Digital Headroom

[mhschmieder]

Many of us old-timers are horrified by the trend towards over-compressing at the mastering-limiting phase of recording these past few years, with the problem accelerating out of control starting around 2004.

At the AES Show in San Francisco this past weekend, there was a paper presented in conjunction with a software prototype for measuring digital distortion, with the goal of eventually reaching industry consensus on metering and calibration standards.

The paper was entitled ••œStop Counting Samples••

[OldTimey]

Pretty interesting stuff here...I knew something was 'off' with Madonna's track "Hung Up"...I actually find it to be a groovy tune, but was really amazed at how awfully distorted and squashed it sounds, almost as if it was done on purpose. I thought surely I had a bad version of the track...but i guess not. Wonder how a mulitmillion dollar record could get to sound so ugly?

[mhschmieder]

One possible answer, which I haven't had time to write up yet, is they used the wrong mix . Probably not in Madonna's case as she's a top star.

But one of the things I learned at AES is that the wrong mix is frequently used these days, due to a combination of poor project archiving standards, miscommunication, rushed deadlines, someone high up hearing a preliminary mix and saying "good enough" and denying the budget for what everyone thought would be a more professional final mixing session.

[ester]

i was looking for that article on the AES website (they post all speakers papers) do you happen to remember his name?

Good stuff!!

i think i read bob katz talking about the increasing compression in music and how soon we will all be deaf as a result!

[David Polich]

While I agree with all of the points in the first post and the replies, I think what's overlooked is that in today's world, loud has won. Period, we are not going back.

Here's why - people expect it. Case in point - I sent a mix of a track I've been working on with the singer of my band. Didn't "master" it too loud. She hated it. Blew it up to insane proportions, sent it again, a-ok this time. That's because she's used to everything she plays in her car being brickwall maximized. I didn't change anything else in the mix, I just made it really loud.

Industry A&R, managers, lawyers - they all expect loud tracks. It can mean the difference between a track or act accepted or rejected. When it c omes down to money, if super-loud is what it takes to get the deal, I'm going to make it super loud. I'll save my "audiophile" masters for home use.

It's a lo-fi world.

[monkey man]

Thank you, mhschmieder, for that fascinating contribution!

There's an issue I expected to be mentioned (perhaps it was in the presentation, but as you said, you couldn't include it all).

I've heard that most (domestic) DA converters distort horribly when not operating in their "linear" zones.
I believe it's been shown that if we master to around -2 or 3dB, good-quality domestic playback systems can deliver reasonable results.
However, in the cases of the vast majority of systems I believe it's been shown that anything above -5 or 6 dB or so, is serverely compromised in its integrity.
A simple experiment that was suggested was to burn tests at say, -1dB, -3dB and -6dB and compare them on a typical domestic PB system.

From what I can remember, the results are profound.
It looks like it could be a major overriding factor in determining PB quality.
I figured I'd take it on board, but with so many influential factors involved, your and other Unicorns' opinions are of great value.
Thoughts?

MM

[monkey man]

Here's why - people expect it. Case in point - I sent a mix of a track I've been working on with the singer of my band. Didn't "master" it too loud. She hated it. Blew it up to insane proportions, sent it again, a-ok this time. That's because she's used to everything she plays in her car being brickwall maximized. I didn't change anything else in the mix, I just made it really loud.

So sad, and yet typical.

...It's a lo-fi world...

That's why we need 384kHz sampling rates.
Pristine crap is where we're headed, and we artists have to fork out for this nonsense.
Mandatory ear-canal cleaning for all with wax. Waxman, you're excepted, of course.
Oh, and we may need to swing some sort of bulk-labotomy deal.

Seems there's hope after all.
MM

[David Polich]

Here's why - people expect it. Case in point - I sent a mix of a track I've been working on with the singer of my band. Didn't "master" it too loud. She hated it. Blew it up to insane proportions, sent it again, a-ok this time. That's because she's used to everything she plays in her car being brickwall maximized. I didn't change anything else in the mix, I just made it really loud.

So sad, and yet typical.

...It's a lo-fi world...

That's why we need 384kHz sampling rates.
Pristine crap is where we're headed, and we artists have to fork out for this nonsense.
Mandatory ear-canal cleaning for all with wax. Waxman, you're excepted, of course.
Oh, and we may need to swing some sort of bulk-labotomy deal.

Seems there's hope after all.
MM

Pristine crap! That is awesome...

Don't get me wrong - I detest the "louder than everything else" norm. I wish there was some sort of standard that could be set. Unfortunately the mp3 and the i-Pod are the standard we've developed.

Most people can't tell the difference between an mp3 and a 24-bit AIFF file of the same tune. They really can't. Perhaps it is best to keep in mind that when you ask people to spot a difference, the first thing they think of is a "musical" difference, not an audio quality difference. Most people are not audio engineers or project studio owners.

[monkey man]

Pristine crap! That is awesome...

Yup, no need to Polich these turds, Davo!

Most people can't tell the difference between an mp3 and a 24-bit AIFF file of the same tune. They really can't. Perhaps it is best to keep in mind that when you ask people to spot a difference, the first thing they think of is a "musical" difference, not an audio quality difference. Most people are not audio engineers or project studio owners.

It's easy to forget that most people don't share our approach in these things.
Nowadays it's entertainment, and not much more, it seems.

We needn't even attempt the bulk-labotomy deal and mandatory ear-canal cleaning, it would seem. Bugger.
Cheers, Davo.
Nicky

[kelldammit]

bulk lobotomy? so i can get more cheaper?!?
YES!!!

do i need my costco card?

[James Steele]

Man... Dave hit the nail on the head with this one. People *expect* loud and you can't get very far swimming upstream. It's the "Yeah-this-cranks!" world of rock and roll... compressed to the hilt.

The point that most people don't hear the difference is important and people are swayed by production quality and loudness. For some reason this brought back memories of some sort of industry seminar thing I went to many years ago... with a panel of A&R people from labels who were almost across the board 20-somethings with attitude. A common (perhaps less common these days) myth that you will hear is that these experts can listen to the "song" and it's the "song" that's important-- production value isn't what they claim to be listening for. However this is bunk, and I humbly submit that you can't be noticed with the fabled demo recorded with just a guitar and cassette recorder. People expect production value and if they don't hear it, their ears and minds close down. I guess to tie it together, *loudness* *punchiness*... whatever you want to call it is what the public and the allegedly musically sophisticated industry types expect to hear... period. Consequently, when my CD is done, I think I won't be blazing a trail on principle, but rather squashing the crap out of it as well. I'll let someone else take the bullet in the name of audio purity.

[monkey man]

Well said, Jimbo.

I can't help but wonder whether allowing progressively more dynamic range back into our mixes and masters could in time reverse the trend.
After all, those who pushed "loudness" in the first place got us into this awful mess.
Don't dynamic mixes sound better after radio compression than those that are already squashed-to-smithereens?

If the market became aware of listening-fatigue to an extent, there may be hope.
This "me" era we live in tends to see folks diving in headfirst, lapping up all they can, then having to rest from "over-filling".
Sooner or later they get tired of this and back off in the particular endeavour.
Of course, they don't notice the pattern is almost ubiquitous in their lives, but that's another discussion.
Perhaps if the sales machine told 'em "softer" CDs sound better with the volume turned up than vice-versa, well, who knows?

Sorry for the rant.
It's monkey-money, but it's worth 2 cents where I come from.
OK, OK; would you believe 1 cent?
M

[David Polich]

James, that was right on.

Totally agree - people cannnot hear a song past the production. I'd go so far as to say these days it is the reverse - they respond to the production first, then the song. Hip-hop, electronica, dance, these are all genres where the production actually IS the song.

We're not going back to the old dynamic days anymore than sci-fi movies are going back to the days of black-and-white Flash Gordon serials with plastic spaceships suspended from strings. The battle for dynamics has been lost.

However, I will say this - it doesn't hurt to point the issue out every now and then. I'm still a firm believer in encouraging people to wear earplugs, listen at lower volumes when they can, and develop better listening skills. Too many people are damaging their hearing listening to distorted audio systems at high levels.

[monkey man]

Crikey; more gold from DP (cool initials, BTW).

I still hold out some hope on the future of dynamics.
B+W sci-fi movies contain less data than modern ones.
Modern graphics systems have been sold on pixel and colour counts, refresh rates etc.
Consumers have had to examine contrast qualities, colour temps and a plethora of new parameters in their quests for quality.

In the case of dynamic music, the quieter the low-level passages are, the more data the mix needs to contain in order to retain acceptable resolution.
Why is it not possible the consumer will one day see his folly (oops, I've over-eaten, I'm deaf from this •••• etc.), put it down to having been sucked in to yet another inferior format, and splash out on the "next big thing"?
If that thing was a market-driven, renewed appreciation of quality?
Reduced headaches, listening fatigue and hearing damage sound like marketable options, starting with parental targeting.

Again, your thoughts are pure gold, Davo.
Dang, gold, gold, gold...it just keeps comin'.
Good stuff, methinks.

Anyway, another 2 monkey-cents enters in the coffer.
M

[Kawentzmann]

I think the consumers can't realize what causes their tiring from certain styles of music. Let••™s hope they don••™t turn entirely away from music, just the excessively loud. Having a sound like a brick means no climaxes after all, and that is like viagra with no orgasm at the end.

KK