Music Into Noise: The Destructive Use Of Dynamic Range Compression part 2

Metal needs a Blue Note/RVG

i think it would be cool if more bands found people who could do nice well-balanced analog recordings and then just transfer that to cd. or just find studios/engineers who are more well-rounded.

there's a guy i know near me who does great analog and digital recording. he's equally adept at both but i love his analog atuff that i've heard. he's recorded friends of mine - bunwinkies and fat worm of error - and he gets a great sound. he does all of dinosaur's stuff and the witch stuff. and i think he did the new thurston album. anyway, there are people out there who do great stuff people just need to think outside the box. he's not a "metal" engineer or producer but i bet he could do an amazing job with a metal band.

Obliteration's Nekropsalms is a beautifully recorded death metal album

we're scaring lex and the others out of the thread by talking about metal

That Fat Worm record sounds amazing. Main problem with analog, especially for bands at that level, is the cost. Tape is pretty pricey.

oh and tarfumes before i forget i have it right here: the who disc is from 1996 and the reissue was produced by jon astley, remixed by andy macpherson and jon astley, and remastered by bob ludwig.

Yep, that's the one a lot of people can't stand. I kept my copy for the handful of additional Moon bits the remix revealed (a cymbal tinkle in "Acid Queen," a tympani fill in "Overture"), but the sound, ugh. I think the 1989 remaster is supposed to be good, though (single disc, faces in the artwork).

its pretty demented, sound-wise.

but, you know, if that's the only one you've ever heard/owned it doesn't matter, i guess. that's what the album is to you. that's how it has always sounded.

Hm, I definitely would not say that "Penny Lane" has flat dynamics on the 2009 stereo remaster of MMT.

― EveningStar (Sund4r), Monday, February 25, 2013 9:49 PM

Listening to the mono mix, I just don't hear a lot of change in the overall track volume level despite constant changes in the instrumentation.

As to the "Da Doo Ron Ron" waveform not being just a big rectangle: Yeah, but the volume peaks are weird. Like the bass drum that's really loud or some vocal peaks or something. Basic backing track is pretty much a sound mash imo.

Re: that chicago mastering article above (which is very good), I took a look at a MP3 of that Radiohead song in Audacity, and my version shows a lot of clipping. I ripped it from the CD ages ago, not sure which program I used (I no longer have the CD). Can an encoding to MP3 introduce distortion?

Pop, rock, and dance music, even some jazz, doesn't need to have dynamics to rival classical music. That's not the point, and I don't think anyone's arguing about it. My main issue is the way that impaired sound quality - through clipping and muddy, over-stuffed mixing and general lack of clarity - is often a side-effect of things being very flat and loud. I'm a lot less militant about it now that I was six years ago, partly cos my tastes have changed slightly and I'm listening to less stuff that's really impaired, and partly cos I think a lot of people have realised they don't like this. But I still think that a lot of modern records sound very, very samey and boring, because they're so loud; things like Aerial and The Drift just sound absolutely bizarre and brilliant and avant-garde to me, and that's wonderful.

Interesting list of 'worst offenders' here:

http://tvtropes.org/pmwiki/pmwiki.php/Main/LoudnessWar

the ability to have infinite tracks of overdubs is by far the worst thing about digital recording

then again on the first cars album they used like 5000 vocal tracks for the harmonies so maybe people should only use 5000 tracks per song.

cuz that album rules.

the ability to have infinite tracks of overdubs is by far the worst thing about digital recording

otm. I feel like digital compression is used most commonly as a way to glue meticulously separated digital sounds together.

that vs. something like "Da Doo Ron Ron," where the wall of sound (iirc) comes from all instruments being played into the same room simultaneously. the "compression" is partially shaped by a real acoustic space.

I guess, whatever they did, I think of the Cars as the opposite of a cluttered band. I'm more talking about the generation of kids who got convinced they were studio auteurs by The Soft Bulletin, ”kitchen sink” indie

this is for tarfumes. sweet early german pressing! yes i know my camera isn't very hi-fi.

https://www.youtube.com/watch?v=m7sp73OlTD4

Beautiful! Thanks, Scott! I have a later German pressing (from the Phases box) that sounds pretty good; I'll have to dig it out.

happy Dynamic Range Day, folks! be sure to celebrate this awesome Day by letting your friends know how music is too loud. It used to be quieter, now it is too loud.

LOL at that "worst offenders" list linked upthread. Somehow a guy named Kevin Gray has finally done justice to Iggy Pop and David Bowie's work. I guess no-one would've ever heard of Raw Power until he decided to turn up the guitar slightly.

Giorgio Moroder weighs in.

"My son helped me to get the screenshots in Audacity" - sounds weird coming from Moroder. You'd've thought he'd be good at computers.

I dunno, I wouldn't expect that necessarily.

Maybe. It's just funny, a rubbing-together of eras.

An addendum

Whatevs. These don't demonstrate anything. You need to be able to compare apples with apples. What are these songs? How long are these waveforms? Only the "Get Lucky" waveform shows the tracklength. 30 secs on Audacity could make the 1977 one look like the 2013 one. Also, the Get Lucky one is too weak. If you put that in a mix you'd have to boost it up. If that's really the unadjusted waveform of the new song then they fucked up.

Also, I have worked with thousands of commercially released songs in Audacity and you rarely, if ever, see something like the 2013 thing (unless you specifically want it to look like that).

you're talking out your arse

No, I'm not.

I guarantee that that will not be the waveform for Get Lucky.

you better check again man, cos you're showing all the signs

Here’s a wavefrom I just made from the mp3 that is going around, seems pretty similar to me:

https://www.dropbox.com/s/fga0lyjjnophl0w/Screen%20Shot%202013-04-18%20at%2023.39.57.png

https://www.dropbox.com/s/fga0lyjjnophl0w/Screen%20Shot%202013-04-18%20at%2023.39.57.png
https://www.dropbox.com/s/fga0lyjjnophl0w/Screen%20Shot%202013-04-18%20at%2023.39.57.png

Well, I can't argue with that but sheesh, that is fucking weak. Only thing I can think of is that sometimes leaked versions have the volumes deliberately lowered in order that people still might want to pay for a proper version.

I mean, it's basically unbroadcastable if you left it at that volume.

some more advanced audio systems allow users the ability to control the volume at which they wish to listen, or broadcast

seriously i don't understand some aspects of the loudness war pushback; that waveform has nothing at all to lose by being brought up 6 or 7 db because it's already v flat dynamically, as it is it doesn't take up the full dynamic range and can even be considered lower fidelity than another track that does - only by a bit (maybe literally, as each bits in yr 16-bits of cd quality represents a certain amount of dynamic headroom and leaving that much unused space at the top is akin to only using 14 or 15 bits)... but still. the way that song is mastered it looks and is quieter but doesn't seem to have any more dynamics than yer average brickwalled whatever. i understand not wanting to change the sound of a song in mastering with hard limiting, but that doesn't mean you have to make a song much quieter than allotted, and it doesn't mean it'll sound any better if you do.

iow, i feel these waveform comparisons are often misleading!

lol contendo

that waveform has nothing at all to lose by being brought up 6 or 7 db because it's already v flat dynamically, as it is it doesn't take up the full dynamic range and can even be considered lower fidelity than another track that does - only by a bit ... but still. the way that song is mastered it looks and is quieter but doesn't seem to have any more dynamics than yer average brickwalled whatever.

sleepingbag otm, was thinking the same thing. nice that it's not completely brickwalled, but it does look p heavily compressed, and if you're gonna do that, there's no reason to limit the available dynamic range by reducing the volume overall.

except, you know, to make a point

It’s already at roughly -1.5db. Just turn it up, I guarantee it will sound better. Radio stations will brickwall it themselves anyway.

The track itself is another matter, love the guitar playing but the rest isn’t doing much for me.

Interesting though that Daft Punk are taking this stance after how much they have abused compression in the past.

I agree that it's an extreme statement, but yes zoomed out to four minutes, waveform comparisons are misleading.

When I import the mp3 and zoom in to the loudest transient I can find, it peaks at -1.7 dB under. It's not impossible there's a slightly louder one in there somewhere if I had the time to crawl for it. So it's basically a zero compromise master; even though most of the drum hits are down around -3.0, they left about one full dB of headroom louder than the loudest ones, so that not even one single drum hit had to be sawed off for the sake of bringing up the overall volume.

I'm for it!

xpost yes this is one of the bands I always used as a counter-example to the more rabid 'compression is evil' arguments, so it's interesting that one of the main bands that mainstreamed the creative use of sidechain compression is throwing down this gauntlet

consistent w the perversity of their stance on edm i guess

just played it back while watching on a loudness meter. levels safely average around -3, leaving safe room for a handful of drum transients to spike, usually around -2.5, though a couple go louder, and the loudest one is that one hit at -1.7

they did not normalize the track

totally conventional cd mastering practice, circa 1985

I was thinking the reason for the headroom would be that so louder tracks on the album could be louder, but being that this is a single edit, that doesn't stand up to reason

So Bob Katz reckons iTunes Radio, which has Sound Check turned on permanently, will win the Loudness War by making people master thing approx 7db quieter. To use VERY simple terms.

More here: http://www.digido.com/forum/announcement/id-6.html

Thoughts?