestion?

[darth_borehd]

I've listened to both and have never been able to tell the difference.

I wonder if they tested it in a double-blind experiment if audiophiles could choose the uncompressed music better than random chance.

No need to

[jones_supa]

We can take a song, whack it with a MP3 encoder and say "there, just like new" and it takes less space. But do we have to go through that process? There is plenty of HDD space and we can use WAV/FLAC to always enjoy the original quality without compromises.

Re:No

[fatphil]

And if you put them up for a test, and told them which source was which in advance, I'm sure they'd be able to tell you the flaws in the one you said was the mp3 (or whatever). Even if you deliberately swapped the cables over.

Yes.

[smadasam]

Next question.

Yes.

[smadasam]

Yes, next question.

Depends on far too many things.

[gmarsh]

Lets see...

- The music being encoded. Some songs have combinations of sounds which don't encode well.
- The encoding format, and the type of artifacts that it produces.
- The bitrate and other encoder configuration.
- The playback gear being used, and the listening environment. A quiet environment and gear with clear treble reproduction will tend to highlight encoding artifacts.
- The listener, and whether they know what to listen for.

I spent most of a decade designing broadcast audio hardware and DSP code, and as a result I've become pretty good at picking out glitches/artifacts/etc - especially with familiar songs. But I'm not most people.

Re:Speakers or headphones?

[Voyager529]

As a DJ, I'll hopefully shed a little light on this:

1.) few DJs of any skill and seasoned experience will use the internal soundcard. Odds are good that they're using an offboard unit like the Audio8, Rane SL-2/3/4, Presonus Firepod...or some similar interface that costs between $500-$1,000 and is built properly with audio isolation, etc.

2.) Songs that an average top 40 DJ plays are likely compressed to hell and optimized for iPod earbuds, not actual speakers. Some people have the time and opportunity to optimize a 64-band EQ for each song they listen to, but as a DJ I'll say that the odds are good that I've got less than 90 seconds to pick and queue my next track; I'll fix the EQ if there's a particularly audible distortion or it sounds like it'd be "piercing" on the floor, but other than that I've got bigger fish to fry when I'm spinning.

3.) Some DJs get songs "the right way" and are properly encoded. Others still traverse Frostwire or whatever the latest P2P music sharing app is. There's no guarantee that the jock didn't get a sucktastic encode and didn't realize it until it was too late...

4.) Even if the audio is clean up to the mixer, there' s no guarantee it doesn't hit an old/horridly configured/poorly cabled preamp, crossover, or amplifier before it hits the speakers. The speakers themselves could be in dire need of a reconing or be mismatched with the amps, etc. There's *plenty* of areas in the post-mixer realm for audio quality to degrade. Even if you're getting a mostly-perfect, amplified signal, the "sizzle" gets louder, too. If the sizzle is -90dbm, and you're amplifying it by 100dbm, it's gonna hit that floor well before headphones will.

There is only one reason to use lossless formats

[Anonymous Coward]

If you are a content producer you MUST have the original records in a lossless format.

This has NOTHING to do with you being able to know the difference, the main reason is because you are going to distribute/edit/re-master/convert your audio data to a different lossy format with different algorithms.

Lossy algorithms "remove" information and different algorithms remove different parts of the original audio data, if you store your originals in a lossy format son or later the audio quality is going to start to degrade until you start noticing the artifacts.

Its like in photography, you took raw pictures not because you can look the differences compared to a high quality JPEG, you took raw pictures because if you want to edit something later you will want all the original data available at the time that you took the picture.

Re:No

[BarfooTheSecond]

ha ha h, what an ethernet cable!

Yes, it's the same kind of "audiophiles" who are chasing that perfect amp with 0.000000000001% of distortion ratio, which would require ultrasophisticated and expensive lab equipment to be measured, while a much higher ratio would be inaudible to them. And anyway, they'd shell out several grands for that.

These guys who are more interested in the specs of their audio system than in truely listening to the performance and musical intentions of the musicians. (I'm a musician and I don't mind if sometimes there is a slight difference. Sometimes it sounds even better!)

sometimes, but lossy audio isnt the worst problem

[AxemRed]

I don't think that lossy audio compression is inherently hurting recorded music. Lossy is fine as long as good encoders and sufficient bitrates are used. At a certain point, no one can tell which is which (lossy or lossless) in a blind test.

I mostly listen to MP3 encoded rock music. The loss of quality is very noticeable to me at 128kbps. The loss of quality is much harder to discern at 192, especially if a quality encoder is used. I use LAME -V 2 when I rip CDs and usually end up with average bitrates from ~190-215, and I can't tell the difference between those MP3s and the original CD.

IMO there are bigger problems facing recorded music anyway. See: http://en.wikipedia.org/wiki/Loudness_war

Re:A lengthy, thorough, and well-explained discuss

[fredrated]

You jerk! I clicked on that link!

Re:No

[osu-neko]

Doesn't matter, the audiophile market is not rational (kind of like the wine market).

Show me a rational market, and I'll have to inquire as to the nature and evolutionary history of the species of aliens participating in it.

Yes, but it's subjective.

[funkyjunkman]

I have engineered and mixed songs for decades. My training over the years makes me very aware of when I'm listening to something compressed. But who cares? Me, of course, but to you it might not matter.

* Can you really tell the difference between *
- A Picasso and a reproduction?
- Genuine marble and simulated materials?
- HD video and Film projection?
- $500 shoes and $50 knockoffs? ...

I'm happy that technology and storage has allowed me to retain my music (previously on CD) as lossless files for my enjoyment today. For some of my friends, they are completely happy with 256k AAC or MP3 files. That's the way it goes!

AIFF?, Flac!, Lossless in General. & Randomnes

[neoshroom]

I've been into compressed lossless audio from the start. First, AIFF is definitely not one of the most popular lossless audio formats for distributing music because the popular formats are compressed lossless audio and AIFF is uncompressed. The top formats are FLAC, APE and ALAC. FLAC is the most popular because it is open-source and versatile. APE was highly popular in the late 90's and early 00's and still is with some because it has better compression than any of the other formats. However, as time went on hard drive space became more plentiful and mobile devices started popping up. APE achieves its superior compression via calculations that are more intensive than FLAC uses and thus more taxing on mobile devices. It is also less cross-platform-compatible. ALAC is Apple's Lossless Audio Codec and is a latecomer onto the scene. It has good iTunes support and slightly better compression than FLAC, but that's about it.

Also, it is definitely possible to tell lossless audio from lossy audio, even at higher bitrates. Around 2002 I had a friend who completely mocked my lossless ways, even though I'm not one of those gold-cable audiophile people -- just a normal guy who likes his music. I just had a decent pair of Klipsh speakers with a subwoofer. My friend was so certain that this was all in my head and I was so certain that it was not that we devised a simple test. He would show me two identical-looking files in iTunes, just showing the titles. One was a high-bitrate AAC and the other a FLAC file. I could click on them to play them as much as I wanted. I was then to decide which was lossless and which was lossy. We did this with 10 files. It was basically double-blind as he didn't know which was which either until he took the computer back to check my answer. He set up 10 files this way. All in all the test took just 5 or 10 minutes.

I got 9 of 10 right. It is hard to describe sounds, but the lossless music is "deeper," especially bass, guitar vibrations and high notes. This makes it obvious for many songs.

However, I expect not everyone has hearing like this. I suspect this because one day I heard this annoying buzzing sound and asked my girlfriend about it. She couldn't hear anything. So, I searched all over for what was causing it. It turned out it was a television that was on, but that was on a non-channel so it was completely black on the screen. However, the CRT television emitted a sound from being on in a silent room that I found annoying and my girlfriend couldn't even hear. My sister could also hear it when I tested her later. I also sometimes find the sounds fluorescent lights make annoying too.

Anyway, lossless is great and, yes, you can hear the difference if you have hearing which can hear the difference. It's sort of tautological, but it's the truth.

My Torture Test

[JBMcB]

The opening of Royal Oil by the Mighty Mighty Bosstones. It starts out with a quiet snare roll that gets progressively louder, joined by a simple bass line. I've yet to hear a lossy codec at any bitrate that doesn't turn it into watery gibberish.

Disk space is cheap. Rip to FLAC or ALAC. For portables, 256kbps AAC seems to do the least amount of damage.

Re:My Torture Test

[evilviper]

For portables, 256kbps AAC seems to do the least amount of damage.

Try your test track with Musepack or MPEG-1 Layer2 (MP2). There's a world of difference between those two temporal domain codecs, and common frequency domain codes like MP3, AAC, Vorbis, Opus, etc.

Re:My Torture Test

[jmv]

And try Musepack or Layer2 on something extremely tonal like harpsichord or (to a lesser extent) 12-string acoustic guitar. Each type of codecs has upsides and downsides. Overall though a freq-domain codec *with* a good encoder should be better because it still has the option of going with a good time resolution (Layer2 can't ever use a good freq resolution).

Re:My Torture Test

[evilviper]

And try Musepack or Layer2 on something extremely tonal like harpsichord or (to a lesser extent) 12-string acoustic guitar.

It'll still work fine, it's just not the most efficient. But with that really only matters with streaming these days.

There's no question temporal domain codecs aren't the best at low-bitrate coding, but at high bitrates (192k+), which are commonly used for music downloads, CD rips, etc., they do a far better job than the more common codecs, and it's very strange how we ended up with the most ill-suited codecs being used far outside of their niche.

And there's always the option of a hybrid like AC3... What do you say about merging Opus and Musepack? It's free. You've already got CELT+SILK, what's one more coder? ... *cough*

Re:No

[mhesd]

digital is objectively more accurate.

but music isn't

Vinyl is hissy and wears down over time.

[Timmy D Programmer]

Vinyl is hissy and wears down over time and tapes distorted the hell out of music. The quality of the new higher quality mp3s is one of the biggest reasons I tend to buy more music these days than ever. I think nostalgia has tainted their hearing.

If lossless is preferred...

[rlwhite]

...then there should be a market for lossless albums on DVD. I'm not an audiophile, but I haven't heard of this happening. Is there one?

Lossy vs Lossless

[David_Hart]

When converting my CD collection I first used FLAC and then converted the FLACs to MP3 VBR 320kbps. I've listened to both and can't tell the difference. With Lossy, a high bitrate definitely is better. I can quickly tell if an MP3 has a bitrate of 192kbps or lower. I've also been buying MP3s from Amazon at 256kbps or higher and I've purposefully stayed away from iTunes (originally due to DRM and low bitrates).

Of course, it makes a difference in the playback equipment. I replaced the manufacturers Bose system in my car with a Kenwood + Infinity Reference speakers. The sound quality difference was like night and day. I'm now hearing a greater range of sound with clear separation, and this is with my Lossy MP3s on an iPod. Personally, I would prefer to use a Creative Labs MP3 player (better quality sound) but Kenwood only offers an iPod connector kit and only the iPod works with playlists, etc. with control from the deck.

lossy?

[Frontier Owner]

this is not a loss of electrical energy. and for me, 90% of the time the ear buds are there to block the office noise more than listen to the music. I keep it low enough to hear when someone is trying to get my attention. the other 10% is my drive home and my truck stereo isn't that good either.

Re:Better question

[Tharkkun]

This is the real point: People are so used to listening to music with no dynamic range, on ear buds, in crappy acoustic environments that they wouldn't know where to start listening for a difference.

Nor can they afford any better so while they are listening to a lesser quality, they couldn't begin to purchase equipment to give them what these artists say they are missing.

Comment removed

[account_deleted]

Comment removed based on user account deletion

People can't tell the difference above 128kbps

[sc0pie]

Coding Horror did a great experiment with their readers where they provided several samples of the same song at different bitrates and then had everyone vote on which they thought sounded best. The result? People could only tell the difference between 128kbps and everything else, and even that was not overwhelming. In fact, 160kbps beat CD!

Re:People can't tell the difference above 128kbps

[locopuyo]

The experiment was done using as song recorded in the 80s... You probably couldn't tell the difference if it was played back on a cassette.

Re:No

[hedwards]

In practice 192kbps variable Lame preset standard is good enough for pretty much anybody. Now, it might not quite beat uncompressed for some people, but it's close enough that I don't bother to worry too much about it. It's better than the head phones that most people use to listen to their music with.

Re:Better question

[coldfarnorth]

Good point. Sadly, my $3k hearing aids don't seem to help either.

Bitrate doesn't matter much if your ears are the lossy part.

Dodged the bullet

[tannhaus]

Thank God my hearing isn't worth a crap and I don't have yet another thing to geek over.

As long as Frank Sinatra doesn't sound like Donald Duck, I'm cool with it.

Re:No

[inputdev]

Yet suckers keep paying money for $500 speaker cables and $1000 bottles of wine. Just stoking ego at that point.

I completely agree about the speaker cables - and while I don't have enough money to spend $1000 on a bottle of wine to know for sure, I do think that there is a psychological phenomenon similar to a placebo effect that actually makes drinking the expensive wine more pleasurable. Here's some cool research: http://www.gsb.stanford.edu/news/research/baba_wine.html
You're still probably right about stroking ego, but if I had billions of dollars, I might try the $1000 bottle. :)

Re:No

[Nikker]

I think what these artists are trying to point out is they spent a lot of time and money making $X, with compression technology and storage being adequate maybe they just want the ~80% of the audio to at least be available somewhere in some format.

Also, why not? Right now downloadable content is sold at the same prices as its "real" counterpart. They don't pay for pressing, printing album art, shipping, depreciation while it sits on a shelf or paying indirectly for the brick stores to pay leases and wages.

So maybe the question is if the artist paid for it and your going to pay for it, why not get it?

Re:No

[PRMan]

The Hydrogen Audio people with "golden ears" did hearing tests on LAME encoding at alt-preset medium (about 240kbps typically) and NOBODY could tell the difference. That's why LAME hasn't changed lately. There's nothing left to do if you leave it at this setting. 320 won't be any different.

Usually no, occasionally yes

[julian67]

If you've actually done some blind testing such as abx then you'll have had to swallow your pride and admit that in general you can't distinguish lossless audio from lossy until the lossy bitrates plummet.

But there are specific "killer samples" that expose the deficiencies in lossy encoders. For example there is a sample called eig_essence on which mp3 encoders completely fail and which ogg vorbis requires very high bitrates to encode without smearing. Modern codecs do a lot better: iTunes AAC encoder or Fraunhofer's AAC encoder will encode of the same sample at moderate bitrates with the sound indistinguishable from original.

eig is an extreme example because most people won't have anything in their music collection that sounds similar (amphetamine addicted techno freaks excepted), but there are other well known problem samples (search somewhere like hydrogenaudio for trumpet and castanets) which are the kinds of music you might own and hear often.

When people say that they can distinguish lossy from lossless they shouldn't be dismissed out of hand but the claim should be able to survive simple scrutiny i.e. a blind test. And if I can't hear any difference between a lossy encode and lossless it doesn't mean that someone else can't, only that I can't. There are irrational people who assert they can identify lossy from lossless 100% of the time, or 44100 Hz from 96000/192000 Hz, and conversely there are irrational people who believe their subjective experience with their $20 ear buds and cellphone music player extrapolates to "everything sounds the same".

Re:No

[zzsmirkzz]

In medical tests, people are given a placebo and yet claim to feel better or feel the same effects as people who are given the real medication.

People don't claim to feel better, they do feel better. There is no incentive for them to lie, in fact, there is a disincentive for them to do so. The reason behind the cause of the "placebo" effect is in the mind of the patient. The patient believes they should be getting better and then they do. Power of thought, belief and, if defined correctly, faith. Really, it is the power of consciousness which no one fully understands.

This can be applied to apparent differences in audio formats. The observer believes that one source should sound better and then it does. Since qualifying better/worse is entirely subjective, objectivity has no place in the argument.

Lossy vs lossless is not the issue

[Anonymous Coward]

The problem is far worse. There are actually two problems with music distribution today:

- Massive compression of every music track to make music as loud as possible, eliminating the concept of dynamics

- The fact that a large majority of consumers of music have grown up listening to it as lossy MP3, and EXPECT to hear artifacts in their music. They think this is normal. When they hear a correct, lossless version of their music, they think it sounds "wrong."

Re:Better question

[mwvdlee]

Look, you want your 0's and 1's to look like stupid Comic Sans 0's and 1's or like high quality, stylish Zapfino 0's and 1's?

I knew this article was gonna be BS

[SD-Arcadia]

"By Young's estimation, CDs can only offer about 15% of the data that was in a master sound track"
And nothing of value was lost in the remaining 85% of the *data* that is inaudible to the human ear.

"Young, in fact, created his own digital-to-analog conversion (DAC) service called Pono. Young has tweeted that the Pono cloud-based music service, along with Pono portable digital-to-analog players, will be available by summer."
There's your cash-in scheme lurking behind all the BS.

"Young's service would increase the quality, or sampling rate, of the music from 44,100 times per second in a CD (44.1KHz) to 192,000 times per second (192KHz), and will boost the bit depth from 16-bit to 24-bit."
I would like to repeatedly hit you over the head with http://people.xiph.org/~xiphmont/demo/neil-young.html

"The sample rate of a digital file refers to the number of "snapshots" of audio that are offered up every second. Think of it like a high-definition movie, where the more frames per second you have, the higher the quality."
NO, do not think of it like that unless you're a charlatan. Refer to rebuttal on xiph.org.

"Millions of people in the world are audiophiles."
No doubt, Millions of people in the world are fools and they have money that could be yours.

"It's just common sense that the higher the resolution -- the more data that's in an audio file -- the better the sound quality, Chesky said."
Too bad this thing called SCIENCE has been trumping "common sense" for millenia now.

"The site also recommends high-resolution player software such as JRiver, Pure Music, or Decibel Audio Player. The software, which basically turns your desktop or laptop into a music server or a digital-to-analog converter,"
HILLARIOUS. I won't even begin to..

"The most popular music server among audiophiles, according to Bliss, is an Apple Mac Mini."
This is beautiful. I am not surprised in the least to see this audiophile-appleophile overlap.

Re:I knew this article was gonna be BS

[ddd0004]

This is good to know. The next time I'll just tell my wife I was searching for the Neil Young audio conversion service "Pono" and I mistyped it.

Re:I knew this article was gonna be BS

[residents_parking]

So many fallacies! It's too beautiful!

Re:I knew this article was gonna be BS

[pauleir]

The rebuttal you link to on xiph.org ignores research that illustrates that humans can in fact perceive frequencies far beyond the classical limit of ~20 kHz. Higher frequencies present essential localization cues. Higher sample rates, like 192 kHz., allow for the reproduction of higher frequencies (assuming playback equipment that can actually reproduce the higher frequencies) leading to recordings which are far more realistic than what is possible with the 44.1 kHz sampling rate.

The difference between 24-bit and 16-bit amplitude resolution is like night and day. As someone that has recorded much contemporary and classical concert music, I can certainly attest to the huge difference between the two bit rates. If you listen to music with a wide dynamic range, then the comparison between the two bit rates is highly noticeable. Quiet sounds can be masked by quantization noise. You want the highest bit rate possible.

Re:No

[bobbied]

Doesn't matter, the audiophile market is not rational (kind of like the wine market). After a certain quality threshold, say 256kbps mp3 or $100 bottle of wine, nobody can tell the difference in a blind test. Yet suckers keep paying money for $500 speaker cables and $1000 bottles of wine. Just stoking ego at that point.

Yea, the audiophile market is full of snake oil sales men too. I laugh when I listen to your average sales person even at a high end store explain why system A is better than B or why your home system is all wrong. Then they resort to the "side by side" test and I can almost ALWAYS guess what "sounds better" before they demo it by looking at the type of speakers. Ported speakers will usually win because they are louder and have more base so switching from A to B and not changing anything louder "sounds better" to most. Problem is, usually the less sensitive speakers are better so suspended speaker designs (without ports) will actually produce better results, you just need more gain/Power in the amp.. Sales guys don't understand *any* of this usually they are just looking to get a fool to part with his money.

This "can you hear a difference" reminds me of past audiophile debates. Tube amps over solid state ones, where the tube guys swear their amps are better and more 'mellow" than that harsh solid state.. Or analog over CD recording where analog just sounded better than that harsh digital stuff. Now we are debating Codecs, sample bits and sample rates in areas where it is generally ridiculous to think *anybody* could hear the difference.

In reality, what you can and cannot hear and what is "good enough" to listen too is probably a lot less quality than you imagine. Unless you have unusably good hearing, do this for a living, have excellent equipment installed in really good acoustically designed listening environment you are unlikely to know the difference between an MP3, CD or High Bit Rate recordings until the compression rate gets pretty high. You might be able to hear a difference, but I doubt you can identify the higher quality material in a double blind test. Just like I bet I can get you to pick the junkiest pair of speakers in the place as the best sounding if you let me "adjust" between the side by side tests.

Re:No

[PRMan]

And if you used the LAME codec, I would wish you good luck at 256. Cause you're gonna need it.

satellite radio

[anyaristow]

That's interesting. I'm not particularly sensitive to compression artifacts, but this is the effect I hear with satellite radio.

YES.

[lkcl]

i can. it's like playing music through grated cheese. it's typically cymbals, trumpets and other complex sounds that i notice particularly are affected.

an associate who worked for a Real-time Audio restoration company - his job was to spot audio discrepancies such as phase errors on old mono tracks that had been incorrectly recorded in stereo - could tell even *more* than i could ever notice.

basically it entirely depends on YOU. if your aural cortex and your ears are sufficiently developed / not-damaged, you WILL notice - it's as simple as that.

Re:No

[Waccoon]

For chiptunes, I can hear a difference between 256 and 320, but just barely.

The biggest factor is how the high frequencies are filtered out before the audio is compressed, because the filtering appears to be the same regardless of the final bitrate. Even ultra-high bitrate audio will sound awful if the stock frequency cutoff is used, and I have to fiddle with the settings in LAME to make my songs sound good, even at 320.

It depends

[Miletos]

The issue for me is that mp3's only sound good if you listen to them "as is".

I do DJ work every now and then. If you use DSP's, as I do, like equalizers, compressors and all sorts of stereo/surround effects, the resolution in lossy audio is SO limited that artifacts become clearly audible to most people.

Even home systems suffer from this, albeit to a lesser extent. Systems that are calibrated for the room with a certain equalizer setting, 5.1 receivers that upscale stereo to virtual surround. They all mess with the audio source in a big way...and when that source is lossy, sometimes you can clearly tell.

Uncompressed audio simply has more resolution to play with. And that's why most of my music collection is now FLAC. Compare it to photoshopping a PNG vs. a JPG.

Re:No

[LordLimecat]

There may be more than ego-stroking to it.

To some degree it may be because doing the research to find the best fit is considered hard or tedious. Monster / Pear / whatever are total rip-offs, but im sure they are as good quality as the best "cheap" brands-- that is, that the $1000 Pear cable is as good as the best Monoprice offers. On the other hand the user may not know enough to avoid the crappy uninsulated cables that truly do introduce distortion or crap out or fail or have faulty connectors. Its possible that they (or the person they contracted out to) knows that there is a budget for the "high end", and its easier to simply pay for it and be done with it than stumbling around in ignorance looking for the non-crappy part.

Re:Let a professional optimize things

[arth1]

Leave it up to the sound engineer to create "optimal" versions using various codecs and compression rates, and let him recommend which versions are "good enough to sell" as a retail full track.

I'd say yes to that, except that many of these sound engineers are the same ones that brought us the loudness wars, and effective dynamics of 2-4 bits instead of 16.
Yes, there are good sound engineers, but they are in minority.

The last CD I bought from a live band, I looked at the sound data, and found that not only did it have a non-existing dynamic range, but it also had a cap (lowpass filter) at around 12 kHz. My guess is that it had been mastered in MP3 by one of the band members.
What can I say - it sounded good when drunk.

Re:AIFF?, Flac!, Lossless in General. & Random

[evilviper]

we devised a simple test. He would show me two identical-looking files in iTunes, just showing the titles. One was a high-bitrate AAC and the other a FLAC file. [...] I got 9 of 10 right.

AAC (like MP3) is a frequency-domain codec, and can therefore never provide transparent audio. It has nothing to do with "deeper". but instead is an inability to represent transients... non-tonal components like percussive sounds and other noise.

If you had performed the test with Musepack/MPC or even MPEG-1 Layer II at high bitrates, you would have failed the test.

http://en.wikipedia.org/wiki/MPEG-1#Quality

Pono Increases Bitrate and Depth?

[Zaphoddd]

A quote from the article.. This smells funny: "Young's service would increase the quality, or sampling rate, of the music from 44,100 times per second in a CD (44.1KHz) to 192,000 times per second (192KHz), and will boost the bit depth from 16-bit to 24-bit." Too lazy to actually go there, but if it means convert your cruddy 16bit 44.1k music into clear 24bit 192k music.. ehhhh...

A continious blast of noise

[FilmedInNoir]

Range compression ( http://news.cnet.com/8301-13645_3-10360787-47.html ) has done more to destroy the subtitles of music than lossey/lossless formats.

100% Yes

[Murdoch5]

If you don't believe a different exists just grab a flac file of your favorite band and that same file as Mp3, Ogg, Wav and anything else you want to try. If you have good hearing and good quality sound equipment you'll hear a dramatic difference. Of course listening to lossless sound files on crappy sound systems will leave you with an experience equal to mp3, this is why it's important for music lovers to always buy the best sound cards possible, at least for computers.

Re:Better question

[Joce640k]

Nor can they afford any better so while they are listening to a lesser quality, they couldn't begin to purchase equipment to give them what these artists say they are missing.

Plus none of them have a special quiet room where they go to to sit down and do nothing but listen to music.

They want music/noise constantly and as a backdrop to whatever else they're doing at the time.

Depends on your ear...

[jd.schmidt]

There was an experiment I heard about on Radio Lab where several pieces of colored paper were given to someone to tell apart, supposedly this would identify someone with 4 color receptors rather than 3 (a small percentage of woman) and was largely analogous to color blindness tests (normally men). In theory people with 3 color receptors would be unable to tell the hues apart.

What they found is that some people with lots of experience working with color could tell the color samples apart fairly easily, while most people literally could not. A lot had to do with training and life experience apparently. So yes, some people really see more colors than you because they are trained to as incredible as that sounds.

Sound could be the same way. Plus, depends on your stereo system I guess.

Re:Depends on your ear...

[residents_parking]

Exactly. Uncommonly for Slashdot, the psychological aspect has been *completely* ignored. It's *obvious* that a great deal of neural training takes place in musicians and other folks working in the field, and this is being broadly dissed, largely due to ignorance, with a bit of prejudice added for good measure.

I hate this question

[holophrastic]

I've hated it for decades now. Whether or not I can consciously discern a difference has always been irrelevant. Similarly, whether or not my speakers can produce a difference is equally unimportant.

To the latter, assuming that I've "purchased" the music, and intend to retain it (as opposed to one-time streaming), at some point in my life I'll be using better speakers. Music lasts a really long time.

To the former, 3-minute listening tests are meaningless. Listen to the same song/album/artist/format for ten hours straight -- something I do recreationally, professionally, as background to work, and for inspirational moments. Some formats produce headaches. Some produce zero inspiration. Some have me "tired of listening to music". Others produce no headaches, tonnes of inspiration, and have me enjoy ten hours of music.

There is a difference. And not all differences are at the top of cognition.

Some music has my cat leaving the room.

It really depends.

[John Pfeiffer]

Lighter stuff and instrumental music you probably won't hear the difference above 192~256kbit MP3, but anything 'busy' always sounds more flat to me. Like comparing a photograph to the original scene. I also have the misfortune of being able to hear compression artifacts...

Probably because the psychoacoustic models used to design these adaptive transform compression schemes are based on the AVERAGE human and I would charitably be described as 'abnormal'. ;)

Re:Better question

[Minwee]

That's a myth. Monster cables are no better than cheaper products from other vendors.

If you can hear a difference, then it's probably because you have your ethernet cable connected backwards.

Re:A lengthy, thorough, and well-explained discuss

[Minwee]

You need to go deeper.

My own complaint

[globalist]

"Major recording artists, such as Neil Young and Dave Grohl, lead singer of the band Foo Fighters, have been publicly critical of compressed file formats and the "significant loss" data, and therefore music quality, consumers are suffering..."

I, on the other hand, have been publicly critical of the crappy musical ideas and songs that Mr. Grohl has been churning out over the years since Nirvana has disbanded and, likewise, no-one will listen to me and just keep on buying Poo Fighters music like it's the second coming of Grunge... So go figure. But seriously, the older he gets, the more I feel Grohl he should just shut the hell up. Who's with me?

Agreed

[PhrostyMcByte]

I agree to a point. Using good headphones and specific songs, I can tell a 192kbps VBR MP3 from FLAC.

You need to train for it, though. For me, 192kbps VBR is transparent in nearly all cases -- I really need to listen for specific things in order to hear it, and only in some songs. Before learning what to listen for, 160kbps VBR was completely transparent to me.

I keep all my music in FLAC at home, but really only because disk space is cheap and if I ever feel like moving from MP3 to Vorbis/AAC on my DAP, I can do so without re-ripping.

Re:Phase

[fa2k]

This is an interesting question. I hope someone else answers it, but I will have a crack at the maths

Re:Better question

[rocketjam]

Shitty black earbuds okay then?

Re:Better question

[Psyborgue]

Have you found there is some music you just cannot appreciate? I ask because my partner has hearing aids and he seems to only appreciate percussive or very plain, melodic, music. If there is a lot of stuff going on, if the music is very layered, it's like he can't hear it, or the hearing aids flatten the music or cut out some important frequencies or... I don't know. Have you found an FM reciever for your aids helps you appreciate music more, if you have one, by bypassing the air gap? We've been thinking about saving up to get them, but they're very expensive. Are they worth it?

Re:No

[Githaron]

Because the publishers want to sell you double HD, triple HD, etc. in the future. They can't do that if they sell you the master quality recordings from the get go.

Why you might care...

[Overzeetop]

No matter how much space is on my current player, I never have enough space to hold my whole collection (well, except on my 160GB iPod classic...but I digress). That means either juggling what is and isn't on my device, or compression, or both. And, its entirely possible that I might choose a player that doesn't work well with the format I've chosen (cough*mp3Pro*cough).

Having a lossless version of everything means never having to worry about re-compression. My perception trails off between 200-230kbps. I can deal with 192 pretty easily, and 128 isn't the end of the world if I'm in my car or am on a cheap pair of earbuds. Heck, on my SwimP3, 64kbps is overkill. But a 200kbps that then gets re-coded to 128 can really end up with some weird sounding shit. So all my old CDs were ripped to FLAC. When I switched from Creative players to iStuff, I just recoded all of my library from FLAC to ALAC. No loss, no worries, no re-ripping. Most of what I buy today gets ripped straight to ALAC, but if I ever ditch apple, I can just recode it back over to FLAC.

It matters that you get a lossless format because then you can convert it to any format that works for you. And if you change formats in the future, just re-code and never worry.

Yes! Undoubtedly!

[frootcakeuk]

The squelch and audible artifacts has been something that has pissed me off ever since the birth of mp3. Admittedly, it has got a lot better over the years but to answer the question, it still very much audible, at least to me, in bit rates up to and exceeding 320kbps. I do accept that mp3 has never sounded better than it does today (depending on the encoder) and to pick out these subtleties does require a bit of concentration to pick it out, but it is still there! The case gets much worse for formats such as aac (eaac+) and other super-high compression formats. They(codecs) do a fantastic job of getting an astonishing amount of high quality audio into smaller and smaller space and as incredible as it is, perfect it is not. (In fact I used to (12-13 yrs ago) use an encoder with a low-pass filter cutoff at 15khz to stop any artifacts above that frequency and to dramatically lower my filesizes as I was trying to fit as much as possible onto a 64Mb Memory Stick Sony Clié. Room for improvement but worked like a charm! I do also suspect that mobile phone manufacturers are fully aware if this and design their included headphones to filter out the squelches and whatever else to make the audio sound that much better using lossy formats. I'm not convinced this is true but it wouldn't surprise me in the slightest.

Are we still talking about this?

[endus]

How many articles have I seen on this on Slashdot?

The answer is that, yes, you can tell the difference and your ability to tell the difference increases with how discerning a listener you are and how good your audio equipment is. We don't need to debate this any more.

Re:Better question

[t4ng*]

I think the real point is that there are known limits to human hearing and many audiophiles fantasize about their hearing being superhuman. It just ain't so. Dynamic range compression is one thing, but perceptual compression, sample rate, and bit depth are a different matter. No audiophile has ever heard the difference between FLAC and 320Kbps mp3 audio in an ABX test at a statistical rate that is better than guessing.

Any time this argument starts, I refer people to this well written article that lays out the limits of human hearing compared to the specifications of recording formats...

The only way to know is to test.

[Dr. Crash]

You won't know until you test. So I did. Here's my results:

With the aid of my girlfriend, I tested myself to see just what I could tell apart. The test music was "Veteran of the Psychic Wars", by
Blue Oyster Cult, listening through some very high end Audio-Technica headphones I picked up in Akihabara earlier that year.
I tested:

16bit WAV (GRIPped right from the CD, 1440 Kbit equivalent)
320Kbit LAME ABR MP3
256Kbit LAME ABR MP3
192Kbit LAME ABR MP3
128Kbit LAME ABR MP3

I found that the WAV and the 320Kbit LAME were "different", but I couldn't tell which was better. So, dead heat. I could tell that the
256Kbit LAME encoding was pretty damn close, but not quite as clean (the snare drums were the giveaway). Anything less was
clearly not as good. 128Kbit was practically unlistenable when I A/Bed it against the WAV or 320Kbit, it was that bad.

So there; now when I rip my CDs I keep the .WAV and encode
at 320Kbit ABR

Cannot hear the flaws, but I can hear its better

[Meeni]

I cannot hear the flaws of high bitrate MP3. Listening to them on decent (not the best, but not some crap computer speakers) gear reveals no defect to me (except for the Xing encoder, that has so many defects that it is painful).

However, the difference between that perfectly adequate MP3 and a real 48kHZ master is stunning. The clarity is just another world. Ironically, I found that the difference is even more marked in a noisy environment, where the music competes with background noise. It is more or less impossible to identify the overtones on the MP3 in such environment, while they remain very audible with the uncompressed format.

I am not an expert, certainly not a superhuman (I have very average scores on blind tests), but to me the difference is like a nose in the middle of the face, even though the MP3s are "plenty enough" quality, already.

Re:Common misconception

[David_Hart]

Thus, the *correct* way to appraise say mp3 is with very good speakers in a treated listening room

No it isn't. At least most of the time it isn't, though that result would be interesting.

If I'm trying to decide whether to archive my CDs to MP3 or FLAC, I don't give a rat's ass what it sounds like with great monitors in a treated listening room, because that's not where I listen to music. If my speakers give a non-linear result that amplifies the distortions from compression, that's what matters; not what it sounds like in an ideal situation.

Do both.... When I converted my CD collection I used FLAC and the converted the FLAC to MP3 VBR 320kbps. I then archived the FLAC on to DVD discs and put all of my CDs in storage. My thinking was that I would use MP3s for now and, when audio players had enough storage, switch over to FLAC. Plus it also gave me a lossless backup in case my MP3 files became corrupted. I would just have to reconvert them again rather than having to run through my whole CD library.

Re:Better question

[Lucas123]

Support your local electronics outlet. I buy all my audio and computer cables from You-Do-It Electronics in Needham, Mass. Not only do they have the least expensive, largest variety of cables, they also have actual experts on hand to help -- and no, they don't sell Monster.

See for yourself with an ABX test

[coldsalmon]

Use an ABX testing program. This will provide a definitive, scientific answer.

Re:Better question

[turkeyfeathers]

Beats by Dr Dre?

Re:Cannot hear the flaws, but I can hear its bette

[Meeni]

And to further my anecdotal experience, contrarily to all people that say that you need "good gear" to hear the difference. I found that on the contrary, an excellent source would render ok-ish on bad gear, while a bad source will just vomit mashed potato through the speakers.

Re:No

[russbutton]

I've been an audiophile since Nixon was president. And no, I've never dropped Large Cash for cables. I run 14 gauge zip cord to my speakers. MOSFET 60 wpc power amp. Pretty pedestrian by hi-end standards, but my home system is the equal of any so-called "reference system" I've ever heard. I agree there's a lot of snake oil out there, but if you know what you're doing, you can get truly superior results with the right gear. I do location recording for my wife's string quartet. I master at 24/96 and when I down sample to 16/44.1, I can hear the difference on my home system. It's not going to be heard on ear buds or a car system, but on something like my home system (Peachtree DAC, Bryston preamp, Linkwitz Orion loudspeaker system), you're going to hear it.

Lossless codecs make autotune sound more real

[bregmata]

It can even make boy bands sound like their voices have changed.

Fourier components & audible components differ

[ODBOL]

A modulated (varying frequency or amplitude) signal with an audible carrier frequency has Fourier components of unboundedly high frequency. These components can, and sometimes do, have an audible effect on the modulation. The value of >44.1 KHz sampling is debatable, but it's not dismissable mathematically.

Put another way, the "components" below 22.05 KHz that are preserved by 44.1 KHz sampling are the infinitely long unmodulated sine waves of Fourier analysis. The "components" that we hear are modulated sine waves. Cutting off the Fourier components above 22.05 KHz changes the modulation of the audible components below 22.05 KHz. Whether that change is perceptible depends on deep study of human perception, not on the mathematics of sampling.

Why don't we do a Blind Study

[RedDeadThumb]

Let's actually test it and find out. This would be a study I could get behind with my tax dollars (unlike this one: http://now.msn.com/duck-penis-study-cost-395000-dollars). If we find out no one can tell the difference it is going to save a lot of disk space.

There are many factors

[thetoadwarrior]

I do think once you go belo 256 bit rate you start hearing issues or at least you do with some music. But then people also listen to these songs on shitty PC speakers, cheap headphons or worse yet their mobile's speaker. Lossless vs lossy doesn't matter as much when playing the music through poor speakers.

Re:No

[evilviper]

No you can't. Not with any reasonably modern encoder and bitrates above 256. Anyone who tells you otherwise is experiencing the placbo effect.

All frequency-domain lossy audio codecs (MP3, AAC, Ogg/Vorbis, others) have inherent limitations that prevent transparent reproduction of audio. Transients will be poorly reproduced, and artifacts like pre-echo are unavoidable.

2+2 does not equal 5. The sky is not green. Water is not dry. Your assertion is wrong, and based in total ignorance of the topic.

Reproductions @ different rates not identica

[ODBOL]

Not if you can mathematically prove that the two sound reproductions are identical

The best possible signal reproductions at different sample rates are not identical, so of course you can't prove such a falsehood mathematically. The argument is that they are indistinguishable in human perception. That's a very difficult thing to study, with many variables that are hard to control.

Re:No

[Iniamyen]

Beyond a certain point, the human ear+brain will NOT be able to tell a difference. CD quality meets this threshold. There have been so many studies on this, you're just spreading misinformation at this point.

Yes there's a difference. Did A-B test years ago

[Anonymous Coward]

Years ago a friend was raving over the MP3 format and said you can't hear any difference. I said you could and we set up an A-B test with my system. The sound stage on the MP3 collapsed. With the original recording, there was a distinct separation of the sound stage, placing musicians at left, right and center. Switching to the MP3 format, everything came from center.

Listening in a car, I'd opt for MP3 because you're unlikely to notice. If you're sitting down for a serious listen, go with the original recording. You'll get a lot more out of it.

Everybody here is missing a crucial difference

[baka_toroi]

Hearing the difference now isn’t the reason to encode to FLAC. FLAC uses lossless compression, while MP3 is ‘lossy’. What this means is that for each year the MP3 sits on your hard drive, it will lose roughly 12kbps, assuming you have SATA – it’s about 15kbps on IDE, but only 7kbps on SCSI, due to rotational velocidensity. You don’t want to know how much worse it is on CD-ROM or other optical media.

I started collecting MP3s in about 2001, and if I try to play any of the tracks I downloaded back then, even the stuff I grabbed at 320kbps, they just sound like crap. The bass is terrible, the midrangewell don’t get me started. Some of those albums have degraded down to 32 or even 16kbps. FLAC rips from the same period still sound great, even if they weren’t stored correctly, in a cool, dry place. Seriously, stick to FLAC, you may not be able to hear the difference now, but in a year or two, you’ll be glad you did.

It depends...

[ebunga]

Yes, a difference is noticeable on many recordings, but most of the time it doesn't degrade the listening experience. On the contrary, lossy recordings played on crap speakers in a crap listening space often sound better than those same recordings in lossless format on proper studio monitors in a proper listening space.

Mumble mumble psychoacoustics.

Re:AIFF?, Flac!, Lossless in General. & Random

[0xABADCODA]

ALAC is Apple's Lossless Audio Codec and is a latecomer onto the scene. It has good iTunes support and slightly better compression than FLAC, but that's about it.

Apple's ALAC lossless codec is only a dozen C/C++ files (C for the actual codec, C++ for the file format). It's easy to understand, port, and include in other software. To build it you type 'make'. So from a source code perspective ALAC is much better... FLAC has many dozens of source files, assembly, uses automake etc so it's annoying to work with the actual source.

Not that any of that matter to users, but to programmers ALAC is *much* better.

Different meanings of signal "below" 20 KHz

[ODBOL]

a 44.1kHz sampling rate can perfectly encode any signal that is =22.05kHz, and nobody can hear over 20kHz.

People keep saying this, but it involves two different meanings of a signal with content below 20 kHz. The Nyquist theorem says (correctly) that, for an infinite number of perfectly accurate samples at S Hz, there is only one signal agreeing with those samples and containing Fourier components all below S/2 Hz. Fourier components are infinitely long sine waves, with no variation in frequency or amplitude. People hear components that are modulated sine waves with carrier frequency below (for most of us, far below) 20kHz. "Modulated" means that the amplitude and/or frequency (usually both) vary. Fourier components of a signal with arbitrarily high frequency affect the modulation of audible components with arbitrarily low frequency. Whether the effect on that modulation is audible is a very subtle thing, quite difficult to measure, and not completely known at present.

Re:No

[Bengie]

Ogg 64Kbps sounds better than LAME 320Kbps in many sounds, and better across the board at 128Kbps. Either that test was about how well LAME can encode a 60hz single note or those people didn't have "golden ears".

When a drummer hits a cymbal, the horribleness of MP3 is apparent. Just flip between the compressed and uncompressed in small samples of sound, like 2-3 seconds. It is night and day.

Re:No

[UnknownSoldier]

But we're not talking about crappy 24-bit (with 8-bits per primary channel: Red, Green, Blue) which is NOT sufficient. There are horrible mach banding artifacts primarily in the primary colors. Dithering the non-primaries colors hides them but doesn't solve the initial problem.

The analogy is more like is 10-bit/channel sufficient or do we need 12-bits/channel? Hint: There is a reason rendering uses 48-bit (16-bits/channel) as it provides enough headroom to completely remove compositing quantization errors.

Another analogy: Why do we need 30 fps, 60 fps, or 120 fps when movies are displayed at 24 fps? Because almost everyone call tell the difference between the 30 and 60 once they know what to look for. However a few of us can tell the difference between 60 Hz and 120 Hz. Is 120 Hz "good enough" or can anyone even tell the difference between a higher framerate?

If you don't understand the difference between 60 Hz and 120 Hz this video will help demonstrate it:

Asus VG278H High Speed LightBoost Video
https://www.youtube.com/watch?v=hD5gjAs1A2s

Human limits mark the end-points

[AlexOsadzinski]

Disclaimer: I'm a skeptical audiophile, i.e. I love high-quality sound, understand the placebo effect and raise my eyebrows (but not my credit card) at $15,000 speaker cables.

What I find most fascinating about all of this is that the human body is not subject (yet) to Moore's Law. Computers catch up to us, fast. So, while 16-bit 44.1kHz audio was impossible in terms of processing power and storage 20 or so years ago, now it's trivial. At some point, anyone will agree that the resolution of music recordings exceeds the listening capabilities of even super-ears folks. 192kHz 24-bit stereo audio, uncompressed, is only ~4GB per hour; a trivial amount of storage and a very low data rate. Make it full 7-channel just for giggles, and it's ~14GB/hour. So a run-of-the-mill (today) 4TB drive can store ~300 albums. A typical audiopohile collection of 5,000 albums will require 70TB of storage, 100TB if you want some RAID redundancy, and that's with no lossLESS compression. Today that's $8,000 of storage, and it takes no imagination, plus 0.6x conservative lossless compression, to get that down to $2,000 or so. Not cheap, but not outrageous. The albums themselves cost said audiophile about $75,000.

Will there still be a debate that 192kHz 24-bit isn't "enough"? Probably. But human ears aren't changing much, and, it could be argued, we hit the limit of perception for most people at 44.1kHz 16-bit. At 6.5 times that resolution, it's probably enough. Problem solved (modulo a lot of infrastructure changes).

The same will eventually happen for video. 8x, if you've ever seen it, is like looking through a window, until the camera pans (which it shouldn't). Many people can't tell the difference between 480p and 1080p, never mind 4x.

It's already happened for photography. The resolving power of modern high-end digital sensors exceeds that of all but the best lenses. Storage of these multi-megapixel images is a non-issue.

There are imponderables, though: passionate, intelligent and sincere people I know wax lyrical over the aforementioned $15,000 speaker cables. Sound reproduction in a home environment, or when hearing headphones, is extraordinarily complex, and nobody has ever achieved "the absolute sound", i.e. a feeling that you're in a live music environment when listening at home. All you can hope for is some of the emotional involvement that the recording artist intended, whether in the studio or live. In the car on on good earbuds, I can JUST about tell that I'm listening to a 192 or 256kbps AAC (or MP3): it's in the sizzle of a cymbal and the like. But it's a fine line. At home, sure, a good audio system can make you cringe when listening to a highly compressed source. I've ABed the same recordings at 44.1/16 and 96/24. On high dynamic range stuff (orchestral, mostly), the quiet passages show a difference, albeit a minor one, because the signal is encoded using only 2-3 LSBs.

What's hardest to understand is the emotional engagement offered by (good, well-recorded, well-pressed) vinyl. It could certainly be a placebo effect, or it could be something yet to be understand in human hearing response when listening to digitally-encoded audio. To my surprise, the difference between a $10 speaker cable and a $500 speaker cable is clear, and in the favor of the $500 cable. There's some science there, but also a lot of voodoo. One or two manufacturers claim to have figured out how to measure the difference, but they're not saying how, for obvious reasons: they're, um, "marketing", or they really have found something to measure and want to keep it proprietary.

Finally, audio technology is really advancing, and fast. The DACs in most iDevices are ok: not great, but ok. The bundled earbuds are bad. But as little as $30 gets you decent earbuds. At $100 you're experiencing perhaps the same quality of sound as from a $1,000 pair of speakers. A decent DAC can be had for $250. A decent headphone amp for $200. So Mac/PC (which you already have)+$250 DAC+$200 amp+$100 earbuds gives you sound far superior to what most people enjoy. Not free, or cheap, but not very expensive, either. Then, I bet, a lot of people will want to move up from 128kbps MP3s.

If you want to defuse this argument

[ddd0004]

Charge less for the lossless format. Audiophiles use a common benchmarking system that is hidden in plain view. You look for the number after this character "$" and the larger that number is the better the component or recording or system will sound.

Re:No

[ozydingo]

In medical tests, people are given a placebo and yet claim to feel better or feel the same effects as people who are given the real medication. These must be the same people who rail against mp3s.

Don't dis the placebo effect, it works (for some limited benefits), even in cases where the subjects were aware that they were receiving a placebo
The most similar analogy would be to say that someone can enjoy lossless music more than lossy music. This could be true even if they can't tell them apart in a blind study. Of course, under these assumptions, they'd also enjoy lossy music more than lossless music if the labels were switched and they believed the labels. It's enjoyed more simply because of what it is believed to be. That may be silly, but hey, who am I to crap on someone's enjoyment?
On the other hand, making the claim that you can tell the difference, i.e. discriminate between then, is more directly challengeable and probably false in most cases.

Re:Better question

[ddd0004]

Thanks for the pointer, I've had my electrons swimming upstream all along. I also rewired my usb mouse after I discovered that it was wired the wrong way around at the factory. You won't believe the warmth of my lefts, the mellowness of my rights, the dynamic ups and well rounded downs.

Re:A lengthy, thorough, and well-explained discuss

[Dennis Flynn]

Who'e the jerk now?

9 out of 10 kids prefer...

[Xenna]

http://www.audioholics.com/news/industry-news/kids-prefer-poor-quality-mp3

(and remember, kids are able to hear frequencies that you can't!)

Re:No

[Anonymatt]

If you think Fiona Apple is a strong vocalist, you can't hear shit!

The difference isn't in the hearing.

[jensend]

It's quite unlikely that you can hear the difference between the lossless original and a 160kbps lossy version from the best modern encoders (e.g. Apple's AAC encoders from the last couple years). If you can, it's going to be for just a few isolated samples tested in ideal circumstances and it won't impact the quality of your listening experience.

People who claim otherwise are either using outdated formats and encoders or they're not doing proper blind testing and their results are dominated by psychological bias.

But if you ever want to encode your music in another format, transcoding from one lossy format to another is like xeroxing xeroxed copies; you get generation loss and are more likely to hear some artifacts. Encoding from the lossless original will never have that problem.

You can think of it like this: when you buy an mp3, you own an mp3. When you buy a FLAC, you own the music- the format becomes irrelevant since you can re-encode it in any other format, past, present, or future, and have the result be just as good as if you re-purchased the music in that other format.

DSD is pretty awesome

[lophophore]

I have a DSD (SACD) Player. I have several discs of the same music in CD (red book 44.1 KHz 16-bit) and DSD. DSD is PWM at 2.8 MHz.

I have done A/B tests with myself, and "blind" tests with friends. Everybody prefers the DSD playback. This is on higher end consumer gear, not high-end audiophile stuff by any means.

I have no doubt the DSD versions were mastered more carefully. Perhaps that is the biggest difference. However, they do sound better than PCM CDs to my ears.

Yes, I still can

[rickb928]

I can easily tell the difference between 128k MP3s and 320k, and from CD. I rip all my CDs at 320 or FLAC, 128 is incredibly annoying to me. the very high end is usually reduced to bacon frying, and the dynamics are hosed.

Of course, now I'm busy normalizing my library on Google so it gets over the gym noise, and pre-eqing it to make my Bluetooth headset 'sound better'. Ick, but it's a bit better.

But, but, but, I really liked ATRAC, especially after 4.x. I had a Sharp Minidisc player/recorder that rocked, and was vastly better sounding than the early MP3 players, iPods included.

ATRAC gets no respect. MD failed. But I still use it sometimes.

On behalf of all Volkswagen owners..

[RyuSoma]

.. I'd like to say that for $40,000 your 'lossy audio file' at least doesn't have a reputation for spontaneously bursting into flames.

My 3 priorities are

[justthinkit]

My 3 priorities are:
(1) What I am listening to -- e.g. I prefer Beethoven to Tchaikovsky
(2) What version is this -- e.g. in general I hate live (vs studio), and in classical works the symphony/conductor is very important
(3) Are there kids on my lawn? -- gray ears don't need more than MP3 has to offer

Audiophiles are, for the most part, gullible

[taustin]

After all, we're talking about people who buy $1,000 Monster cables, even though in a blind test, they can't tell the difference between those and wire coat hangers.

Re:Audiophiles are, for the most part, gullible

[the eric conspiracy]

Audiophiles usually consider themselves to be in either the objectivist or subjectivist camps.

The subjectivists generally don't care what measurements say, all they are concerned about is what their reaction to the sound is. It's these people who buy $1000 cables.

Objectivists are into the idea the quality of the sound is something that can be predicted from objective measurements. These are the people who do double blind tests of equipment to eliminate confirmation bias and the placebo effect.

I like the objectivist idea simply because I think it leads you to buy less snake oil.

Why even use compression now?

[ironicsky]

Back in the 90's when people had 56k modems, a WAV file was huge (10Mbit per 1 minute of audio) but since then our connections are pushing 100Mbit/second+ (Canada on Shaw) , with a nice average of about 15-20Mbit/second - With this you could download a 3 minute wav (30Mbit) in 1-5 seconds. Yes, I know some people have poor quality providers, or slower connections around 1Mbit a second but still 30-60 seconds isn't bad.

The other limiting factor back then was small hard drives averaging around 50-80Gb which were around $500 in 1999/2000. Now, you can buy 3Tb of storage for $130. 3Tb is enough to store 104,857 - 3 minute WAV files.

Even 64Gb iPod Touches have enough storage for 15,000+ songs in WAV format.

Re:Why even use compression now?

[Legion303]

"Even 64Gb iPod Touches have enough storage for 15,000+ songs in WAV format."

Close! It's actually more like 1,300 songs, depending on song length.

Banausen!

[jurgen]

Lossy, lossless---recordings are for the rabble! Anyone who would settle for anything less than lying under the piano or sitting in the middle of the live orchestra does not deserve to hear the works of the great masters of the classics!

It doesn't matter.

[MouseTheLuckyDog]

Since if you store in some lossy format say MP3, and the world switches to some other lossy format, you won't be able go convert your files to the new format without hearing a big difference. If they are still good enough quality to listen to.

You need to store your stuff in a lossless format so when formats change you can convert. Even if you have versions in a lossless format for listening.

Re:Better question

[DragonTHC]

when audio was analog, they were better.

Now, being digital, it no longer matters. If it touches, it talks.

Re:Better question

[blind monkey 3]

it doesn't matter how lossy or lossless the file is if you're listening with shitty white earbuds.

Dude, use some alcohol wipes on them before you get an infected ear.

"Modern music" is recorded with much higher gain than previously so having higher quality equipment probably won't make much of a difference if your taste is mainstream. A couple of links:

The loudness war.
This made me smile. Why? I was listening to a youtube clip on pc speakers to pick up the affect on sound quality after clipping occurs...... he does explain it well though.

Disclaimer: Most of my digital music is in flac format - sounds brilliant through my main system at home, not so crash hot on my phone using black earbuds.

Re:No

[Hatta]

Show me the data. Produce a properly blinded ABX test of LAME -V0 vs PCM audio where the subject was able to identify lossy audio at a rate that is significantly (p>0.05) different from chance.

Lossy codecs do have artifacts. But that doesn't mean those artifacts are perceptible. The only way to know is to do a blind test.

Re:Better question

[methano]

This may not be germaine, but it is well known that if you test a new antidepressant in the clinic, you get an average 30% positive response in the control group. It's also known that if Neil Young tells you that your music is crap, 30% will believe him. However, few will do anything about it.

Re:No

[joelpt]

'The more that pseudoscience goes unchecked in the world at large, the harder it is for truth to overcome truthiness... even if this is a small and relatively insignificant example.' ~ http://news.slashdot.org/story/12/03/06/0048259/why-distributing-music-as-24-bit192khz-downloads-is-pointless

'For me, it is far better to grasp the Universe as it really is than to persist in delusion, however satisfying and reassuring.' ~ Carl Sagan

Asking for people to behave rationally may not always be the easy way, but in my experience it is almost always worth doing. I think as a species we'd be a lot better off if everyone valued rationality highly, so I think we should encourage that in everyone.

Re:speak for yourself

[Hatta]

After I installed it, I found that a lot of the MP3s I have were actually not sounding as good as some others and the flac albums I have

Did you ABX? Were you able to distinguish lossy from lossless at a rate greater than chance? What was the p value you used to determine statistical signifcance?

If you didn't do a properly blind test, and you didn't do the appropriate statistics, then your observations are meaningless.

Re:No

[Hatta]

Tetrachromats can pass blind tests. Audiophiles cannot. That's the difference.

Re:Better question

[DragonTHC]

It ain't so for most humans.

My hearing actually is better though. I can hear a wider range than 20-20k

I can hear bats at night. I can hear ultrasound tests on my abdomen. So, yes it matters to me. Try listening to Jeff Buckley's version of "Hallelujah" in 320k vs flac.

My headphones go up to 80k. My sound cards can't go above 18K. The difference is all in the high range though. Most people will be able to tell the difference.

Re:No

[noh8rz10]

Asking for people to behave rationally may not always be the easy way, but in my experience it is almost always worth doing. I think as a species we'd be a lot better off if everyone valued rationality highly, so I think we should encourage that in everyone.

glass houses, my friend... could your life undergo such scrutiny? Would you want to be faced with the determinations of such an evaluation?

128K MP3 is good enough for me.

[Lost Race]

When I was much younger and had better hearing and MP3 was a new thing, I ripped my CD collection and encoded everything with whatever the state of the art was back then (bladeenc? mp3enc? this was pre-lame). After a while I started hearing artifacts in MP3-encoded music so I did some A/B testing against the original CD the music was encoded from. Turned out those same artifacts were in the CD.

MP3 encoding has matured and improved since then, so whatever degradation there may have been, it's less now. I've only ever used 128Kbps stereo encoding, and I've never been able to detect any difference from the CD in any kind of music. This is with fairly high-quality sound cards, amps, and speakers.

Of course, my high-frequency hearing is pretty much gone now so I sometimes worry that my music collection might sound horrible to anyone with fully functional ears.

Re:128K MP3 is good enough for me.

[Lost Race]

Ha ha, no, pressed retail CDs bought from a record store.

Re:128K MP3 is good enough for me.

[Legion303]

"Of course, my high-frequency hearing is pretty much gone now"

Which means the fact that your hats and cymbals sound like they're underwater probably doesn't bother you...

Re:No

[evilviper]

The only way to know is to do a blind test.

Not true at all. The physics of human hearing is extremely clear. And I really don't care if you want to be a flat-earther and refuse to believe it.

Re:Better question

[DragonTHC]

I just redid this test and I always think the same thing. The flac sounds more alive and less recorded.

Re:Better question

[Pieroxy]

The linked article features $500 for some simple cables. But people can spend MUCH MORE MONEY than $500 on simple cables. For example:

$699 for 3M of speaker cables: (look for STEREOVOX Firebird Speaker Cables, 3M): http://www.gcaudio.com/products/steals.html

Ironically, the products are labelled "steals". Very true indeed.

But there's more. Not all products are "steals". "The next step up is the LectraLine cables priced at $295 for the 1M" http://www.gcaudio.com/products/newArrivals.html

But it gets better. At musicdirect.com you have power cords for $2,699.99 !!! Obviously it's "The Absolute Sound Golden Ear Award Winner!" Of course. http://www.musicdirect.com/c-650-power-cables.aspx

But it gets better, again. At nordost they build power cables made out of "99.99999% oxygen free copper conductors." I let you imagine the cost of production. A mere 1.25M of power cord is 8,795.00 (and these are UK pounds, worth more than a dollar). For 5M count 20,495.00 pounds. Yes, that's about $31K !!! http://www.highendcable.co.uk/Nordost%20ODIN%20Power%20Cords.htm

But it gets better, so much so that it gets boring. But still. Can you spend more than $31k on a simple pair of wires? Well, yes, you can. Look at the bottom of that page, 6M of speaker cable for only $50k. A bargain, really. http://www.audiofederation.com/dealership/prices/nordost/index.htm#prices

It is astonishing to say the least. That said, it some people have the money...

Re:Phase

[fa2k]

When you have at least two speakers, and signals from both speakers hit both ears of the listener, it makes a difference.

You can generate two high frequency signals, one from each speaker, such that the difference of the frequencies is in the audible range. The sum of the two signals includes a modulation with the difference of the frequencies (sin(x)+sin(y) = 2*sin([x+y]/2)*cos([x-y]/2). The phase depends on the distance from the speaker times the frequency. If the signals come from different sources, the phase of the low frequency modulation signal (which is audible in this example) depends on the distance from each speaker times the frequency of the signal from that speaker.

It is not possible to produce such a spatial variation with only low frequency signals. If both speakers instead produced a low frequency signal with a different phase each, the sum would have a phase which varies in space with a wavelength corresponding to that frequency. There would also be a sinusoidal spatial modulation resulting in places where the amplitude goes to zero (this doesn't happen in practice because of reflections in the room and the finite size of the source).

So the spatial variation of the sound will be affected by higher frequency information. Headphones are not affected, and can equally well be fed 22 kHz signals, but technologies like Dolby Headphone would theoretically make it equivalent to speakers. The variation of the phase in space is not reconstructed correctly by stereo speakers, or any number of speakers for that matter, except for at a single listening position if set up correctly, so it is not clear to me that the higher frequencies would improve the realism or the perception of space in music.

Lossy compression also makes recompression worse.

[Samuraid]

If I'm familiar with the song, then it typically takes a bitrate of 320Kbps or higher before I cannot hear a notable difference. However, the often-ignored problem with lossy file formats is what happens when you attempt to edit the audio you've licensed/purchased. For example, say you want to re-encode a lower bitrate version of a song for a mobile device, or maybe adjust the volume, or trim a song down in length for personal listening preferences. (I do this quite often, actually.) Trying to re-compress lossy source material again after editing just makes things sound far worse. This remains another reason I try to get lossless audio files whenever possible.

Music Lovers and Audiophiles

[Frankie70]

Music Lover - Someone who loves music
Audiophile - Someone who loves his stereo equipment

May be we can add "music Formats and Containers" to that definition.

Re:Music Lovers and Audiophiles

[multimediavt]

Music Lover - Someone who loves music Audiophile - Someone who loves his stereo equipment

May be we can add "music Formats and Containers" to that definition.

I don't know. I know a lot of music profs that are crazy audiophiles! As well as more than a few handfuls of musicians.

Re:No

[Hatta]

The physics of human hearing is extremely clear

Then it should be easy to experimentally verify your predictions. Produce an ABX, and you'll have a point. Show me the data, and I'll change my mind.

Re:Better question

[rochrist]

It doesn't take massively expensive equipment to hear differences. Just a decent amplifier and decent speakers.

Re:No

[jovius]

The difference can be heard. All you need to do is to produce and mix your own music at 24bit/48khz (real instruments) and then at the end of the session compress it to any level mp3. The mp3 sound field is full of holes, phase distortions and other typical artefacts. In any case the lossless formats are more archivable.

Re:No

[Hatta]

How did you set up your ABX, what was your success rate at distinguishing lossless from lossy, and what statistical test did you use to demonstrate that your success rate was significantly better than chance?

but it doesn't matter for *listening*

[Chirs]

Go ahead and record in high quality, but for listening CD-quality is fine.

so record in high def, play back as CD-quality

[Chirs]

As others have said, there are valid reasons to record/mix in high def. But you should be able to downsample the final result to CD-quality with no audible loss in quality.

Re:Better question

[Anonymous Coward]

Wanting decent sound does not mean you are an audiophile with strange habits and expectations. For example. I do not like listening to Pink Floyd in compressed format. I can hear the difference even with my $30 Sony headphones and in my car.

I'm not an audiophile at all but I did pick up some really good used equipment over the years.

Re:Better question

[Anonymous Coward]

"I think the real point is that there are known limits to human hearing and many audiophiles fantasize about their hearing being superhuman"

No. The difference between a live acoustic instrument or human voice and a recording is immediately obvious, even to people with significant hearing damage. Waving paper cones around in boxes is not a great way to reproduce sound, it's just all we have with today's technology.

Audiophiles are not trying to get the last few percent of reproduction quality, they are trying to get some improvement on the terrible quality we have today.

I say that as a studio engineer with 30 years experience. I do my best, but we are still in the very early days of recording and reproducing sound. Matters have not improved for so long that many people have forgotten how much of a compromise audio reproduction currently is.

As ever, the hard part is the transducers. Wide bandwidth storage is practical now, but microphones and speakers generate huge amounts of distortion, and have bizarre phase responses and radiation patterns.

Matter of principle

[TeknoHog]

I rip CDs to flac, because I don't want to keep worrying if I could have made a better rip.

Living rooms are not double-blind ABX environments

[hacksoncode]

My biggest problem with the way that both audio researchers and audiophiles approach this is that human perception is not scientific or rational.

It doesn't matter whether the difference between sample A and sample B is real or perceived, because when I'm actually listening to music, that is 100% perception, and I *do* know (or think I know) a priori which sample I'm listening to.

The scientific approach is great (mandatory, really) when you're doing science. I will go beyond saying that it doesn't actually help much at all with determining what you will enjoy. I assert that more often than not it actively *decreases* your enjoyment of the experience itself.

Of course, one can certainly enjoy understanding and appreciating the science behind it, leading to more enjoyment overall... I'm speaking purely of the perceptual portion of the experience.

If...

[multimediavt]

This is my position, and I have recorded a lot of 24/96 audio and know how good it sounds compared to other digital versions. I am not a fan of the RIAA or the mainstream music business machine, but I do support artists' rights.

If I am actually buying a license to listen to an audio track and not the track itself, the way the RIAA wants it, then I want the highest possible quality digital version of that track. I feel I am entitled to access to it because my money is not only going to pay for the license to the track it pays for that original recording. If the original was analog, then a different license would need to be obtained for a physical copy of that. One thing I do not have a disagreement with is the separation of digital and analog rights. I do believe (from some experience) that high-end studio analog equipment is better than even the current high-end digital systems (24/192).

Basically, if I can buy an AIFF of a file I will! If I could get a 24/96 or better version of a song I would feel better about plunking the money I now do for digital music. It requires more effort to make the crappier versions anyway, so why? Let the masses do it themselves to put on their music players. Plus huge hard drives will drop in price! Win-win really. Hehe

Re:No

[evilviper]

Show me the data, and I'll change my mind.

Find the data for yourself. I don't take orders. There's tons of research out there, and it's not hard to find, and always getting easier.

You're the one who started off by making baseless bald-faced assertions. I'll leave the onus with you.

Re:Better question

[Cillian]

Genuinely not sure whether joking or audiophile.

Re:Better question

[Gr8Apes]

I don't know about that - there's a difference between 128mbs and lossless that I can tell on some music even in my car going down the freeway. Granted, not all music is inherently made unlistenable by compression, but some definitely is. Breakfast at Tiffany's can probably be listened to fine at 64mbs, while something more rich, say Tchaikovsky's 1812 overture will have noticeable artifacts or drops. (Yes, extremes, but I do know some other modern music doesn't compress well at even 320mbs with lame)

Buy a good DAC and a good set of headphones

[Anonymous Coward]

Then download your favourite album in mp3 format, and in FLAC format; there is big difference, but which depends heavily on the mp3 bitrate.

Since a FLAC album is typically between 200-500mb, and space/bandwidth is cheap, why would you want lossy formats anyway?

Even with a good DAC/headphones, get a DAC where you can change the opamps inside it; you would be amazed at how different opamps completely change the character of what you're listening to, and it is fun experimenting with them (so long as you don't get knock-off opamps that fry your DAC...).

Summary by Bad Analogy Guy?

[GrahamCox]

settling for a Volkswagen instead of a Ferrari.

Have you looked at a Volkswagen recently? They are actually quite high-end cars, and vastly, immeasurably, more practical than a Ferrari.

By this analogy, a Ferrari is the oxygen-free amorphous copper $500/metre speaker cable of the auto world.

Music lovers and audiophiles

[PhunkySchtuff]

Music lovers love to listen to music. Audiophiles on the other hand would rather listen to their equipment.

Having weighed in with that bombshell, I've got a fairly decent sound system (Rotel preamp/processor, Rotel power amp, VAF speakers) and I can't hear the difference between MP3 V0 (VBR at around 220kbs) 256kbs AAC (my preferred format, simply because it's what I get from iTunes) and redbook CD audio.

Re:No

[Hamsterdan]

In the DJ world vinyl is prefered because it's much easier to manipulate, not for sound.

That's why an almost defacto standard is dual SL-1200 turntables for mixing vinyl.

Besides, people listening to that kind of music don't really care about *quality*

Re:A lengthy, thorough, and well-explained discuss

[AaronLS]

That's what she said.

No.

[Zed is not Zee]

If that was an actual request for data, my answer is no. But then I only migrated from cassette to CD because I was tired of having to turn the tape over.

Yes. Next Question.

[residents_parking]

All you need is entry-level prosumer monitors, like KRK's RP6 G2. The difference in staging and bandwidth is obvious, even if - like me - you're over 40.

Of course you can.

[GrumpySteen]

You just have to be listening while sitting motionless in an isolated room with nothing else to produce background noise.

Sounds a lot like a coffin to me, though, and I prefer to mix my music with the rest of my life. I won't hear whatever loss there might be in an MP3 because I'll be making too much noise dancing.

Don't ask me, I'm in IT

[ALeader71]

After nearly 20 years of life in data centers, NOCs, etc with the constant droning of fans in my ears, I couldn't tell the difference at all. Heck my ears are already ringing just thinking about it.

Re:No

[Artifakt]

There are some really strange cases of the Placebo effect, for example getting subjects to use Placebo opiates for pain relief for a week or more and then giving them an opiate inhibitor without telling them, and having it turn out to inhibit uptake of the fake opiates. There's some old research that can't be replicated now because of modern ethical guidelines (Which raises the question, is it still science if it becomes irreproducable, not because of a technical limitation but because of increased moral standards?). After reading up on some of these oddities, I've come to the opinion Science does not understand the general Placebo effect nearly as well as individual researchers think they understand how it applies to their special cases. It's fair to say double-blind studies have proven many times that people are claming auditory abilities they simply don't posess, but that really doesn't necessarily mean we can jump from that point to conclude there's some aspect of the Placebo effect involved.

Re:Better question

[Lagmo]

Yea i personally am a big fan of the 'broke student 5.1 Surround Special'

Using a 4 Channel Car-fi amp(4x100W typical) ~$25-$50 used, usually powered by a PC ATX PSU or similar (practically free as in beer).

A 5.1 USB soundcard($20ish) hardwired directly to amp + Guitar/Bass amplifier($50) used as subwoofer. Decent speakers bought used can be had for around $50 a pair, good speakers last for decades unlike most crap made nowadays, so just spend whatever you can reasonably afford, since it's a fair investment in the long run, just make sure to check they still work(no damaged tweeter/woofer units).

Use standard Cat-5 network cable for signal cable AND speaker cable(double up), failing that regular mains cable works too, again free as in beer.
Use media player/OS driver to mix center channel into front speakers.

Bonus points for:
Soldering the connections.
6 ch. car amp, bridge one pair for center speaker, or get a powered full range PA mono speaker(easily found used).
Braiding the Cat-5 cables, with honors if the speakers you got can be bi-amped and the braid uses it(2 bass and one for treble/tweeters)
Using a studio grade soundcard, firewire ones tend to be common, useful if you have the connector anyway.
If you can find a mains power noise filter and/or have proper ground connections.
Using the guitar amp for it's intended purpose.
Optional mounting hardware.

I've had people compare such setups to $10K+ supposedly high-end gear, they can be amazingly high fidelity for what can be done on a very frugal budget.

Don't get frightened by how it looks, just consider it 'ART' =)

Audiophily vs. Classical Music

[billstewart]

I was in college before CDs came out, so the audiophile types had vinyl, fancy-for-the-time turntables, high-quality cartridges and needles, etc. One of my housemates liked classical music, and said that once he had a medium-quality stereo system, it didn't make sense to spend more money upgrading the audio quality - it was a lot more important to get records from better orchestras with better conductors. His system was good enough that he could pretty much hear what they were playing, and if you were listening to Beethoven you wanted the Berlin Philharmonic, not the 101 Strings, and you probably had opinions about whether you wanted Furtwangler or von Karajan conducting, and getting rid of that next-to-last bit of distortion wasn't going to fix a lousy recording.

I mostly listen to music in my car. A decent MP3 is close enough to CD quality when played over road noise, and it doesn't skip when you go over bumps.

MP3 is quite outdated

[UpnAtom]

Test 320kps with Apple's AAC.

If you can tell the difference, submit yourself to Hydrogen Audio's blind testing because nobody else can.

Yes, mp3 has a limit

[Artemis3]

Some people think a 320kbps mp3 is perfect, but its not.

Yes, sometimes it is possible to achieve transparency with mp3, but not with complex samples, especially those involving percussion in high frequencies. It is a format limitation, and can not be fixed without abandoning mp3.

Other lossy formats such as vorbis, are not limited in this regard; so if a passage (sample) is too complex, it can simply bump the bitrate as much as it needs until transparency is achieved. Of course, this needs extensive encoder tuning, but the format is no longer a limiting factor.

Unfortunately with mp3 you can't put frames above 320kbps, and the samples that fail, will fail and 320kbps cbr can't help you, so if you use mp3 you might as well use a more cost effective vbr choice such as lame -V2; otherwise you are simply wasting space and not achieving transparency anyway.

Furthermore, different lossy formats have different properties, and some can actually achieve transparency, given enough tuning a lots of abx testing and data gathering.

While lossless might be wasting some space compared to a perfectly tuned lossy, it allows you to have a safe, clean source to test all those existing and emerging formats to begin with. Think of it as archival quality; from which you can then lossy compress in whatever you need.

Also you should not transcode something already lossy compressed into another lossy format, every time you do this you introduce artifacts and reduce quality.

Those deeply interested in the subject should visit: http://www.hydrogenaudio.org/ and read the many years of quality discussion archived there.

Pychoacoustics (used to tune lossy encoders) introduce another factor. Aside from different people having different hearing abilities; you are supposed to equalize your listening environment for a "flat" response. People not only rarely ever do this, they bump the settings to make it sound how "they like"; ie lots of bass or treble getting away from the average perceptual "flat" eq curve; which is what lossy encoders strive to keep; resulting in poor perception. Raw/lossless has more data able to help this real time audio modification. Ie, a lossy encoder discarded something you would have normally never listened to, but with your misuse of eq, you were expecting to hear.

Note: its not actually flat, but "Equal loudness contour" which is an average of how humans actually perceive tones.

i would prefer a 50 gig blueray disk

[ralphaostrander]

With one song or album on it that has all the quality of the master.

"Lossless"

[GWBasic]

My understanding is that 256 or 360kbps AAC is, for all intents and purposes, highly accurate.

Something that the discussions don't really cover are "lossless" formats like DTS and Dolby Digital HD. These formats tend to use about 6-1000 kbps, yet don't incorporate the phase-changes that MP3 and AAC do. From what I understand, the resulting sound is more accurate than merely decimating 24bit to 16bit.

Re:A lengthy, thorough, and well-explained discuss

[thegarbz]

Scroll down a bit. I think someone in that linked discussion posted a link to another long discussion among very qualified individuals.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:No

[thegarbz]

40 years of exposure to loud music has probably damaged their hearing enough that they really don't know what they are hearing.

Hearing damage presents as a loss of range for human hearing, not that the ear suddenly starts scrambling what it can hear. Maybe they can't hear anything about 13kHz. Doesn't matter. That has no bearing on what they thing for sound below the 13kHz range.

The difference here is they are musicians. They have spent much of their lives dedicating themselves to perfecting their sound. Tweaking things subtle picking different instruments or slightly different strings to get the exact sound they want. Chances are you don't even know they are playing different guitars in different tracks. They do.

My mother is a classic example. She can't tell if my guitar hasn't been tuned (I can) but she can hear me if I fart in the back yard so clearly her hearing is just fine. I'll take Dave Grohl and Neil Young's opinion any day over anyone I see wearing a set of Dr Dre Beats heaphones, or anyone wearing white ear buds.

With out a doubt

[Slashfart]

For the tone deaf, don't listen for it in the high end. Listen for in the kick drum. If you go back and forth between ANY lossy comp and a CD, for example it will sound like the CD drops an extra octave. Also in the echo. Half the echo is gone on lossy files. Very noticeable.

Re:Better question

[mathew7]

Please stop comparing analog reproduction with digital artifacts. Lossy compression brings artifacts. For example take a single note on an instrument which produces a range of frequencies, something like a gauss curve. The lossy compression in this case would just cut the ends of the range. This is related to low-amplitude (low-volume) frequencies. This can be heard even on cheap gear provided the ambient noise is low enough. But the question is: do you actually know how the instrument has to sound?

If you listen to just mp3s, then ok...you say mp3s are as good as the original. But did you listen to the original as much as the mp3? Did you listen to it in a low-noise ambient (something like goind-to-sleep enviroment)? Even the cheapest gear can show you the artifacts. And once you found them, you will always hear them.
Ok...maybe I sound like I say that cheap vs. quality equipment makes no difference. That is not true. Reproduction quality does make a difference of how you spot the artifacts, but the idea is that the difference is not that big. I think most artifacts can be spotted in low-ambient noise enviroment, which is usually not associated with cheap equipment. Compact PC speakers? Gaming/office computers (where focus is on the screen, not audio). Earbuds? Portable on the (noisy) street (to exchange enviroment noise with music). Hi-fi gear? Dark, silent HT room.

As a side note, while many people cannot differentiate lossy vs lossless (or did not have the right enviroment), I noticed very annoing artifacts on recompression: I made a local-network radio station (because I got tired with 3 devices with it's own playlist) with 320Kbps MP3 streaming, and when I got to hear a 128Kbps song, I had to stop it after 10 seconds (I'm talking about rock which some say it's hardest to spot). The original 128Kb song was good enough, but recompressed (even to higher bitrate) was terrible. Now I have 2 streams, one flac for devices that support it, and one 320Kbps mp3 for those that don't or when I want to listen away from home (well, Logitech insists on checking my private streams, so I had to make them available outside). And I avoid mp3 files as much as I can.

Re:Better question

[mathew7]

I wonder who selected the test music? Like many already commented, some music sounds better than other at same rate of lossy compression. Not to mention the encoder itself.
Just make sure you listen to one tune from your original CD is a quiet room and then switch to it's mp3. Difference comes from comparison, and if you can't accurately remember one sample while playing the other, you can't make a comparison. Oh...and you can test them with you 2$ earbuds. If there is something to be found, you will find it.

Do remember that any statistics can be skewed to show one result or the other, when infact the truth is right in the middle.

PS: the article you point to has only a few paragraphs for lossy compression, with no conclusion. Nice read (I already agree that CD-quality is enough)...but....

Re:Common misconception

[ChrisMaple]

A 2 TB hard drive costs $100 and will store 3000 CDs uncompressed. If you're archiving, you don't want to lose the ability to have good quality as your playback equipment changes.

Re:Better question

[flyneye]

Agreed, your equipment is going to be a LARGE factor in your ability to hear what you've been missing.
Even good phones aren't always a cheap answer. Nice home unit and set of full range speakers light quite a lot of studio artistry up, previously unavailable to cheap iPod docks or most expensive ones, for that matter.

          Secondly, there are just some badly ripped music out there that will gnaw on your cranium like a starved rat with progressive tail mange. Thou Shalt Not Rip to Small Audio Files! (carved in the side of a mountain for all the world to see for all time)

          Third, are YOU capable of hearing in the higher ranges of human ability? Sadly, many in their 30s and up lose the tickle cones in the ear necessary for the high end sostenueto to reach your gray matter. That'll teach you to stand next to the speakers and bang your head, Beavis! Yes, what isn't lost naturally gets wrecked from excessive indulgence in Concerts, Industrial Equipment, Stadium Noise, Jet engines and all the other things in life that don't kill us, yet fail to make us stronger. Although there actually is some music to be listened to at massive volume in order to create the sparkly hearing anomolies that attract us to Hi-Fi studio artistry in the first place. I present Glenn Branca: http://www.youtube.com/watch?v=xdLhRB4dJJI So FWIW all is not completely lost, if you can develop the taste.
Perhaps a little Boyd Rice/ NoN http://www.youtube.com/watch?v=TrInnSXaQ0o
I like to think of it as the future of classical music representing the late 20th century.

So all in all there are going to be a wide range of listeners of music getting varying degrees of art from it anyway. Common sense is to default to the highest denominator and go with quality while a majority wouldn't mind listening to Dark Side of the Moon from stereo bullhorns. You can still please all the people all the time. Or just deny the dissatisfied exist and ignore them, til they go away.

Re:No

[jovius]

As I said the setup is highly subjective and the differences are certain when you are very intimate with the source material. The degradation in the quality and sound field can be heard.

Mp3 for example is anyway and ancient format created for the bandwidth and storage limits of the time. Suddenly back then the linear progression in the quality of mass produced consumer sound went backwards (vinyl - c-cassette - cd - mp3). Now it seems quality is coming back to at least where it was left.

Yes. Definitely and definitively.

[Anonymous Coward]

Are you suggesting some people can't?

Lossless v lossy

[Xylene2301]

Lossless sounds better and analog/vinyl better still on good equipment.

I knew cables were bunk but really learnt it when

[caveat]

I was working at Brookhaven with a fancy-pants laser-driven time-of-flight mass spectrometer. We needed (ideally sub-)nanosecond resolution with as close to zero variability as possible (at that point you need to account for things like cable length to the cm and "refraction" across connectors). The kind of technical minutiae that gives audiophiles hard-ons and makes One Billion Dollars for the cable companies.

We used...high-quality oxygen-free copper cables with gold-plated BNC connectors. Certainly not the $5.99 bargain bin from Staples, but we were paying ~65 for a one-meter stretch. It's not like I was dumb enough to fall for Monster's schtick before that, but it made the point in a way I don't think anything else quite ever could.

Some of the best cans you can get are ~$100

[caveat]

Sennheiser HD280 Pros. No fancy bells and whistles, no big-name musicians attaching their names, and especially a dead-flat frequency curve so you get out of them exactly what you put into them - they don't have some "inherent warmth" but if you play with your EQ you can make them sound however you damn well please. There's a reason they're in just about every recording studio on Earth.

Re:A lengthy, thorough, and well-explained discuss

[Keybounce]

I almost did -- I actually went to look at it, I figured it went to an older, archived discussion.

Speaking of which, have you heard about TvTrope's forray into producing a TV/Podcast series? http://tvtropes.org/pmwiki/pmwiki.php/JustForFun/AvatarAndTheAirbendingFellowshipOfVampireSlayers

It's actually several years old now, with spin-off "books" and "comics" (web-based, of course)

Re:Better question

[leenks]

I'd hope a 320mbit stream would sound *incredible* - maybe not 230 times the quality of CD such a bitrate could offer, but enough to put even the analogue audiophile purists in their place!

Yes.

[dradler]

You can be trained to hear the lossy compression artifacts. But trust me, you don't want to be. Once you can hear them, you can't unhear them.

Without a doubt

[cundare]

But there are qualifiers, some of which others have mentioned. - Your playback equipment has to have sufficient resolution to be able to reproduce the differences. On my high-end Quad ESL / Quad II system, it's not all that hard to hear differences between even FLAC and WAV, all things equal. But over my car system, the only difference I can hear between a 128Kbps MP3 and uncompressed CD audio is the MP3's digital clipping on overmodulated peaks (mostly percussion hits). - You have to understand what you'd be listening for, and that is sometimes a function of the type of compression. Even lossless FLAC can compromise audio in subtle ways, but most people would not be able to put into words differences that they hear in parameters like soundstage or ambient detail. MP3 is easier because low bitrate MP3 produces pretty gross distortion that's hard to miss and you can train yourself to hear the difference by comparing the same pice of music ripped at a range of bit rates, starting as low as you can. Take ten minutes to give your ears a chance to hear what's going on and you'll never ask another question like the one that started this discussion - Remember that most people who have opinions on this topic are full of merde. Maybe me too, but I've tried to many years to confirm that I'm not merely believing my own BS (degree in physics, a life time of audio engineering readings, a lot of listening to high-end sources, 10+ years writing audio reviews for two mainstream print magazines, and hundreds of hours of benchmarking audio gear & comparing measurement results to what I hear). People like Neal Young, OTOH, I believe, generally overestimate their own competence in this field. For example, wtf does he mean by "15%"? Although I'm not a big fan of the commercial CD (and that's an understatement), I can play you CDs on entrepreneurial music-lover lables like Chandos & dmp that make any analog Neil Young release sound muddy and artificial. And don't start me on Dr. friggin' Dre and his devil-spawn headphones. So before you believe what anybody tells you -- including me -- about audio formats, analog v. digital, tube v. solid state, vinyl v. CD, just take the time to investigate yourself. If you can't hear the difference, it doesn't matter. If you want to be able to better hear the differences, take the time to investigate yourself. - And maybe the most important point: There's an increasingly popular impression today that concerns about audio quality is elitist technobabble motivated by cultish geeks with too much money. Nothing could be further from the truth. The bottom line for any audiophile I know has always been the music. A great sound system with well-recorded source material allows great music to shine. Even the most compelling source material can be uninvolving, or even fatiguing, when played over crappy headphones or recorded with too much bass and compression. If you don't have the experience or vocabulary to describe the inaccuracies you hear in reproduced music, you would likely assume that the music itself is not so good. It's no coincidence that general complaints about how "artists are simply not producing new music as good as what we used to hear" begain to gain traction at about the time that the industry started seriously screwing around with commercial recordings. But even if you don't like classical music, listen to one of the superb 1960s Living Stereo orchestral recordings ($10 all over eBay) on a good set of headphones and you may find yourself amazed by how lifelike the experience is.

Re:I knew cables were bunk but really learnt it wh

[Pieroxy]

And I can understand why it could - in theory - matter for speaker cables. I mean, there's analog signal running through so if our ears had the resolution - and the speakers the capability to reproduce it - I'm sure it all could make a difference. Not at $50k, but still. Not mentioning our ears don't have the resolution. But power cords? Really?

Some people have too much money.

It's not just what you hear; it's what you don't.

[BuffaloBill666]

It's not just what you hear, and what you consciously notice, it's what you DON'T hear and don't consciously notice. Subharmonics and Ultra Harmonics, above and below the range of human hearing, and the complex harmonics that are left out [and of course the smooth analog wave, as opposed to the stacked cubes of of digital with varying degrees of density]. I'm no expert [very far from], but from the little I know about, this is part of the subject of psycho-acoustics.... and yes, while you may not be able to CONSCIOUSLY tell the difference, your nervous system certainly can, and it is DEFINITELY not the same experience. You do not experience music solely with your frontal-lobes and language centres. I love dig. for it's convenience, but many acoustic worlds and experiences are lost.

One Sound Mixer's perspective

[Airon]

I'm a trained mixer with many thousands of hours of listening experience. Here's my record of telling apart 320 kbps MP3 files from the original CD audio, which is what losslessly compressed audio is as well :

Once, in a specific enviornment.

The material was high dynamic range mix of orchestral and eletronic music. Well mastered IMHO and pleasant to listen to over longer periods of time. I listened to it in an accoustically well-treated room that housed the working gear of a composer and music mixer. The speaker system was a set of professional monitoring speakers, namely two Dynaudio BM15A full range speakers. These are the kinds of speakers the folks use to make the material all you audiophiles listen to.

I didn't expect to be able to tell a difference but I did. It wasn't very subtle but in this excellent listening environment with an excellent reproduction system, it could be heard by trained ears. I seriously doubt I could say the same for material with the dynamic range of a sine wave, as is the case so often today.

For most material, I'd only trust a proper ABX test, but I do not see the point in doing so. It's just one batch of material I listened to and it's pointless to argue when so much music is beyond screwed in terms of distortion and dynamic range.

Also, it's a fact that most rooms people listen to music in are anything from ok to bathroom-terrible. The accoustics are by far the biggest influence you'll ever encounter in a listening environment. Even in a good environment or with very good headphones does it take training to even detect those differences when comparing the original audio to a 320 kbps MP3 encode of it.

Really, there's nothing to gain from this but "I feel better using Flac".

Re:Multi-track is what we want

[ebunga]

Oh hell no. Listening to raw tracks is a bit like seeing a first thing in the morning. Have you actually taken a song from raw tracks to a final mix? It's not exactly an easy task. If you really want to try it, take a look at http://www.shakingthrough.com/stems and have fun with your favorite DAW.

Re:Better question

[coldfarnorth]

I actually like lots of music, even though my hearing's not so great. I do tend to prefer music that has strong vocals or strong instrumentals, but not both - it can be hard to distinguish between them.

What I find helps me appreciate music the most is a quiet room with a comfy chair and a glass of wine. Distractions or background noise makes things tough.

As to whether I could recommend a particular device for your partner, the answer is no. The details of this totally depends on the state of your partner's hearing loss. I would suggest that (if you have not already done so) you go visit an audiologist and make sure you have a good understanding of the loss, then go shopping for devices with an MP3 player full of your favorite music. Find out what works by empirical testing.

Depends on the bitrate.

[screwdriver]

I have a decent sound system in my car and if the bitrate on a lossy compression is high enough, I can't tell the difference. I'd say 192 KBps and above and I'm good. Streaming Pandora on the same audio system is very "tinny" and I can definitely tell the difference. I still think that a high quality digital system is better than vinyl, even though I might be considered the audiophile's version of the antichrist for saying that. Our ears just aren't as good as we think they are, especially when we get older. Digital audio has the potential to sound exactly the same from one playback to the next, whereas vinyl doesn't.

Re:AIFF?, Flac!, Lossless in General. & Random

[God Of Atheism]

I've got the same problem with CRT televisions, less so with monitors. Because of the annoying high-pitched tone coming from them I'm glad they're going the way of the dodo.

I did do a test of which I don't know whether it counts as double blind, with a track in wav and 44.1/16, 48/24, 96/24, and 192/24 flac formats (I think from Linn records as testcase). All from the same high quality master, I played them a number of times in random order, and then checked whether I recognised them correctly. I found that 96/24 was noticably better than the lower rates, but I could not discern any difference between 96/24 and 192/24.

That said, I did do a hearing test a few years ago, which surprisingly told me my hearing was very good for my age (then low thirties). I did suffer from a sudden prolonged bout of tinnitus about seven years ago, and have repeatedly had short episodes of tinnitus after attending metal concerts (before this happened).

Re:Better question

[lsatenstein]

Monstor cabels are a marketing gimmick to lighten the weight of the consumers wallet. If you took the local Dollarama store cables ($#2.00) vs the Monstor cables, the only difference you would measure would be the gullibility of the consumerr and the greed of the vendors.

In fact, analog audio is not carried on the cables, but digital signals. Digital signals are clocked ones and zeros. When would the signal propagation in a 6 foot cable be serious enough damaged by skewing to prevent the signal being being reachedand decoded by the TV or monitor. Digital sigals are faithful, unless there is a flaw in the connection. (broken connection). Gold plating is great if you are tansporting kilowatts of power as 60 cycle 120 volt residential connections and dont want the wire to get warm. Copport is the preferred conductor.

To kill digital signals, put in a low frewquency bandpass filter, considting of choke coils and filter capacitors. ..

Re:Better question

[MiG82au]

No audiophile has ever heard the difference between FLAC and 320Kbps mp3 audio in an ABX test at a statistical rate that is better than guessing.

That's a bold claim. You're going to be really disappointed with this thread.
http://www.hydrogenaudio.org/forums/index.php?showtopic=70598&start=0

Audiophiles can be idiots

[Dishwasha]

I'll give kudos to HDtracks for offering a service I would actually pay for, but I find their 192kHz/24bit service a bit idiotic and the people who buy from it are gullible.

When you purchase an HDtracks file, it is the same quality as a store-purchased CD.
-https://www.hdtracks.com/index.php?file=staticpage&pagename=faq#1

So what you're saying is that somehow I get more fidelity when a 44.1kHz/16bit digital audio source is upconverted to 192kHz/24bit? Perhaps they should give more detail on where their actual music is sourced from. Saying how the artist originally mastered their work is just snake-oil unless they guarantee that was the source used to derive the hidef copy they are selling.

A difference may be heard...

[PCeye]

At higher bit rates I'm not bothered by lossy compression. I can be bothered by the results at lower bit rates and if I am aware how the track is supposed to sound.

Back in early 2000, I ripped much of my earlier collection using 224 kbps ABR. I was a big Maximum PC reader, and one of their multimedia issues recommended using VBR for MP3 encoding. Not understanding too much about encoding, I used ABR for compatibility, and "stereo" as I found "joint stereo" butchered cassette rips. I played these tracks mostly through my PC & laptop so I didn't notice any issues. I used CD's on my main system anyways and never used headphones.

When I purchased a Grado headset for a new (and first) iPod, I found differences in many of my CD rips. This bothered me to no end. For example, Thievery Corporation albums had distorted flaws in echo decay, and highs were harsh. Strings in some classical music seemed butchered while piano had detectable warbliness. High hats seemed wrong in my rock recordings. Choral music vocals sounded harsh. Similar experience when iTunes finally came to Canada, I bought a couple of Iggy Pop tracks. They were aac's encoded at 128kbps. The tracks were clean but the guitars and cymbals were so harsh, I had to stop listening after only a short while. Once I bought the "New Values" album on CD, I didn't experience the fatigue with the same tracks.

Most of the issues mentioned above dropped with properly set command line in Lame with significantly higher bit rates. I do notice a difference. Once I set up a media server all the old rips had to go. I re-ripped my collection. I notice very minor differences with Lame V0 tracks, and 320 kbps CBR, compared to CD's but they don't prevent me from enjoying the music. If I buy tracks from eMusic, ignorance to the original recording is bliss. At their low prices and with tracks ripped mostly with Lame V0 to V2, I can also accept the cost/quality trade off.

I have a couple of DVD-A and SACD discs. Aside from a slightly better sound stage, I would probably fail an A/B comparison test. Either my 40 year old hearing or my equipment would fail me. If I'm being charged iTunes prices, I would still opt for the CD or FLAC equivalent for my music, and hope sound engineers trend back to recording quality.

Re:No

[Hatta]

As I said the setup is highly subjective

In other words, it's not real.

Re:No

[Hatta]

There is tons of research out there, and no one has been able to distinguish LAME -V0 from PCM in blind studies. You're the one who is claiming it's possible. Produce the data.

Re:A lengthy, thorough, and well-explained discuss

[telchine]

There is a long discussion among very qualified individuals on this subject. You can read it here

Hmm, I'm worried that the HTTP compression used for that discussion may have resulted in the diminished quality that I can see. Can you please provide a copy of it on pen and paper for me so that I can read it in the way the authors originally intended?

Don't knock the placebo effect

[Trashcan Romeo]

I personally can't tell the difference between V0 mp3 and FLAC. But I'll still spend the download time and storage space on FLAC just because it gives me a warm feeling to know that I'm getting all the bits. An irrational pleasure is still a pleasure.

Re:Better question

[t4ng*]

The article is talking about the scam trying to get consumers to buy uncompressed (as in data compression, not audio compression) 24-bit 192KHz sampled audio as opposed to 16-bit 44.1Khz sampling. I don't care how long you've been a studio engineer or how perfect your hearing is, the human ear is incapable of hearing the difference between the exact same audio sampled at 16bit, 44.1KHz and sampled at 24-bit, 192KHz in a blind ABX test. And even if your system is actually capable of playing back the ultrasonics that a 192KHz sample rate can capture, that is just wasted energy and storage space. It makes about as much sense as making video cameras and TVs that could reproduce ultraviolet light and claiming that somehow improves picture quality.