Hacker News new | past | comments | ask | show | jobs | submit login

I only buy lossless music. I usually end up buying WAV/AIFF/FLAC direct from the artist, from their record label, or from Beatport. Once in a while, if I can't find something, I'll buy it on CD. Earlier today, I actually bought vinyl because I couldn't find a lossless digital version of this track anywhere: http://www.youtube.com/watch?v=AwsUfEvmKwI

Lossy formats are useless to me since I can't remix them without it sounding like crap. I also feel I'm getting ripped-off when they charge near CD price for lossy tracks.

Also, 20 years from now I'm going to want to convert my music to whatever fancy lossy codec has supplanted MP3 and AAC. By archiving in an open lossless format like FLAC (or now ALAC), I can do this without introducing artifacts.




> Earlier today, I actually bought vinyl because I couldn't find a lossless digital version of this track anywhere

Two things strike me as astonishing. The first is that anyone remains unconvinced that vinyl has any redeeming technical aspects whatsoever, other than the touchy-feely emotional stuff, and self-deception.

The second is that the music you referenced is so unashamedly digital, that the notion of purchasing it on a vinyl record is like buying a copy of Angry Birds printed on tree bark.

A 256 kbps AAC file (as sold by iTunes) is so very nearly transparent, only a small fraction of 1 percent of people would be able to identify the vanishingly minute differences. The difference is so small, it would be comprehensively outweighed by meaningful factors such as your choice of speakers, the shape of your room, and the presence of any ambient noise.

Whereas vinyl has crackle (unless you're playing a pristine copy in a dust-free environment), clicks, pops, rumble, wow distortion, and intentionally limited dynamic and tonal range. Fidelity progressively reduces as you move to the inner grooves, and high frequencies are literally scratched away as the stylus scrapes past -- every time you play a vinyl, it will sound poorer than the last. You can mitigate some of these problems, but generally at great effort or expense.

Of course, to make that music useful, you'll need to rip it back into your computer. You'd have to be mentally ill to believe that a [Lossless > AAC-256 > Lossless] conversion is more detrimental than [Lossless > equalisation > analogue mastering > lathe cutting > vinyl stamping > stylus scraping > pre-amplification > ADC > Lossless].


one thing lossy formats do really bad at: speeding it up and down. (because they remove 'unhearable' sound artifacts -- that become hearable at different speeds)

big slowdowns also sound 'ugly' with lossless formats (a 44kHz sampled track at half the speed is only 22kHz).

but who plays a track at halfspeed..? :)

i dont want to make a case for vinyl -- sjwright has given a nice overview of arguments against it, that i can all underline.

but there is one thing that a physical soundwave, pressed in a disc of vinyl, never needs to do: anti-aliassing. but since the tracks usually get delivered digitally to the vinyl-press; it is the vinyl press that does the anti-aliassing for you :)


There's nothing the vinyl pressing process can do to a digital sound source that can't be modelled in a purely digital environment.


> Digital music is technically superior to bumps on a plastic disc

What you completely failed to realise, is that technical superiority is not the only factor as to how music sounds.

CD mastering these days is almost invariably awful. The real advantage to vinyl (which comes partly because it is a niche format, and partly because of how it is made) is that it has better mastering. This is not subtle on anything better than PC speakers.


> What you completely failed to realise

...is that you're going to invent quotes? Nice.

> [Vinyl] has better mastering

A common myth -- google it. For music produced in the past decade, it's exceptionally rare for a vinyl mastering to be superior to the 16/44.1 digital release. Of course there's always exceptions to prove the rule, but the rule is definitely not as you assert.

> partly because of how it is made

Because of how it is made, a vinyl release is often the same master as the CD but with an additional pass through a multi-band limiter. This unquestionably degrades the quality of the vinyl master, and is done to stop the cutting head coils from literally burning up.


I paraphrased. It was done for brevity. Did i misrepresent the intent of your post?

I do not need to google it, becauase i can hear it myself. Like i say, not subtle. I haven't got vinyls/rips of everything, but i can assure you it is not uncommon. Must i provide examples?

Destructive 'loudening' of music is not useful with vinyl music (IE. compressing the waveform 'up' does not make for a louder output, as what happens with the digital loudness wars-- thus is 'pointless'), because there is no 'reference point' (that is, you can't have a waveform offset, because the needle is already moving wherever to meet the waveform), you can only increase/decrease the dynamic range. Because of a greater difference in amplitude, it is easier to cut a record with a larger dynamic range.


> I do not need to google it, becauase i can hear it myself

Try doing an ABX comparison of whatever it is you think you can hear, and then talk to me.

> Must i provide examples?

Prove it to yourself first with a representative sampling of vinyl records and a blind waveform analysis.

> Destructive 'loudening' of music is not useful with vinyl music

That's true in theory. It's also true that destructive loudening of music is not useful with digital music, but they still do it. In reality, most vinyl masters are the same mastering plus an additional pass through a multi-band limiter. This is not a guess, by the way.


You are stubborn as a mule.

I have done ABX. Like i said 1000 times, it wasn't subtle. I don't need 'a representative sample', i wasn't arguing 'all' or 'most', but 'not uncommon'.


> Like i said 1000 times, it wasn't subtle.

Yes, the distortions and noise added by even the most pristine vinyl reproduction aren't subtle.


I'm fairly certain that devices in twenty or fifty years will still be able to play MP3. It's too pervasive and supporting it is too easy for it to disappear just like that. It might not be supported in hardware (like it is today in many devices to save battery life) but there will be plenty of software.

You tend to not be able to use formats because hardly anyone used them in the past. MP3 is too popular for that to happen. (As is JPEG, PDF and probably AAC.)


The other factor is that the patents that serve as a disadvantage for MP3 will be expiring sometime around 2015-2018.


Do you really think you could hear the difference between 256kbps AAC and Lossless? Even if your hearing is way above average, most audio players and earphones aren't that good.

(Also, given the less-than-subtle music you linked to, I doubt that it really matters.)


No, I can't; and I wasn't claiming that I can. However, as I stated, if I use an MP3 or AAC in a remix, then compress my remix using a lossy codec like AAC or MP3, the result will sound like crap. It's similar to working with JPEGs in Photoshop (rather than RAW), then trying to re-compress to JPEG: you usually get artifacts.

I've been playing classic violin since the age of 3, and I'm very sensitive to details. Just because you don't find a particular genre of music "subtle" doesn't it's not susceptible to the compression artifacts I'm talking about.


> if I use an MP3 or AAC in a remix, then compress my remix using a lossy codec like AAC or MP3, the result will sound like crap

Or far more likely, it won't sound like crap.

> trying to re-compress to JPEG: you usually get artifacts.

This argument is akin to suggesting that a very very very minimally JPEG compressed screenshot of a website is inferior to plonking a film camera in front of your LCD and exposing a frame. After all, big analogue artefacts don't matter nearly as much.

The reality is, AAC at 256kbps is such a light touch that it could probably survive four or five rounds of offset re-encoding before the artefacts would compound meaningfully.


Again, you forget he's also remixing. That changes the whole game. You may be right if re-encoding was all he does, but it isn't.

You take the 256kbps high quality AAC, and you time stretch and repitch it by a few percent (to match it up with the track you're mixing it into) and probably fiddle the equalizer a bit, to "drop the bass" :)

If you do that a few times, you're losing information, even on a lossless digital medium.

If you use even high quality audiocodecs in between, it will sound like crap. And not the kind of super subtle near-impossible to hear difference between lossless and 256kbps AAC crap, but actual crap. I know this from experience myself as well. The sound becomes really flat, hard to describe, but you notice it when you're playing one right next to another track that hasn't been through such a process. It's like you want to turn it up louder but it's already loud enough. Hard to describe, but it's a real problem if you're playing these mixes on high quality speakers in a club--suddenly the sound is not as crisp anymore, and the crowd notices it too.

If you want to make the analogy with JPEG, imagine re-encoding at 95% quality, but with a few percent scaling and a few degrees rotation in between, before saving again as 95% quality JPEG.

You do that twice, with a quality high res digital photograph, and you're going to wind up with a photo that looks pretty much like the original ... until you make a high res print of it (or zoom in on the screen). Edges that were crisp at first have gone fuzzy, and at places you can spot the typical blocking and ringing JPEG artefacts.

You'd probably get away with it, most of the time, too. But not always, and if you're a digital graphics professional, this is your craft, and the mark of a good craftsman is that they put in the work for the little details that you don't really notice, until they're not there, and how you tell you're holding a piece of quality work.


Codecs like AAC apply psychoacoustic rules that were only correct for the overall piece being encoded, not samples taken from it. If a song has two overlapping sounds and the human ear physically can't hear both, a lossy codec looks for that and conserves space by completely omitting the sound that was masked, leaving you with no way to recover it if your postprocessing takes away the sound that had been masking it. A lossless codec preserves the sounds you might be able to hear after sampling or remixing, though of course it takes more space to do that.


My guess is that a lot of the psychoacoustic models used by lossy codecs aren't tuned for electronic music. The resonant distortion on the bassline of the linked track, for example, might sound different in the original. I have one song in which I can tell a difference for sure: Who Said by Planet Funk, about 1:30 in, has a rising modulated treble sound that sounds awful in 320kbit MP3 compared to the original.


Psychoacoustic models are very good, but MP3 has unavoidable coding weaknesses. Use a modern audio codec and you won't have this problem nearly as much.


I can... easily. And it matters to me. My hearing is average, and I have HD280Pro Seinheisers which are pretty standard headphones. Maybe it has something to do with the music genre that I like - techno, trance, dubstep, elecronic, etc. This stuff is delicately mastered in the studio and I can tell when the sound isn't reconstructed well upon playback.

I'd love to do a blind hearing test if you can set one up - but you have to do it with a song that I enjoy. I suppose I could even set one up myself as a study. It would be pretty interesting.


> This stuff is delicately mastered in the studio

And football is the sport of delicately manoeuvring your opponent onto the grass.

And heroin is a drug that delicately shifts your brain chemistry.

And a chainsaw delicately tickles tree trunks.


Ok so you don't like electronic dance/club music. We get it.

I gotta say, I disagree with the guy below me being downvoted, because when I read this "troll" was the first thing that sprang to my mind as well.

I don't disagree with you on the matter that you can't tell the difference between quality lossy and lossless recordings, btw. As long as they're the end-product and not being used for further processing.

Which is one of the rare scenarios in which is actually does make sense to prefer a vinyl copy over a high quality lossy compressed file, if you can't find the lossless digital version.

No you won't be able to hear the difference in the version you bought, but it will make a difference if you intend to work with the data and transform it into something else, because then it is possible that sounds that were thrown away by the lossy codec because no human "golden ear" could hear them, contexts shift and they come into hearing ranges again. But the data is lost, so there's something missing instead.

With a digital lossless copy, or with vinyl, you know you're also getting the inaudible bits, even if you can't hear them.

And if you're just going to listen to the track, sure, get the quality lossy codec.

But if you're doing remixes, you're going to want the inaudible bits too, because you don't know if you'll need them or not.

Lossy compression is for end products, not half products.


"Ok so you don't like electronic dance/club music. We get it."

I think what he meant was that the act of mastering popular music often isn't exactly a subtle process. During mastering they try to make the music as loud as possible, reducing dynamic range in the process.

"With the advent of the Compact Disc (CD), music is encoded to a digital format with a clearly defined maximum peak amplitude. Once the maximum amplitude of a CD is reached, loudness can be increased still further through signal processing techniques such as dynamic range compression and equalization. Engineers can apply an increasingly high ratio of compression to a recording until it more frequently peaks at the maximum amplitude. Extreme uses of dynamic range compression can introduce clipping and other audible distortion. Modern albums that use such extreme dynamic range compression therefore sacrifice sound quality to loudness. The competitive escalation of loudness has led music fans and members of the musical press to refer to the affected albums as "victims of the loudness war"."

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


> Ok so you don't like electronic dance/club music

On the contrary, I really love it.

I also love the cognitive dissonance which suggests that even a slight loss of fidelity in the 'inaudible bits' is far worse than a distinct loss of fidelity that occurs with an astonishingly imprecise trip through a multitude of lossy analogue processes.

Or that the addition of audible hiss, wow and surface noise doesn't matter.

You seem to suggest that algorithms like MP3 and AAC work primarily by throwing out the inaudible bits. A bit of that does happen, but most of the compression occurs by removing redundant data. After all, over 50% of the data can be removed before even a single bit has to change -- you can remove a shitload more if you're willing for it to be a very very very very very very close approximation instead. Psychoacoustic modelling is used to ensure that priority is given to sound that matters more over sound that matters less.


In 2011, it doesn't matter much any more, audio is not very big. In 1996, I guess it made sense. For the few cents of extra space you get the peace of mind of perfect copies, the ability duplicate, remix, and reencode ad infinitum.

Notice you won't see the visual effects industry using JPG's to build their shots ... no, they use very high resolution 4k ~48-bit color depth files, that are then combined and then scaled down in resolution/color to theater/blu-ray output. This maintains the highest quality throughout.

You could also think of it the same way as free software (although not the perfect metaphor). Software may pick up compatibility bugs over the years. You'll want the community to have the ability to make fixes or risk losing it.

All these are examples of "keeping your options open," which people value. Now that the cost is not prohibitive and only getting lower there's not much reason not to.


I can.

It depends on how well you know the album, what the music is.

Try Paper Tiger by Beck off of Sea Change... strings and bass at the same time is challenging to lossy compression.

Try Page One by Charlatans off of Between 10th and 11th... same problem.

I don't use an iPod for that reason. I have an external sound card and use Fidelia on my Mac for that reason. I own good headphones for that reason.

When you listen to music you think you know and your heart skips a beat as you hear something new and it moves you... why compromise ever again?

Isn't this the superlative experience that Apple used to deliver? The ability to make you experience something that moves you. Music really can, but when you compress you deaden it, you kill the sound stage, some of the noise is lost.

Listen to a compressed Double Dutch by Malcolm McLaren the skips sound like a synth, listen to the lossless on a capable system and you can hear the distinctiveness of every rotation and hit of the rope on the floor. The latter puts you there, you picture it more.

Sound is even more important than vision. We are all rushing to 1080p and hopefully better, but sound is what fires the imagination more. Film directors have long known this, music producers have long known this... sound is the thing to get right, and when it's right, you fall in love every time you hear the music.

If you're buying any music equipment, find a good retailer and do a blind test of the equipment. I would be very surprised if most consumers would pick compressed audio and poor equipment... more likely they will settle for the point at which they no longer perceive a noticeable difference, but that point is after you've discovered lossless.

The main reason compression is accepted is that on crappy computer speakers, when listening on headphones whilst you have high background noise (commuting)... you could not hear the difference. Everywhere else, the difference is almost black and white, chalk and cheese.


Try doing a double-blind ABX test and then you can discuss your ability to do what has flummoxed all but the most insanely acute 'golden ears'.


Do you really think you could hear the difference between 256kbps AAC and Lossless?

All the people who claim to be audiophiles and think that thousand dollar wooden knobs make their music sound more rich certainly do.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: