Is There Any Merit To Neil Young's Plan To Improve The Quality Of Digital Music?
from the the-(record)-needle-and-the-damage-done dept
Neil Young has been unhappy with the state of digital audio for a while, and he’s made various overtures about fixing it. Now, some trademark applications found by Rolling Stone suggest his plans are in motion, though details on those plans are scarce. The only real clue comes from a tangential mention in an unrelated press release:
A press release issued last September by Penguin Group imprint Blue Rider Press, which is publishing Young’s upcoming memoir, may have revealed the working title of Young’s entire project. In addition to the memoir, says the release, “Young is also personally spearheading the development of Pono, a revolutionary new audio music system presenting the highest digital resolution possible, the studio quality sound that artists and producers heard when they created their original recordings. Young wants consumers to be able to take full advantage of Pono’s cloud-based libraries of recordings by their favorite artists and, with Pono, enjoy a convenient music listening experience that is superior in sound quality to anything ever presented.”
But does Young actually have a new idea? There are already lossless formats like FLAC that some audiophiles swear by, not to mention uncompressed formats like WAV and AIFF. But there is theoretically room for improvement: most uncompressed digital audio is sampled at a rate of 44,100 Hz, but some pro studio equipment can record at twice that, and technologies like DSD can go much, much further. Moreover most consumer audio consists of 16-bit samples, which could be bumped up to 24-bit. So on the technical side, there is the potential for new formats to popularize higher-quality digital audio. Who knows if that’s what Young has in mind.
That, however, leads to the bigger question: is there really a market for such a format? The digital audio debate has been raging for years, and it has a lot of contours—not just the strengths and weaknesses of digital and analog formats, but also changing approaches to sound engineering and the debates over loudness, audio compression and overprocessing. While some audiophiles insist they can tell the difference, blind listening tests have proved they rarely can. For the average listener, convenience, selection and price surely trump such a negligible (and possibly undetectable) quality difference—and since it sounds like Young hopes to develop a proprietary, cloud-only format, I’m guessing those other factors aren’t high priorities. Moreover, since most people are listening to their music on earbuds and other low-definition systems, the quality bottleneck exists much further down the line than the file format—and since an increasing amount of music is recorded with consumer tools like GarageBand that operate at the standard sampling rates for uncompressed AIFF/WAV files, there’s another bottleneck above the file format too. Though, in theory, these factors are part of what Young wants to change with his push towards higher quality—and there may be some potential in that direction over time as bandwidth and storage space increases, and even some sort of immediate market among audiophiles. But it’s hard to see what he could offer that existing formats don’t already provide.
I know some people will insist that digital audio sucks, and that they can tell the difference—but frankly that’s a meaningless assertion if they haven’t done a controlled test. There are simply too many biases to account for. But even if it is a real problem for some people, it is likely to be a very small niche market, not a cultural sea-change like Young seems to envision. Some of his proclamations about the effect of music sound eerily close to Prince’s insane ramblings about how audio interacts with the brain, which is hard to swallow. Music may create transcendent human experiences once it’s inside your head, but your ears are still made of flesh and bone, not magic. And evidence suggests that most people’s ears can’t tell the difference.
Filed Under: digital music, flac, neil young, quality
Comments on “Is There Any Merit To Neil Young's Plan To Improve The Quality Of Digital Music?”
More power to him
It wouldn’t be the first time someone was parted with their money over a perceived value of a product, even when there was no actual value. (Monster Cables, I’m looking at you) If he can get someone to pay for it, good for him.
Re: More power to him
Yup, although he might want to reconsider his naming choice – pono?
Re: Re: More power to him
I liked the name…
Re: Re: More power to him
I suggest “pwned”.
Re: More power to him
The real value with Monster cables is in the return policy. If the cable gets wonky, I take it to Guitar Center and they exchange it with no questions asked. Otherwise for permanent installations I make my own cables.
Neil Young’s pronouncements are pretty roundly dismissed over at the Hydrogenaudio forums, but I think everyone can agree to the question posed here: of course there’s a market for it! Give someone a knob that goes to 11 and they’ll believe it has to be better than one that goes to 10, it just has to! Even if they’re proven to be indistinguishable, better safe than sorry, you know! And then show them one that uses different technology and has numbers going to 1000, well, that’s the best of all, right?
Granted I am no audiophile but I have never seen anything wrong with the digital downloads sold by Amazon. They sound good to my humble ears so I don’t see the need for anything ‘better’.
Digital is pure sound
CD-quality audio (44100 Hz, 16 bits) is almost an exact match for the limits of human hearing. Going beyond that has essentially no benefit, even with the absolute best headphones in a sound-proof chamber. And if you don’t have the best headphones or if you aren’t in a sound-proof chamber, CD-quality is more than adequate. FLAC and other lossless formats will take CD, WAV, or AIFF audio in and produce a smaller file that contains precisely the same sound as the original. MP3 does introduce defects, but at the higher bitrates the introduced defects are mostly unnoticable. WMA and AAC and other more recent formats are even better than MP3.
There is room for higher-quality sound, but only in pre-production situations. An audio engineer might want to take one very quiet sound and amplify it, and to do that he needs more than 16 bits. And certain effects degrade the sound quality by a certain percentage, so if you want a high quality result, you have to start with an even higher quality source. But for delivery to the consumer, WAV, CD, AIFF, MP3, WMA — any of the above is perfectly fine.
Those who prefer analog are telling the truth. However, it isn’t that analog by itself is better than digital. Any sound that can be produced by any analog source can be produced equally well using CD audio. The issue is that analog introduces some noises or distortions that aren’t in the original recording. Some people become accustomed to those noises and end up preferring to hear them. You could make them happy by adding those extra noises to the CD audio and they wouldn’t be able to tell the difference.
Re: Digital is pure sound
There’s a pirate that takes classic vinyl records and meticulously digitizes them to FLAC, so you get the vinyl noise in a high quality digital file (albeit precisely the same “noise” every single time, which is a bit of a downer). I’ve enjoyed his/her work.
Re: Digital is pure sound
Doug is pretty much right on all counts.
The bitrate of a file determines its signal-to-noise ratio (the dynamic range). A 16-bit file can give you a S/N ratio of about 96dB, while a 24-bit file can give you a S/N ratio of about 144dB. In nature, 120dB is the “threshold of pain,” and humans generally can’t hear sounds below 20-40dB (depending on frequency), so 16-bit audio is perfectly acceptable.
Humans cannot consciously perceive sound frequencies above 20 KHz (in children – this starts to go down once you reach eight years old), and anything below 20 Hz is felt as a beat rather than a tone. A 44.1 KHz sampling rate can reproduce sounds up to 22.05 KHz, so no problems there.
Some audiophiles say humans can unconsciously perceive sound frequencies above 20 KHz, but I don’t think this is backed up by empirical tests.
Now, a lot of this has to do with the actual equipment that is used to translate analog sounds to digital signals. In the early years of digital audio, this equipment was pretty terrible – lots of classical music digitally recorded in the 80’s doesn’t sound very good. But this was pretty much fixed twenty years ago.
That doesn’t mean there isn’t room for higher audio quality, but as Doug said, it’s relevant in “pre-production” situations, like when a musician is first laying down tracks. It much more relevant when it comes to signal processing and digital synthesis. These are done using floating-point arithmetic, and using a higher number of bits can reduce rounding errors; a higher frequency rate can also reduce certain artifacts in digital synthesis. Of course, it also takes much more storage space and processing power, but that’s not really a problem nowadays.
With the explosion in digital technology came a concurrent explosion in “prosumer” recording equipment. And this is where Mike is a little off the mark:
In fact, even most “prosumer” recording equipment has been able to record at much more than “twice that” for at least a decade.
I have a digital card in my computer right now that can record stereo audio at 24-bit/96KHz. I bought it in the early 2000’s (possibly even the late 90’s, I don’t remember). It cost me $150 back then, and the modern version of that card can do 24-bit/128KHz and costs $100.
I also own a digital recording console that I got about ten years ago. It can do 8 tracks of recording at 24-bit. It was made for an ADAT, so it can only record at either 44.1KHz or 48KHz (and due to rounding errors during conversion, recording at 48KHz actually isn’t much of an improvement). Even that was too low of an audio quality for digital boards at the time, so they were letting those go for $600.
So, this type of technology has been around for a long time now. In fact, if you wanted to, you could release an audio DVD in 5.1 surround sound in 24-bit/96KHz, and because it’s part of the DVD standard, it would be playable on most DVD players on the market. If you’re only talking stereo, you can go up to 192KHz.
You know what’s really hobbling the high-end market? You guessed it – copy protection.
Because of concerns about piracy, DVD-audio is not allowed to send uncompressed digital audio that is higher than 16-bit/48KHz to any analog device.
Re: Re: Digital is pure sound
he modern version of that card can do 24-bit/128KHz
That’s a typo: it can do 24-bit/192 KHz (in stereo).
Re: Digital is pure sound
Carver of Carver amps did exactly this (add distortions) to his amps to address those who claimed only tube amps could really reproduce sounds correctly.
It was successful in that the tubies praised the sound…
Things really never change 🙂
Re: Digital is pure sound
I think he wants to put the sound waves generated through naturally changing air pressures when recording the sound from source to mic back into an analog/digital mesh. Digital doesn’t capture those air pressure differentiations. Distortions and imperfect sound captures are indicative of live performances and add a personal charter to his music that I don’t think he was able to get in the early pioneering of his digital transformation. Go Neil!
Re: Digital is pure sound
That’s almost true, but not quite. Human hearing (like all of the human senses) works in logarithmic space, in that if you double the sound pressure, then double it again, it sounds like the same increment in volume.
The samples in CD audio, however, are in linear space. The dynamic range for CD audio was chosen to give good fidelity across a wide range of sounds, but the choice of linear space meant that something had to give. As a result, the human ear can resolve quiet sounds to higher fidelity than CD audio can encode.
You don’t notice because good mastering engineers don’t go there. Modern music tracks are engineered with full knowledge of the limitation of CD audio in mind. It’s usually only classical music which ends up distorted, though only in quiet passages, and even then, there are tricks to make the distortion fairly subtle.
Once the labels decided to ditch DRM on music, I’ve never quite understood why they didn’t merely offer at least FLAC if not fully uncompressed audio for their digital catalogues. Sure, it’s a niche market, but allowing audiophiles the chance to buy high quality files (presumably at a premium) should be a no-brainer. It’s not particularly complicated on a technical level, it can be marketed quite easily and I don’t believe there would be much of a licencing problem.
To me, this is just another lesson the industry have truly failed to learn from the last decade. Sites like AllOfMP3 showed that people wanted a choice of product, and that they were willing to pay higher prices for higher quality files. They were actually willing to go the grey market route rather than pay for lower quality files through iTunes, etc. Rather than heed the lesson there, they of course freaked out and attacked the “OMG piracy” aspect rather than the actual customer demand – demand which, of course, is still only addressed by pirates on many occasions (though many indies at least understand the demand).
I don’t particularly care about the sound quality in the way Young does, but fair play to him if he’s successful. Whether I agree or not, there’s certainly people out there who want a higher quality file. However, I can’t help but feel that this is another example of a “problem” that should have been fixed a long time ago if only the labels bothered to listen to their customers.
Re: Re:
They don’t offer FLAC because the iPod won’t play them.
Who wants to deal with all the people who download a FLAC file and are clueless about what it is and complain when it won’t play in their ipod?
Re: Re: Re:
Erm, somehow I doubt that most people who are bothered enough to demand FLAC are going to be playing them back on the iPod…
Either way, iTunes isn’t the only retailer, iPod isn’t the only device, and FLAC isn’t the only lossless format. Give the customers the choice, rather than imposing your preferences on them. That’s really my point, and the demand for lossless formats is proven by the demand for them on pirate sites, even if they’re not the majority. Given the relative ease with which this can be done, why not offer it – and charge a premium if necessary?
Re: Re: Re: Re:
I agree with you, I’m just imagining why the RIAA members don’t offer FLAC. They’re scared people will be confused when the people that buy them can’t play them on their iPod, and they’re probably right. It was hard enough getting them to offer mp3s.
Re: Re: Re:2 Re:
Given their history, I personally doubt that compatibility issues are putting the RIAA off – after all, they had no problems fragmenting and stunting the market when they were insisting on DRM. Then complaining when they essentially handed the market to Apple as a direct result, of course. No, I think it’s more to do with the fact that it’s a niche market and they don’t know or don’t care about it.
As for customer confusion, shouldn’t this be something the retailers should be doing? Whether it’s an iTunes or Amazon offering the service or a new player offering only lossless formats, it would be up to the retailer, not the supplier to provide. They’d be the ones being complained to if they don’t make compatibility issues clear, after all, not the labels.
Anyway, forget I mentioned FLAC specifically – Apple have their own lossless formats and most devices will play .wav files or similar. There’s a market for lossless audio, and there should be no reason why it’s not already being addressed apart from the usual “we don’t listen to what our customers want” mentality that leads to demand for piracy. FLAC is the normal default option, it seems, but there’s others if that’s not directly suitable.
on sampling frequency
You don’t have to be an engineer to understand the fundamental Nyquist?Shannon sampling theorem. Humans can hear sounds up to ~20 kHz (decreases with age). This is the reason of using 44 – 48 kHz sampling everywhere for final audio products. There is no point in increasing the rate because we cannot hear the difference.
Re: on sampling frequency
Just some further technical insight for others:
Audio frequencies/sampling rates are incredibly slow compared to other electronics. A halfway decent oscilloscope can sample in the GHz range. The reason audio is not sample that high is that the computation required to run some of the effects processing grows exponentially compared to the benefit distortion introduced by the processing.
This debate is so overdone. Everything is relative when listening to sounds. There are plenty of times when I’ve noticed something subtle when listening to songs on terribly cheap speakers vs. high quality stereo that make me appreciate the song more. I think that different variety of quality actually benefits the appreciation of music rather than hinders it.
Re: Re: on sampling frequency
You are right in saying this debate is overdone. For better or worse, though, it’s going to be with us for a long time.
That said. Sampling above the human audible range does restore side tones that are present in analog recordings and live performances in the upper range which we’re aware of even if we can’t detect them. The same applies to frequencies below 20Hz which we, generally, can’t hear but our bodies can detect as vibration. Sampling too much above the human range is pointless because there are no side tones we can detect or be aware of affecting lower frequencies. Though it does have interesting affects on our cats and dogs who can hear it. (The domestic cat, apparently, can hear below our bass frequency cut off of 20Hz which is how they can detect the movement of prey in it’s underground burrows.)
You’re also right in that differing quality does benefit the appreciation of music. At least you can hear the difference between narrow and wide band recordings and such.
Not true anymore...
“Moreover, since most people are listening to their music on earbuds and other low-definition systems”
Ehm. It’s simply not true anymore that earbuds are automatically “low-definition systems”. There are a few brands (I’ve bought a pair of Etymotics myself) that are truly aiming for high definition quality and that gives a level of detail that’s impressive. My ER4’s are a good example, especially when you consider they’re “just” earbuds, and I haven’t heard much (speakers or headphones) that’s been better.
After over 2 years I still hear details in some recordings I’ve missed before. Details I’ve missed even in my speakers that admittedly isn’t the high-end cutting edge no expensives saved kind of speakers but are still very much a decent pair of speakers (definitely not “low-definition”).
Apart from that I fail to see a reason to use anything else but flac since it’s lossless, can handle virtually any sample rate or bit depth, anything from mono to surround and is fairly spread by now or, for maximum compatibility with mobile devices etc, mp3 since pretty much anything today can play mp3. I definitely agree that inventing a new format is more or less pointless (and probably will fail anyway).
Moore’s law of music, does this mean this is the end point of digital files. Companies love changing format as people encouraged to buy the same things again. How can this be justified with digital format that there is no evidence of significant benefit to the user of change.
The difference between what?
You were quite unclear about what it is that in blind tests people can’t tell the difference between. I’m of the opinion that it’s quite easy to hear the difference between 22k and 44k sampling, or 128 kbs, 320 KBS bit rates and FLAC. I’m assuming you mean it’s difficult to tell the difference between “high quality” digital and “high quality” analog (of which I prefer digital). I believe the later distinction can be difficult make, so you’d be right. Otherwise I’m unclear how it’s difficult to distinguish.
Re: The difference between what?
The blue words are things you can click on.
Most people is not everyone… Many get much more enjoyment as quality improves…
Re: Re:
Funny thing about how enjoyment works…..even if it’s just in the user’s head, it’s still enjoyment!
Monty of Xiph.org says it’s rubbish.
He might be in position of knowing few things about digital audio.
Anyway it’s a good post.
http://people.xiph.org/~xiphmont/demo/neil-young.html
Re: Is There Any Merit To Neil Young's Plan To Improve The Quality Of Digital Music?
And Monty has a good answer to the post title:”Misinformation and superstition only serve charlatans”.
Not feeling enough like a unique butterfly in a global information culture? There’s a charlatan for that.
Proprietary formats are dead
And as much as I like Neil Young as a musician, he’s wrong about this.
Moreover, cloud storage is exclusively for the stupid and gullible. Everyone equipped with sufficient computing clue knows that it’s not only insecure, but it’s being monitored by the relevant governments. If you want something kept securely and kept securely, you need to use strong encryption and keep it yourself.
Re: Proprietary formats are dead
Moreover, cloud storage is exclusively for the stupid and gullible. Everyone equipped with sufficient computing clue knows that it’s not only insecure, but it’s being monitored by the relevant governments. If you want something kept securely and kept securely, you need to use strong encryption and keep it yourself.
1. I can upload encrypted files to the cloud.
2. Things I upload to the cloud are not things I’m trying to keep secure. They’re things I want greater access to. I’m willing to trade the convenience of greater access for the risk of less security.
Lumping anyone who uses cloud storage as “stupid and gullible” is stupid and gullible.
Re: Re: Proprietary formats are dead
Lumping anyone who uses cloud storage as “stupid and gullible” is stupid and gullible.
It depends upon if the Pono distribution system is cloud-only (no backups). If so, that’s at least unwise, though I wouldn’t go so far as to say “stupid and gullible.”
Pro?
most uncompressed digital audio is sampled at a rate of 44,100 Hz, but some pro studio equipment can record at twice that…. Moreover most consumer audio consists of 16-bit samples, which could be bumped up to 24-bit.
I have in front of me a cheapo digital recording device – 96KHz and 24 bit recording – it’s not expensive “pro equipment”…
“is there really a market for such a format?“
SACD? Flop. DVD-A? Flop.
If that doesn’t answer your question, the latest digital radios transmit music in “CD quality.” The technology behind the Red Book CD format is 32 fricken years old.
The vast majority of people don’t give a shit about how music sounds. Good enough is good enough. CDs didn’t take over because they sounded better, they took over because they were more portable than than LPs and you didn’t have to worry (as much) about scratching them.
Re: Good enough
I was an audio snob for quite a while, but after years of live performances and listening to recordings at levels sufficient to appreciate a wide dynamic range, my ears are permanently ringing.
At this point a transistor radio would sound about the same as anything else to me.
I will note that the ringing in my ears is incredibly detailed and has amazing fidelity…
Re: Re:
True, but the costs of selling hi-fidelity and making them available the whole world of niche listeners is much lower online.
And cassettes were more portable and durable than CDs. CDs did sound better, esp. with classical music (and most people think they sound better than LPs, including me).
The audio snob misses the point.
The audio snob misses the point.
A crap piece of music
@ 32bit/floating point , 192khz
is….STILL a crap piece of music.
Sure , bad mp3 encoding can ruin a great sound , but good encoding on a suitable source can be 100% undetectable.
yes 100% eg…
Square waveform oscillating @ 1000hz ( properly encoded )
Issue that rarely get’s talked about but are reality today for the remix and connected culture we live in.
Data transfer via the net.
eg…
I have a 64 track mix that I want to share with someone so they can remix.(nowadays there is no reason to limit the amount of tracks so… 200 tracks is not abnormal )
That’s 64 wav files ( most containing lot’s of blank space including headroom space)
a 5min song @ 50mb per track X64 is 3.2GB.
Flac enables us to virtually cut out the “blank space” for transfer , shrinking the data transfer size dramatically.
We need the headroom and high sample rate when working/editing/processing/mixing the files.
But when it comes time @ master, we usually…..
1.shrink the dynamic range via compression
2.cut out the inaudible frequencies ( ironic attacking mp3 then ? )
3.adapt the sound to sound “”good”” on as many systems as possible ( like MONO bass for vinyl )
The audio snob misses the point.
Re: The audio snob misses the point.
Truth spoken
Re: The audio snob misses the point.
The audio snob misses the point.
I hope you don’t mean me 🙂 Because I totally agree with everything you just said
Re: Re: The audio snob misses the point.
not directed at you.
Was agreeing with your article , from a different location of thought.
Format / tool / audio snobs do my bap in : )
They are everywhere in the old boy (mostly “rock type” recording ) industry.
Re: The audio snob misses the point.
Mostly you are exactly right. The one thing I take exception to is: “1.shrink the dynamic range via compression”. Yuck!!! That is the one thing that has killed the fidelity of most recent recordings. It has way way more effect on the quality of the recordings you hear than any bit depth or sampling rate change. I bought the Springsteen Pete Seeger tribute CD and only listened to it once because the compression was so bad. You can see examples of this here: http://en.wikipedia.org/wiki/Loudness_war. Even the first example here is pretty baddly clipped.
I think changing this might be what Ms Young is up to. Currently mediocre quality has become the norm (audible or not). If he can inspire some interest in higher quality recordings and get a few manufacturers to build players to support them then maybe he can reverse this trend. He even said as much in an interview I read (not the slightest idea where so no link). So, it’s not about the absolute format but trying to increase quality across the board and he’s using “connect with fans, reason to buy” etc.
BTW, while sample rate etc. doesn’t make much difference, the move from stereo to 5.1 is incredable. Check out the DVD Audio version of “Seven Brigges Road” on the end of the Eagles “Hell Freezes Over” DVD on a decent home theater system.
Re: Re: The audio snob misses the point.
agree on compression usage trends
I too prefer large dynamic range on most music I listen to and make.
( i usually listen on at least ns10’s in sound proof studio , which twists my reality )
BUT reality is that such range is unwanted in a majority of listening environments. ( car , radio , club etc…)
Disagree on sample rate *
*When working with audio.
Double the sample rate… It’s like drawing with a pencil that is half the thickness. ( more detail to edit/draw with)
I Always drop the master down to 44.1kHz
( we cant hear faster than that , like a 500 fps video is also too fast to see )
Sorta agree on 5.1
I make music sometimes in 5.1 and while it does add spacial awareness in the listening field. Move…. you lost it.
Great if you don’t move.
but….
Most of the music I make is club music.
Some clubs ( lot’s in Europe/UK ) have great sound systems where….
tweeters are in the ceiling pointing down
mid range’s are on the walls/ pillars
bass+subs are on the floor
So what I am saying is….
5.1 is great in it’s place , but that place is very restrained.
Good sound systems in clubs can beat the best 5.1 system hands down , in a very large area with movement of listeners.
Re: Re: Re: The audio snob misses the point.
At least *NS10s*?! In a “sound-proof studio?” Be careful there, chief. Yeah, you’re reality sounds twisted. Is this “studio” also your bedroom?
Almost no one with a serious interest in making music uses “at least” NS10s. You use those pieces of crap to check how your mix will sound on crappy speakers. I guess if you’re in a “sound-proof studio” you wouldn’t be able to tell the difference.
Ethan Winer, is that you?
Just, quickly read that again:
“SOUND PROOF studio”
Nothing strikes you as, I don’t know, WRONG with that statement?
I’ll throw you a bone so you can continue playing at being an audio guru on the ‘nets: THEY’RE CALLED *TREATED* ROOMS.
Some Audio Tech Comments
I am not paying much Attention to this.I have been working with Audio since I started fooling With Audio Tech in 1977.My Older Recordings started in Boston’s Original Punk Scene.
I have some really quality gear and still like to do stuff with Audio.Here are my thoughts.
1.Either Using Good (My Fave is Tube-Based Vintage) Analog Gear or Really Good Digital Gear will result in a Decent Recording.Either way it is a matter of preference.The old tinny digital is long gone.
2.If Digital Source should be done at 24BIT 88.2k which when downsampled for CD Master will be an exact half of 88.2………….44.1k is CD Audio
3.Using higher bit depth and higher sample rate gives a Recording the space to breathe.You may not “hear” some of these frequencies of course yet they are still around you and recording high sample rate gives a recording more Space, fullness, and clearer sound.
4.We currently have solutions which are rather good for anyone who wants to “carry” their music around.Using a format like FLAC will satisfy 99% of the people and those not satisfied will probably be the more Audio Anal Crew who are lucky they happen to have ears or maybe an ego to hear more than the rest of us.
5.Bottom Line for me is I am curious in a Tech kind of way but at the same time it just seems like a waste of time.I am of the thought this is not really needed.
6.Would of been cool to have built a Tube-Based Portable Digital Player with internal Audio Circuitry based on 1950’s – 1960’s Awesome Analog Gear.A quasi-Analog/Digital futuristic Player just for playing Audio anywhere with that awesome old “Tube” Sound.
“Tube Sound” vs. Digital Sound does sound a lot different.I prefer old Warm Sounding Tube on Good Audio.
All the technical talk about encoding is interesting, but the real question is, how does Young’s proposal address piracy? It’s easy enough to encrypt the data between the cloud and the player software, but how does he prevent the sounds from being captured between the player and the audio driver? If his player software intends to supersede my operating system then No thank you please!
Re: Re:
It definitely sounds like he’s thinking of some sort of proprietary player/server system where only “enabled” devices will work with the format – but that’s largely a guess since there are so few details
The thing is, I really like Neil Young, so I think I might be overly generous in assuming he even has a plan… I’m giving him the benefit of the doubt that he has a genuine idea in mind (even if it turns out to be a bad one) and that he has considered the kind of factors we’ve discussed here. But, if I’m being honest, it’s equally likely that he was just rambling one day about the low quality of music, made a vague commitment to do something about it, and filed some trademarks — at which point we’re all discussing nothing, because there’s an excellent chance the plan will never come to fruition anyway. I hope that’s not the case but it may well be.
Re: Re: Re:
That said, there were reports that he was discussing this idea with Steve Jobs before he died, so there’s reason to believe that they may have answers for some of the core questions…
Imagine today’s overproduced manufactured pop music in much higher resolution where every nuance of compression and auto-tune can be heard. Hooray.
Well, I hope Neil Young will remember...
A Southern man don’t need him around anyhow.
Seriously, truly high fidelity audio is a Sisyphean effort that goes beyond simply increasing the accuracy in every step of the recording and reproduction process. Our perception is changed by everything from the acoustics of the venue in which audio is reproduced to the way our brains interpret those sounds. (Auditory synesthesia is an extreme example of this.)
There are some exciting developments in audio reproduction that could ultimately make far more difference in overall experience than simply trying to add another 9 behind the decimal place somewhere. Unfortunately, pervasive greed within the music industry or incompatibility with popular technology will likely kill most of these ideas.
Here is an interesting demo I ran across a few months ago. They are not using super-complex processing, but the realism is exciting. Be sure to use headphones or ear buds for the full effect.
http://www.qsound.com/demos/virtualbarbershop_long.htm
DJ's
Well I know of at least one set of folks who enjoy lossless formats and that would be DJs. Especially with bass-heavy music like old school jungle or dnb (or even dubstep, i know boo). With tunes with rolling basslines, it is easy to tell if you are listening to a compressed mp3 vs. wav. Mid range stuff can also sound a little strange once you pump up the bpm’s. The nice thing though is that most of the foreign sites that sell electronic music will give you an option of purchasing the wav file. I think here in the US, we don’t see it as much because a lot of popular music (a lot, not all) isn’t meant to be played loud in a club. Doesn’t mean I wouldn’t mind being able to pick up old school hip hop tracks (DJ Premier, etc.) in a wav format.
Confusion
It amazes me how many people get this wrong.
Yes humans can only hear a waveform around 22 kHz.
But sampling at 44kHz is NOT the same.
The waveform is the actual sound wave.
The sampling is how many times a second the analog soundwave is captured digitally. Therefore, more samples = closer to what the sound wave looks like in reality. A good analogy would be the old 3d video games and jagged lines. Since a curve is simply a TON of small lines put together, the more lines you have, the smoother the curve looks.
I guarantee that if you had a 24 bit file played properly on just mediocre equipment you can tell the difference. Unfortunately, you have to do special things on most computers even if they are equipped with a 24 bit sound card to have it bypass the windows audio driver which downsamples things back to 16 bit (see WASAPI). iDevices can’t play 24 bit properly either.
Peter Gabriel’s Scratch My Back in 24 bit, played back in 24 bit is amazing.
Cassette Tapes
Yes, there’s a market for high-end audio, but how much of a market? Back in the ’80s, cassette tape sales overtook LP sales despite the fact that high-speed duplicated tapes had terrible audio quality. Their big selling point was portability. Everybody had a boom box or a Walkman and people wanted what was more convenient. Today, MP3s are damn convenient!
I think the only way you could get people to switch to a higher audio format is to sell the songs for a LOWER price than standard quality. Everyone would go for the lower price and trade in their old hardware for newer hardware. If you try to charge a “premium” price, it will fail like SACD and DVD-A. (Remember half-speed mastering and virgin vinyl releases? Few people cared.)
The bottom line is most people can’t hear the difference, and with cheap electronics frequencies over 10k are just so much distortion anyway. In a lot of situations, low-fi sounds better than hi-fi.
Improving more than just bitrate
I’ll concede, for the moment, the argument of bitrate / sample rate and the quality of the audio.
What I want to see improvement in is the area of soundfield reproduction. If I’m listening to a live performance of a Chopin piano concerto, I want to be able to pinpoint the idiot with the cough in the 5th row.
Well, maybe not that level, but something that close to being there.
Current stereo mp3/flac/format-de-jure do not give me that capability.
I thoroughly enjoy listening to the few DVDs I have with score only tracks. I get drawn into the music much better than simple stereo.
There is a lot more to the quality of the listening *experience* than bits.
If Neil Young wants to get my money, give me a better experience, not just numbers on a page.
Scott
The Godfather of Grunge
Am I the only one to see irony in ?The Godfather of Grunge? mounting an effort to deliver pristine audio quality?
My guess is that Mr Young is trying to grown a new lawn under the “digital world”, and then he will turn and order the kids to get off his lawn.
Re: Re:
Maybe he will just order you off his lawn.. >:-D
Confusion
Actually, it isn’t at all like the jagged edges on a screen. Audio works differently. Your eyes actually do perceive images as a bunch of dots, so if there are too few dots to see, you see them as dots instead of as a smooth image. Ears are different and instead detect sets of frequencies — a tone at 220 Hz, a louder tone at 330Hz, and one more at 1440 Hz. Digital sound never produces a jagged tone (that would sound awful). Instead, good playback equipment produces a tone that almost exactly matches the original tone with all of the higher frequencies removed (bad playback equipment might introduce distortion here, but that’s not based on sample rate or bit depth).
To summarize: higher sampling rate doesn’t make the “shape” of the resulting wave any more accurate (assuming good playback equipment). Higher sampling rate only affects the cutoff point (frequencies higher than the cutoff point are eliminated from the music).
When properly mastered and with the playback volume correctly adjusted, a bit depth of 16 is sufficient to match the accuracy of human hearing. It can accurately reproduce sounds that you can just barely hear in a sound-proof booth (granted, they’re reproduced at poor fidelity, but double-blind tests have verified that people can’t tell the difference between low-fi and hi-fi sounds at that volume, so the sounds are perfectly accurate at the limit of human hearing). And it can accurately reproduce sounds at the pain threshhold (above which no increase in volume has any noticable effect on the perceived quality). Admittedly, if you turn the volume all the way up during the quiet part, you can definitely hear the limitations of 16-bit audio. But that’s a mastering issue (if people want to turn the volume up during the quiet part, you should have mastered the audio at a higher volume instead of having it be so quiet).
Admittedly, 24-bit audio gives you some leeway for mastering errors. So if there was an inexperienced audio engineer doing the final master on a 16-bit CD and he didn’t follow the correct procedure for creating a good 16-bit final product, you would be able to tell the difference between his 16-bit version and his 24-bit version.
Couple of additional notes. Most modern sound cards can record and play 192Hz 32bit audio data. That doesn’t actually mean anything. It is almost impossible (even at the professional level) to produce audio hardware that accurately records or plays at more than about 19 bits of accuracy. Anything beyond that is just essentially invented by a random number generator on the sound card during recording and is thrown away during playback. That’s not to say the additional bits aren’t worthwhile (see earlier remarks about volume knobs), but they aren’t very important during recording or playback. They are definitely worth having during audio processing, but they have no effect on recording or playback.
Finally, WASAPI’s native internal format is currently 32-bit floating point. The accuracy numbers given above are for fixed-point, which doesn’t exactly translate into floating point, but it is safe to say that WASAPI is at least accurate to 24 bits of fixed-point precision. If there is downgrading going on, it is happening in your app (because your app doesn’t know how to directly provide 32-bit float data to WASAPI) or it is happening in your sound card’s driver (because your sound card doesn’t actually handle more than 16 bits natively). You can skip the native internal 32-bit float format by opening the device via WASAPI’s exclusive mode, in which case you’ll get access to the device’s real native format.
Re: Confusion
To summarize: higher sampling rate doesn’t make the “shape” of the resulting wave any more accurate (assuming good playback equipment). Higher sampling rate only affects the cutoff point (frequencies higher than the cutoff point are eliminated from the music).
Actually, that’s not exactly true.
For example, say you recorded some sound in nature without providing any processing. You record it as-is with a 44.1 KHz sample rate.
Now, there are sounds that are certainly higher than 22.05KHz in nature. Our ears can’t process them, and we don’t need them, but obviously they exist.
What will happen is that if you just sample those sounds as-is, you will get aliasing around the Nyquist frequency. Say, for example, that you have a sine wave at 48KHz. If you don’t filter it out, and sample it at 44.1KHz, you will also get sidebands at 20.1 KHz, 18.15 KHz, etc. (In digital imaging, it is known as a Moire pattern.)
Here is an (exaggerated) example in Ogg format.
So, in order to not have this sideband noise, you need to filter the original signal so that it does not contain any frequencies above 22.05KHz.The steeper the filter, and the closer it is to the Nyquist frequency, the better the audio will sound. In fact, you need the same type of filter to play it back, or else the “square” peaks of the digital waveform will also result in aliasing.
This is where digital audio can go wrong. Crappy DAC’s and ADC’s can result in absolutely horrific artifacts. In both pro and “prosumer” equipment, this is taken care of (both by better filters and by “oversampling”), but the soundcard that is built into your motherboard probably cheapened out on the DAC’s (not surprisingly, since most people don’t listen to digital audio above CD quality).
Dog-like hearing?
Several issues here, I feel. Does everybody (especially in their latter years) have hearing like a dog? The top end drops off with age and some older folk may not be able to tell the difference between bit-rates, different formats, etc. I’m 62 and (thankfully) can still just about hear the difference between a high and low rate mp3. Also, I don’t think it’s the format that the music is in that matters so much (higher bit-rates preferred, obviously), as the actual production philosophy and whether it contains hideous amounts of audio compression and processing at the production stage. This is one of my pet hates (especially on radio broadcasts). If a current so-called re-mastered re-issue is loaded into Audacity, it looks considerably flatter than the original issue. Just a load of tweaking applied to make it sound “louder” with the result that it sounds worse. Early CD producers seemed to throw everything in the pot “just because they could”. A well-produced digital recording is fine. Just ban the processing!
I Think Neil Young's Ideas Have Possibilities ....
I read an article about Neil Young’s idea for audio. I came away with a different thought, he wants to create a format that allows you to hear what the artist intended. I get a sense of some sort of spatial audio, and allowing it to be customized to sound like what the singer heard, guitar player heard, etc.
If this can be combined high resolution audio (24 bit or higher) .. I think this has very good possibilities.
Problems
First I’d like to make a couple observations:
Most MP3s suck!
We already have vastly superior formats to the CD and they have gone no where.
SACD and DD-A haven’t taken off because of two problems:
The old chicken and egg. There isn’t enough good material because there aren’t affordable players; there aren’t affordable players because there isn’t enough good material. In part because a lot of the stuff if just reformatted from material recorded at 44kHz and 14 bits.
Secondly most listening is done in unfaorable environments: subways, cars?. Even most home theater systems don’t provide a good enough environment to allow the best DVA-A to sound any better than a CD.
As a long time audio editor and ex studio owner I would make one other observation; most music genres are crap and would not sound any better live under ideal conditions than on a low grade MP3. Classical and some world/folk are the exceptions.
So where is the market?
Digital versus analog
This is a bunch of BS. Digital recorded at 16 bits and 44k is indistinguishable from live for most material (hammer dulcimer, pipe organ and orchestra being the exceptions). The guys who love analog are liking distortion provided by tape or vinyl not clear sound. If you like that distortion OK but you are not listening to the real thing. Yes the early digital sucked. Most A/D only really did 12 bits of resolution and until the millennium most A/D was 14 bits. 16 bits resolution can be differentiated from 24 bits on some difficult material. So 24 bits is preferable. Today there are no true 24 bit A/D (well except for for lab gadgets cooled to liquid nitrogen temperatures) but 22 bits sounds better than 16.
Re: Digital versus analog
Emphasis mine. This is just plain silly. First of all, no, recording at 16 bits produces a noticeably different track. I’ve never met a single audio professional who has said otherwise. The noisefloor for 16 bit recording is ridiculously high. Clarity, depth, headroom, and focus of sound suffers.
Second, even if you meant listening to a 16 bit recording (i.e. a CD), there is zero truth to the claim that a CD ever actually sounds like a live band. The reflections from the venue walls alone are enough to quash that notion. A live recording may convey the emotion and artistic intent of the live show, but it should be patently obvious that two speakers cannot possibly sound like 3-4 amps, a full drum kit, and a vocalist. Or hundreds of people playing their own instruments in a properly built concert hall.
This is typical tech silliness. Computer engineers make asinine statements like “a recorded hammer strike sounds like a live hammer strike, therefore a recording of a full 5 piece band (or 120 piece orchestra!) sounds just like being there in person.” That’s total hogwash. This is the kind of nonsense that gave us improperly calibrated digital equipment so that for the rest of time young engineers will have to be taught that -18db in digital needs to be “mentally” calibrated to 0dbVU, when it should have been calibrated thusly to begin with. But did anyone actually bother to reference a professional audio engineer when designing these interfaces? Of course not! Those engineers are just big bad audiophiles! Also, they’re gatekeepers! And monopolists! Waaah!
No. The guys who love analog love it because TAPE WAS DESIGNED FROM THE GROUND UP TO CAPTURE SOUND. The old mics that sound so sweet are designed for delivering sounds to tape. Vinyl was designed specifically to reproduce the sound captured by that tape. Etc, etc, you get the point.
Digital recording is a simple byproduct of computer science. It’s an afterthought. And it’s lacking in many ways beyond simple sound reproduction. For instance, someone mentioned earlier that it’s ‘not at all odd to have 200 tracks.’ Hell yes it is. Even a full orchestra isn’t recorded with so many tracks (that would be one mic for every player and then some. Much more “and then some”). It’s madness. I can only imagine what kind of mess such a steaming pile of tracks would sound like. Pink Floyd didn’t even approach that amount. But once again, this idea that a person can make something coherent out of 200 individual performances/tracks is a byproduct of digital production. Ever wonder why modern music sounds ‘overproduced’? There’s your answer; people think that layering a vocal track 30 to 40 times is ‘normal’ and, even more confounding, they believe it’s somehow more ‘realistic.’
Etta James did her vocals in one take. It was the limitations of analog recording that forced such amazing performances. Katy Perry does her vocals in hundreds of takes, because digital storage is much cheaper than tape, and the engineer layers 6 or 7 overdubs, since tracks are infinite and you don’t need to leave some open for bussing or drum kits.
Which method would you say has most successfully captured performances which are timeless?
Whatever Neil Young is doing
~ Is alright with me. Go Neil!!
digital vs. analog
some may not be able to tell the difference in blind tests using one recording back to back, but I could tell the difference, using analog tape at 3.75 IPS against a CD, back to back simultaneously, using the tape/source switch on my home rack system. The recording was “The Rolling Stones- Get Your Ya-Ya’s Out” live. The tape sounded much better, deeper, more little cymbal-high end-crowd sounds, better separation. No contest the tape sounded better.
Where anyone will be able to tell, is how LONG you can listen to CD vs. analog tape. You can listen to analog tape all day 8 am to 10 PM and not tire of it, your ears will not fatigue. If you listen to CD that long, you’ll have to shut it off after about 2 hours, it will just aggravate your ears.
ask me how I know, I grew up on vinyl and tape in the 1960-70’s, and when I bought my first new CD player in 1986, it sounded worse, not better. The only good thing was no background noise.
but in real life, live music, or studio music, HAS back ground noise. It’s never totally quiet. So why are they making it this un-natural and quiet ?
if they want better sound, increase the tape width and speed of the reel to reel deck.
Author You Need New Ears
After reading the closing of your article I knew for sure your simply an underprivileged consumer of McDonald’s like products of sound.
While I enjoy the portability of listening to music on my iPhone and the music I play though my computer at work there is nothing finer than listening to music off a record and the alternative CD from a decent stereo system.
Our ears work in amazing ways as equalizers. They’ll adjust and focus making many sounds seem clearer and filtering out some frequencies.
Simple home AB tests prove that when you take a recording (preferably a recording of hi fidelity) a vinyl record sounds superior. Larger, more dynamic and fuller.
While true at some point in human history Americans were pouring money into large stereo systems to experience musical performances today we generally pour monies into compact computer devices like phones, pads and desktops; generally speaking to enjoy a low quality youTube video.
Times are great and times are strange and the times well, they are in a flux. Once the dust settles and portable devices / computers become not as changing as they are the focus may return to the quality of sound and image.
Look at both digital audio and video. Neither come close to their analogue counterparts. Although many may argue…. “The Alexa camera…. Blah blah blah.” I argue that’s it’s simply a vision of tomorrows cell phone camera and nothing more as was the RED, Canon D5, 24p and so on.
For audio I experience just the opposite. While recording with a Pro Tools system I can record some rather wonderful instrumentation. Ill argue that it is not Pro Tools or my super Mac (which I love) that creates the sound. It’s everything that leads up to the computer. MIc Pre’s, ADC’s, instrument quality and the musician. An experienced engineer helps too. For audio’s failure with the playback accoutrement. Fitting a huge recording with multiple instruments through tiny speakers just plain old sucks. Is that why simple hip hop recordings have been dominating????
Think about? How many REAL musical instruments go into your typical hip hop recording? Generally vocals are the only organic instrument in hip hop. And for general pop music the tracks are so compressed I find they lack dynamics. And that is what a compressors job is to do, squish the dynamics.
To this day there is no better recording than Dark Side of the Moon. AB that recording.
About 10 years ago I spent several days listening to tons of records in my studio and after several stacks of LP’s came to the conclusion that there was nothing that came close to Dark Side of the Moon.
Since day 1 my ears have rejected every digital audio recording I’ve ever heard. I cannot accept the sound of it… sounds horrible to me.