The "Loudness war" is old news...

Alvaro Martín Gómez A. · #1 02-01-2009, 10:20 AM

...and it's not that I didn't notice it before, but I really felt it yesterday. My band is planning a Rush tribute concert for an special event next March, and I told my bandmates: "Hey, we've played '2112' before, but we only did the first two parts. I'd like to add the suite's Grand Finale". They asked me to send them a recording of "2112" as I wanted to play it for them to check it out, and that was easy: I got the idea from the R30 concert, where the guys from Toronto play it exactly that way. So it was just a matter of sending them that version.

But when I listened to it again, I found it so dense and not so pleasant-sounding that I preferred to edit the studio version. No problem doing that job, but I couldn't avoid thinking how absurd is the fact that something that should be for best (the current recording and mastering techniques) ends up being for worse. This great and well-known (I think) article shows the problem very clearly, but anyway let me show you the sonic difference between the original version of 2112 (after editing it - parts 3, 4, 5 and 6 removed) and the R30 version (with the same parts - Just six seconds longer):

Not to mention that the studio waveform was taken from the remastered version, so it was even quieter back in 1976, for sure. Of course there's crowd noise added on the R30 version, but anyway it shouldn't affect the dynamic range of the recording that much.

Again, these are old news, but this particular situation made me feel so sad about how quality music is ruined by technology.

Pilgrim · #2 02-01-2009, 04:25 PM

I'm not sure which is which, but the lower screenshot shows a waveform that's so badly overmodulated that it should sound absolutely terrible. The top waveform looks good to me - gain optimized but not off the edges of the graph. You can always amplify a clean signal on playback, but a signal that was overmodulated or clipped while recording will sound like crap no matter what you do with it afterwards.

Alvaro Martín Gómez A. · #3 02-01-2009, 04:28 PM

Quote:

Originally Posted by Pilgrim

I'm not sure which is which, but the lower screenshot shows a waveform that's so badly overmodulated that it should sound absolutely terrible.

That's the live version from the R30 audio CD.

Pilgrim · #4 02-01-2009, 04:37 PM

Interesting. I can't imagine that the gain was that high on the original recording - it looks like there was some additional gain on the input to your device. I suspect that was the case because it appears that you were successful in reducing the gain and maintaining a clean waveform.

It's hard to know how to interpret the screenshot unless one knows the complete equipment configuration between source and scope, including the gain settings and adjustments on that input. It does certainly look badly overmodulated in the origonal screenshot.

MaxRossell · #5 02-01-2009, 04:47 PM

Quote:

Originally Posted by Pilgrim

Interesting. I can't imagine that the gain was that high on the original recording - it looks like there was some additional gain on the input to your device. I suspect that was the case because it appears that you were successful in reducing the gain and maintaining a clean waveform.

It's hard to know how to interpret the screenshot unless one knows the complete equipment configuration between source and scope, including the gain settings and adjustments on that input. It does certainly look badly overmodulated in the screenshot.

If it's just a straight rip from the CD, that waveform is a true reflection of the levels on it.

It's not that odd. I grabbed a Deftones track from their latest last weekk for my drummer, and the thing was just blocks of constant -.1db brickwall.

Mastering's stupid trendy. These days I deliberately master all my stuff quiet so people have to reach for their volume knobs if they want it loud. I'd sooner that than kill all the dynamics I spent hours mixing. Like I give a crap if it isn't as loud as the latest Britney single.

Alvaro Martín Gómez A. · #6 02-01-2009, 04:53 PM

Quote:

Originally Posted by MaxRossell

If it's just a straight rip from the CD, that waveform is a true reflection of the levels on it.

Exactly. My only editing job on the studio version was cutting the unneeded sections of the suite. I didn't alter the waveforms' amplitude in any way.

Pilgrim · #7 02-01-2009, 04:53 PM

A direct rip - no kidding.

Wow, that's contrary to every rule of mixing I've ever used. But I've done a lot more in analog than digital, although I've found that digital responds even more negatively to overmodulation than analog.

I'm with you, MaxRossell - the original mix doesn't make sense to me.

And my compliments to Alvaro on what I think is a considerable improvement!

MaxRossell · #8 02-01-2009, 05:02 PM

Well, if it's a live I suspect that the original mix probably wasn't the best to begin with, and then as Alvaro is saying, blasting it through a compressor/limiter has just messed it up more.

It's so funny, all of these advances we've made in music production and audio fidelity technology over the years, all wasted because people turn up their music so loud you can't even hear it properly, and then their fans take it and squish it down into mp3 files, losing five-sixths of the information in the process.

Bah. Gimme vinyl any day.

Alvaro Martín Gómez A. · #9 02-01-2009, 05:06 PM

Hmm... this is a very interesting conversation, but I think I'm confused: The upper image is the studio version of 2112 from 1976 (remastered) and the lower one is the live version of the same tune recorded in 2004. Does the studio waveform still seem too loud for an acceptable standard?

MaxRossell · #10 02-01-2009, 05:10 PM

Quote:

Originally Posted by Alvaro Martín Gómez A.

Hmm... this is a very interesting conversation, but I think I'm confused: The upper image is the studio version of 2112 from 1976 (remastered) and the lower one is the live version of the same tune recorded in 2004. Does the studio waveform still seem too loud for an acceptable standard?

Not at all. Well, I haven't heard it, but looking it, it would seem that the dynamic range has been more or less preserved, which is what you want.

Basically what I always look for is a balance between a track that's loud enough for a good signal-to-noise ratio, but that isn't so loud I'm starting to lose definition or dynamic impact.

Pilgrim · #11 02-01-2009, 05:15 PM

The upper image (studio) looks very good to me. It's as far as I'd dare push the gain on a signal without fearing that it would sound overmodulated on playback.

The article is very good as well - I haven't finished it, but I think the author is right on. Perhaps compression is indeed one of the culprits - otherwise real-world dynamics would push the top end of the gain into fuzziness.

But that also implies extremely heavy-handed compression, to the point where there are few if any dynamics left. NOT what real music sounds like!!!! I think Max and I are of similar mind in that regard.

Edit: here's the most pertinent paragraph from the article Alvaro cited:

"However, the latest trend is LOUDER IS BETTER. This approach basically ignores any distortion caused by limiting and seeks to make the audio as loud as possible. The idea is to peg the meters and keep them pegged. As a result the signal is just ruined."

Alvaro Martín Gómez A. · #12 02-01-2009, 05:33 PM

I'd like to know what Geddy, Alex and Neil think about that situation. I know that they realized their mistake on "Vapor Trails", but the audio releases of "Rush in Rio" and "R30" (for instance) don't seem to be a big improvement IMO.

Pilgrim · #13 02-01-2009, 06:34 PM

Quote:

Originally Posted by Alvaro Martín Gómez A.

I'd like to know what Geddy, Alex and Neil think about that situation. I know that they realized their mistake on "Vapor Trails", but the audio releases of "Rush in Rio" and "R30" (for instance) don't seem to be a big improvement IMO.

That's an interesting web page - but although they whine a lot, they don't seem to have gone back, remixed and corrected the problem. That's obviously what they need to do.

hbarcat · #14 02-01-2009, 06:38 PM

Quote:

Originally Posted by Pilgrim

A direct rip - no kidding.

Wow, that's contrary to every rule of mixing I've ever used. But I've done a lot more in analog than digital, although I've found that digital responds even more negatively to overmodulation than analog.

I'm with you, MaxRossell - the original mix doesn't make sense to me.

And my compliments to Alvaro on what I think is a considerable improvement!

It's not the mixing, it's the mastering. Whoever did it compressed the crap out of it in multiple stages using specialized compression devices to eke every last fraction of a dB out of the waveform.

The waveform on top is heavily compressed, although it's just about average by today's standards. The bottom one is compressed about as far as it will go before it completely dissolves into noise.

MaxRossell · #15 02-01-2009, 06:39 PM

It's worth noting that the loudness issue is not usually down the record company, even though that's the assumption. Except in commercial corporate pop, that is. In rock music, it tends to be the artist who wants their album as loud as the latest Metallica. Which by the way clips audibly when you play it on a good stereo.

How surprised was I to find, when listening to Guns N' Roses' latest album, that it was quiet, dynamic, detailed, really beautifully mixed and only mastered up to levels and EQ, not really pushed in any significant way. Of course, that record doesn't have many other redeeming features, but it would be nice if it marked the beginning of a movement towards quieter, better-sounding records.

hbarcat · #16 02-01-2009, 07:13 PM

I think the latest Metallica album, Death Magnetic may have been a turning point. It has been generally much better received than St. Anger, from a musical standpoint, but there are so many people complaining about the ludicrous amount of compression on it that it has been kind of a wake-up call to the consuming public.

If over compression bothers you as a music fan then the way to stop it is to complain loudly and publicly and demand that they go back to having some dynamic range as a higher priority in the final product.

MaxRossell · #17 02-01-2009, 07:16 PM

Quote:

Originally Posted by hbarcat

I think the latest Metallica album, Death Magnetic may have been a turning point. It has been generally much better received than St. Anger, from a musical standpoint, but there are so many people complaining about the ludicrous amount of compression on it that it has been kind of a wake-up call to the consuming public.

If over compression bothers you as a music fan then the way to stop it is to complain loudly and publicly and demand that they go back to having some dynamic range as a higher priority in the final product.

Irony...

ehque · #18 02-02-2009, 08:37 AM

The waveforms in Vapor Trails don't look compressed, they look simply digitally limited, like someone had upped the gain too high in Audacity.

It's good that changes are being made.

jnevi9nr · #19 09-02-2011, 09:45 PM

Same problem with the new rhcp album. Alot of the vinyls are ending up sounding better than CDs. I like that pink Floyd is coming out with remasters on blu ray! Time to go vintage.

iamlowsound · #20 09-03-2011, 07:37 PM

Quote:

Originally Posted by jnevi9nr

Same problem with the new rhcp album. Alot of the vinyls are ending up sounding better than CDs. I like that pink Floyd is coming out with remasters on blu ray! Time to go vintage.

Vinyl always sounds better than the CD version. I have a good stereo, if I want it to be loud, I can make it loud; you don't have to do it in studio for me.

lowsound