NFT Bone
April 27, 2022 1:22 AM Subscribe
“An Ionic Original is the pinnacle of recorded sound,” Burnett said in a statement. “It is archival quality. It is future proof. It is one of one. Not only is an Ionic Original the equivalent of a painting, it is a painting. It is lacquer painted onto an aluminum disc, with a spiral etched into it by music. This painting, however, has the additional quality of containing that music, which can be heard by putting a stylus into the spiral and spinning it.” T Bone Burnett Debuts New Audio Format That’s “The Pinnacle Of Recorded Sound” With Bob Dylan Re-Recordings
Has T-Bone Burnett Invented the Indestructible Lacquer?
T Bone Burnett Reveals New Analog Reproduction Technology
Bob Dylan and T Bone Burnett Create 'Ionic Originals' Debuting New Technology
Has T-Bone Burnett Invented the Indestructible Lacquer?
T Bone Burnett Reveals New Analog Reproduction Technology
Bob Dylan and T Bone Burnett Create 'Ionic Originals' Debuting New Technology
Er. That's looks suspiciously like it's just a dubplate? Not exactly "new technology", even if someone has invented a way of making them last a bit longer with some fancy plastic, sorry, "lacquer".
But as with most things marketed to audiophiles it is an efficient means of parting idiots from their money, so the NFT parallel is apt I suppose.
posted by tomsk at 2:01 AM on April 27, 2022 [14 favorites]
But as with most things marketed to audiophiles it is an efficient means of parting idiots from their money, so the NFT parallel is apt I suppose.
posted by tomsk at 2:01 AM on April 27, 2022 [14 favorites]
I bet it still sounds like crap till you buy the gold power cords.
posted by zompist at 3:00 AM on April 27, 2022 [44 favorites]
posted by zompist at 3:00 AM on April 27, 2022 [44 favorites]
came here to say something snide - but the duck has it, fur-lined twaddle indeed
posted by mbo at 3:27 AM on April 27, 2022 [3 favorites]
posted by mbo at 3:27 AM on April 27, 2022 [3 favorites]
five-star, fur-lined, ocean-going nonsense
now that is something to aspire to.
The second link, the Analogplanet one, leads down a brief but fantastic rabbit-hole that swings by The Supersense where you can buy a 'lacquer' made from 'the original tapes' etc etc etc... Supersense also make 'polaroid' film... which I think is a great / possibly bad / idea. Similarly with these 'super' recordings. Go for it! The technology is there, make something exceptional! Line it with fur and make it not merely waterproof, make it ocean-going. My god it makes my heart sing. It's like the guy who makes the off-road Porsches, or this tooled-leather belt my sister gave me once that has a Molson bottle-cap buckle. Really tasteful, maybe not necessary. True luxury. (Just yesterday I spent a heady twelve minutes contemplating a pair of woolen... pants (?) sweats (?)... that were knit like a fair-isles sweater: Like as though you were wearing a sweater for pants. It was hard not wasting money on them (but, well, chafing? and, hoo-hoo! expense!)
Reading about developments like this I can't help but think how good god damn! the future is bright.
Excelsior!
posted by From Bklyn at 3:49 AM on April 27, 2022 [4 favorites]
now that is something to aspire to.
The second link, the Analogplanet one, leads down a brief but fantastic rabbit-hole that swings by The Supersense where you can buy a 'lacquer' made from 'the original tapes' etc etc etc... Supersense also make 'polaroid' film... which I think is a great / possibly bad / idea. Similarly with these 'super' recordings. Go for it! The technology is there, make something exceptional! Line it with fur and make it not merely waterproof, make it ocean-going. My god it makes my heart sing. It's like the guy who makes the off-road Porsches, or this tooled-leather belt my sister gave me once that has a Molson bottle-cap buckle. Really tasteful, maybe not necessary. True luxury. (Just yesterday I spent a heady twelve minutes contemplating a pair of woolen... pants (?) sweats (?)... that were knit like a fair-isles sweater: Like as though you were wearing a sweater for pants. It was hard not wasting money on them (but, well, chafing? and, hoo-hoo! expense!)
Reading about developments like this I can't help but think how good god damn! the future is bright.
Excelsior!
posted by From Bklyn at 3:49 AM on April 27, 2022 [4 favorites]
"You listen to these modern records, they're atrocious, they have sound all over them. There's no definition of nothing, no vocal, no nothing.... CDs are small. There's no stature to it." - Bob Dylan
posted by Foosnark at 4:26 AM on April 27, 2022 [3 favorites]
posted by Foosnark at 4:26 AM on April 27, 2022 [3 favorites]
On further reflection, I am amused to be able to say every one of my releases has sold more copies than a Bob Dylan record.
posted by Foosnark at 4:56 AM on April 27, 2022 [6 favorites]
posted by Foosnark at 4:56 AM on April 27, 2022 [6 favorites]
Audiophiles being audiophile. There's always such a rockist slant. I mean, like if you're listening to Autechre, warmth is the last thing you want.
posted by treblekicker at 5:26 AM on April 27, 2022 [7 favorites]
posted by treblekicker at 5:26 AM on April 27, 2022 [7 favorites]
April Fools was 26 days ago.
Wait, he's serious? Oh my.
posted by tommasz at 5:48 AM on April 27, 2022 [3 favorites]
Wait, he's serious? Oh my.
posted by tommasz at 5:48 AM on April 27, 2022 [3 favorites]
You know what the main use of this is going to be, though:
Mark Knopfler solo album remasters. It's always Mark Knopfler solo album remasters.
posted by prismatic7 at 5:57 AM on April 27, 2022 [16 favorites]
Mark Knopfler solo album remasters. It's always Mark Knopfler solo album remasters.
posted by prismatic7 at 5:57 AM on April 27, 2022 [16 favorites]
Isn't lacquer what old 78s were made out of?
posted by Bee'sWing at 6:13 AM on April 27, 2022 [2 favorites]
posted by Bee'sWing at 6:13 AM on April 27, 2022 [2 favorites]
I'm so tired of this boring bullshit. I want an actual Bob Dylan record. I want Bob Dylan, naked, with a groove etched into his belly, permanently strapped to a giant turntable in case I want to listen to him.
posted by snofoam at 6:15 AM on April 27, 2022 [18 favorites]
posted by snofoam at 6:15 AM on April 27, 2022 [18 favorites]
I am a composer and I have nothing right now to say specifically about the audiophile claims here
but in general the more someone tries to equate any piece of music to a single physical or digital form that is the best, most purest, most one-of-a-kind, most definitive, etc. etc.
(as opposed to an ineffable moment in shared listening experience that can't be captured or reproduced, only lived)
and they claim that (conveniently) you can buy this intangible musical experience as an object, as a commodity, as a thing you can horde away after paying a lot of money
the more I say
fuck those people and fuck that thinking about music in general, because even if people recognize this as a grift
it's still got them thinking that music is a commodity, not a vital human activity, that only the most perfect and polished and recorded and packaged music counts
and so fuck all that
let's make weird hideous shit that no one in their right mind would pay for, let alone try to capture the experience on a lacquered disc
let's set a piano on fire etc. etc.
That is my thought for the day
posted by daisystomper at 6:19 AM on April 27, 2022 [39 favorites]
but in general the more someone tries to equate any piece of music to a single physical or digital form that is the best, most purest, most one-of-a-kind, most definitive, etc. etc.
(as opposed to an ineffable moment in shared listening experience that can't be captured or reproduced, only lived)
and they claim that (conveniently) you can buy this intangible musical experience as an object, as a commodity, as a thing you can horde away after paying a lot of money
the more I say
fuck those people and fuck that thinking about music in general, because even if people recognize this as a grift
it's still got them thinking that music is a commodity, not a vital human activity, that only the most perfect and polished and recorded and packaged music counts
and so fuck all that
let's make weird hideous shit that no one in their right mind would pay for, let alone try to capture the experience on a lacquered disc
let's set a piano on fire etc. etc.
That is my thought for the day
posted by daisystomper at 6:19 AM on April 27, 2022 [39 favorites]
gold power cords
Well of course but you also need a power conditioner for each of your components because standard household electricity creates barriers from you truly enjoying the sound of a musician scratching themselves as they play their instruments.
Mark Knopfler solo album remasters
Years ago, when I worked in media retail, my boss asked me to be nice to this rich tech guy who'd come in telling me that he was a big audiophile. So curious I started to chat with him about his (of course he was a man!) set up and what kind of music he liked to listen to. I forget the components he had but they were very expensive and impressive but when I pressed what he listened to he told - "Oh no I don't listen to music. I have several music demonstration discs that I will put on to enjoy." So all he listened to were the discs you'd use to calibrate your speakers with... I knew another guy who built all his own components and amps and all he listened to was CBC radio!
posted by Ashwagandha at 6:27 AM on April 27, 2022 [13 favorites]
Well of course but you also need a power conditioner for each of your components because standard household electricity creates barriers from you truly enjoying the sound of a musician scratching themselves as they play their instruments.
Mark Knopfler solo album remasters
Years ago, when I worked in media retail, my boss asked me to be nice to this rich tech guy who'd come in telling me that he was a big audiophile. So curious I started to chat with him about his (of course he was a man!) set up and what kind of music he liked to listen to. I forget the components he had but they were very expensive and impressive but when I pressed what he listened to he told - "Oh no I don't listen to music. I have several music demonstration discs that I will put on to enjoy." So all he listened to were the discs you'd use to calibrate your speakers with... I knew another guy who built all his own components and amps and all he listened to was CBC radio!
posted by Ashwagandha at 6:27 AM on April 27, 2022 [13 favorites]
Dewds i've got a bootleg audience recording of the Dead from the Philly Civic Arena on 420 '84 on a maxell metal cassette tape that's only been played about 10,000 times that is the actual pinnacle of recorded sound, but it's priceless.
posted by OHenryPacey at 7:02 AM on April 27, 2022 [6 favorites]
posted by OHenryPacey at 7:02 AM on April 27, 2022 [6 favorites]
this is five-star, fur-lined, ocean-going nonsense
Yep. Five-star, fur-lined, copper bottomed ocean-going nonsense. Fuxache.
The "stature" thing is something I could get behind, though. I do hanker nostalgically for huge gatefold album covers, with their big-enough art and their comprehensive-enough sleeve notes. CD packaging has always been completely miserable by comparison, even the weird US version where the boxes are twice the height they need to be.
posted by flabdablet at 7:28 AM on April 27, 2022 [3 favorites]
Yep. Five-star, fur-lined, copper bottomed ocean-going nonsense. Fuxache.
The "stature" thing is something I could get behind, though. I do hanker nostalgically for huge gatefold album covers, with their big-enough art and their comprehensive-enough sleeve notes. CD packaging has always been completely miserable by comparison, even the weird US version where the boxes are twice the height they need to be.
posted by flabdablet at 7:28 AM on April 27, 2022 [3 favorites]
So all he listened to were the discs you'd use to calibrate your speakers with...
You're modding more than eight; you're going to get wow on your top. You try to bring that down through your rumble filter, to your woofer, what do you get? Flutter on your bottom!
Soundtrack for the thread
posted by flabdablet at 7:34 AM on April 27, 2022 [4 favorites]
You're modding more than eight; you're going to get wow on your top. You try to bring that down through your rumble filter, to your woofer, what do you get? Flutter on your bottom!
Soundtrack for the thread
posted by flabdablet at 7:34 AM on April 27, 2022 [4 favorites]
Not that I have ever tried it, but I have a feeling that if you strap somebody—anybody—to a giant turntable, you're going to be getting an earful whether you've switched it on or not.
posted by Flexagon at 7:37 AM on April 27, 2022 [2 favorites]
posted by Flexagon at 7:37 AM on April 27, 2022 [2 favorites]
No, Mr. Bond. I expect vocal fry.
posted by flabdablet at 7:53 AM on April 27, 2022 [19 favorites]
posted by flabdablet at 7:53 AM on April 27, 2022 [19 favorites]
So did no one tell them about the RIAA EQ curve used in the phono preamp for almost every turntable built in the last 40 years or what?
And anyone who is an archivist and knows how recording technology works, isn't a snake-oil believing audiofool and buys one of these things - I know, really tiny microscopic Venn diagram segment, here - they're probably going to rip it to a digital file on the very first and possibly only play so they can listen to it without damaging it further.
posted by loquacious at 7:59 AM on April 27, 2022 [2 favorites]
And anyone who is an archivist and knows how recording technology works, isn't a snake-oil believing audiofool and buys one of these things - I know, really tiny microscopic Venn diagram segment, here - they're probably going to rip it to a digital file on the very first and possibly only play so they can listen to it without damaging it further.
posted by loquacious at 7:59 AM on April 27, 2022 [2 favorites]
let's make weird hideous shit that no one in their right mind would pay for, let alone try to capture the experience on a lacquered disc
I can't remember who it was, but I think it was an IDM artist on WARP records who released an album that was... sandpaper.
I've only seen it once back in the 90s.
And yes, we did try to play it on a very old bad needle we didn't care about. It sounded amazingly horrible. Even better it was different every time you tried to play it.
posted by loquacious at 8:02 AM on April 27, 2022 [9 favorites]
I can't remember who it was, but I think it was an IDM artist on WARP records who released an album that was... sandpaper.
I've only seen it once back in the 90s.
And yes, we did try to play it on a very old bad needle we didn't care about. It sounded amazingly horrible. Even better it was different every time you tried to play it.
posted by loquacious at 8:02 AM on April 27, 2022 [9 favorites]
Warp Records Are Reasonable People, u kno.
From Bklyn: Excelsior!
We're upgrading to ExcelsiXOR but can't yet afford ExcelsiNAND.
posted by k3ninho at 8:08 AM on April 27, 2022 [1 favorite]
From Bklyn: Excelsior!
We're upgrading to ExcelsiXOR but can't yet afford ExcelsiNAND.
posted by k3ninho at 8:08 AM on April 27, 2022 [1 favorite]
Come on Bob, just remix your tracks in Dolby Atmos with Dan the Automator and release a corresponding Dogfish Head beer like everyone else.
posted by credulous at 8:18 AM on April 27, 2022 [3 favorites]
posted by credulous at 8:18 AM on April 27, 2022 [3 favorites]
This is an oldie but a goodie, explaining why all the arguments that CDs sound "cold" or "grainy" or whatever is, to use the scientific term cited by the duck upthread, flapdoodle. This links to text, but that page links to a long video explainer/demonstration.
I can absolutely believe that there are differences in the signal path that produce a sound that some people like better. I've got some early CDs with questionable mixing that produced painful high notes (my guess is the recording engineers were mixing for vinyl that would mute the highs, or maybe they were just showing off that "CDs can make your ears bleed").
posted by adamrice at 8:19 AM on April 27, 2022 [2 favorites]
I can absolutely believe that there are differences in the signal path that produce a sound that some people like better. I've got some early CDs with questionable mixing that produced painful high notes (my guess is the recording engineers were mixing for vinyl that would mute the highs, or maybe they were just showing off that "CDs can make your ears bleed").
posted by adamrice at 8:19 AM on April 27, 2022 [2 favorites]
As much as I appreciate a good title pun, this doesn't seem to actually have anything to do with NFTs, thank goodness, although it's still deeply dumb, its creator's staggering credentials notwithstanding.
posted by Halloween Jack at 8:24 AM on April 27, 2022 [3 favorites]
posted by Halloween Jack at 8:24 AM on April 27, 2022 [3 favorites]
I propose that the artist should be a physical component in the medium. Each 13 inch (odd numbers for diameter favor both odd and even harmonics) lacquer disk will be .05 % of the ashes of the artist. For living artists, the disks will be fashioned with the dust in their bedrooms as this will be primarily dried skin flakes of the artist. When the disk has been gently lowered by electrostatic hydraulics on to the turntable, it will be Bob Dylan spinning around in your home.
posted by njohnson23 at 8:59 AM on April 27, 2022 [1 favorite]
posted by njohnson23 at 8:59 AM on April 27, 2022 [1 favorite]
Slow, respectful clap for the post title. Would slowly clap again.
posted by riverlife at 10:07 AM on April 27, 2022 [2 favorites]
posted by riverlife at 10:07 AM on April 27, 2022 [2 favorites]
I'm no expert. But in my listening experience 99.9% of the "inferior sound" comes from the mixing and compressing of all the sounds... the production of most music these is what makes it so much of it sound like garbage. The trend in music these days is ultra compressed, mixed to a click track and auto tuned to make everything "perfect" and it comes out sounding soulless and "digital" sounding.
There's been a recent resurgence of acoustic and electric guitar in pop music which I think is great. But as long as they compress the sound of every instrument and vocal track to cater to people listening with earbuds, it's just never going to sound like human beings performing music, IMO. I miss dynamic range.
But of course, here I am with no $600 rosewood cable holders to keep my speaker cables off the floor, so what do I know?
posted by SoberHighland at 10:09 AM on April 27, 2022 [2 favorites]
There's been a recent resurgence of acoustic and electric guitar in pop music which I think is great. But as long as they compress the sound of every instrument and vocal track to cater to people listening with earbuds, it's just never going to sound like human beings performing music, IMO. I miss dynamic range.
But of course, here I am with no $600 rosewood cable holders to keep my speaker cables off the floor, so what do I know?
posted by SoberHighland at 10:09 AM on April 27, 2022 [2 favorites]
OK, that's the most poorly written comment I've ever made here. Apologies. Need some NFT coffee or something.
posted by SoberHighland at 10:15 AM on April 27, 2022 [2 favorites]
posted by SoberHighland at 10:15 AM on April 27, 2022 [2 favorites]
I know these people know this, but how do audiophiles think music is recorded? Especially today? Good luck keeping things analog. Even back in ye olden days that so called “warmth” was just overdriving compressors or shit ribbon mics. I listen to a lot of 78’s and wax cylinders (Irish music nerd) and there’s your warmth and purity. Is that what you want? Yeah, let’s go back to wax. I hate audiophiles.
And here’s my Mark Knoefler story. About decade ago my wife and I got into an Uber in LA and the drivers first words to us upon pulling onto the road was “are you familiar with the solo works of Mark Knoefler?” I was quite the ride. We still not entirely sure it wasn’t Mark Knoefler.
posted by misterpatrick at 11:18 AM on April 27, 2022 [7 favorites]
And here’s my Mark Knoefler story. About decade ago my wife and I got into an Uber in LA and the drivers first words to us upon pulling onto the road was “are you familiar with the solo works of Mark Knoefler?” I was quite the ride. We still not entirely sure it wasn’t Mark Knoefler.
posted by misterpatrick at 11:18 AM on April 27, 2022 [7 favorites]
Previously: Bob Dylan sells entire songwriting catalog in nine-figure deal.
posted by box at 11:28 AM on April 27, 2022 [1 favorite]
posted by box at 11:28 AM on April 27, 2022 [1 favorite]
I know these people know this, but how do audiophiles think music is recorded?
There is, of course, still a lot of "analog gear, preferably vintage, is best" on the production side. Which, I don't know, I'm never even going to see a $50k console, and producers can do whatever they like as long as it sounds good at the end. But most of it doesn't pass the smell test. In any event, when cassettes started making a comeback I gave up on ever understanding the world.
posted by sinfony at 11:56 AM on April 27, 2022 [1 favorite]
There is, of course, still a lot of "analog gear, preferably vintage, is best" on the production side. Which, I don't know, I'm never even going to see a $50k console, and producers can do whatever they like as long as it sounds good at the end. But most of it doesn't pass the smell test. In any event, when cassettes started making a comeback I gave up on ever understanding the world.
posted by sinfony at 11:56 AM on April 27, 2022 [1 favorite]
Even back in ye olden days that so called “warmth” was just overdriving compressors or shit ribbon mics.
If anything “warm” means “not a lot of high frequencies.”
posted by atoxyl at 11:58 AM on April 27, 2022 [4 favorites]
If anything “warm” means “not a lot of high frequencies.”
posted by atoxyl at 11:58 AM on April 27, 2022 [4 favorites]
But in my listening experience 99.9% of the "inferior sound" comes from the mixing and compressing of all the sounds... the production of most music these is what makes it so much of it sound like garbage.
TBF, Dylan has objected vociferously to the "loudness war" in general, and does have good basis to argue (you can compare dynamic range numbers on this site) that the CD masters of most of his self-produced 21st century albums have markedly lower dynamic range than the vinyl versions of the same mixes, and do not reflect his intention. I can sort of understand how that kind of frustrating experience might drive someone to want to do something like this, to make a statement about the definitive versions of works. On the other hand, it does seem like he's getting his way about masters these days. In particular, I'd note that the streaming/download version of Rough and Rowdy Ways sounds absolutely fine to me, and measurements suggest it's the same as the vinyl version, so I'm not convinced this is a particularly timely or necessary response at this point.
posted by howfar at 12:31 PM on April 27, 2022 [4 favorites]
TBF, Dylan has objected vociferously to the "loudness war" in general, and does have good basis to argue (you can compare dynamic range numbers on this site) that the CD masters of most of his self-produced 21st century albums have markedly lower dynamic range than the vinyl versions of the same mixes, and do not reflect his intention. I can sort of understand how that kind of frustrating experience might drive someone to want to do something like this, to make a statement about the definitive versions of works. On the other hand, it does seem like he's getting his way about masters these days. In particular, I'd note that the streaming/download version of Rough and Rowdy Ways sounds absolutely fine to me, and measurements suggest it's the same as the vinyl version, so I'm not convinced this is a particularly timely or necessary response at this point.
posted by howfar at 12:31 PM on April 27, 2022 [4 favorites]
For sale: 24 track analog mixing deck, 2 tracks or more don't work but that only increases the purity of the effort. Musicians should have obstacles; learn to play around the difficulties, because your sincerity will show up generations later on some other format for some other endless corporate entity, or Bob Dylan.
$ 45b or BO (no twitter accepted)
posted by winesong at 1:33 PM on April 27, 2022 [1 favorite]
$ 45b or BO (no twitter accepted)
posted by winesong at 1:33 PM on April 27, 2022 [1 favorite]
I always think of this [Flanders and Swann A Song of Reproduction Played On A Garrard 301 Turntable]
The 301 is a nice old turntable as idler wheel drive designs go. But anybody who honestly believes that they can achieve better than digital sound quality by spending 25 grand on a well built retro turntable that they then operate without a dust cover is somebody whose opinions on audio gear I simply cannot bring myself to respect.
posted by flabdablet at 7:07 PM on April 27, 2022 [2 favorites]
The 301 is a nice old turntable as idler wheel drive designs go. But anybody who honestly believes that they can achieve better than digital sound quality by spending 25 grand on a well built retro turntable that they then operate without a dust cover is somebody whose opinions on audio gear I simply cannot bring myself to respect.
posted by flabdablet at 7:07 PM on April 27, 2022 [2 favorites]
I know these people know this, but how do audiophiles think music is recorded? Especially today? Good luck keeping things analog. Even back in ye olden days that so called “warmth” was just overdriving compressors or shit ribbon mics. I listen to a lot of 78’s and wax cylinders (Irish music nerd) and there’s your warmth and purity. Is that what you want? Yeah, let’s go back to wax. I hate audiophiles.
Yeah, I've hashed this out with die hard analog audiophiles dozens and dozens of times now - and to be honest I fucking love ranting about this topic because it makes me feel smart and it's taken me decades to learn a lot of this stuff and debunking audiophile woo.
I'm discovering there's possibly even more disinformation, myths or magical thinking going around today then there was during the absolute peak of the pure analog studio album.
And lets get this part out of the way first - one of the reasons why this myth persists is the extreme analog quality of recordings in this absolute theoretical peak of analog recording - the era in the 70s of the studio album.
Those recordings do indeed sound good, but it's because they had huge, nearly unlimited budgets to throw at the problem. They could afford the endless reels of very high quality tape. The analog mixing consoles were the best they would ever be. They could afford to do things like slave together multiple 12 to 24 track tape decks to avoid "bouncing down" stems or tracks to a single track to add and mix more tracks. They had analog mixing desks with huge amounts of channels. And it was a sweet spot of newly developed analog, tube based or solid state transistorized amps, pre-amps, audio effects, microphone technology and so much more.
The main factors here are less about audio technology and more about nearly unlimited amounts of time and money to throw at recording and producing music.
Let's also address and put aside the problems with the loudness wars and the whole concept of audio compression - and I mean dynamic range compression, not data compression codecs like MP3 or MPEG stuff. Audio compression is a complicated topic but simplifed it means reducing the dynamic range so quieter audio signals are louder and louder signals are - sometimes - quieter. Like a limiter.
In fact a dynamic range compressor is effectively a kind of limiter, see the wikipedia link above:
Because of this one of the myths that refuses to die especially among hard core magical thinking audiophiles is that somehow there's magical extra dynamic range from infrabass to ultrasonics being encoded in an all analog recording path and captured and reproduced on vinyl.
Which is absolutely not true at all. There are absolutely no magical inaudible tones making it to any kind of mass produced vinyl record, especially not the recordings that analog audiophiles hold up as gold standards.
For starters, the high and low pass filter has been a standard part of the recording and mastering tool chain since, I don't know, at least the 60s or even earlier, and it's used specifically to filter out any of those frequencies on purpose.
It's not just an aesthetic choice, it's an essential part of good mastering and recording for audio reproduction, and I can totally dive into the nuts and bolts of why mastering and recording techniques like high/low passes are used.
Part of this myth and misconception is that analog recording media have "infinite" resolution compared to digital sampling which is also absolutely not true at all.
Every single recording, playback or reproduction technology has inherent limitations whether it's analog or digital.
For analog tape the dynamic range and frequency bandwidth is determined by a number of factors and limitations. One of these is the grain size of the magnetic coating. Another - and related to grain size - is tape speed. There's also the factor of tape width, too. But there's another major limitation and that is tape bias.
And another limitation that's co-related to all of the above is the actual physical size of the coils in the tape head - roughly speaking, the larger your coils, the worse it does at high frequency recording. The smaller or finer the coils are, the better they do at high frequency recording. This is head gap.
The tape speed, grain size and size of the coils of the recording head all inform the total frequency response and dynamic range. There's also the issues of tape speed consistency, called wow and flutter.
This is why metal oxide tapes were the absolute pinnacle of magnetic tape recording. They had extremely fine grains and magnetic domain sizes. As a reference point the peak of audiophile quality for magnetic tape in the 1/8" compact cassette format happened in the late 80s through the mid 90s with metal oxide tapes, notably on Nakamichi decks combined with lossy noise reduction schema like Dolby Noise Reduction.
Combine all of the above and there's a general frequency response and dynamic range gamut (just like color reproduction on analog film or digital imaging) that is clearly defined by all of these factors.
Ok, well, what about skipping tape and recording straight to a lathe and lacquer? What about "extra" information in the infrasonic or ultrasonic spectrum there?
Well, in theory? Sure. Sure, maybe you could capture a much wider dynamic range and frequency response with a theoretical unrestricted lathe head cutting on the finest lacquer coating ever invented.
In reality and in practice for any professionally recorded, mastered and cut album this is also definitely not true.
Trying to record ultrasonic tones on a record cutting lathe head would probably break it and make it explode.
Also, good luck finding any phonograph needle or preamp capable of reproducing any of these ethereal ultrasonics, especially if an RIAA EQ compatible phono pre-amp is involved.
You also can't record super deep bass without cutting extremely wide grooves and drastically reduce the available run time of the groove on a 12" record. And even if you did cut a record with huge fat grooves and wobbles it would make all known consumer record player needles and tone arms jump right out of the groove and your recording time at 33.3 or 45 RPM would be measured in, oh, seconds or maybe even up to a minute or two instead of dozens of minutes per side.
This is the whole reason why they invented the RIAA EQ curve and phono pre-amp system. This is how LP - Long Play - and the earlier EP - Extended Play - records actually work.
It's a form of analog data compression that reduces the bass end in the cut master lacquer so they can fit more music time on a 12" record, and it's reconstituted by applying the curve in reverse via the phono pre-amp to reconstitute most of the frequency response and dynamic range.
There is some flexibility here to cut a record with reduced run time to add extra bass to a recording and this was indeed used in vinyl for bass music, dance music and so on, trading some of the available run time for more bass. But this is still within the parameters of the RIAA EQ and phono preamp, and it was a subtle and black art.
See also sonic artifact.
But wait, there's even more limitations!
Most microphones, guitar pickups and other forms of transducers to record with can't even sense ultrasonics in particular anyway. They also have limited frequency response and dynamic range! They also all have their own "tone" or highs and lows in the gamut of frequency response, and will distort or become lossy if trying to record anything that's too quiet to move them or too loud to record accurately.
Wait, we're not done yet! Let's talk about audio reproduction. And by that I mean speakers and amplifiers.
These also have their own limitations and gamut of frequency response and dynamic range. Too quiet and the coils or transducers of a speaker don't move. Too loud and they distort or clip or otherwise degrade.
The current, practicable theoretical maximum frequency response of even the most expensive, unobtanium grade sound system of any kind whether it's pro audio for an arena sized concert or the most expensive magical thinking audiophile grade sound system is about 5-10 hz to about 50 khz.
For the infrabass and super low end we're talking super exotic subwoofers that use servos instead of voice coils or even rotating paddles like a fan to get that low. The kind of thing where you're tuning your whole house or listening room to become a subwoofer and using resonance and constructive feedback to even attempt 5-10 hz. Yes, this is possible. No, you can't hear it. Yes, you could possibly feel it, but not hear it.
At the high end of the frequency response you can get ultratweeters and ribbon tweeters capable of creating 50 khz tones, if not even higher tones. And they're not particularly exotic or expensive, you can go to your nearest hardware store and buy ultrasonic pest repellers. Or go to Mouser or Digikey and buy ultrasonic transducers used for things like range finding or proximity sensing. Or you can even buy super high frequency ultrasonic transducers used for industrial processes like ultrasonic welding of plastics.
No, most people can't hear this, either, and if they're "hearing" anything above about 20-22khz even with the best ears they're actually hearing second or third order harmonics and resonance of something, anything in the environment sympathetically vibrating and acoustically coupling with those inaudible tones.
In reality and practice?
Most people don't own and/or have never even heard any kind of audible speaker that can do anything lower than about 30-50 hz or above about 18-20 kilohertz. 40-50 hz is REALLY good for a home subwoofer, and you need about a 12" to 18" driver and about 1000 watts of power in a nicely tuned and ported speaker enclosure to get it or down to about 30 hz. You also need high powered amps for it. And the lower you go on bass and sub bass the more energy you need to move that much air to hear or feel it. You can throw thousands and thousands of watts or dollars at this problem to break past about 30-40 hz. Or alternative driver/transducer types like servo-drive speakers.
This is exactly why the audio engineering and mastering uses low/high pass filters, to eliminate unneeded frequencies outside of the range of human hearing because it makes everything within that range of human hearing sound cleaner and louder and takes less energy for amps and speakers to reproduce.
This is a known problem in professional audio reproduction even for something like a really massive speaker array for a very large electronic music festival, and this is why these kinds of professional speakers use things like digital signal processors and crossovers to eliminate any frequency information below about 30 hz because anything below that takes exponentially more and more electrical energy and power to move the speaker coils or transducers to try to recreate those super deep bass tones.
And paradoxically you can get cleaner, more articulated and responsive bass tones by eliminating the inaudible portions of it below human hearing levels, because that means more of the available amp power is dedicated to audible frequencies instead of trying to reproduce deep infrasonics so low that not only can you not hear it but you could barely feel it because it's just too slow for human perception.
On the high end of the frequency response being able to hit about 20 khz is also really good and needs really good tweeters or ultratweeters. The power requirements for trying to reproduce this are much less troublesome because you're not trying to move as much air mass, but in practice most people are working with speakers that peak out at about 12-18 khz before being useless or ineffective.
But wait, there's even more limitations!
The final ingredient into this sonic puzzle is the limitations of human hearing itself. Most people are lucky if they can hear from about 40-50 hz to about 16-18 khz, if not much less. There definitely are people who can hear outside of this range, especially above 18khz with upper extremes pushing 20-24 khz in the most rare, and unique edge cases.
I've had my hearing tested, and despite my long history with super loud and weird music I've taken pretty good care of my hearing and have been using ear plugs religiously for most of my adult life, I have really good ears and last time I checked I can hear tones from about 20 hz to about 20 khz, which is not normal or average.
But even then, most of our music is well within about 50-60hz to about 12-18 khz. This is especially true if we're talking about acoustic or traditional instruments, whether it is a full symphonic orchestra or an acoustic guitar, a stand up bass or any vocalist. Whether the vocalist is Tom Waits, Bob Dylan or Diamanda Galás.
Note: All of the above frequency ranges or values are estimates and I'm shooting from the hip based on my experiences and recollection of these ranges. Please don't @ me. They should be close enough for horseshoes and hand grenades.
So, let's talk about the most extreme end of audio reproduction, and we can discuss this in two parts - live acoustic instrument performances or analog (or digital!) synthesizers.
First, let's look at synthesizers with an entirely analog amplification path, tuned and open crossovers, no low pass or high pass, not a recording, played through exceedingly good speakers.
Yep, you sure can really push the gamut of frequency response and dynamic range and it's not at all expensive or even that difficult, especially with really good speakers with very wide frequency responses. You can noodle around with sine waves and tones right on down to 0 hz and up into the ultrasonic ranges. You can straight up light speakers on fire and melt them into slag with tones in the 0-30 hz range with enough amplitude.
I've personally seen speaker cones and coils melt or smoke or tear out their support baffles moving slow enough with deep excursions to count the cycles per second on your fingers with a stop watch. I've done it on purpose with a synthesizer or oscillator. It's fun.
You can blow the caulking and sealing right out of speakers doing this and walk the nails or screws holding the speaker box right out of the wood and break all of the things. You can do terrible, terrible things to speakers with little more than a single oscillator circuit down in the 5-30 hz range and a really strong amplifier.
There's still going to be limitations in frequency response and dynamic range in your gain and amplification path, whether it's a tube amp or solid state transistor. There's still going to be physical limitations to your amplifiers and speaker drivers.
But it's going to be much cleaner with a much wider range of frequency response or dynamic range of any recording media of any kind due to all of the limitations of recording technology. A pure analog synthesizer hooked up to really good speakers is something special. But it has limitations.
Second, let's look at live performances whether it's purely acoustic, whether it's a major symphony orchestra or a live acoustic folk band or electrically amplified rock band through tube amps and a direct connection and relation between the mics or pickups to the amps and speakers, drivers, coils and etc.
Yep, there's all kinds of magic that can happen there from subsonic bass to ultrasonics. The tone of strings vibrating on wood, the whisper or buzz of fingers on fretboards, the sound of a human breathing between vocalization, the natural reverberation of a space, the sympathetic vibrations of our physical world being energized and resonating with the tones of a live guitar or a piano or a vocalist.
Every grain of dust, every plank of wood, the pressure, temperature and humidity of the air, every acoustically reflective surface in the environment contributing to this magic of live music, the complex tones of all that live delay, reverberation and the magic of constructive and destructive waves of acoustic energy.
Yes, there is *magic* there and so many acoustic/psychoacoustic variables and the unmediated immediacy of live music that can vastly exceed and transcend the limitations of audio recording, duplication and reproduction that invokes all of the magic and complexity of real world physics that we know.
That part of the magic of music and sound is all very real and very difficult to quantify or measure whether it's an acoustic guitar or an analog synthesizer.
Can we accurately capture and record it? Perfectly? Hell no. And that's a beautiful and amazing thing.
All of that being said and you still want really good speakers and sound in your life?
Well, the good news is that you really don't need to spend hundreds of thousands of dollars on bespoke audiophile speakers the size of a refrigerator and monoblock amplifiers or solid gold speaker cables or any of that gold plated steaming load of fur-lined horseshit.
The people that spend that kind of money on those kinds of speakers are deluding themselves and drinking the snake oil flavored Kool Aid.
Well, with some caveats. There's some really exotic speakers and sound technology out there and that exoticness ranges from musical instrument grade woods to tube amplifiers to omnidirectional drivers and all kinds of black magic acoustics that are just as valid as the difference between a Stradivarius violin compared to a cheap Mendini.
But you can get to within 1% of those high end systems with much, much less money, with just a little homework. Dive into the the Budget Audiophile subreddit at www.reddit.com/r/BudgetAudiophile and check out their FAQs and guides.
There are incredible speakers out there today where you can put together a 5.1 home theater system for much less money than that fur-lined ocean-going utter nonsense.
You can pick up extremely capable speakers and amps for well under $1000 USD that get you within 10% or 1% of the best speakers and amps and audio equipment in the known world.
I've found audiophile grade speakers at thrift stores in the bookshelf speaker size range for like $5 per speaker.
If you ever see any KLH 45 model 3 way bookshelf speakers at a thrift store, buy them and get a cheap T-class digital amp for like $10-30 off of Amazon, like the famous Lepai 2020A+ and go with it. Pair it with a decent subwoofer like an 8-10" Audiosource box and you have a really nice, compact sound system going on a shoestring budget.
You can also get really amazing speakers and studio monitors from companies like Focal, KRK, Mackie, Yamaha, EV or even vintage Bose and so many more with prices ranging from 200-300 a pair on up to 2000-3000 a pair.
Want to splurge and jump straight into something that's exceedingly high quality and value for the price for a home theater or music system for something that not only competes with wacky unobtanium grade audiofool speakers?
Check out ADAM speakers. Yeah, they're rather pricey compared to something you could pick up at Costco or whatever, but those speakers are some of the best in the world without diving even deeper into pro audio PAs or audiofool snake oil nonsense. They have ultratweeters rated to 50 khz, like I was talking about above.
Those ADAMs are exactly the kind of speakers that pro audio engineers use for mixing blockbuster movies or albums. I have a friend with a pair of the AX7s with a matching sub and they are absolutely fucking incredible speakers, and I can compare them to speakers and sound systems I've heard at 10x or 50x the price.
They're so dynamic, responsive and forgiving that you can hook up an analog synth to them and make horrible duck fart noises through them and it's paradoxically pleasant to listen to, well, anything from the sublime and melodic to the horrible and dissonant. I've put some incredibly terrible, unpleasant noises through my friend's home studio setup and they don't even blink at it and reproduce it faithfully.
Doing a 5.1 home theater or even a full surround Atmos system with those ADAM speakers would be incredible and would blow your favorite high end Dolby/THX certified movie theater right out of the water if only because you could turn them down so they don't clip and distort like your local AMC megaplex.
Want something bigger and louder but relatively affordable? Pick up a pair of QSCs self powered speakers and a matching sub and you're half way to a very high quality dance/club rave sound systems. You can get a pair of QSC K10s to K12s for under 100-2000 or even less on sale or from Sweetwater or other pro audio purveyors and wake up the whole neighborhood with crystal clear action movie noises or the wubbiest of wubby dubstep noises.
There's a whole range of options ranging from well under 50-100 USD or even cheaper to well under 10,000 USD for a full surround theater system that would compete with if not destroy the whole concept of massively overpriced speakers and amps worth more than an average middle class home.
If you're going to spend time and money on good audio for your home and life, spend it on doing some research and homework. Spend it on acoustically treating your listening space or home media room and understanding how speaker placement and reflections work.
And if you're into DIY projects, woodworking and electronics there's a whole world of open source DIY speaker plans out there.
And all of that being said? For goodness sake, good audio is more affordable than it has ever been. I've heard tiny little bluetooth speakers and boomboxes that sound so good that it makes me feel things. Good home audio is more affordable than it has ever been.
This nonsense about phonographic records of any kind still being the peak of high quality audio? It's chrome plated copper-bottomed self-fellating complete and utter horseshit.
We're living in an age where you can pick up a pocket sized field recorder with embedded microphones, with 32 bit dynamic range and 192 kbits per second sampling rate with a frequency response capable of recording the whisper of butterfly wings and the incredible ear-splitting roar of an orbital rocket taking off ranging from ultrasonic bat chirps to the rumble of an earthquake all in the same take and you could play most of it back on speakers less then the cost of a decent mid range bicycle.
And they want to take a swing at all of that with a fucking lacquer lathe cut? Get the fuck out of here.
And all that being said? Sure, buy some vinyl or dubplates or lacquer cuts as an art object to support your favorite artist. That's fine. Large format album art and printing is great. There's worse things to spend your money on if that's what you want to do. Analog media is cool, I get it.
But it isn't better or otherwise more accurate than a competent, professional digital master and reproduction. No fucking way. If you really want to talk about extended dynamic range and frequency response digital recording is exponentially more accurate, more pure and has far less reproduction issues as long as we're talking about raw pulse code modulation WAV file without lossy audio data compression codecs.
PS, streaming sucks. Not your files, not your music. FLAC is cool, AAC is cool, and even MP3 at 320k is more than enough for almost everyone. Fuck Tidal. And fuck Spotify in particular because the CEO is absolutely awful and terrible.
posted by loquacious at 7:53 PM on April 27, 2022 [63 favorites]
Yeah, I've hashed this out with die hard analog audiophiles dozens and dozens of times now - and to be honest I fucking love ranting about this topic because it makes me feel smart and it's taken me decades to learn a lot of this stuff and debunking audiophile woo.
I'm discovering there's possibly even more disinformation, myths or magical thinking going around today then there was during the absolute peak of the pure analog studio album.
And lets get this part out of the way first - one of the reasons why this myth persists is the extreme analog quality of recordings in this absolute theoretical peak of analog recording - the era in the 70s of the studio album.
Those recordings do indeed sound good, but it's because they had huge, nearly unlimited budgets to throw at the problem. They could afford the endless reels of very high quality tape. The analog mixing consoles were the best they would ever be. They could afford to do things like slave together multiple 12 to 24 track tape decks to avoid "bouncing down" stems or tracks to a single track to add and mix more tracks. They had analog mixing desks with huge amounts of channels. And it was a sweet spot of newly developed analog, tube based or solid state transistorized amps, pre-amps, audio effects, microphone technology and so much more.
The main factors here are less about audio technology and more about nearly unlimited amounts of time and money to throw at recording and producing music.
Let's also address and put aside the problems with the loudness wars and the whole concept of audio compression - and I mean dynamic range compression, not data compression codecs like MP3 or MPEG stuff. Audio compression is a complicated topic but simplifed it means reducing the dynamic range so quieter audio signals are louder and louder signals are - sometimes - quieter. Like a limiter.
In fact a dynamic range compressor is effectively a kind of limiter, see the wikipedia link above:
Compression and limiting are identical in process but different in degree and perceived effect. A limiter is a compressor with a high ratio and, generally, a short attack time.That's a known problem, and it's very real. Almost all of the music audio you've ever known or experienced uses some kind of compression or limiter somewhere, whether it's old analog circuits used to make vocals punchier or louder on a single track or entire albums or mixdowns.
Because of this one of the myths that refuses to die especially among hard core magical thinking audiophiles is that somehow there's magical extra dynamic range from infrabass to ultrasonics being encoded in an all analog recording path and captured and reproduced on vinyl.
Which is absolutely not true at all. There are absolutely no magical inaudible tones making it to any kind of mass produced vinyl record, especially not the recordings that analog audiophiles hold up as gold standards.
For starters, the high and low pass filter has been a standard part of the recording and mastering tool chain since, I don't know, at least the 60s or even earlier, and it's used specifically to filter out any of those frequencies on purpose.
It's not just an aesthetic choice, it's an essential part of good mastering and recording for audio reproduction, and I can totally dive into the nuts and bolts of why mastering and recording techniques like high/low passes are used.
Part of this myth and misconception is that analog recording media have "infinite" resolution compared to digital sampling which is also absolutely not true at all.
Every single recording, playback or reproduction technology has inherent limitations whether it's analog or digital.
For analog tape the dynamic range and frequency bandwidth is determined by a number of factors and limitations. One of these is the grain size of the magnetic coating. Another - and related to grain size - is tape speed. There's also the factor of tape width, too. But there's another major limitation and that is tape bias.
And another limitation that's co-related to all of the above is the actual physical size of the coils in the tape head - roughly speaking, the larger your coils, the worse it does at high frequency recording. The smaller or finer the coils are, the better they do at high frequency recording. This is head gap.
The tape speed, grain size and size of the coils of the recording head all inform the total frequency response and dynamic range. There's also the issues of tape speed consistency, called wow and flutter.
This is why metal oxide tapes were the absolute pinnacle of magnetic tape recording. They had extremely fine grains and magnetic domain sizes. As a reference point the peak of audiophile quality for magnetic tape in the 1/8" compact cassette format happened in the late 80s through the mid 90s with metal oxide tapes, notably on Nakamichi decks combined with lossy noise reduction schema like Dolby Noise Reduction.
Combine all of the above and there's a general frequency response and dynamic range gamut (just like color reproduction on analog film or digital imaging) that is clearly defined by all of these factors.
Ok, well, what about skipping tape and recording straight to a lathe and lacquer? What about "extra" information in the infrasonic or ultrasonic spectrum there?
Well, in theory? Sure. Sure, maybe you could capture a much wider dynamic range and frequency response with a theoretical unrestricted lathe head cutting on the finest lacquer coating ever invented.
In reality and in practice for any professionally recorded, mastered and cut album this is also definitely not true.
Trying to record ultrasonic tones on a record cutting lathe head would probably break it and make it explode.
Also, good luck finding any phonograph needle or preamp capable of reproducing any of these ethereal ultrasonics, especially if an RIAA EQ compatible phono pre-amp is involved.
You also can't record super deep bass without cutting extremely wide grooves and drastically reduce the available run time of the groove on a 12" record. And even if you did cut a record with huge fat grooves and wobbles it would make all known consumer record player needles and tone arms jump right out of the groove and your recording time at 33.3 or 45 RPM would be measured in, oh, seconds or maybe even up to a minute or two instead of dozens of minutes per side.
This is the whole reason why they invented the RIAA EQ curve and phono pre-amp system. This is how LP - Long Play - and the earlier EP - Extended Play - records actually work.
It's a form of analog data compression that reduces the bass end in the cut master lacquer so they can fit more music time on a 12" record, and it's reconstituted by applying the curve in reverse via the phono pre-amp to reconstitute most of the frequency response and dynamic range.
There is some flexibility here to cut a record with reduced run time to add extra bass to a recording and this was indeed used in vinyl for bass music, dance music and so on, trading some of the available run time for more bass. But this is still within the parameters of the RIAA EQ and phono preamp, and it was a subtle and black art.
See also sonic artifact.
But wait, there's even more limitations!
Most microphones, guitar pickups and other forms of transducers to record with can't even sense ultrasonics in particular anyway. They also have limited frequency response and dynamic range! They also all have their own "tone" or highs and lows in the gamut of frequency response, and will distort or become lossy if trying to record anything that's too quiet to move them or too loud to record accurately.
Wait, we're not done yet! Let's talk about audio reproduction. And by that I mean speakers and amplifiers.
These also have their own limitations and gamut of frequency response and dynamic range. Too quiet and the coils or transducers of a speaker don't move. Too loud and they distort or clip or otherwise degrade.
The current, practicable theoretical maximum frequency response of even the most expensive, unobtanium grade sound system of any kind whether it's pro audio for an arena sized concert or the most expensive magical thinking audiophile grade sound system is about 5-10 hz to about 50 khz.
For the infrabass and super low end we're talking super exotic subwoofers that use servos instead of voice coils or even rotating paddles like a fan to get that low. The kind of thing where you're tuning your whole house or listening room to become a subwoofer and using resonance and constructive feedback to even attempt 5-10 hz. Yes, this is possible. No, you can't hear it. Yes, you could possibly feel it, but not hear it.
At the high end of the frequency response you can get ultratweeters and ribbon tweeters capable of creating 50 khz tones, if not even higher tones. And they're not particularly exotic or expensive, you can go to your nearest hardware store and buy ultrasonic pest repellers. Or go to Mouser or Digikey and buy ultrasonic transducers used for things like range finding or proximity sensing. Or you can even buy super high frequency ultrasonic transducers used for industrial processes like ultrasonic welding of plastics.
No, most people can't hear this, either, and if they're "hearing" anything above about 20-22khz even with the best ears they're actually hearing second or third order harmonics and resonance of something, anything in the environment sympathetically vibrating and acoustically coupling with those inaudible tones.
In reality and practice?
Most people don't own and/or have never even heard any kind of audible speaker that can do anything lower than about 30-50 hz or above about 18-20 kilohertz. 40-50 hz is REALLY good for a home subwoofer, and you need about a 12" to 18" driver and about 1000 watts of power in a nicely tuned and ported speaker enclosure to get it or down to about 30 hz. You also need high powered amps for it. And the lower you go on bass and sub bass the more energy you need to move that much air to hear or feel it. You can throw thousands and thousands of watts or dollars at this problem to break past about 30-40 hz. Or alternative driver/transducer types like servo-drive speakers.
This is exactly why the audio engineering and mastering uses low/high pass filters, to eliminate unneeded frequencies outside of the range of human hearing because it makes everything within that range of human hearing sound cleaner and louder and takes less energy for amps and speakers to reproduce.
This is a known problem in professional audio reproduction even for something like a really massive speaker array for a very large electronic music festival, and this is why these kinds of professional speakers use things like digital signal processors and crossovers to eliminate any frequency information below about 30 hz because anything below that takes exponentially more and more electrical energy and power to move the speaker coils or transducers to try to recreate those super deep bass tones.
And paradoxically you can get cleaner, more articulated and responsive bass tones by eliminating the inaudible portions of it below human hearing levels, because that means more of the available amp power is dedicated to audible frequencies instead of trying to reproduce deep infrasonics so low that not only can you not hear it but you could barely feel it because it's just too slow for human perception.
On the high end of the frequency response being able to hit about 20 khz is also really good and needs really good tweeters or ultratweeters. The power requirements for trying to reproduce this are much less troublesome because you're not trying to move as much air mass, but in practice most people are working with speakers that peak out at about 12-18 khz before being useless or ineffective.
But wait, there's even more limitations!
The final ingredient into this sonic puzzle is the limitations of human hearing itself. Most people are lucky if they can hear from about 40-50 hz to about 16-18 khz, if not much less. There definitely are people who can hear outside of this range, especially above 18khz with upper extremes pushing 20-24 khz in the most rare, and unique edge cases.
I've had my hearing tested, and despite my long history with super loud and weird music I've taken pretty good care of my hearing and have been using ear plugs religiously for most of my adult life, I have really good ears and last time I checked I can hear tones from about 20 hz to about 20 khz, which is not normal or average.
But even then, most of our music is well within about 50-60hz to about 12-18 khz. This is especially true if we're talking about acoustic or traditional instruments, whether it is a full symphonic orchestra or an acoustic guitar, a stand up bass or any vocalist. Whether the vocalist is Tom Waits, Bob Dylan or Diamanda Galás.
Note: All of the above frequency ranges or values are estimates and I'm shooting from the hip based on my experiences and recollection of these ranges. Please don't @ me. They should be close enough for horseshoes and hand grenades.
So, let's talk about the most extreme end of audio reproduction, and we can discuss this in two parts - live acoustic instrument performances or analog (or digital!) synthesizers.
First, let's look at synthesizers with an entirely analog amplification path, tuned and open crossovers, no low pass or high pass, not a recording, played through exceedingly good speakers.
Yep, you sure can really push the gamut of frequency response and dynamic range and it's not at all expensive or even that difficult, especially with really good speakers with very wide frequency responses. You can noodle around with sine waves and tones right on down to 0 hz and up into the ultrasonic ranges. You can straight up light speakers on fire and melt them into slag with tones in the 0-30 hz range with enough amplitude.
I've personally seen speaker cones and coils melt or smoke or tear out their support baffles moving slow enough with deep excursions to count the cycles per second on your fingers with a stop watch. I've done it on purpose with a synthesizer or oscillator. It's fun.
You can blow the caulking and sealing right out of speakers doing this and walk the nails or screws holding the speaker box right out of the wood and break all of the things. You can do terrible, terrible things to speakers with little more than a single oscillator circuit down in the 5-30 hz range and a really strong amplifier.
There's still going to be limitations in frequency response and dynamic range in your gain and amplification path, whether it's a tube amp or solid state transistor. There's still going to be physical limitations to your amplifiers and speaker drivers.
But it's going to be much cleaner with a much wider range of frequency response or dynamic range of any recording media of any kind due to all of the limitations of recording technology. A pure analog synthesizer hooked up to really good speakers is something special. But it has limitations.
Second, let's look at live performances whether it's purely acoustic, whether it's a major symphony orchestra or a live acoustic folk band or electrically amplified rock band through tube amps and a direct connection and relation between the mics or pickups to the amps and speakers, drivers, coils and etc.
Yep, there's all kinds of magic that can happen there from subsonic bass to ultrasonics. The tone of strings vibrating on wood, the whisper or buzz of fingers on fretboards, the sound of a human breathing between vocalization, the natural reverberation of a space, the sympathetic vibrations of our physical world being energized and resonating with the tones of a live guitar or a piano or a vocalist.
Every grain of dust, every plank of wood, the pressure, temperature and humidity of the air, every acoustically reflective surface in the environment contributing to this magic of live music, the complex tones of all that live delay, reverberation and the magic of constructive and destructive waves of acoustic energy.
Yes, there is *magic* there and so many acoustic/psychoacoustic variables and the unmediated immediacy of live music that can vastly exceed and transcend the limitations of audio recording, duplication and reproduction that invokes all of the magic and complexity of real world physics that we know.
That part of the magic of music and sound is all very real and very difficult to quantify or measure whether it's an acoustic guitar or an analog synthesizer.
Can we accurately capture and record it? Perfectly? Hell no. And that's a beautiful and amazing thing.
All of that being said and you still want really good speakers and sound in your life?
Well, the good news is that you really don't need to spend hundreds of thousands of dollars on bespoke audiophile speakers the size of a refrigerator and monoblock amplifiers or solid gold speaker cables or any of that gold plated steaming load of fur-lined horseshit.
The people that spend that kind of money on those kinds of speakers are deluding themselves and drinking the snake oil flavored Kool Aid.
Well, with some caveats. There's some really exotic speakers and sound technology out there and that exoticness ranges from musical instrument grade woods to tube amplifiers to omnidirectional drivers and all kinds of black magic acoustics that are just as valid as the difference between a Stradivarius violin compared to a cheap Mendini.
But you can get to within 1% of those high end systems with much, much less money, with just a little homework. Dive into the the Budget Audiophile subreddit at www.reddit.com/r/BudgetAudiophile and check out their FAQs and guides.
There are incredible speakers out there today where you can put together a 5.1 home theater system for much less money than that fur-lined ocean-going utter nonsense.
You can pick up extremely capable speakers and amps for well under $1000 USD that get you within 10% or 1% of the best speakers and amps and audio equipment in the known world.
I've found audiophile grade speakers at thrift stores in the bookshelf speaker size range for like $5 per speaker.
If you ever see any KLH 45 model 3 way bookshelf speakers at a thrift store, buy them and get a cheap T-class digital amp for like $10-30 off of Amazon, like the famous Lepai 2020A+ and go with it. Pair it with a decent subwoofer like an 8-10" Audiosource box and you have a really nice, compact sound system going on a shoestring budget.
You can also get really amazing speakers and studio monitors from companies like Focal, KRK, Mackie, Yamaha, EV or even vintage Bose and so many more with prices ranging from 200-300 a pair on up to 2000-3000 a pair.
Want to splurge and jump straight into something that's exceedingly high quality and value for the price for a home theater or music system for something that not only competes with wacky unobtanium grade audiofool speakers?
Check out ADAM speakers. Yeah, they're rather pricey compared to something you could pick up at Costco or whatever, but those speakers are some of the best in the world without diving even deeper into pro audio PAs or audiofool snake oil nonsense. They have ultratweeters rated to 50 khz, like I was talking about above.
Those ADAMs are exactly the kind of speakers that pro audio engineers use for mixing blockbuster movies or albums. I have a friend with a pair of the AX7s with a matching sub and they are absolutely fucking incredible speakers, and I can compare them to speakers and sound systems I've heard at 10x or 50x the price.
They're so dynamic, responsive and forgiving that you can hook up an analog synth to them and make horrible duck fart noises through them and it's paradoxically pleasant to listen to, well, anything from the sublime and melodic to the horrible and dissonant. I've put some incredibly terrible, unpleasant noises through my friend's home studio setup and they don't even blink at it and reproduce it faithfully.
Doing a 5.1 home theater or even a full surround Atmos system with those ADAM speakers would be incredible and would blow your favorite high end Dolby/THX certified movie theater right out of the water if only because you could turn them down so they don't clip and distort like your local AMC megaplex.
Want something bigger and louder but relatively affordable? Pick up a pair of QSCs self powered speakers and a matching sub and you're half way to a very high quality dance/club rave sound systems. You can get a pair of QSC K10s to K12s for under 100-2000 or even less on sale or from Sweetwater or other pro audio purveyors and wake up the whole neighborhood with crystal clear action movie noises or the wubbiest of wubby dubstep noises.
There's a whole range of options ranging from well under 50-100 USD or even cheaper to well under 10,000 USD for a full surround theater system that would compete with if not destroy the whole concept of massively overpriced speakers and amps worth more than an average middle class home.
If you're going to spend time and money on good audio for your home and life, spend it on doing some research and homework. Spend it on acoustically treating your listening space or home media room and understanding how speaker placement and reflections work.
And if you're into DIY projects, woodworking and electronics there's a whole world of open source DIY speaker plans out there.
And all of that being said? For goodness sake, good audio is more affordable than it has ever been. I've heard tiny little bluetooth speakers and boomboxes that sound so good that it makes me feel things. Good home audio is more affordable than it has ever been.
This nonsense about phonographic records of any kind still being the peak of high quality audio? It's chrome plated copper-bottomed self-fellating complete and utter horseshit.
We're living in an age where you can pick up a pocket sized field recorder with embedded microphones, with 32 bit dynamic range and 192 kbits per second sampling rate with a frequency response capable of recording the whisper of butterfly wings and the incredible ear-splitting roar of an orbital rocket taking off ranging from ultrasonic bat chirps to the rumble of an earthquake all in the same take and you could play most of it back on speakers less then the cost of a decent mid range bicycle.
And they want to take a swing at all of that with a fucking lacquer lathe cut? Get the fuck out of here.
And all that being said? Sure, buy some vinyl or dubplates or lacquer cuts as an art object to support your favorite artist. That's fine. Large format album art and printing is great. There's worse things to spend your money on if that's what you want to do. Analog media is cool, I get it.
But it isn't better or otherwise more accurate than a competent, professional digital master and reproduction. No fucking way. If you really want to talk about extended dynamic range and frequency response digital recording is exponentially more accurate, more pure and has far less reproduction issues as long as we're talking about raw pulse code modulation WAV file without lossy audio data compression codecs.
PS, streaming sucks. Not your files, not your music. FLAC is cool, AAC is cool, and even MP3 at 320k is more than enough for almost everyone. Fuck Tidal. And fuck Spotify in particular because the CEO is absolutely awful and terrible.
posted by loquacious at 7:53 PM on April 27, 2022 [63 favorites]
Jesus Christ, loquacious. That's longer than Thick as a Brick.
posted by Halloween Jack at 8:53 PM on April 27, 2022 [10 favorites]
posted by Halloween Jack at 8:53 PM on April 27, 2022 [10 favorites]
In fact a dynamic range compressor is effectively a kind of limiter
Makes more sense to say a limiter is a kind of compressor. Yes I’m responding to this one line out of that comment.
posted by atoxyl at 9:56 PM on April 27, 2022 [2 favorites]
Makes more sense to say a limiter is a kind of compressor. Yes I’m responding to this one line out of that comment.
posted by atoxyl at 9:56 PM on April 27, 2022 [2 favorites]
That's longer than Thick as a Brick.
Gross! MODS!
Pardon the huge rant. Parts of it are clumsy. I've written this dumb rant about analog vs. digital at least a dozen times now, and I was thinking about this one all day.
There's a whole lot going on with all of this, the science and art of music, recording it, playing it back and listening to it but it doesn't need any mystification.
Music - and sound recording and reproduction - is already naturally mysterious and magical.
Even with the worst speakers and recording technology. I have had some truly horrible music listening devices from my first pocket radio to really terrible modern bluetooth speakers, and even those were better than nothing.
It's like pizza. Bad pizza is usually better than no pizza at all.
It's not the technology. It's the music.
I really don't care if any artist wanted to cut a thousand different takes of the same song direct to lacquer and individually sell ridiculous prices because each one is truly one of one or a limited run like a lithograph. That would be cool.
The part that bothers me about these kinds of endeavors in the link is the elitist and purist shtick, any claims about any kind of ultimate audio fidelity, especially from a phonograph.
posted by loquacious at 10:27 PM on April 27, 2022 [6 favorites]
Gross! MODS!
Pardon the huge rant. Parts of it are clumsy. I've written this dumb rant about analog vs. digital at least a dozen times now, and I was thinking about this one all day.
There's a whole lot going on with all of this, the science and art of music, recording it, playing it back and listening to it but it doesn't need any mystification.
Music - and sound recording and reproduction - is already naturally mysterious and magical.
Even with the worst speakers and recording technology. I have had some truly horrible music listening devices from my first pocket radio to really terrible modern bluetooth speakers, and even those were better than nothing.
It's like pizza. Bad pizza is usually better than no pizza at all.
It's not the technology. It's the music.
I really don't care if any artist wanted to cut a thousand different takes of the same song direct to lacquer and individually sell ridiculous prices because each one is truly one of one or a limited run like a lithograph. That would be cool.
The part that bothers me about these kinds of endeavors in the link is the elitist and purist shtick, any claims about any kind of ultimate audio fidelity, especially from a phonograph.
posted by loquacious at 10:27 PM on April 27, 2022 [6 favorites]
Now that is an amazing comment for the MeFi hall of fame. While I'm all about quality in general including audio (but forget furlined, etc.) I have to say thanks to loquacious for that clear and comprehensive covering of all the basics in sound recording and reproduction as relating to frequency response.
posted by blue shadows at 10:54 PM on April 27, 2022 [1 favorite]
posted by blue shadows at 10:54 PM on April 27, 2022 [1 favorite]
Want to know something else (something I learned about eyes working on colour correction hardware in the early desktop publishing days) - everyone's ears are different - at a gross level the older you get (or the more loud concerts you go to) the less you can hear at the high end - if you can afford those ultrasonics you probably can't hear them .... but at a smaller level it's different - those tiny hairs in your ears than detect sound, they're different to the ones in your neighbours' just like the hair on your head
There is no perfect sound, no golden ears - tweak that equaliser to make it sound good to you
posted by mbo at 11:01 PM on April 27, 2022 [1 favorite]
There is no perfect sound, no golden ears - tweak that equaliser to make it sound good to you
posted by mbo at 11:01 PM on April 27, 2022 [1 favorite]
FLAC is cool, AAC is cool, and even MP3 at 320k is more than enough for almost everyone.
There's scarcely any source material of any kind where anybody will ever pick up the difference between a 200kb/s MP3 and and 320kb/s MP3 made with the current generation of LAME.
That said, I still archive all my CD rips in FLAC because audio files take up a lot less storage space than movies so why not? But the versions I listen to most often are the -qscale:a 2 LAME MP3 dubs that live on my phone.
Gross! MODS!
I will not stand by and allow Thick As A Brick to be casually impugned in such a fashion. Thick As A Brick is a great album and having something you've written described as being like it in any way should be taken as the highest of compliments.
As for self-styled Golden Eared Audiophiles who opine that vinyl or lacquer or any other bloody thing captures more musical nuance than a properly mastered 16 bit 44100Hz sampled CD ever could:
Really don't mind if you sit this one out
My words but a whisper, your deafness a shout
posted by flabdablet at 5:54 AM on April 28, 2022 [1 favorite]
There's scarcely any source material of any kind where anybody will ever pick up the difference between a 200kb/s MP3 and and 320kb/s MP3 made with the current generation of LAME.
That said, I still archive all my CD rips in FLAC because audio files take up a lot less storage space than movies so why not? But the versions I listen to most often are the -qscale:a 2 LAME MP3 dubs that live on my phone.
Gross! MODS!
I will not stand by and allow Thick As A Brick to be casually impugned in such a fashion. Thick As A Brick is a great album and having something you've written described as being like it in any way should be taken as the highest of compliments.
As for self-styled Golden Eared Audiophiles who opine that vinyl or lacquer or any other bloody thing captures more musical nuance than a properly mastered 16 bit 44100Hz sampled CD ever could:
Really don't mind if you sit this one out
My words but a whisper, your deafness a shout
posted by flabdablet at 5:54 AM on April 28, 2022 [1 favorite]
get a cheap T-class digital amp for like $10-30 off of Amazon, like the famous Lepai 2020A+
Are there any switching-style power amps available that are fully digital all the way to the final switching transistors, so that there are no analog voltages in the low-level signal chain whatsoever? Because that strikes me as the right way to make nice clean audio. Should be really easy to design, too: all it would need is a cheap microcontroller that accepts a range of digital input formats and emits a nicely noise shaped, inherently switching-friendly delta-sigma encoding like DSD, then sends that straight to a set of switching MOSFETs driving the speaker output via a passive LC 20kHz lowpass filter. The only analog-signal feedback needed in such a design would be for stabilizing the power rails that supply the output switches.
Theoretically better still would be speakers with digital servos for voice coil positioning, but given that almost all the distortion added by speakers comes from driver flex and cabinetry and listening room resonances, with almost none due to the tiny nonlinearities in cone positioning vs applied analog signal voltage, I doubt that the considerable extra complexity would make an audible difference.
posted by flabdablet at 6:32 AM on April 28, 2022
Are there any switching-style power amps available that are fully digital all the way to the final switching transistors, so that there are no analog voltages in the low-level signal chain whatsoever? Because that strikes me as the right way to make nice clean audio. Should be really easy to design, too: all it would need is a cheap microcontroller that accepts a range of digital input formats and emits a nicely noise shaped, inherently switching-friendly delta-sigma encoding like DSD, then sends that straight to a set of switching MOSFETs driving the speaker output via a passive LC 20kHz lowpass filter. The only analog-signal feedback needed in such a design would be for stabilizing the power rails that supply the output switches.
Theoretically better still would be speakers with digital servos for voice coil positioning, but given that almost all the distortion added by speakers comes from driver flex and cabinetry and listening room resonances, with almost none due to the tiny nonlinearities in cone positioning vs applied analog signal voltage, I doubt that the considerable extra complexity would make an audible difference.
posted by flabdablet at 6:32 AM on April 28, 2022
No Flanders and Swann aficionados here I guess.
posted by anguspodgorny at 6:35 AM on April 28, 2022
posted by anguspodgorny at 6:35 AM on April 28, 2022
Given that A Song Of Reproduction has already been linked here independently by two of us: I beg to differ. O gno, gno, gno, as it were. That duck on the lake's a decoy.
posted by flabdablet at 6:37 AM on April 28, 2022
posted by flabdablet at 6:37 AM on April 28, 2022
I will not stand by and allow Thick As A Brick to be casually impugned in such a fashion. Thick As A Brick is a great album and having something you've written described as being like it in any way should be taken as the highest of compliments.
I was teasing. Mostly.
posted by loquacious at 10:31 AM on April 28, 2022
I was teasing. Mostly.
posted by loquacious at 10:31 AM on April 28, 2022
There's scarcely any source material of any kind where anybody will ever pick up the difference between a 200kb/s MP3 and and 320kb/s MP3 made with the current generation of LAME.
I have a pair ambient/noise albums I made that breaks MP3 compression and makes it freak out, but part of the source material and production process involved using encoding artifacts from MP3/MPEG encoding, so when I rip it from a full res PCM/WAV it tends to make even the latest versions of LAME freak out just like you were encoding and re-encoding a stepped chain of encoded files.
The problem was so bad I couldn't release it on Soundcloud like I originally planned to because of their default 128k bit crushed encoding for streaming just couldn't deal with it and turned it all into a weird, muddy sounding soup. I ended up switching to Bandcamp to publish it, and keep the full res original WAVs in cloud storage to send to anyone who happens to buy it or want it as intended.
Encoding it with extreme settings at 320K CBR is... acceptable and tolerable but not ideal because it loses a lot of the harmonic and noisy details and presence of the originals.
I also DJ fairly often and mainly play/use MP3s which is considered a huge faux pax in the digital DJing world, and I have to vet and preview all of my tracks I intend to play out on larger sound systems because the encoding can be all over the place and sometimes isn't up to par, but this is more of an encoding setting issue than an issue with high bit rate MP3s itself.
It's about 2-3 tracks out of 10 tracks or so where the encoding isn't good enough.
It's odd and kind of a paradox but certain kinds of highly detailed or complex electronic music like experimental noise, IDM and Ambient seems to be particularly difficult for LAME and other encoders to get right. Stuff like Aphex Twin and Boards of Canada is an easy example of this.
I can easily hear the difference between MP3 rips of Boards of Canada and the original full resolution 44.1k WAVs or original CDs, but there's a lot of tiny details in Board of Cananda that are basically musical noise, hisses, crackles and other intentionally lo-fi touches that are too far in the background or close to the noise floor of the dynamic range of a CD, and they can get lost even in the highest, most extreme settings for LAME.
posted by loquacious at 10:46 AM on April 28, 2022 [1 favorite]
I have a pair ambient/noise albums I made that breaks MP3 compression and makes it freak out, but part of the source material and production process involved using encoding artifacts from MP3/MPEG encoding, so when I rip it from a full res PCM/WAV it tends to make even the latest versions of LAME freak out just like you were encoding and re-encoding a stepped chain of encoded files.
The problem was so bad I couldn't release it on Soundcloud like I originally planned to because of their default 128k bit crushed encoding for streaming just couldn't deal with it and turned it all into a weird, muddy sounding soup. I ended up switching to Bandcamp to publish it, and keep the full res original WAVs in cloud storage to send to anyone who happens to buy it or want it as intended.
Encoding it with extreme settings at 320K CBR is... acceptable and tolerable but not ideal because it loses a lot of the harmonic and noisy details and presence of the originals.
I also DJ fairly often and mainly play/use MP3s which is considered a huge faux pax in the digital DJing world, and I have to vet and preview all of my tracks I intend to play out on larger sound systems because the encoding can be all over the place and sometimes isn't up to par, but this is more of an encoding setting issue than an issue with high bit rate MP3s itself.
It's about 2-3 tracks out of 10 tracks or so where the encoding isn't good enough.
It's odd and kind of a paradox but certain kinds of highly detailed or complex electronic music like experimental noise, IDM and Ambient seems to be particularly difficult for LAME and other encoders to get right. Stuff like Aphex Twin and Boards of Canada is an easy example of this.
I can easily hear the difference between MP3 rips of Boards of Canada and the original full resolution 44.1k WAVs or original CDs, but there's a lot of tiny details in Board of Cananda that are basically musical noise, hisses, crackles and other intentionally lo-fi touches that are too far in the background or close to the noise floor of the dynamic range of a CD, and they can get lost even in the highest, most extreme settings for LAME.
posted by loquacious at 10:46 AM on April 28, 2022 [1 favorite]
loquacious: Also consider that MP3 is pretty ancient now. Both AAC and Vorbis, and especially Opus perform far better than MP3, even if newer MP3 implementations are a lot better than they used to be. 256kbps AAC is basically flawless, and you can do 320kbps if you really want to.
posted by Joakim Ziegler at 12:06 PM on April 28, 2022 [2 favorites]
posted by Joakim Ziegler at 12:06 PM on April 28, 2022 [2 favorites]
their default 128k bit crushed encoding for streaming just couldn't deal with it and turned it all into a weird, muddy sounding soup
As you mentioned, streaming sucks. Nobody should be streaming music as 128kb/s MP3s, that's just horrible. It's fine for talky podcasts, or playback via phone speakers, but that's about it.
I ended up switching to Bandcamp to publish it
I do like me a bit of interesting noise and I can't find a Bandcamp link in your profile. Hint. Hint.
and keep the full res original WAVs in cloud storage to send to anyone who happens to buy it or want it as intended
Is there some good reason why I wouldn't be able to reconstruct those bit-for-bit from one of Bandcamp's FLACs?
posted by flabdablet at 1:44 PM on April 28, 2022
As you mentioned, streaming sucks. Nobody should be streaming music as 128kb/s MP3s, that's just horrible. It's fine for talky podcasts, or playback via phone speakers, but that's about it.
I ended up switching to Bandcamp to publish it
I do like me a bit of interesting noise and I can't find a Bandcamp link in your profile. Hint. Hint.
and keep the full res original WAVs in cloud storage to send to anyone who happens to buy it or want it as intended
Is there some good reason why I wouldn't be able to reconstruct those bit-for-bit from one of Bandcamp's FLACs?
posted by flabdablet at 1:44 PM on April 28, 2022
Tomsk your linked Wikipedia article says that "Despite their name, professional grade "acetate" discs do not contain any acetate. They consist of an aluminum disc with a coating of nitrocellulose lacquer."
Good call
posted by MengerSponge at 5:35 PM on April 28, 2022
Good call
posted by MengerSponge at 5:35 PM on April 28, 2022
There's scarcely any source material of any kind where anybody will ever pick up the difference between a 200kb/s MP3 and and 320kb/s MP3 made with the current generation of LAME.
I did an online test once where I was able to distinguish 3 out of 4 high bitrate vs. standard bitrate samples...
posted by blue shadows at 6:14 PM on April 28, 2022
I did an online test once where I was able to distinguish 3 out of 4 high bitrate vs. standard bitrate samples...
posted by blue shadows at 6:14 PM on April 28, 2022
Is there some good reason why I wouldn't be able to reconstruct those bit-for-bit from one of Bandcamp's FLACs?
I don't know, I haven't tried it. I'd be curious to see if it works fine or not. Theoretically FLAC shouldn't freak out the way JPG/MPEG encoders do because it's a totally different codec scheme.
posted by loquacious at 4:02 PM on April 29, 2022
I don't know, I haven't tried it. I'd be curious to see if it works fine or not. Theoretically FLAC shouldn't freak out the way JPG/MPEG encoders do because it's a totally different codec scheme.
posted by loquacious at 4:02 PM on April 29, 2022
FLAC is indeed completely unfreakoutable, because its role is limited by design to squeezing out only such redundancies as actually exist in the original PCM-encoded bitstream which, for most of the audio I've ever encoded, results in FLAC files somewhere between a third and a half the size of their input WAVs. If you feed FLAC something designed specifically to be indigestible - perhaps a bitstream that wasn't even sampled audio to begin with - the very worst that can possibly happen is that the resulting .flac file needs more storage space than the raw original. None of the original bits are ever dropped under any conditions.
Lossy codecs like MP3 and AAC, by contrast, are designed to guarantee a pre-specified degree of storage savings while still being able to reconstruct an approximation to their input bitstream that should ideally sound although not be indistinguishable after a round trip from PCM. Given at least 200kb/s of post-encoding bandwidth to play with, both of these codecs almost always do a much better job of reproducing CD-grade 1.4Mb/s PCM audio than the traditional analog parallel of dubbing vinyl to cassette. At 320kb/s they do an unambiguously better job than dubbing analog master tape to vinyl. The kinds of artifacts left behind by a good MP3 encode are way less distracting than those caused by dust or scratches or suboptimal turntable or tonearm setup, because the engineers who designed these codecs were really good at their jobs.
At 320kb/s, source material that reveals audible differences between MP3 and newer codecs like AAC or Vorbis or Opus in the course of a properly conducted A/B/X test would be very thin on the ground. Almost all the performance improvements that newer codecs have to offer appears, again by design, at lower encoded bandwidths.
Just bought Edge Detection and Ansible - thanks, loquacious - and downloaded both in FLAC (after noting with amusement that Bandcamp also offers downloads in WAV format for those of us who prefer our bits more thoroughly bubblewrapped). Playing now, and I like what I'm hearing. It's doing good things with my breathing.
Are there particular passages in which I'm likely to be able to hear an MP3 re-encode make a mess of your intentions?
posted by flabdablet at 7:07 PM on April 29, 2022 [1 favorite]
Lossy codecs like MP3 and AAC, by contrast, are designed to guarantee a pre-specified degree of storage savings while still being able to reconstruct an approximation to their input bitstream that should ideally sound although not be indistinguishable after a round trip from PCM. Given at least 200kb/s of post-encoding bandwidth to play with, both of these codecs almost always do a much better job of reproducing CD-grade 1.4Mb/s PCM audio than the traditional analog parallel of dubbing vinyl to cassette. At 320kb/s they do an unambiguously better job than dubbing analog master tape to vinyl. The kinds of artifacts left behind by a good MP3 encode are way less distracting than those caused by dust or scratches or suboptimal turntable or tonearm setup, because the engineers who designed these codecs were really good at their jobs.
At 320kb/s, source material that reveals audible differences between MP3 and newer codecs like AAC or Vorbis or Opus in the course of a properly conducted A/B/X test would be very thin on the ground. Almost all the performance improvements that newer codecs have to offer appears, again by design, at lower encoded bandwidths.
Just bought Edge Detection and Ansible - thanks, loquacious - and downloaded both in FLAC (after noting with amusement that Bandcamp also offers downloads in WAV format for those of us who prefer our bits more thoroughly bubblewrapped). Playing now, and I like what I'm hearing. It's doing good things with my breathing.
Are there particular passages in which I'm likely to be able to hear an MP3 re-encode make a mess of your intentions?
posted by flabdablet at 7:07 PM on April 29, 2022 [1 favorite]
Just for fun, I tried encoding 200MiB of cryptographic randomness formatted as CD-quality audio with FLAC, the entire point of cryptographic randomness being that any given chunk of it is too unpredictable to compress by any method:
Now let's abuse an MP3 encoder the same way:
All of the above files sound like white noise when played, exactly as expected. Anybody who claimed to be able to hear differences between them when reproduced on the same equipment would have to be fooling themselves.
posted by flabdablet at 12:25 AM on April 30, 2022 [1 favorite]
stephen@jellyshot:/tmp/sox$ sox -t raw -r 44100 -c 2 -b 16 -e signed --endian little <(dd if=/dev/urandom bs=4M count=50) yow.wav 50+0 records in 50+0 records out 209715200 bytes (210 MB, 200 MiB) copied, 1.7388 s, 121 MB/s stephen@jellyshot:/tmp/sox$ flac -8 yow.wav flac 1.3.3 Copyright (C) 2000-2009 Josh Coalson, 2011-2016 Xiph.Org Foundation flac comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it under certain conditions. Type `flac' for details. yow.wav: wrote 209877052 bytes, ratio=1.001 stephen@jellyshot:/tmp/sox$ flac -d -o yow2.wav yow.flac flac 1.3.3 Copyright (C) 2000-2009 Josh Coalson, 2011-2016 Xiph.Org Foundation flac comes with ABSOLUTELY NO WARRANTY. This is free software, and you are welcome to redistribute it under certain conditions. Type `flac' for details. yow.flac: done stephen@jellyshot:/tmp/sox$ ls -al total 614604 drwxr-xr-x 2 stephen stephen 4096 Apr 30 16:29 . drwxrwxrwt 19 root root 20480 Apr 30 16:20 .. -rw-r--r-- 1 stephen stephen 209715244 Apr 30 16:28 yow2.wav -rw-r--r-- 1 stephen stephen 209877052 Apr 30 16:28 yow.flac -rw-r--r-- 1 stephen stephen 209715244 Apr 30 16:28 yow.wav stephen@jellyshot:/tmp/sox$ shasum *.wav d8722314d4ac96f3bb9b3ec3210f1473a683ae8a yow2.wav d8722314d4ac96f3bb9b3ec3210f1473a683ae8a yow.wavFLAC does well here, increasing the storage required over WAV by only 0.08%. Decoding the FLAC file back to another WAV yields one that's bit-for-bit identical, as shown by the SHA1 checksum results.
Now let's abuse an MP3 encoder the same way:
stephen@jellyshot:/tmp/sox$ ffmpeg -hide_banner -i yow.wav -c:a mp3 -q:a 0 yow.mp3
Guessed Channel Layout for Input Stream #0.0 : stereo
Input #0, wav, from 'yow.wav':
Duration: 00:19:48.86, bitrate: 1411 kb/s
Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 44100 Hz, stereo, s16, 1411 kb/s
Stream mapping:
Stream #0:0 -> #0:0 (pcm_s16le (native) -> mp3 (libmp3lame))
Press [q] to stop, [?] for help
Output #0, mp3, to 'yow.mp3':
Metadata:
TSSE : Lavf58.76.100
Stream #0:0: Audio: mp3, 44100 Hz, stereo, s16p
Metadata:
encoder : Lavc58.134.100 libmp3lame
size= 36395kB time=00:19:48.85 bitrate= 250.8kbits/s speed=27.1x
video:0kB audio:36395kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.000679%
stephen@jellyshot:/tmp/sox$ ffmpeg -hide_banner -i yow.mp3 yow3.wav
Input #0, mp3, from 'yow.mp3':
Metadata:
encoder : Lavf58.76.100
Duration: 00:19:48.91, start: 0.025057, bitrate: 250 kb/s
Stream #0:0: Audio: mp3, 44100 Hz, stereo, fltp, 250 kb/s
Metadata:
encoder : Lavc58.13
Stream mapping:
Stream #0:0 -> #0:0 (mp3 (mp3float) -> pcm_s16le (native))
Press [q] to stop, [?] for help
Output #0, wav, to 'yow3.wav':
Metadata:
ISFT : Lavf58.76.100
Stream #0:0: Audio: pcm_s16le ([1][0][0][0] / 0x0001), 44100 Hz, stereo, s16, 1411 kb/s
Metadata:
encoder : Lavc58.134.100 pcm_s16le
size= 204800kB time=00:19:48.85 bitrate=1411.2kbits/s speed= 304x
video:0kB audio:204800kB subtitle:0kB other streams:0kB global headers:0kB muxing overhead: 0.000037%
stephen@jellyshot:/tmp/sox$ ls -al
total 855812
drwxr-xr-x 2 stephen stephen 4096 Apr 30 16:46 .
drwxrwxrwt 19 root root 20480 Apr 30 16:20 ..
-rw-r--r-- 1 stephen stephen 209715244 Apr 30 16:28 yow2.wav
-rw-r--r-- 1 stephen stephen 209715278 Apr 30 16:46 yow3.wav
-rw-r--r-- 1 stephen stephen 209877052 Apr 30 16:28 yow.flac
-rw-r--r-- 1 stephen stephen 37268565 Apr 30 16:46 yow.mp3
-rw-r--r-- 1 stephen stephen 209715244 Apr 30 16:28 yow.wav
stephen@jellyshot:/tmp/sox$ shasum *.wav
d8722314d4ac96f3bb9b3ec3210f1473a683ae8a yow2.wav
23a52a1ffd4c2e4d76e15861ba8ddfaf63fa394c yow3.wav
d8722314d4ac96f3bb9b3ec3210f1473a683ae8a yow.wav
Using the highest quality variable-bit-rate encoding supported by the LAME MP3 encoder, the resulting MP3 file is a little under a fifth of the size of its input. The decoded result, though, is not at all the same: it's a smidge bigger, MP3 being a bit vague about the beginning and ending of what it encodes, and it has a different SHA1 checksum.All of the above files sound like white noise when played, exactly as expected. Anybody who claimed to be able to hear differences between them when reproduced on the same equipment would have to be fooling themselves.
posted by flabdablet at 12:25 AM on April 30, 2022 [1 favorite]
loquacious
We are not worthy. *bows in respect*
I am nearly 60, but have taken care of my ears and still have very good hearing. Did test them a while back and have the standard age-related high frequency loss, but otherwise they are still fine.
Music is a big part of my life, rarely less than 3-4 hours a day, and I am a total digital convert. Analog audio served its purpose well enough, but its time has passed, and good riddance to the fiddly temperamental noisy crap. The first time I heard a CD I knew analog was history, and the full DDD process was the future for hi-fi.
256AAC is my audio standard now, and my CD collection was converted to it and then sold off.
posted by Pouteria at 6:20 AM on May 1, 2022 [1 favorite]
We are not worthy. *bows in respect*
I am nearly 60, but have taken care of my ears and still have very good hearing. Did test them a while back and have the standard age-related high frequency loss, but otherwise they are still fine.
Music is a big part of my life, rarely less than 3-4 hours a day, and I am a total digital convert. Analog audio served its purpose well enough, but its time has passed, and good riddance to the fiddly temperamental noisy crap. The first time I heard a CD I knew analog was history, and the full DDD process was the future for hi-fi.
256AAC is my audio standard now, and my CD collection was converted to it and then sold off.
posted by Pouteria at 6:20 AM on May 1, 2022 [1 favorite]
« Older We are fed pabulum laced with little hits of... | "we stand for Mourner's Kaddish for 11 months... Newer »
This thread has been archived and is closed to new comments
What absolute flapdoodle. Burnett has his talents but this is five-star, fur-lined, ocean-going nonsense.
posted by the duck by the oboe at 1:30 AM on April 27, 2022 [64 favorites]