The Steady Demise of Recording Quality
[Sorry, but I've had to disable the audio links in this post.]
[This is a long, long article which my brother wrote for the site. It's a fascinating and really slightly scary piece of writing and I seriously recommend you make the time to read the whole thing. He explains very neatly how the quality of recorded music is taking a massive nosedive, and how really high-definition may be lost to us forever quite soon.
This is something I am genuinely proud to have on this site. After all the fanny jokes and random bollocks, it's nice to publish something with genuine depth of knowledge and thought, despite the depressing inevitability that it took someone else to write it. Little bastard. He's handsomer than me too, the little fucker.
The other thing is that he illustrates his points with mp3s and I am not sure if the quality of the songs I provide will make his argument properly, but click on the pictures and you'll get the point. I know it doesn't fit the narrow column width format too well, but there's not a lot I can do about that, I don't think.]
My name is Ben. To give a brief background I am the Brother of Toad. You may remember me from my brother’s tales of my wedding where he represented his homeland wonderfully by firstly only telling embarrassing stories about my wife, leaving me unscathed; secondly using the word ‘fuck’ and thus reducing a small army of middle aged American women to giggles of twittering joy and shock. I live in the United States where I am attempting to live my dream of recording classical music. In my time I have recorded orchestras, worked with conductors described as “the best in their field“, designed sound for opera and ballet tours (regional and international), been flown to Malaysia and Copenhagen to teach audio, and done Aladdin the Panto in Halifax with Darren Day (no really). From this I have learned that music is best heard in grand old theatres, if you mic opera you can end up making ears/arses bleed, and that Darren Day is one of the most considerate, professional and genuinely nice men I have ever met (No really, the man‘s a Prince. Who knew?).
Anyway, I was driving to a wedding in New Jersey, which is painful enough in itself because it means going towards New Jersey, but also because unless you put your foot down quite firmly before you set off on road trip with an American from the suburbs (my lovely wife) you can quite easily find yourself listening to four hours of Matchbox 20. So as I trundle down the lazy back roads of Connecticut listening to Billy Bragg sing his heart rending classic ‘Saturday Boy’ and chatting to my wife I felt truly, truly happy.
Billy Bragg – The Saturday Boy (Clip)
Then as the song wound down the stereo suddenly erupted in a great racket and I jumped a foot in the air almost taking the car off the road. You see what had happened was that the excellent song ‘Natalie’ by The Killers had come on and the shock to my system from the simple purity of the one, to the garbled audio mess of the other was enough make me want to puncture the very ear drums that have served as my livelihood these last ten years. Not that the Killers’ first album is bad. It isn’t. It is, however, so much that is wrong with modern pop music.
The Killers – Natalie (Clip)
Let me go back a bit…
A hundred years ago (or so…) people started recording music. The problem with this is that music is not meant to be recorded. It is meant to be played to you by musicians in a concert hall. In this concert hall the quiet bits will echo wistfully around the room and the loud bits swell around you wrapping you in a blanket of sound that flows over and through you. The thing is, you and I can’t really afford to assemble the London Philharmonic whenever we want to, so we buy the record, and so recording music gave music to the people, and was thus a Good Thing. The problem is that when you stick music through a mic the quiet bits are too quiet and the loud bits are too loud because your room is, sorry to say it, not the Albert hall. That is until some clever bugger invented tape which took all the loud bits and squished them down in way that was warm and pleasing, and things became good. How good? This bloody good:
Bob Dylan – Like a Rolling Stone (Clip)
Listen to it. It’s loud, but the guitar is coming out of one speaker, the organ from another and slap bang in the middle is the scratchy distorted voice of Bob Dylan. The instruments don’t sound clean but they sound perfect enough you don’t need to do anything to them. This is called tape compression. The really rough bits have been smoothed, it’s got enough reverb on it that it sounds like it’s in a room – a room where you, Bob and the band are sitting around with a bottle of sippin’ whiskey making great music. Its natural successor is this:
Bruce Springsteen – Erie Canal (Clip)
Pretty good, but what this I hear you say? “Ben that was good but it was a bit less clear and I couldn’t really hear where anything was coming from anymore? It was loud and great, but the instruments blended into one another…it was more of a…mess. “. That is one of my favourite recordings of the last few years but the problem is that music has gotten louder and louder.
One of the reason music got louder is to make it stand out. Let me play you a clip:
Radio Mix
Those last two stand out a bit eh? I cheated too: I Fought the Law is generally considered to be one of the defining songs of the Wall of Sound where people started to really experiment with compression as a means of getting your song to sound louder. You see, once you compress it and get rid of all those peaks and troughs you can turn the whole lot up louder and really strip people’s ear drums. The problem is you lose the clarity. Listen to that first one, considered senseless noise at the time. It’s garbled and raw, absolutely, but you can hear it. All of it. By the Oasis track it’s a mess, but God Almighty that last one is rubbish. It‘s musical barf. Not the song, the quality. This is what you were listening too:
We all like to really get blown away by a track every now and then and I wouldn’t deny you that, but here’s a trick, turn your own damn stereo up. Your speakers have loads of head room on them, but in order to stand out this music is being crammed into all the space that is available, so it arrives in you record player turned up.
Herein lies the problem. Non-technically, if you look at the images, you can see there is only a finite amount of space to fit all the information in. The more you cram in, the more garbled the rest becomes.
Think of it like this. If I record Tom Waits playing his piano, there are a lot of sounds. There is Tom’s voice, there is the sound of Tom’s voice reverberating around the room, there is the sound of the hammer hitting the string, there is the sound of the string vibrating, both of these sounds hit the piano’s sound board so there are those sounds as well, then all of the sounds echo around the room and arrive at the microphone a bit late so there are those sounds too. This last bit is crucial because all rooms sound different. The main room at Abbey Road is famous because it sounds great. The hammer hitting the chord, the chord vibrating and Tom’s voice are the loudest. If you turn up the recording loud enough you drown out all the other sounds, which is why Tom isn’t loud.
Tom Waits – Martha (Clip)
And if you don‘t believe me about background noise really turn that clip up and tell me what those clicks are at the end? Tell me that! It‘s like being taken to see the Mona Lisa and the guard grabbing your head and jamming it up to her eyes. They are the best bit, no doubt, but without the rest of the painting they are just accurately painted eyes. Thing is, not even Tom Waits is immune.
Tom Waits – Road to Peace (Clip)
Et tu Tom? So what I said about loud music being enjoyable still stands. Turn the Martha clip up as loud as you can. Loud, eh? So they aren’t doing it because music needs to be louder are they?
Here’s the other effect that it has on us. If you don’t have quiet bits, people don’t sit forward in their chair to listen. When a song goes a bit quiet, people pay more attention. The audience, and please forgive me the expression: listen up. If they don’t listen up, and if they aren’t sitting forward how can you blast them back in their seat and really blow them away? This is a long clip but listen to the difference in the volumes: when the crescendo hits, by god does it hit. And it’s not even a great bit of music. I mean, it’s no Mozart.
Sylphide Overture
So this is what the opening minute of that looks like:
.
Now we shall compress it. It looks like this:
Then the wave form looks like this:
Sylphide Overture with Compression
Besides the now clearly audible mobile phone, it lacks punch, the music doesn‘t pull you around as much. Now compare to this recording travesty:
Yann Tiersen & Neil Hannon – Les Jours Tristes (Clip)
Call that a crescendo? My wife has greater crescendos than that when I’ve had 8 pints! But it’s louder. But that doesn’t matter because the musical escalation looks like this:
So not only can‘t you hear all the instruments like the previous bit, you also don‘t get the actual effect of music like the composer intended. If he had wanted it louder he‘d have had all the instruments playing, wouldn‘t he? Do you see, if the composer wants it quiet he plays one violin, if he wants it loud he uses all 18. But no, some yahoo music executive knows better eh? Penises.
Maybe now I have made it clearer why when I went from Billy Bragg:
To the Killers:
I was shocked by volume, and horrified by the sound quality. I was sitting in a room while Billy Bragg sang a song just for me one moment and the next I was huddled in a corner surrounded by The Killers, being subjected to their song like I was in the aural version of that scene from Clockwork Orange. So why over-compress and garble when the record label could just trust that you will turn it up yourself? Why ruin art? Why rob a musician of his voice, and a composer (alright, writer) of his ability to express himself? Because no one gives a shit about the quality of product when you can peddle another fucking Killers album. Well fuck you! I want my music back.
More importantly, why would I bore the shit out of you with this unfunny rant on a very funny site? Well because I don’t know if the studios teach the listener what to listen to, or if the listener demands drive the studios to get louder and louder. Maybe a little of both. To re-iterate: this is the Dead Kennedys:
The Dead Kennedys – Kill the Poor (Whole song – yay!)
This is the Decemberists:
The Decemberists – The Engine Driver (Whole song – yay!)
The Decemberists? In whose fucking corporate wet dream to the Decemberists ever need to be that loud. They play fucking folk music! Folk music! Either way, you are a huge market (sorry, evil way to describe the purity of love that the Toadlings feel for their tunes but face it, a market is all you are to these people). You have to start listening to music, and demanding quality or quality will go forever. I can record an orchestra at four times the audio quality of a CD. You can hear everything, just like you were there. I can sit you next to the conductor, and have him say ‘Listen to this sonny, this is how Beethoven heard this when he dreamed. But will anyone ever hear it?
When Shane McGowan sings I can hear the upper register of his voice wobble as he strains to hold the high notes, although less so on the newer ‘enhanced’ version, and certainly not on the mp3 version. They are taking your music from you. Please, for all of us, take it back. Will no one save us in this horrible world? Well, there is one…
The Divine Comedy – A Drinking Song (Whole song – yay!)
Thanks Neil, don’t mind if I do.
Excellent post Brother Of Toad!
Nice turn, Brother Toad. The compression of music (much like the compression of TV adverts) rattles the fuck out of me. Do you have any idea how fucking frustrating/annoying it is trying to accurately judge/control the levels of widely varying compression rates song by song on a radio show? Add to that the mixing of live vocals… does my fucking lump in.
nice one, brother of toad. Am finding similar things on the iPod, and it’s not getting any better…
See – at least one of us has some useful skills.
You’ve touched on a topic near and dear to my heart, and indeed one that, in some cases, you might have the opportunity to work around.
Firstly, the compression thing – when you’re producing a radio show that routinely segues between current and twenty-five year old songs, which is what I do with my spare time, you trip over this a *lot*. The culprit isn’t digital media, either (though it has its own faults.) The culprit is the record exec who wants to make damn sure their disc isn’t the quietest one in the carousel. It has resulted in a steady escalation of volumes and, I agree, it has resulted in some awful releases of otherwise compelling music.
On the other hand, I like to buy music on vinyl if at all possible, and I find the escalation doesn’t necessary propagate to LPs. The new Springsteen, for instance, sounds scads better on vinyl, and it’s not because of the infinitesimally “truer” reproduction vinyl imparts, it’s because they haven’t compressed the bejeezus out of it. (It even sounds better on the CD I digitized off the vinyl – not louder, mind you, *better*.)
I learned alot from this blog. I copyrighted and produced 5 songs. I did not do anything but sing…but what you are saying makes me want to go back and listen to the quality of my own music.
Thanks for the education.
I am going back to vinyl now for similar reasons. I don’t mind at all having the bulk of my collection in high bitrate mp3 format, but the really good stuff, the stuff I want to sit down with a glass of wine and really appreciate, I would like to have on vinyl now.
…even though it became obsolete fifteen years ago?! *tongue-in-cheek*
Well you know what I mean. I never bought a record player after I left home, and all my vinyl stayed behind, so this will be the first record player I’ve had since 1993. Happy Christmas to Toad, love Toad.
Wonderful, the sort of well prepared, worked at and crafted piece of writing that makes me ashamed that I don’t at least try harder, even if I couldn’t ever have done that. I particularly like the fact that towards the end you can kind of hear Matthew’s voice in the writer’s head saying ‘needs more fucks’.
I had no idea this didn’t translate to record. The problem with vinyl is that it won’t be able to handle the high resolution recording that, I hope, are coming our way. It also though has that natural warmth that is so pleasing, like tape, but better. The reason Dylan got used in this by the way is because I remember listening to those albums when Toad had them on record and we were both nowt but Toad-poles.
I remember listening to records, and being able to hear the difference between tom-toms on drum fills, and between an acoustic guitar and a ‘clean’ electric guitar.
I thought I was going deaf.
Thank you Ben.
Right on. One irony in all this is that when CDs first came in, they were touted as being better than LPs because they could handle wider dynamic range without having to be compressed to fit as LPs were. That was why the first generation of CD remasters sounded so good. Now classic rock is being remastered again with even more compression than on the old LPs. Apparently the new Led Zeppelin compilation is one example (I haven’t heard it yet) – but though Led Zep are regarded as the grandfathers of heavy metal, their music was never just about heaviness – a song like “Stairway to Heaven” is all about light and shade, going from a quiet start to a thunderous crescendo then back to a gentle finish. If the whole damn thing sounds loud, that subtlety is lost.
hm, let’s see. slighting new jersey- what would the Boss say?
couple things in the article i want to comment on:
the difference between ‘Closing Time’ and ‘Highway 61 Revisited’ and the Killers or whatever is way more than the delivery format- you can go grab either (or are they up to three) of the Killers records on vinyl and you will hear very much the same effects you describe. the problem is inherent in the production style of modern music, and isn’t really strictly a by-product of the volume wars, as they’re called. on the dylan record you have a live recording, everyone playing at once- even vocals- all recorded to either 2 or 4 track. tube microphones, tube console preamps, tube limiters, going of course, to tape. and then theres great musicians playing a great song, but we’ll leave that out in the name of objectivity.
things have changed so much in the modern era- even one obsessed with retro sonics- that pretty much, in short, add up to shitty sounding recordings. you’ve got- at the top of the shit heap- pro tools, which sounds mediocre at best. it offers mediocre sound quality at best, and the potential for manipulation after, during, and _in lieu of_ the songwriting process (‘fixing ‘ takes, comping tracks, reliance on quantization, the temptation of effect after effect, the temptation to just loop parts) is the one factor, in my opinion, that makes modern music really fucking suck. you’ve also got isolated recording as a norm- instruments are recorded one at a time, drum hits replaced one at a time and slid into time, bass notes corrected, vocals autotuned, etc. you don’t have paul butterfield’s searing lead guitar, al kooper’s thundering organ, and bob dylan flailing away on a telecaster- you’ve got hair model #3 playing one verse and one chorus, and some pro tools dude copy-and-pasting the Killers’ way onto the charts.
it’s confusing to refer to a Clash tune as a signpost for the ‘Wall of Sound’- for one, Phil Spector scored a stunning number of hits by doing exactly what you decry- jamming the speakers with as much sound as possible. and this was in the early 60s, so maybe that philosophy was always part of the mix (no pun intended). massive compression was the name of the game around the same time at Stax and Motown as well, you listen to those early soul singles and they’re distorted and generally low fi, but the kick and snare are in your face. they were engineered to cut-the-fuck-through on mono AM radio, and they still kick ass today on an ipod or television or whatever.
and seriously, the loss in quality inherent in overcompression is nothing compared to the 10:1 data compression affecting the quality of mp3s.
as for some of these comments:
vinyl is no magic bullet. vinyl, if poorly mastered or played on a worn needle, can sound way way way worse than even the worst cd. records wear out, and as your needle wears out you wear out your records even more- and very very slightly, you barely notice until your record is pretty much thump and sibilance. rebuying and rebuying and rebuying the same albums is a fact of life for the vinyl devotee. i used to be one, and really, well-mastered cds are all kinds of better. sacd/dvd audio leaps over both- when i heard an sacd for the first time, there’s no way i was going back to vinyl. under optimal conditionals it can be better, sure, but it’s not universal.
“The problem with vinyl is that it won’t be able to handle the high resolution recording that, I hope, are coming our way.”
what? CDs have required a downsample from 24bit/96khz and up masters for like, a decade, since those rates have been available on even low grade home studio equipment. the latest version of Sound Forge supports sample rates in the Gigahertz!! if anything vinyl represents a better capture of better-than-CD-quality program material as you could master the record without dithering.
“It also though has that natural warmth that is so pleasing, like tape, but better.”
i take this to mean you’re saying an LP has more of a positive effect on the final product than if it had been recorded on tape? uh, no.
all in all i think you make some good points, but you can’t explain ‘why current music sucks’ in the nutshell of mastering levels and compression ratios. really, there’s bigger fish to fry- namely a public that has thrown any sort of audio quality out the window by embracing the mp3.
Ho hum. Poor Bruce.
The Clash are interesting, and it’s why I said I was cheating when I used them. They were a sort of upper limit for ‘good’ compression, and what’s bizarre is that we have surpassed them several times over in terms of cramming. But, what they did was record a song, them compress the whole. Which means your mix is intact, as is the stereo separation. Now a days you stick an effect on every track, reverb on every track and then on the mix so on and so forth. It’s one of the reasons I used it. It’s a wall, but so much less garbled than other stuff our there today. As I said in the post, the tool itself is good, but with unlimited ability to use it comes unlimited ability to bugger things up.
Two quickies. First, the Bruce Springsteen album I used was recorded in single takes, in a cabin in the woods no less, in single takes but on multiple tracks. So while the audio is a bit still a bit overprocess because everything is treated, it maintains the ‘live’ feel. That’s sort of what saves it.
Second Pro Tools (and Cubase, and Sonar, and Logic…) does not sound crap. The audio quality of computers has far and away surpassed anything that has ever existed up until this point. The amount of information that a high quality mic can send into the computer has quadrupled. This has all sorts of benefits for recording room resonance and tonality. It lacks the same natural compression of tape, simply because if you use them right there is no need. And there in lies the rub:
Because the tools have become more powerful, the ability to use them has become more pertinent. And as we rely more and more on difficult to control technology we are also slicing into fractions the amount we train. Pro Tools makes good engineers better and bad onces worse. If you want tape sound, you can run any audio signal through a tape machine and back into your computer. Done. Or you can buy emulators and use those. The issue is that the imperfections were pleasing, and now they are gone, so you are either leaving perfect audio quality in your mix (not interesting), or retro engineering it in, which takes a LOT of skill (no one is trained), great hearing (we all listen to such loud music no can hear any more), and time (for which you can read ‘money’, and we all know how that ends). The emulator choice hence becomes the one where bad people make worse records. If you are going to go after a program go for garage band that teaches children in the suburbs that they have what it takes to make a record when in fact to master these tools still takes years of practice.
Lastly I take your point about MP3’s. Thing is that the same was said of tape. They are horrific, but as an encryption they will be replaced by something that takes the same room but sounds better. Someone will take that technology and trim to a worse sounding even smaller file it and the cycle will continue. MP3’s scare me, but I would imagine they will vanish. More important is that people don’t only own MP3’s of something, and continue to listen to the CD’s/ vinyle. However, let’s not forget that MP3 technology is very young.
Good response though.
Deary me I’m silly. I forgot to mention the subtle differnce that tape an vinyl are a somethign to imprint a signal on to, where as MP3 plays and CD’s (or DVD’s) are places to store information to be decoded. As such you can stick whatever sample rated signal you like onto tape or vinyl, it still handles it as best it can. A DVD can handle as nice a sound quality as will fit it. I actually play wav. files on my MP3 player, which sound better than the lot of ‘em. So I wasn’t saying that tape makes the sound, tape just led to the discovery of a sound that was good.
why does it matter how well it’s compressed. i think it still sounds good. maybe ur all just touchy faggets.
[...] links the state of the music industry today is disappointing for many reasons, here is one and the other is here. [...]
[...] is something that is only going to get worse, I presume. If you remember my brother’s article on what compression is doing to indie music which I published on Song, by Toad a couple of months ago, it will be most detrimental to people [...]