I’ve watched Neil Young’s Pono project with a jaundiced eye. When it comes to audio quality, I’m all for it. I am not a voracious purchaser of downloads for many, disparate reasons. First and foremost, for the inconvenience of having a file to play for a start! My ecosystem is built around the compact disc, which offers me the perfect harmonic balance of quality and convenience. It allows me to listen to music without tape hiss or surface noise, with 96 dB of dynamic range. It allows me to listen to music without the time or expense of repurposing one format to another. It saves me hours of building and maintaining playlists and synching them between another device.
Another reason why I don’t care too much for downloads is that said files must be backed up for the rest of my life, at cost of time and expense. They are ephemeral, and the most fragile of music formats. If all of my music were in FLAC format I can’t begin to estimate how many TB of storage that would be required, but it would be very costly. I barely have the money to buy music, much less pain what would amount to exorbitant maintenance fees… just to listen to what I already own!
Finally, the fact of the matter is, that 95% of all downloads are in compromised, compressed file formats. First 128 kbps files predominated a decade ago, but now that we are living in the broadband era, 320 kbps is more common. This still dispenses with more than 75% of the data density that a compact disc offers the listener. That’s so uncool. Herein, is where the Pono system actually addresses a lack in what the marketplace is offering. Neil Young wants to sell you soul… for a commensurate price and a first class pass into the safety of his walled garden. Let’s view this helpful graph from Neil’s Pono website as it clearly shows the ways in which most of what are sold as downloads are clearly wanting.
To the left, you can see the vast majority of file formats [green, purple] are but slivers of what the ancient compact disc can offer your ears. The compact disc was designed by Philips and Sony in 1980, and the sampling rate of 44.1 kHz was specifically designed with the Nyquist Theorem in mind. This theorem states that in order to accurately reproduce any given frequency, we need a sample rate of at least twice that frequency. The theoretical range of human hearing is 20 Hz – 20,000 Hz. Thus we need a sampling rate of 40,000 Hz (or 40 KHz) to capture all frequencies that humans can hear. Due to the inclusion of a necessary part of the conversion system called an anti-aliasing filter, which is far too complex to get into here, we need to bump the sample rate up to 44.1 KHz. The CD has two channels of 44.1 KHz audio, which (because of the “extra” 4.1 Hz sample rate) reliably reproduces the entire range of human hearing, and just a bit more above the high end. The 16-bit depth was designed to be the bleeding edge of what semiconductor circuits and computers were capable of at the time, and since each bit of resolution gives us 6.01 dB of dynamic range, it is these 16 bits of data per sample that give us the 96dB of dynamic range that the CD offers (ok, 96.16 dB, technically). These parameters and many more collectively dictate the contents of what audio engineers call the Red Book Standard.
I have experimented with high-resolution audio files in my stabs at remastering from vinyl. My DAW software is capable of resolutions of up to 24-bit/96 KHz, which corresponds to what the typical high-res audio of DVD-Audio is capable of delivering. Keep in mind that I am mastering from vinyl, which has a headroom of 70 dB at the outer edge in mint vinyl, and that these high resolution formats more than double that dynamic range (24 bits x 6.01 dB per bit = just over 144 dB of dynamic range).
Bearing in mind that I have treated my ears as my valuable friends and have worn earplugs at rock concerts for most of my concert-going life, I found that as a middle aged man [a decade ago] who has coddled his ears, that bit depth makes a difference that I can hear more than sampling rate. I tried 24-bit/44.1 KHz, as well as 24-bit/96 KHz from vinyl sources, and my ears barely perceive an improvement in sound at 24-bit/44.1 KHz versus the CD rate of 16-bit/44.1 KHz, whereas at 24-bit/96 KHz I could hear no commensurate improvement in sound at all in spite of files that were 217% bigger. So when I remaster vinyl, I make my raw files at 24-bit/44.1 KHz and denoise them by hand before downsampling and bit truncating to 16-bit/44.1 KHz for eventual burning to gold archival CD-R and years of listening pleasure. Sure, 24 bits sounds somewhat better, but until I migrate to a media server [not bloody likely, see above] I keep the 24-bit files archived when I complete a project and listen to the CDs. It is also important to note that due to a factor called quantization error, each process applied during the restoration process has the potential to degrade the audio, even while trying to improve it. Working at the highest possible bit depth throughout the production process significantly mitigates quantization error and the resulting audible distortion, even if the final product will be 16-bit, or even MP3.
Mr. Young says that you can hear music’s soul when the FLAC files you are listening to are not shackled to outmoded 16-bit/44.1 KHz files. What Mr. Young is proposing is a music playing device that can play files that are vastly larger than 16-bit/44.1 KHz. The Pono store will feature files that are 24-bit/192 KHz! That means that the sound is sampled 192,000 times per second, and each one of those slices will be 24 bits of information… as opposed to CDs which are sampled only 44,100 times per second in 16 bit chunks. So each sample will be 1.5 times the amount of data compared to a CD, with 442% more samples per second. Pono files will take up over six times the amount of data that a CD holds… or about 24 times the amount of data in a 320 MP3 or AAC file!!!!! Is you hair on fire yet? That sounds great… except for one thing. Human hearing does not have a frequency response of 20-96,000 Hz, which is what a 24-bit/192 KHz file is capable of reproducing. Even if our ears could hear frequencies that high (some animals can), the microphones used in the studio can’t capture those frequencies, so they’re not even making it to the original recording medium (analog or digital, it makes no difference).
At this point, a most cogent argument can be made that Neil Young is young… in name only. He’s 68 years old as of today. He’s been around the block quite a few times. The figure above, which was almost liberated from Pono labs, shows the spread of Neil Young fans across the age spectrum from 15-85 years of age. Not surprisingly, the bulk of Neil’s fans to possibly follow his lead in this new venture are for the most part in the 45-75 years of age spectrum, or post-middle age. This is important, because most hi-fi enthusiasts who would have the $400 laying around for a player that holds 128 GB of music at up to 24-bit/192 KHz, and would then be able to afford the mooted $24.95 cost of an album on download, would simply have to be in the, er, prime of life. Hi-fi is not typically a young man’s game. And therein lies the achilles heel of this new platform.
If one is going to deliver six times the amount of data that the human ear is capable of receiving, the ears that will most readily be appreciative of this data fall on those younger than 15 years of age, where the full spectrum of human hearing can resolve sound across at the full 20Hz to 20KHz frequency spectrum. The average human experiences high frequency hearing loss with age, generally beginning at early adulthood and continuing until death, with middle aged males experiencing the most rapid loss of ability to hear about 10 KHz (this figure varies widely). That’s nothing to sneeze at, but when the life of Neil Young is examined, as the above graphic not quite stolen from Pono labs reveals, it tells a dramatically different story.
For Neil Young has been playing amplified electric guitar since the age of 15 and his time spent in bands like Buffalo Springfield, Crosby, Stills, Nash & Young, not to mention the incredibly loud Crazy Horse, have meant that Neil has spent over 55 years onstage standing in front of Marshall Stacks® that have wreaked havoc on his hearing ability. Sure, there have been spells of acoustic folkie playing, but those have been battered into splinters by the years he’s spent rocking in the free world. His decision to tour relatively late in life with Sonic Youth opening up for him probably killed off what little hearing he might have had in 1991 and was, in retrospect, a very bad idea for his ears.
And yet this man, who has complained about the specs of the compact disc for as long as journalists have given him the inches to do so, hardly seems like he might be capable of hearing a gramophone record, much less exotic high-res digital formats he’s hoping to sell to the unwashed masses. Garden variety digital music players usually top out with the ability to play 24-bit/44.1 KHz files, which by no coincidence, is also the sampling resolution/rate that I have found that I can perceive the slight difference in quality as a post-middle aged man who had coddled his ears. If you want to pay many times the going rate for the horror of a portable digital music player that is horribly bulky to boot [we’ll reserve judgement for the device’s UI until it reaches the marketplace, but even now, it’s pocket resistant size provokes astonishment], then be my guest.
– 30 –
p.s. A tip of the hat to regular commenter JT, who makes his living engineering sound, for the crisp rewrite [above] of my sloppy, hacked out post from yesterday. He made sure that my coloring stayed within the outlines this time. I understand his pain. When I see civilians given graphic design software, I also get very picky… so turnabout is fair play.