Neil Young Pushes Pono, Says Piracy Is the New Radio

Hugh Pickens writes "Kia Makarechi reports that Neil Young isn't particularly concerned with the effects of piracy on artists but is more concerned that the files that are being shared are of such low quality. 'It doesn't affect me because I look at the internet as the new radio,' says Young. 'I look at the radio as gone. Piracy is the new radio. That's how music gets around. That's the radio. If you really want to hear it, let's make it available, let them hear it, let them hear the 95 percent of it.' Young is primarily concerned about whether the MP3 files we're all listening to actually are pretty poor from an audio-quality standpoint. Young's main concern is that your average MP3 file only contains about five percent of the audio from an original recording and is pushing a new format called Pono that would be 'high-resolution' digital tracks of the same quality as that produced during the studio recording. Young wants to see better music recording and high resolution recording, but we're not anywhere near that and hopes that 'some rich guy' will solve the problem of creating and distributing '100 percent' of the sound in music. 'Steve Jobs was a pioneer of digital music, his legacy was tremendous. But when he went home, he listened to vinyl.'"

Comments

FLAC

[Trintech]

Isn't FLAC already lossless? What makes Pono better?

Re:FLAC

[robmv]

Looks like the name is trademarked, so this looks like a way to request money for "Pono compliance"

Re:FLAC

[ant-1]

FLAC is lossless from an audio CD perspective. There is a huge difference between studio recordings and CD content (you lose a lot on both ends of the spectrum, among other changes). That's why audiophiles prefer vinyl, because it captures more sound from thestudio recording. Pono is a try to capture like 100% of what the musician get on the studio tapes.

Re:FLAC

[fuzzyfuzzyfungus]

Based on TFA's (somewhat fragmentary) description, this 'Pono' nonsense appears to be some dreadful 'ecosystem' that includes high quality recordings in some format; but also "Pono's cloud-based libraries" and quite probably some sort of 'social' crap.

Also, given that TFA has some stuff from Young about how CDs ruined everything, the plan presumably also includes using a lossless, or less-lossy, format on the same sources from which CDs are generated, rather than to sling CD audio around.

FLAC would certainly be capable of being the compression system for such a scheme(and, let's face it, all lossless compression systems are going to sound the same, even wholly unoptimized ones like 'well, just gzip the .wav file...', so the only real question is whether somebody's patented magic sauce math will save you enough bandwidth to be worth the licensing fees, or whether using FLAC(possibly with some ghastly proprietary data fields or DRM wrapper) is easier and cheaper).

Re:FLAC

[characterZer0]

FLAC can handle up to 8 channels, up to 32 bits per channel, and a sampling rate up to 655350 Hz.

Redbook CDs use 2 channels, 16 bits per channel, and 44.1kHz sampling rate.

FLAC is lossless from perspectives of much higher quality that CDs.

Re:FLAC

[TheGratefulNet]

flac is fine and it works great. not sure what else you want to improve on it.

it does not have 'trailing garbage' like mp3 does (mp3, without hacks, does not know the *exact* duration of the song; the last block could have unknown padding). this is what causes all the issues in gapless playback.

flac *does* count to the last sample, and so you can append flacs and get gapless playback as the source cd intended.

flac supports 24/192, which is pretty much the highest you'll find for anything commercially buyable. I build and test dacs and i2s systems and all my 24/192, 24/176, 24/96, 24/88.2 stuff is in flac format. and I have to keep testing with all the samplerates and word depths (16, 20, 24, even 32) on my hardware.

flac is even seekable (old shorten (shn) was not as easily).

flac has tags and they are rich enough to be useful.

just not sure what's wrong. probably nothing. what's going on neil? why isn't flac doing it for ya?

"music man, better keep your head,
don't forget what the redbook said."

Re:FLAC

[Tapewolf]

What do the studios record at?

I believe it's 24/96 or 24/192 mostly.

Re:FLAC

[TheGratefulNet]

more and more, they are doing recording to 24bit and 192k (as opposed to 44.1k which we get on cd). most studios agree that there is no need to record/capture higher than 24/192. you keep that format ALL thru out and mix-down to 16/44 only at the last step.

much like you want to capture photos in .raw mode, keep them in 16bit color ALL the way thru pshop, then dither-down to 8bit jpg when you do a 'web save-as'.

studios capture at 24bit (a/d converters are mostly junk beyond 23rd bit anyway) and at 192k or even 320 if they want bragging rights.

some use 'odd' samplerates like 176.4 and 88.2 in addition to the more standard 96. I have files from 'high res' music sellers in pretty much all those formats. its a PITA for DAC and spdif chip guys, let me tell you ...

Re:FLAC

[Desler]

Audiophiles prefer vinyl because they think it gives better sound because they prefer the analog artefacts. In the real world, on average vinyl records and LP players were of pretty low quality and could be easily beaten by a properly mastered CD and even a mid-range CD player hooked to a decent AMP. The benefit is you save thousands of dollars on snake oil audiophile gear.

Re:FLAC

[Anonymous Coward]

You are full of nonsense. First of all, FLAC supports 24-bit samples and up to 655kHz sampling rate. Nobody can hear the difference between 16/44 and 24/96+ if they don't know which is which and vinyl is inferior in every measurable way. You don't have bat hearing so there is no need for high sampling rates and the dynamic range of music fits well within 16 bits. Very few people even have a room/system that can reproduce 16-bits of dynamic range even if there was that much to listen to on the recording. Which there isn't, because almost all music, which didn't have 96dB range in the first place, has had the dynamics mercilessly crushed out of it. The quality of filtering algorithms is such now that has eliminated the any benefit at all to higher sampling rates, as revealed by double blind tests.

Perhaps you meant the music sounded better in the studio right after the musicians/producer finished mixing it and before it was sent off to some jackass who calls himself a "mastering engineer" and crushed the life out of it and clipped all the peaks.

Re:FLAC

[Desler]

It's not rubbish at all. You've fallen for audiophile myth..

Re:FLAC

[roc97007]

Vinyl, um, no. For a lot of technical reasons that no doubt many others will point out, so I will try not to be redundant here.

Directly from studio to flac, then you'd have something substantially better than CD, and substantially better than what vinylphiles think vinyl buys them. Not necessarily better than the studio's original, but at some point one has to say "that's good enough". Vinyl even at it's best (I used to collect the Sheffield Labs disks) is not as good as it could be. CD as a medium *could* be pretty good, (Example: Sheffield's "I've got the music in me") but the mix-down of commercial CDs is often horrible.

I agree with your comment on photographing in raw mode, though. As you said, like audio, you stay at full depth and resolution with no compression artifacts all the way through the process, dithering/compressing at the end. But even there, people will tell you that jpg in-camera is the only way to go. One school of thought is that the camera has all these algorithms for compression and noise reduction and color correction that you aren't using if you shoot raw. To which I respond, "yes, that's true. And that's a good thing." With adequate tools and knowledge of your subject and intended goals, you can always do better than some all-purpose algorithm produced by the manufacturer. I strongly suspect that this is all true for audio production also.

Re:FLAC

[hairyfeet]

For playback yes that's dumb but I think most would agree that for recording guitarists NEED tubes. The distortion from solid state, even $2k+ modeling amps, just sounds like shit. As a bass player i prefer solid state, my Trace Elliot is insanely clean at any volume (weighs a fricking ton though, those 2 ten inches have magnets that are HUGE, its like moving a fricking safe) but I have yet to hear a guitarist, live or recorded, that didn't sound like shit with solid state. In fact the ONLY time I heard one sound really good the guitarist had a pedal using tubes for his distortion and only used the solid state for amplifying the siignal.

Everyone may make fun of tubes but frankly we just haven't figured out how to get a nice fat distortion that responds to the player with solid state, its too brittle and is all or nothing whereas with tubes you can roll off the volume and get dozens of levels of distortion without the harshness. Of course if you WANT harshness, like Black Flag or Pantera then solid states work great, most people don't want their guitar to sound like a chainsaw.

Re:FLAC

[jenningsthecat]

Don't forget to add lots of harmonics to make it sound like tubes were involved. It's not High Fidelity without the characteristic tube distortions.

Enough of this 'lots of harmonics from tubes' BS. From a comment I posted in May of this year:

"By the way, in the 'tubes vs transistors' debate, good triodes have the advantage of being more intrinsically linear than transistors. This means that they require less negative feedback to tame their distortion, and often sound wonderful with NO negative feedback. The THD figures of amps built this way are often quite poor, but look at their spectra and you'll see predominantly second- and third-order, with a smooth and rapid falloff of higher order harmonics. Occasionally solid-state amps can give this kind of performance, but tubes have an easier time of it. Designing a good-sounding, (as opposed to good-measuring), audio amp, requires a lot of skill, and a lot of knowledge about distortion mechanisms and how to counter them. Unfortunately the prevailing practice in HiFi is to add more gain, throw most of it away with additional NFB, get a nice low THD figure, and call the job done. Amps designed this way generally sound like shit, if not initially, then after 20 minutes or so of listening, at which time listening fatigue sets in."

Solid-state amps, (most of which require lots of negative feedback to get anything close to a listenable output), and poorly-designed tube amps, (primarily pentode-based), are the ones that tend to have "lots of harmonics". Good triode amps have some second harmonic, a little bit of third harmonic, and not much at all in the way of higher-order harmonics. Although their THD figures may be higher, the amps sound better because they don't have the higher-order harmonics characteristic of high-negative-feedback designs.

THD is a very poor metric for audio amplifier performance. This was widely recognized in the 1940's - audio engineers such as the BBC's D.E.L. Shorter proposed calculating THD by weighting harmonics by the square, or even the cube, of their order, based on the recognition that higher-order distortion is more audible and more objectionable. But marketing forces won out, and now we have a THD metric that is very poorly correlated with actual listening preferences.

I think a lot of this talk about tube distortions comes from the guitar world, where amplifiers are designed to distort in controlled ways in order to impart a characteristic sound. Designing guitar amps and designing audio amps for home listening are very different disciplines.

Re:FLAC

[Anonymous Coward]

Nobody who understands digital audio and oversampling converters would ever record at 192kHz. No double blind test ever confirmed the claims of the golden ear brigade. Anyone recording at 192kHz is a "sound engineers" in the pejorative because they clearly do not know what they're doing.

48kHz sample rate and 24bit resolution already exceeds the capabilities of human hearing. Any difference people think they can hear (nobody can reliably identify higher sample rates in a double blind test) is undoubtedly due to poorly designed converters.

Low Quality = Radio

[Jason Levine]

Back in the days of Napster, I thought that the recording industry's best course of action would have been to purchase Napster. My idea would have been for them to limit the bitrates of freely shared music (say, to 128kbps) while selling higher bitrate versions of these songs. Listening to a 128kbps copy would have been the equivalent of hearing it on the radio and would have guided people to buy the full-quality version.

Remember, this was before P2P sharing and before Apple/Amazon/etc opened online music shops. The recording industry would have turned piracy into a source of revenue. More than that, though, they would have gotten ahead of Apple/Amazon/etc and would have been the main source for legal digital music purchases.

Yes, some people would have complained and found other ways to freely share MP3s greater than 128kbps, but if they did it right, I think most people would have remained. Instead, they shut down Napster and from its corpse sprang the P2P programs that the recording industry played Whack-A-Mole against for the next decade.

Psychoacoustics and perceptual coding

[Omega Hacker]

He clearly doesn't understand the first thing about the human ear or brain. The *bitrate* is 5% of the original CD, but the human-effective *datarate* is ~95%. That last ~5% of the signal is various harmonics, twitchy bits, and other stuff that the human ear is simply incapable of hearing, but in terms of actual spectral data it's pretty incompressible. Lossy audio compression makes the perfectly legitimate trade-off that you can completely skip that incompressible chunk of the audio signal that the human ear can't actually hear, and save bandwidth.

Modern psychoacoustic models take into account both the physical and mental limitations of the human body. A prime example is "masking", where a louder sound will completely overcome a quieter sound, and do so for a period *longer* than the loud sound. Think of the ear as having an AGC with a slow response: it has to adjust the "gain" for the louder sound and ends up missing the quiet bits before it, then has to adjust the gain back down before it can pick up the quiet bits after. Simple compression trick: toss the quiet bits cause you can't hear them anyway.

What's clear is that he's just fronting for the latest in a long line of "we're better at this than the entire rest of the world combined" snake-oil audio companies with a nifty little lock-in strategy. Just read the list of trademarks.....

Re:Psychoacoustics and perceptual coding

[mcgrew]

He clearly doesn't understand the first thing about the human ear or brain. The *bitrate* is 5% of the original CD, but the human-effective *datarate* is ~95%. That last ~5% of the signal is various harmonics, twitchy bits, and other stuff that the human ear is simply incapable of hearing,

I think you're the one who doesn't understand sound or hearing. The human ear/brain is indeed capable of hearing the difference between a pure sine wave at 440 Hz and a middle C tone played on a piano. The only difference between the pure sine and the piano is harmonics. If you couldn't hear harmonics you couldn't tell the difference between a guitar or a piano -- the only difference is the harmonics.

Human hearing ranges from ~20-30 Hz to 15-22kHz. Most teenagers can hear all the way to 20 kHz. If you take three electronically produced 15kHz tones, one a sine wave, one a sawtooth wave, and one a square wave, that teenager can tell the difference between them. But record those three tones at 44.1k samples per second and you have only three samples per crest; not nearly enough to discern what the shape of the wave it is. That's why vinyl sounds better than CDs if your turntable and speakers are good enough -- and most aren't. In fact, with your average non-super speaker system, the lossy compression will indeed sound no different than the CD it was ripped from.

You are correct, however, about the ear/brains' inability to hear a soft sound when overlaid by a louder sound; IOW its lack of dynamic range. Another lossy compression trick does the opposite; introducing gain on softer parts to make up for a lack of dynamic range compression introduces, and this is clearly audible on samples of LPs and cassettes; the scratches and hisses are magnified. You'll hear tape hiss on the MP3 that you couldn't hear on the original cassette (that said, by the time CDs came out, Dolby had made hissing cassettes a thing of the past).

I also agree that he's shilling for a lock-in; when FLAC or SHN are uncompressed, you get an exact copy of the original digital signal. There's no need for a new, proprietary codec.

This concept has been thoroughly debunked

[MetalliQaZ]

The concept of distributing 24-bit studio recording data directly to consumers in order to get some kind of increase in sound quality has been thoroughly debunked. See this Slashdot article: Why Distributing Music As 24-bit/192kHz Downloads Is Pointless

That article is well worth the read.

Not a step up.

[pavon]

No, it's not a step up. No-one has ever been able to reliably distinguish a 24/96 recording from it's downgraded 16/48 version in a properly conducted double-blind test.

It is absolutely necessary to oversample when acquiring data (since all analog filters have some roll-off), and it is good to use higher dynamic range when mixing to keep the repeated rounding errors below the noise floor. But once the final recording it is mastered, there is no benefit to distributing or listening to the result at higher than 16/48.