Music producers mixing for MP3
In a fascinating article about trends in sound engineering, Rolling Stone notes that producers are now specifically mixing tracks to compensate for the failings in MP3 -- it seems to me that as a society, we're happy to sacrifice fidelity for ease of use, flexibility and low-cost (see, for example, the trend from landlines to cordless phones to mobile phones to Skype). Designing for that, as opposed to lamenting it -- is a damned good and realistic thing to do.
Producers also now alter the way they mix albums to compensate for the limitations of MP3 sound. "You have to be aware of how people will hear music, and pretty much everyone is listening to MP3," says producer Butch Vig, a member of Garbage and the producer of Nirvana's Never- mind. "Some of the effects get lost. So you sometimes have to over-exaggerate things." Other producers believe that intensely compressed CDs make for better MP3s, since the loudness of the music will compensate for the flatness of the digital format.Link (via /.)
the latest
latest episodes
In the studio, sometimes bands would ask to hear the final mix through a car stereo, since they assumed/hoped their music would make it to airplay. We used 2 different "systems" for these requests, a woofer/whizzer cone pair and a 6 x 9 3-way system that was popular at the time. Once in a while we might go back and tweak the EQ settings but usually the final mix was left as is.
I heard an anecdotal story that Aerosmith used this "technique" for some of their work.
Don't know if that was true, and not at all sure with such a wide variety of quality in reproduction available now, that mixing for the format will make any appreciable difference to the average end user. For audiophiles, and anyone with a trained ear it will make some difference, but a "lossless" format like .ape or .flac, among others, would be a far better choice when ripping tracks than trying to overcome the substantial gap between the original recording and the .mp3. In my experience, most users today seem only care about 0-11 volume and whether or not the windows rattle from the bass.
See also this Wired article from 2001 that explains how Celine Dion is louder than AC/DC.
http://www.wired.com/wired/archive/12.01/play.html?pg=2
go figure.
sorry, issue 12.01 - from 2004.
"Designing for that, as opposed to lamenting it -- is a damned good and realistic thing to do."
Mediocrity is of course a sign of the times we
are living in, i don't think there is anything to
be happy about that, though.
The phone example is meaningless, a phone call
is something strictly functional, music is not,
it's just beauty.
Bottom line, reading the article i found myself
with the same opinions of the musicians, opposed
to the music industry.
I suggest everyone looks around and se who he's
with, that may be a good way to evaluate his own
ideas.
IFELTCOMPELLEDTOCOMMENTINSATIRICALLOUDNESSWARSTYLE.
ITMAYBEREALISTICTOCOMPRESSANDEQFORTHEMULTITUDEOFMORONSWITH128KBMP3ANDMICROSPEAKERS,
BUTITSUREISPAINFULTOLISTENTO/READ.
sorry, it had to be done.
I was commenting with a -5db RMS dynamic range.
I actually think the phone thing is a very good example. You started out with a landline, technology built over the last 60+ years. When it's on it works, clear and simple. Then we progressed to cell phone. While still getting better, you do get static and now that irritating wh-- d-d -ou s--?
And most recently moving on to voip...where if your kid is torrenting you sound like you are using a cell phone in a farady cage... All that for "ease of use", "cheaper rates", more convience. You know we spent the better part of last century building the phone network, and honestly it works damn good.
If they want to accomodate users of MP3 players, why don't they make CDs that have MP3s included -- ones that are properly encoded and just as good as the uncompressed[*] versions for all practical listening purposes? (As opposed to letting inexpert users rip the CDs themselves with desastrous results; e.g. the default MP3 encoder that comes with iTunes is far from optimal and may well produce the 'flattness' etc. Butch Vig et al. are talking about.) Now that would indeed be a damn good and realistic thing to do. Do they really believe that the MP3 format as such is that flawed? It has its failings, but it is capable of storing practically CD-equivalent audio (again: for listening purposes) in files that are small enough.
[*] 'Uncompressed', here, as in full bitrate, not as in dynamic range compression, which is actually the main point of the above linked article.
I'd much rather have some degradation of sound quality on my MP3s than to have my CDs sound distorted. If I'm listening to music on my iPod it's for the sake of convenience and I understand that it's not going to sound as good as the CD. I don't expect it to. But I sure don't want someone to come along and ruin my CDs just to make MP3s sound better. Then everything sounds like crap. Or am I a dinosaur for even buying CDs anymore?
ive never understood this loudness war. if a track doesnt have compression on it, it instantly sounds unprofessional weak and lame! theres no going back now.
It seems that there is more interest from independents for offereing highter fidelity. Daniel Lanois is releasing his latest album as uncompressed wav files as well as non drm mp3s.
http://redfloorrecords.com/HereIsWhatIs.htm
and places like
http://zunior.com/
http://burningshed.co.uk/
are offering up flac downloads. Not to say that any of these albums are going to be produced with less compression, but at least the sound will be more accurately reproduced.
Hmm. I've always assumed that mp3 was just a passing phase, and that, as storage and processing hardware got cheaper we'd all move to some as-yet-unknown lossless format.
@9 "ive never understood this loudness war. if a track doesnt have compression on it, it instantly sounds unprofessional weak and lame! theres no going back now."
But now that most prefessionaly produced CDs sound the same, there is an emerging market for analog studios, and I have had clients ask for "that 70's" sound. Just like once vinyl records gave way to CDs, digital samples of vinyl scratches and pops were produced to add that "warm" sound to digitally produced music.
Another way that engineers are changing the way they mix for today's clients, is that they are adding a high-frequncy boost in the kick drums so that the kick can be heard through notebook speakers.
This is hardly news. Producers, or at least, good producers have always kept a crappy speaker on the mixer desk. When trannies first became popular (transistor radios!), they had to mix the pop single to sound good out over a crappy AM signal and out of a crappy little speaker. Wow, you've discovered something people have been doing for over 40 years.
Hey, Michaelangelo, most people are just going to see that ceiling in magazines and art books so don't bother putting any detail into it. Oh, and that statue is way too big. Most people ar just going to see little replicas. Might as well make it a foot tall.
Art is not VOIP. It's not a faster car. Working on a smaller lossless audio format is a "damned good and realistic thing to do," not halfassing the recording process.
Normally i agree with a lot of your leanings, but this was just ignorant. Since when is art about convenience? Maybe you should reevaluate the music you listen to if you don't mind it sounding like tinny garbage.
"we're happy to sacrifice fidelity for ease of use, flexibility and low-cost... Designing for that, as opposed to lamenting it -- is a damned good and realistic thing to do."
Not sure that I can agree with this blanket statement. What about HD video Cory?
People are tossing aside their old TV's to make room for the new increased fidelity of HD.
Cory, designing music for MP3s is ignorant. It's like designing movies for VHS - sure, works great in the short term. But then Laserdisks come out and your movie doesn't look any better, and then DVDs come out.... and then.....
If nothing else, it's like back when LAME had a "hiccup" of a particular frequency when encoding at 160. If you designed your albums around that, your album would be forever crippled.
@JJJJ...
That Wired article also explains the damnable producers of television (and probably radio...) commercials.
Folks complain about them being louder than the programs they buttress, and we keep being told that there are FCC rules in place about how "loud" they're allowed to be, but anyone even tangentially acquainted with audio engineering knows that they are often compressed and limited within a few microdecibels of being illegal.
This annoys the living piss out of me, and millions of others as well, I'm sure.
I saw some interesting comments not long ago from the guru of sound, Brian Eno, about the inevitability of mixing music for mp3. His take was that music evolves along with the most popular storage medium -- that music before 1985 or so was mixed to sound best on vinyl, for example (which is why the first CDs sounded so terrible... you were listening to music designed to be played with a record player).
Designing music so that it'll take advantage of its storage medium and playback method (digital music and small earphones) makes sense.
Despite the negative press, mp3 is surely a great thing for music as an art form. With a few mouse clicks listeners can now check out a great new local band on Myspace, listen to the artists that influenced them, buy their music, send the tracks to their friends. In bypassing the traditional financial hurdle of physical distribution costs, mp3 has heralded a potentially more meritocratic age in music.
A far greater problem, in my opinion, is the labels themselves: signing celebrities who can't sing and sending them into a studio for 2 weeks to record songs even the songwriter doesn't like, that are then tweaked out of all recogntion by an under-paid producer with Autotune. They sound bad when they're mp3'd because they sounded bad in the first place!
To some people music is apparently art. Good for them, they can go and listen to their uncompressed wavs over their deoxygenated wires and la-di-da twaddly nonsense.
But to me and I suspect the majority of consumers, music is not artistic, it is functional. Its function is entertainment, just like movies and sport and comuter games.
Why are MP3s popular? Why does nobody listen to lossless formats? Because MP3 is more functional.
I am puzzled by the skype analogy though: unlike all the other comms methods, Skype gives decent fidelity rather than sounding 8-bit. The difference between talking on a telephone handset and a decent voip system is quite marked.
I consider myself an aficianado, though I'd never call myself an audiophile because most of the time, those people are idiots. ;)
But who the hell uses 128 bit encoding anymore, other than 8 year olds? When I rip a new CD for digital use, I still use mp3 but go with a high bitrate -- 192 for years, though recently I've started going up to 256 after seeing how nice Amazon's files sounded. (Yes, I still buy from them, Cory... they're the lesser evil than iTunes or the rest. ;)
And always always use standard stereo rather than joint, and always a static rate, not variable.
What we're really talking about is mastering for different formats, not mixing. And it makes perfect sense. When mastering the audio for vinyl, you need to carefully control the bass or the tone arm will fly. When mastering for CD, you're trying to manage the dithering and samplerate conversion, etc.
Make perfect sense to me that someone would master differently for an MP3. And there's no risk as technology changes. Mix to get a good balanced sound, then master as appropriate. If the technology changes and you need to revisit things, remaster it again.
jw
As a self confessed anologue whore, this is End of Dayz talk, seriously, the whole point with music is to make something that sounds amazing, to deliberately over produce something to allow for poor playback media hardly encourages devolpment of said media or allows people to experience music the way it was meant to be.
Down with this type of thing!
ok, #20, that makes perfect sense. I am standing down my analogue armies.
gets coat>
M2KEY makes an excellent point. Listening to MP3s hardly marks "the decline of civilization."
A 128K on an iPod was far better than what most people were listening to regularly before then. The hits of the 50s and 60s all became hits for people listening to AM RADIO! Limited by regulations to 5KHz high-end.
Ever heard a collection of vinyl 45-rpm records? Most people played them on machines that quickly chewed the tracks into vinyl dust. And kept listening for years.
128K MP3s aren't that far off the mark. In the typical listening environment, you're not hearing the highs ... and not caring. Most speakers for apartment dwellers these days -- including the expensive ones -- are bass-retarded. Seen many 12-inch woofers? And the audio at most concerts is still far from high-fi.
ugh, storage space and bandwidth have become cheap. We should be moving to flac files, not taking all the dynamics out of the mix for mp3s.
Whether the story #1 relates bout Aerosmith is apocryphal or not I don't know, but I've heard the same about The Raspberries back in the day, concerned not only about how their records would sound coming out of a car stereo, but how they'd sound broadcast over an AM transistor radio, to boot. To me that always sounded smart as hell. Besides, they are not Michealangelo, they are The Raspberries! Big difference.
All I know is that I love listening to mp3s of pre-hi-fi music. If it sounded good enough for AM radio, it sounds good enough for my iPod.
mp3 - as all compression schemes - satisfy the need to reduce bandwidth and storage consumption in part due to a demand for portability
as these constraints fall away (due to bigger and cheaper transmission/storage media) and no longer pose an impediment we will see a proliferation of the higher fidelity formats
imagine 96kHz/24bit instead of mp3
I know I do
Strangelow @ 20 makes a great point and there actually is a perfect video analogy (or three). Before DVDs the vast majority of the home video market came in a 4:3 aspect ratio but major motion pictures were all wider (1.85:1 being the most common flavor). The translation from rectangle in theaters to (almost) square on TV was an ugly process of re-cutting and pan & scan techniques. In the 80s some filmmakers (Coppola and Spielberg in particular) started to compose their shots in such a way that the reduction to a 4:3 frame would not be quite as devastating, realizing that most people would be viewing the films that way ultimately. Was that compromising their art? Maybe, but since it improved the film-watching experience for the more numerous TV viewers, it was a net increase in quality. The films still looked great on the big, wider screen, they were just optimized for the subsequent conversion process.
Anyone who uses a TiVo now is seeing a digitally recompressed version of their TV signal with a noticeable degradation in image quality, but the benefits of the device are so powerful, it's a fair trade off, IMO.
On several occasions, I have commented to friends that the death of the audiophile (along with the death of the public pay telephone) is a sign of technological regression. It is my opinion that the quality of sound recording has regressed to the level available in 1955. (The quality of pay-phone technology has regressed to the level available in 1925.)
Is there any excuse for producing for only one format in today's age of digital recording, mastering and editing? Can music be recorded once and remixed or remastered to accommodate the strengths and weaknesses of vinyl, CD and various digital formats? If so, how much more would it cost?
Strangeglow nailed it.
Forget about how long mp3s or any other mediums are going to useful... The music is being released in the here and now, so we work with that.
When mixing an album, a good producer should always keep the intended media in mind.
If I was releasing an album on vinyl, cd, and online, I would definitely have three different mixes. It's just subtle tweaks, but if you were buying the album on vinyl, you'd feel a little gipped if it was mastered for for cd sound and not analog.
Most people can't tell the difference between MP3 and uncompressed audio and after a year or two of abusing their ears with ipod ear plugs they loose even more of their ability to hear high frequency audio.
MP3 isn't the problem. The listeners and what they want is the the problem. There are also some other problems like shit headphones.
Go get a CD that was mastered in good quality with proper dynamics. There are lots of them out there, but Steve Hoffman (of DCC fame) masters pretty much everything really well. Rip that CD with a good program like EAC. Encode it with a decent MP3 encoder (like LAME) as a 192kbps or 256kbps joint-stereo MP3 file. Play it back through good equipment. You will find music that is WAY more detailed and dynamic than what these producers are creating.
The file format has nothing to do with this issue. Nothing, period. I'm not an enormous fan of the concept of lossy encoding, but I'll defend any scapegoat when I see it.
"Producing for MP3" has been around for a while.
I'v heard it from a music producer, I can't recall his exact words, that it's bullshit. He said he wouldn't know how or what to do specifically to "produce for mp3". The Rolling Stone article only talks about boosting the volume. If that's what they mean, okay then.
I've also read a study that came to the conclusion that practically _all_ teenagers these days suffer from hearing loss.
(Like losing eyesight, you don't really notice, untill you're fitted with proper glasses.)
Which makes me wonder where this is going to end.
As others have said, it isn't designing for 'MP3' but more for the lifestyle associated with them. Loud mobile places that need the audio (not data) compression.]
That said, the vast majority of musicians I've worked with that weren't complete nerds have embraced MP3s years ago. I can safely say, having sat in on a few highly influential sessions of the last decade, the minute the tape starts rolling back what was played, you've already lost. I've had ear drums nearly burst standing on the side of a stage listening to bands and heard the thing played back as a live concert recording and was disappointed. Even recording 192khz 32bit with the best mics and the best engineers -- the sound pales next to what you hear live.
I sat in with a friend a few nights ago who has a speaker setup worth in the hundred of thousands of dollars (he himself is an up and coming designer in this industry and several pieces were still under construction waiting to replace the last set). Sure it sounded great, but for my money I'm going to rely on imagination because nothing is EVER going to reproduce what I've heard in person.
Beyond that, music isn't the audio that you hear, it is and has always been what is in your head. It isn't the ones and zeros, it is the song writing and the performance. A great performance can be captured on a portastudio and still be relevant 100 years ago. I have audio from old wire recordings that I've digitalized for posterity. A great song writer isn't what is on tape, it sounds good and is just as relevant on paper.
I could care less about MP3s or the loudness wars. Great music doesn't suffer from either. And shitty music will be just as shitty on this medium.
The following happened about 5 years after the introduction of CDs.
A friend of mine ordered the CD version of a favorite album, because he wanted the absolute best sound available. To compare the two, and to see how much better the CD was, he first played a selection from the vinyl album. Then, he played the same song on the CD, on the same system with the same speakers.
In comparison to the vinyl, the CD sounded thin, insubstantial, kind of transparant, no "oomph". The record had sounded rich, solid, and more "there". We were puzzled as to why the more advanced CD with it's greater capabilities just didn't sound as good as the record. After reading the article I'm starting to understand why.
Hi-fidelity's ass.
The very best of rock'n'roll was engineered for a cheap dashboard speaker... and it can be again.
We NEED a Phil Spector for the earbud.
...now let's just make the damn music a little better and forget about if I can hear a pin drop.
War is over if you want it,
Mike
Sorry for the very long comment, I hope to clear up some misunderstandings here.
Their parents, and everyone else who happens to have software with modest default settings and has doesn't change them. There's a great number of such people; most users simply have no clue what bitrate, sampling rate, bit depth and quantization are exactly. Luckily some GUIs for encoders offer a slider that simply goes from "OK audio quality, small files" to "excellent audio quality, larger files", but when all you want to do is listen to your CDs on your crappy computer speakers or MP3 player, you don't have an incentive to even crank that slider up.
People who prepare rips for P2P are a different story, but I can't imagine the producers quoted in the above-linked article were thinking about P2P. At least that's the charitable explanation; the alternative is that they're idiots who really think that when some 1337 warez d00d prepares a 256kbits VBR MP3, it'll sound any worse than the original CD.
Could you please explain why? Joint stereo and VBR both are tricks to make files smaller at the same sound quality, provided your encoder is up to handling the slightly more sophisticated encoding technique. Are you saying that the developers of, say, LAME are so incompetent that they can't implement that correctly, or just that your MP3s sounded worse when you played with these settings? I suspect that you'd overlooked something and didn't do a fair comparison. For example, when you encode two separate channels at 200kbps each, you've actually got 400kbps, which may well sound superior to 200kbps joint stereo in demanding situations. But I have absolutely no idea what could have caused you to find CBR better.
Yeah, right, when the SACD format became available, all the studios remastered half of their catalogs, and all the consumers ran out to purchase SACD players and all the albums they already had ... That's nonsense, music lovers who care about audio quality mostly listen on OK systems that sound pretty much as the band/producer intended, and they expect the mastering to complement the music instead of being horribly distorted. On the other hand, the 'mastering tricks' the Rolling Stone article talks about are directed at the much larger audience of people who just want something they like playing in the background while they work, drive, play, read, jog, under suboptimal circumstances. They're meant to boost short time CD sales. In ten years, no one will say, gee, I sure wish they'd remaster that old Avril Lavigne album for holosurroundovision! And to the people with OK stereos, the stuff they're ripping to Flac right now will still sound just as good in 10 years. Your advice is good, but you've fallen victim to the myth that is being perpetuated here: That dynamic range compression, the main subject of the article, has got anything to do with mastering for the specific failings of the MP3 format (as opposed to the failings of the systems through which MP3s are typically heard; which don't change simply beacuse you switch to Flac). There is absolutely no relation. MP3s handle dynamic range just fine, especially the kind mentioned in the article: having different parts of a track at different volume levels. MP3 handles that perfectly. That is a fallacy that is widely spread, but it's still wrong. CDs and digital formats can handle any audio signal with a fidelity that is good enough for most listening purposes. Vinyl has some limitations, but they aren't all that great for a brand-new record on a good system. Mixing for a worn-out, dusty record playing on your grandparents' stereo is silly, as you're destroying the sound for all the listeners in the aforementioned situation -- and these are the ones that actually care about audio quality! What's at issue here is not the fidelity of the different storage formats, but, for example, as another poster mentioned, that you can't make out low kick drum sounds on tinny notebook speakers. You can compensate for that in the studio, but the storage format hardly enters into that equation. The fallacy is this: When producers listen to their tracks on a bass-tard car stereo or a 10$ Chinese MP3 player, they just do that to check how it sounds. However, they don't do their actual work in that manner. They use amps and speakers with the flattest possible frequency response, so they can hear what the audio actually sounds like. It is actually not possible to deliberately use crappy speakers in order to make the music sound OK on all crappy speakers. Instead, using crappy speakers to do your work means you can't hear all the details of what you're doing, and that leads to a worse sound. You make mistakes without knowing it, e.g. you'll never catch that nasty 50Hz hum that'll annoy the heck out of all owners of good systems who actually care about fidelity. But, on the other hand, not all sub-par speakers are flawed in the same way, so, apart from degrading the quality of your recording, you're doing nothing about increasing anyone's listening experience, except for the tiny minority of listeners who happen to have the same crappy system you used. Please please believe me when I say I'm not trolling; I'd like to say that I doubt you're understanding much. A CD can reproduce anything a vinyl record can reproduce. And 'about 5 years after the introduction of CDs', the loudness wars hadn't really started yet, so nothing in the Rolling Stone article applies to your tale. So why did the record sound better to you? Maybe that particular CD really was badly made. A more likely explanation, something that has happened countless times, is that the CD was cut from the same master tape as the record, and much closer to what the engineers in the studio actually heard than any record ever was. That is, when you listened to the CD, the fidelity really was greater, all the 'failings' of the record, such as its horribly biased frequency response, and the distorsions typical of vinyl, were gone. And here's the thing: Many people are used to the 'vinyl sound' and like it. And that's fine. If you say the record sounds better to you than the digital release, then it just sounds better to you, period. However, that doesn't mean that digital sound is inherently less capable of delivering 'richness' or 'solidity'. It's just that the producers had valued 'transparency' more and, with the CD, had finally had the means to deliver that to the listeners as they intended it. These endless 'digital vs. analog' arguments in the 90s were just silly.Why not leave all the twiddling to the rendering device and leave the source media intact? It seems that the content could be mastered to some reasonable level of fidelity and let the device and its environment sort out the details. Back in the day, they even had a button for it and what a coincidence - its label was "loudness".
@ BRAVO: Hear hear! (sorry)
It sure makes sense to me.
I understand that some people want to argue that recording for inferior sound quality cheapens the quality of artistry, but the "artistry" involved in making music is best heard live, not recorded. In fact, you could argue that any recorded music is a degradation of artistry.
Old news: http://online.wsj.com/article/SB118953936892024096.html
I've noticed a big difference in the way the more recent CDs sound compared to how they sounded when they first came out in the early 80s. Heck, I still have the first CD I bought back in '84.
I rarely listen to things through earbuds or cans. Instead, I play stuff through my 21-year old analog class A/B amp, into a pair of equally old Klipsch kg-4 speakers.
Bliss.
I wish the producers and engineers would quit over-compressing their music, and leave the tweaking to the ones who want to tweak on it. I can hardly stand to listen to music produced today, because it's just noise to my ears.
Do I miss vinyl or tape? Heck, no. The creation of the CD was literally music to my (still) sensitive ears. I stopped using the equalizer to 'fix' what I heard when CDs came out. But no equalizer can fix the over-compressed music that is foisted on us today.
I noted that the stereo on my new car has a compression mode to compensate for road noise. It helps- a bit. But vehicles are a very hostile audio environment. My old cassette deck even had a 'car mode' in its recording selections to compensate. It's too bad that with all the advances in recording today, producers couldn't offer us that sort of selection, so those of us who like to listen to music in ideal environments can really enjoy it.
We're at a stage at the moment where we have quite got enough bandwidth. Digital broadcasting is also at fault here: both TV and radio are broadcast at remarkably high compression rates. Radio 4 as a speech station is broadcast at 80kb: the commercial station Oneword is 64kb. Music stations are encoded at 128kb, which is the maximum bitrate currently available on the network (the original intention was for 192kb or 256kb but economics prevailed as usual.
Similarily the quality of many DVDs and now indeed HD and Bluray discs is just plain bad. It could well be that the things that I buy, particularly old British comedy series, are either badly transposed to digital or compressed heavily to fit on DVDs but it's not reassuring to put a commercial DVD in a decent player and see blocks flicker across the screen. I saw a report of the HD version of 'Casino Royale' on Bluray which said that the print (or whatever you call it) looked garishly bright and artificial on a 40 inch LCD screen and wasn't a good advertisement for the format at all.
We accept this: MP3s are convenient because we can carry our music collections around in our pockets, which is fine, but we then start listening to the MP3s on our nice stereo systems (guilty as charged there - I want a hifi separate with a big hard disk that will play FLAC, and one of my part time jobs this year is to re-digitise all of my CDs) and it becomes the norm, so music starts to be recorded so that it stands out or 'sounds punchy' at 128kb like records in the 60s were made to sound good on Dansette record players or in the 80s sprawled sonically to impress the emerging CD market.
One of the better results of the growth of the DVD format has been the way in which it makes available archive material that would have been otherwise left in libraries, however, limitations in processing power mean that the transcription process has to be very meticulous or the resulting DVDs will look bad. This won't go away with Bluray/HD. Similarly if IPTV picks up, we will spend a few years watching degraded video because there isn't the bandwidth to distribute it: the producers' response would be to simplify sets and the compression methods to simplify motion.
Hopefully Moore's Law does still apply and the bandwidth and processing power will become available to improve fidelity but I think the point I'm trying to get to here is that we adapt personally and culturally to our inputs and that we respond to the limitations by working around them, so hopefully dynamics will come back again. It was interesting that the article singled out the Arctic Monkeys and Lily Allen LPs: they are both particularily loud LPs, Lily's to the point of distortion in some places. I'm glad I'm not a young person any more, I don't have to listen to them unless they're on the radio.
[#40, Nex]Thank you!
I was about to make a couple of the many points you made, but not as eloquently or succinctly. You clearly understand audio and compression.
Yes, MP3s can deal with dynamic range. :)
@D3:
One of the issues with early CDs and their "thin" or "cold" sound was the nature of the early filters used to prevent strange noises from entering the recording under certain conditions. Since CD data is a pair of arrays of 16 bit integers moving at 44.1 KHz, if any sounds higher that about 22 KHz get into the recording, the sound recorded is half out of phase compared to the original and sounds lower as the frequency moves higher.
Therefore, a low-pass audio filter needs to be used to prevent these artifacts, known as aliasing. Trouble was that the hard cutoff of some of the early filters used introduced their own artifacts into the recordings, thus the tinny sound of some of the early CDs.
Another quick misconception to clean up (now that no one will be reading this post any more):
The loudness button on many stereos is not magic, or in any way related to the aforementioned 'loudness war': it is a high and low shelf with anywhere from +6 to +12dB gain meant to compensate for the Fletcher-Munson (et al) Inverse Equal Loudness Curves (hence "loudness") by boosting treble and bass so that material can be more easily perceived at low listening levels.
Granted, you say "let the device and its environment sort out the details" but this only works for low listening levels because the Curve disappears (or, flattens, rather) around 80 or so dBSPL for most people and most stereo systems do not 1) know when they're throwing 80 dB SPL at you or 2) change the gain of the loudness function's filters as voltage through the system increases, so if loudness is on and you turn your stereo up because your roommates aren't home or rock music is better loud or whatever, the bass and treble gain is still in place and will distort.
So, while I agree that the source material should remain "intact" and producers/mixers/mastering engineers should have fidelity and artistic integrity being their highest concerns, since 'loudness' and other 'enhancement' and end-user 'mastering' schemes (like iTunes' harmonic excitement and volume normalization (through dynamics limiting) grossness) are pretty passive things, they are definitely not solutions to making better sounding MP3s (which I guess is what the producers in TFA are aiming at). They are last ditch efforts, like using umbrellas as parachutes.
The only solution is knowledge and unfortunately digital audio is scary/technical for a lot of people.
re #32
Sometimes different mixes are done for different purposes. The same track might be mixed differently for normal release than for inclusion in a videogame or film. As long as the client is paying, the engineer will make as many mixes as they want!
re #40
actually, one does have to mix (or master) differently for vinyl than for digital due to how records work physically. See http://www.urpressing.com/advice.html and http://www.urpressing.com/tips.html
Also, I've mixed & produced alot of music in my day and can attest to the crappy speaker check.
No one here has yet mentioned the "club speaker check". Back in the day that meant getting a dubplate of your tune cut and taking out to local club DJs to hear how it would sound, lately I've seen people do as many as 10 different club mixes since burning them to CD-R costs almost nothing.