The importance of mastering and the masterer.

I've assumed that he thinks the final record is overcompressed for radio-friendliness. I don't think the album really sounds that compressed, but since I haven't heard it before mastering, I don't know. I can IMAGINE "heart shaped box" sounding different w/out the need for remixing, though; the chorus really roars in. maybe it was quieter.

anyway didn't someone work on the mixes post-albini?

Mastering Engineer

http://www.nirvana2.com/system/archview.php?thread=5788&forum=4

So my guess is you don't hear a lot of raving about mastering because that would be like cheering the high resolution of the type on the cover.

I heard Rape Me the other day for the first time in years and was struck by how dynamic it seemed.

And not to the engineer who "polishes" the mix into a high resolution, so to speak?

The example of Bill Ludwig is a good one, and the idea that a perfect mastering maintains the artist's vision.

It just strikes me as odd that someone who (I'm assuming) didn't have any direct interaction with the artists and the creation of their music could make those kinds of decisions, and actually affect the final version of their art. What if they wanted those imperfections or undesirable artifacts?

I would say "pft" but I remember I heard some of the unmastered versions of the In Utero songs on those multi-volume Nirvana boots that were everywhere back in the 90s and was shocked by how much more dynamic and loud they were. I think HSB was an exception because it was remixed by Scott Litt with the expectation that it would be mastered without headroom and so the chorus still hits hard. Albini presumably mixed it with the assumption that it wouldn't be compressed all to hell and so there was supposed to be a much greater absolute contrast between the quiet and loud parts.

The one mastering faux-pas that still makes me sad to this day was the awful awful weak ass mastering of the first album by Last Of The Juanitas. There were some mp3 leaks of that album that sounded incredible, so I assumed the LP would be a megawallop, and it sounded like a warped budget cassette on a dying boombox. That probably got fixed with their second album and beyond thankfully, but what a way to kill the momentum of a great rock band's first release.

Ideally, artists and producers would have a whole lot of leeway in picking mastering engineers who were sympathetic to the style they were shooting for -- and some modicum of input in terms of telling the mastering engineer what their intentions are. Even leaving aside the aesthetic element, doing it any other way means a lot of wasted effort, miscommunication, and inefficiency: there's no point in having an expensive producer log hundreds of hours shooting for X sound, and then having an expensive mastering session log hundreds of hours trying to turn it into something else. It makes for a worse product either way, because the producer may be making decisions that won't work properly with the mastering, and the mastering engineer may be working with raw material that wasn't made for his/her purposes. It's like having someone film a romantic comedy and then trying to turn it into a thriller in post-production.

But most major-label artists tend to be in the same position, mastering-wise, that a screenwriter is with regard to who'll direct: the people making decisions will pay for the guy who has a track record of hits and money, and that'll outweigh any considerations about who's best suited to the work. I don't doubt that every mastering engineer in the world goes out of his or her way to try and work in the spirit of the material, but if the labels that are paying them decide that mastering is an opportunity to "fix" stuff they don't like about the mix, or make something more accessible than it was before ... it's just not likely to turn out as well.

While I love the new Beyonce album B'Day, this album was mastered such that it sounds best when encoded to Mp3/M4a then listened to on a iPod -- and probably not much more. The bass beats in the raw waves are too loud and distorted, but when encoded, most of the distortions go away, and then the iPod compression obfuscates what's left.

Exhibit A of a mastering job where marketing dictates all from start to finish.

An honestly strong reason to not sign to a major label for folks who want complete control of their music... Sure, "artistic control" may be common enough or was back in the 90s, but there are many other peripherals to the major label machine that bands forget -- technically oversight of postproduction being one of them. Nabisco explains this gracefully above.

Incidentally, that goes for artists and A&R and producers and everyone else, too. No matter what kind of art you're making, it can be really hard to know, in the middle of the process, what you're shooting for, and what's working, and what kinds of decisions you want to make. That goes double for something as detailed as recording: at some point you've been hearing this music for weeks and weeks, and someone's asking you whether the cowbell should sound like this or like this, and it's got to get hard to keep firm on what you're looking for in the big picture. So I'm sure there are plenty of instances where everyone involved comes around to the mastering process and starts second-guessing things, or trying to find opportunities to turn around things they don't like, or whatever. And that's how you get these weird recordings that sound like someone built a perfectly good hearse and then, at the last second, decided to paint it red.

-- Geir

actually i wouldn't mind knowing more about why Geir thinks this.

I'd imagine so. I think AleXTC means "exciting, fast, brash, rhytmic", as opposed to "the distance between quiet and loud" or suchlike. And the reason that rock/indie for the last five years has sounded "less dynamic" to AleXTC's ears is probably because it's being mastered to try and compete with hip-hop r'n'b, and thus losing it's actual real dynamics in the process.

http://scotthullmastering.com/

edie brickell, skinless, he'll figure it out.

True, but there are two main reasons why it doens't. 1; Few bands are happy to keep it minimal or sensible enough not to go overboard with overdubs, string sections, multiple loops and synth parts, and 2; it fucks up natural recorded room acoustics. Conversely, it's the reason why 3-pieces often have a really full, loud sound without losing it, because they're stripped to just guitar, bass, drums and vox - Nirvana, early Stereophonics, pre-total-bombast Muse.

the thing is, if you keep rock's arrangements/instrumentation minimalist (beat/bass/guitare or synth and voice) there's no reason it couldn't work, is there ?

Nope. Somewhat unintuitively, a mix will generally be perceived as louder and bigger when it has fewer elements. Mixing is a zero-sum game; you only have so much space (volume, frequency) to fill up. If you fill that whole space with two or three things, those things will sound huge. If you fill the same amount of space with 30 things, they start to compete with each other and sound tiny.

Anyway, the effect of mastering can be minimal or it can be extreme. The importance of mastering is proportional to the amount of tweaking done by the mastering engineer.

I don't have time to go into it, but the difference in the mixing and producing is that hiphop is meant to be public music that is played loud and is written, mixed and mastered to move a speaker first and sound good in an audiophile sense second. All the sounds are meant to work a bassbin or a trunk speaker box.

A "good" recording of an acoustic guitar does not do this. When you try and make an acoustic guitar or any other instrument do this artifically it doesn't sound all that great.

If you write and mix electronic signals/sound for hard limiting, it can sound good. Trying to get a mic'ed room sound to work the same way gets mixed results.

So surely some artists have cottoned on to this and are deliberately trying to make more subtle records? The new Sonic Youth sounds dynamic and full of space to my ears, and Neil Young's Living With War is very noticeably quieter, especially considering it was released in 2006.

And what happens to fidelity now that ipods (and playing ipods through stereos) + purchasing music on-line as mp3s is becoming the norm (for joe public anyway)? Is it something we live with until smaller file sizes are achieved with even greater bitrates?

Whereas commercial rock seems to be experiencing a bit of a crisis in terms of figuring out what its production and mastering aesthetics really even are: there's a definite modern style, but it seems fairly ill-suited to what lots of rock bands actually do and play, and it's not entirely clear how much people like it or not. (Obviously plenty of people are fine with rock's current compressed-blaring-overload sound, but I don't think it's necessarily evident that that style is a big success.)

Well, there are different meanings of "bigger." You can have a solo violin playing a line, or you can have a section of 16 violins all playing the same line in unison. Imagine listening to each one at the exact same volume. How do they sound different? When you hear the violin section, it might sound "bigger."

But imagine a recording with a drum kit, four to six guitar tracks, bass, three or four vocal tracks, plus a string and brass section, for instance. Each element is going to sound "smaller" than it would if you just had the guitar, bass, drums, and vocals taking up the same amount of sonic space.

Fun fact: Until the late 70s, there was one mastering curve--that is, the sort of default EQ and compsression settings--in the US and a very different one in the UK. Plus, UK studios were still using these fabulous old RAF radio compressors.

the UK one rocked like crazy but I'm sure Albini would hate it, as the "fidelity" or dynamics or whatever vague terms you wanna use was qustionable at best, but God, that stuff sounded great, in your face and crunchy. Think UK era Sparks vs US Sparks.

The worst mastering I've heard in some time in on Low's "Trust". It's like a textbook example of oops. Someone took discrete parts of the bass and just gutted them while leaving in this troublesome sub-bass thing that makes the CD rumble when you crank it, and makes it sound too quiet when it's at a reasonable volume. There's no proper way to listen to the damned thing (and I love Low.)

Aside from classical CDs, thee's no set-in-stone way to master (aside from avoiding obvious screwups)--there's only the effect you're after.

It stands to reason (and empirical evidence!) that the same kind of thing is happening with rock/indie music. Kids grow up listening to anything from Nirvana to Zep to The Beatles to whatever else, on dodgy radio broadcasts through inadequate radios, on crunched and squashed MP3s through earbuds with shocking frequency responses, and they assume that this is how music's meant to sound, because as far as they're concerned, it's how it DOES sound. And who can blame them? The only thing is that once you degrade the quality at point of manufacture, because why keep it high, and then degrade it once again at point of consumption, you end up with Keane, or The Killers, or whatever.

I'm not sure what you mean. Compression = perceived loudness. You can't literally make CDs louder - 0 dB is the limit, and CDs have been peaking at 0 dB for a long time. Loudness is mostly an issue of how close to 0 dB the level sits on average.

http://vassifer.blogs.com/alexinnyc/tinnitus/index.html

Mr Sick - IEMs? In ear ...?

Continue to drift off topic but I have some shure ec2s and an i river h120 - can never get it loud enough. Care to advise?
Are my ears fucked?

Loved your Earl Brutus account Mark E, btw. Heady days.

This is really interesting; anyone care to point me in the direction of more info on this subject?

Also: what are some examples of big, busy, multi-instrument recordings done well?

Brian May's guitar orchestration thing was preety strictly based on comnventional orchestration frequency use. So he'd have a couple harmonized guitars tweaked like crazy EQ-wise to eat up the space where the reeds might be, have a hollow, flanged guitar in a flute register and so on. Roy Thomas baker--and the engineer, I forget his name--were brilliant in how selective they were with the drums--the attack on the toms, for example, was painstakingly NOT in any range too-used by guitars.

Unsurprisingly, you hear a similar technique used on the newish Muse. But despite all the really neat, helpful things digital analysis of the frequency spread can get you, the engineer in this case can't quite find a space for the drums, which sound a bit lackluster, what with the zillions of guitars eating their top frequencies.

It's no coincidence that one of the moments on the CD that sound truly ginormous come at the end of "Nights of Cyclodia" when it's just this hugeass guitar, bass and drums and room sound.

I love nabisco.

I guess Muse suffer from the current trend of too much compression too.

trust sounds better than the one that came after it (great destroyer) which is way, way too bass-heavy.

anyway I just sat in on the mastering of my own album yesterday and it was a bit revelatory. I have to say when I listen to the record now, I can't say I can necessarily tell that much difference, but when I a/b it next to the unmastered mixes, I sure can; levels are more standard, the highs are brighter (except in a few cases where they were way too sharp before), everything is warmer.

1) A lot of this talk about mastering still kind of paints it as a 'dark art' where people don't entirely know how it's done or what tools are used. So, what does a mastering session look like today, and is it vastly different from what it looked like 30 years ago? Is it a guy sitting in front of a mixing desk tweaking dials and faders, or using some computer program with a ProTools-type interface, or something else entirely? I have a friend who used to have a home studio and is about to open a small professional studio, and I know he's done his own mastering for some self-released CDs, and from what I understand he just does this with a program on his home computer (after transferring the files from a DAT recorded with a proper mixing board). Can an educated amateur who knows what sound they want do as good a mastering job at home with a high-powered Mac as a big time major label pro? Does it take an assload of money to reach a certain plateau of mastering quality?

2) Geir's crack about remastering recent over-compressed albums in 20 years made me wonder: exactly how good are new remasters of 60s/70s albums? Those kinds of things always get rave reviews, but that seems to be based more on the fact that the music itself is good and there's some cool bonus material and liner notes and it sounds better than the critic's old scratchy vinyl or dubbed cassette bootleg from when it was out of print. But to an audiophile who doesn't immediately value vinyl over CDs, do they really sound great, or do they tend to be as over-compressed and needlessly digitized as something recorded today?

1) I gather from the research I did about this that there are two schools - people who do it by ear, and people who do it by visual waveband. Pretty much everyone uses software as opposed to analogue equipment, but not necessarily exclusively. There are, I gather, mastering software suites that don't require you to actually listen to what you're doing at all. Which seems scary to me.

2) Some remasters are great - Love's Forever Changes or the Can ones, for example. Some are loud but OK - Talking Heads - and some can just be nasty - Heaven Or Las Vegas I found unpleasant.

engingeer opened up 24bit aiffs in protools. those were run through a variety of outboard gear, a digital/analog convertor, a compressor, a digital eq, and a huge analog eq, and something else, can't remember. he listened to the song, tweaked a bunch of settings,and then played the song again with the settings set appropriately, recording it to peak. when it was all in peak, he did things like adjusted fades, set spacing b/w tracks, did the sequencing, then burned a cd. the end. it took four hours.

Disclaimer: I have a lot more experience with mixing than mastering.

The thing is, "mastering" is a vague term. It may or may not involve applying any number of tools. It might be subtle, it might be extreme. It simply refers to anything done to the stereo mixdown of a track before it gets to production. Note the distinction - mixing is done with the multitrack data. You can work on the drums, the guitars, the vocals, etc., separately. Mastering is done once it's all been mixed down to a single stereo file.

If you're doing everything digitally, as is common these days, there's nothing weird about it - you're just using ProTools or whatever and applying plugins to the audio. Sometimes they will be some of the same plugins used during mixing, but there are also specific plugins that are meant for mastering. The most common mastering tools are compressors/limiters and equalizers. Usually the compressor or limiter is multiband, meaning it breaks up the audio into discrete frequency bands and can apply different compression settings to each. Keep in mind that the mixing engineer has probably already used compression (maybe lots of it) on the individual tracks in the mix, and now the mastering engineer is using it again on the mixdown.

A mastering engineer might use a spectral analyzer to visualize the frequency distribution, and will probably have many sets of monitors and listening environments in order to fine tune the audio for its intended application. Mastering also tends to involve finalizing the track order, the transitions/pauses between songs, etc. But basically mastering is what you make it.

Geir's crack about remastering recent over-compressed albums in 20 years made me wonder: exactly how good are new remasters of 60s/70s albums?

Check this out:

http://en.wikipedia.org/wiki/Loudness_war#Remasters

This is all good stuff to know about, but I hope it doesn't make people think that "compression" is a dirty word. It's a valuable and necessary tool for most pop or rock mixes. Like most tools, it can be used tastefully or not.

Sometimes a song will have a longer intro if you listen to the song before it, whereas if you were to skip right to the song, you wouldn't hear the long intro. If you watch your counter in some such cases, you'll see that it will switch to "negative seconds" before the true start of the next track.

Also, I've heard of secret songs buried before the first track, so that you'd have to rewind into the "negative seconds" of track one to hear it.

Anyway, I think that's what he's talking about.

CD audio is still 16 bit.

it was mastered by Myles Boisen.

dave marsh's springsteen-bio-part-2 glory days runs through the fascinating pain in the arse it was to master nebraska. it's not all that technical an account, but it's illustrative of the many pitfalls of vinyl mastering (and the pitfalls of trying to master an LP from a second-generation cassette source...not that that's a common problem, but still). in fact, after the first attempt or two it was deemed impossible to master, to the point where they considered putting it out as a cassette-only release.

Yeah I would, but I don't suppose it's easy to get into. I've had a few people, off the back of the first article, contact me to ask if I'd listen to the test masters of their records and give my opinion, which is both weird and also nice.

kinda obvious, but they are as good as the people doing the work on them. and how much money/time they have to work on them also plays a part.

sorry if someone already said something similar. haven't read whole thread.

With home recordings, I tend to start with one of the presets on my Akai DPS16 and gradually tweak until the thresholds start to bite in every frequency range, fiddling with the gain to get the tone I want and - hey presto - everything is suddenly more coherent, richer and less lumpy. I A/B, pulling the faders down on the MBC'd mix, to ensure that I haven't merely made everything louder. In Cool Edit 2000 on the PC, I have more of a blank slate to start with but, thanks to the "offline" nature of effect application on the computer (I don't have to bounce an entire stereo pair before I can A/B), I can just Ctrl-Z, tweak and apply again.

An experienced mastering engineer will immediately be able to identify what they can and can't tease out of a mix with MBC, whereas it's total trial and error with me. I'm very wary of audio/photography analogies (!), but my use of MBC is a little bit like playing with the RGB levels in Photoshop (or "auto contrast") - suddenly that washed-out colour cast disappears and it's a much better picture.

As for doing this kind of thing professionally - I was actually looking for this kind of work four or five years ago and everyone I spoke to in a studio/mastering suite seemed to have been doing it since they were 16 (or they'd started as a tape op/tea boy/general dogsbody at that age, or had come from a gigging muso/live sound background). There didn't seem to be a route in for a 30-something non-musician with a decent academic background - that's not how it works (which is fair enough - an understanding of acoustics and sampling theory is no substitute for a decade of setting up mics and desks, regardless of keenness!).

I don't see what this has to do with it. Unless you're saying there's a (counterintuitive) link between the switch from vinyl to CD and the reckless use of limiting. Because CD gives you way more dynamic range than you ever had with vinyl, it's just that major label pop/rock doesn't use it.

Fair enough but your comment made it sound like "hey, you buy CDs, eventually everything released sounds crap". I didn't follow the logic.

i take everything on a case by case basis. have you ever heard the original u.k. vinyl release of mezzanine? it's staggering.

No, but I have the CD and that sounds pretty great. I don't doubt that the Massive Attack guys take (or took) very special care with their vinyl mastering.

Is there a vinyl version of the Best Of, I wonder?

Bump...

4 other threads wasn't enough huh?

http://www.xtranormal.com/watch?e=20091002201357158