Loudness Wars - The Answer??

[Brad Blackwood]

Originally Posted by cerberus

i am not intimidated by my clients.

It's not about intimidation, but rather respecting their wishes for their art...

[cerberus]

they want it loud as possible. i am asked to determine what is possible for the mixes and report back; then if my proposal is approved ("yes, i can make this seem as loud as american idiot, or kelly clarkson..etc.."), i execute it. they never ask for extra audible distortion. if they did, mastering would be easy... more like a plug-in instead of a skill.

jeff dinces

[Brad Blackwood]

Originally Posted by cerberus

they want it loud. never ask for extra audible distortion.

Not true. Cut a record just last week where they weren't happy untill it was audibly distorted. And they weren't asking for louder, they were asking for 'dirtier' - their exact word. When I pushed it into distortion, they smiled ear to ear.

It's not like I started mastering records last week - one of the things that comes with experience is knowing that every client is different and everything I do (processing-wise) is subjective. Not everyone wants 'clean', just as not everyone wants 'loud'...

This thread is getting tiresome - I really can't believe that my repeated mantra of 'please the artist, it's their art' is so easily misconstrued or misunderstood. It's really a very simple philosophy. So instead of continually repeating the same thing over and over and over, I'm just going to move along.

Have fun.

[cerberus]

this fits with the "picture framer" school of thought? ok i give up too.

jeff dinces

[bob katz]

Originally Posted by Masterer

Fascinating. Did you hear the tracks before they were mastered? Were you privy to what the artists wanted during the mastering session? Do you think its possible the clients may have approved this master? Maybe they even loved it? To admit this would make things less convenient. You couldn't blame the mastering engineer.

You know, before I "outed" Livin La Vida Loca at the AES convention I called my friend Ted Jensen and told him I had a dilemma. I told him that I felt he had just made the loudest CD on record and I was doing an AES seminar and I didn't want to point any fingers, just describe a technical situation. Ted told me that he had transferred the DAT exactly as he received it, saying, "I can't make it any lower than I received it, and I certainly don't want to turn this up!" I thought, "whew", sigh of relief, at least we can't blame the mastering engineer.

At that time I was so pissed to learn that a raw mix would sound this distorted, and I didn't think (mea culpa) to call and discuss this with the mixing engineer, Charles Dye, who has since become an acquaintance if also a friend. Charles has since realized that the best thing to do is to leave the mastering processing for the mastering suite. Anyway, I hoped that if in my demonstration I did not identify the song per se it would be best as it was an "academic demonstration", but let's face it, everyone knows the song, and so it goes...

[bob katz]

Originally Posted by dcollins

all the previous dBm measurements were relative to 600 Ohms in power-coupled interfaces.

ie. 0.775V in 600 Ohms was 0dBm. There's something for the second edition..................


DC

Beat you to it, a footnote in the first edition covers the "dBm story" very well, except for the exact origin of the term "dBu", for which I thank you. I first saw dBu in Rupert Neve schematics in the early 70's, never before that time.

The AES Journal, by the way, does not accept any dB suffixes as having sufficiently general meaning unless they are explained by the author in a footnote. They want all dB references explained.

[cerberus]

i am back to apologize to chris athens and his co-workers because i did not hear the raw mixes; and probably have blown any chance i may have had to do so by being disrespectful.

jeff dinces

[minister]

Quote:
Originally Posted by dcollins

When you look at what it took Dolby to make a meter that approximates human perception

DC

Originally Posted by bob katz

Do you mean the meter that does LEQ-A?

It didn't take Dolby much, actually. Dolby's LEQ-A meter is a simple, quick approximation designed to try to meet the needs of broadcasters and some film issues.

There's a lot more work to be done to get closer to human perception....

sorry to bring this back ....

but, Bob, DC is referring to more that just the LEQ-A meter. they have done a tremendous amount of research into this area and the LM100 barely scratched the surface of that they can do... The equivalent-loudness method standardized by the IEC (IEC 60804) which is the core measurement algorithm used in the LM100, Leq(A), and it is also recognized and recommended in Advanced Television Systems Committee (ATSC) standards and in CEA Bulletin CEB11: NTSC/ATSC Loudness Matching. so...for expediency, they adopted it. but the algorithm is a LE version of the main speech recognition algorithm that they developed which is being eclipsed by some new developments.

DOLBY recognizes that there is much more work to be done and the journey is just being embarked upon. you are correct that the LM100 was done to satisfy broadcast needs... but more is being done now in the form of further research and several new algortithms...

[bob katz]

Originally Posted by minister

sorry to bring this back ....

but, Bob, DC is referring to more that just the LEQ-A meter. they have done a tremendous amount of research into this area and the LM100 barely scratched the surface of that they can do..

snip

Thanks for clarifying, Minister. Well, as long as Dolby do not try to ram a standard and design the refinements in a vacuum without further input from other known researchers in the field, I'm comfortable!

[minister]

Originally Posted by bob katz

Thanks for clarifying, Minister. Well, as long as Dolby do not try to ram a standard and design the refinements in a vacuum without further input from other known researchers in the field, I'm comfortable!

...well, i can't speak for them i am not sure about that. the other issue is that it would be bad if they became the monopolizing purveyor of "standards" machines.... if you mix for Discovery now, you HAVE to use an LM100. now, the LM100 is unique in its abilities, but tying standards to one company's products strikes me as a situation we need to be mindful of.

from a scientific audio point of view they are doing interesting things. and, hey, Dolby A and Dolby SR are pretty good. now, we need to improve the AC3 of SR•D and for DVD..... they do have a new AC3-II standard.

[UnderTow]

Waow. What a thread. A few comments: (All in my humble opinion of course).

Just as many others here, I don't like the sound of alot of modern CDs because of the way things have been processed during tracking, mixing or mastering. (The cause doesn't matter for this discussion). I'm all for calibrated monitoring, good practise guide lines and education education education!

On the other hand, just like Brad, I am against any kind of rules or legislation to enforce anything that will dictate an acceptible max (RMS) level beyond the limitations of the media [1] and I don't believe that any replay-gain metadata will give any real solution to any real or perceived problems.

This deserves some explaining:

It seems that so far, except for the sound quality argument which, being purely subjective, can't be used as an objective measure, the arguments are 1. Artists are being driven by others, in the form of market or peer pressure or whatever, to make their albums loud so they are being bereft of their artistic choices. 2. The advent of CD changers (or similar devices) mean that music with vastely varying levels will be played randomly causing large variations in playback level. If I have missed any other arguments, please correct me!

I'll start with the first point: No artists are forced into making their records loud by anyone. Peer pressure is a BS excuse. Here is an analogy: If a kid on a playground beats up another kid because his friends are cheering him on, does that bully suddenly loose responsability for his actions? Of course not! I'm sorry to say but anyone that claims that artists or MEs or anyone is being driven to do what they do because of what the rest is doing is out of their minds. Everyone has personal
responsability for their actions and behaviour. Period.

Maybe it is an American thing to want to legislate responsability away but I don't buy it. If you (artist/mixer/ME/A&R/Whatever) are responsable for a loud record, the responsability is entirely yours. Blaming anyone else is pointless and rather weak. Fear of not competing level wise with other CDs is entirely the problem of the person with the fear. No one else. Trying to make rules to stop the OTHERS because you feel inadequate compared to them is no different than trying to enforce your morals or religion on another. It is wrong. Plain wrong.

The CD changer complaints to me amount to pure lazyness. As others have said, grab the remote and turn the volume down or up at will. If you want constant music for a party, either get a DJ to do it for you or prepare the material before hand. Yes this takes work but that is the price it costs to have your pie and eat it.

Come to think about it, I don't even like the idea of some kind of constant RMS. I change the volume of the music I listen to all the time and every single person I know does exactly the same thing. I am not talking about audio professionals. I am talking about everybody. People will turn up the volume to a song they like and turn it back down when they loose interest or if it gets too loud or the song is finished. Everyone does this and it is the way things should be because the alternative is a flat bland background of sound with no real long-term dynamics.

The idea that stuff should have some kind of constant RMS during standard replay for practical reasons (read lazyness) is to me in contradiction to requesting more dynamic music.

The track I am listening to right now is a good example: It starts off between -40 to -35 dB FS (That is for the first minute or so of the track) then goes up to about -20 dBFS after about 1m30s and slowly increases in level to about -10 dB Fs at about 8m15s. Varies up and down in level between -14 and -6 db FS for the next 9 minutes and then slowly goes back down in level to about -25 dB FS at 18m30s and dies out to about -45 dBFS after about 30 seconds. This is dynamic music. (And no it is not classical, jazz or any form of accoustic music. It is electronica).

If you had any kind of replay-gain metadata based on the average RMS of the whole track, the beginning and end of this track would be way too soft compared to whatever comes before or after. In other words, there is no solution to this problem that both keeps the dynamics in the music and gives you constant replay levels. These things can never go together because they are contradiction in terms. I see no way to take care of both.

Last but not least, I would like to revive Brad's analogy albeit a bit adapted so that it, IMO, fits this discussion a bit better:

Imagine if painters had rules that only so much bright paint is allowed to be used in a painting because otherwise, the painters with the bright colours have an unfair advantage over other works in a galery and thus force everyone to use bright colours. Afterall, the eye will tend to be drawn to the brighter paintings. To avoid this unfair advantage, the folloing rules are proposed: No more than 30% of the canvas is allowed to be covered by grade A colours. No more than 50% of the canvas is allowed to be convered in grade B colours. And overall, the whole canvas may not be convered by more than 70% of grade A+B+C colours. Further more, as respects to lighting, no spot lights with more than X lumen are allowed to be placed closer than Z meters from the canvas. Etc etc etc

Insane? Indeed.

So to come back to the begining, education is the only solution IMO. We can not force people to comply to our view on esthetics nor should we. If that means that most of what we hear won't sound good to us, so be it. (Hey nothing new there anyway).

Alistair

[1] I do think there is a real technical problem with inter-sample peaks overdriving (cheap) converters but that IMO is a different topic and a technical one. Not an esthetics one.

[PatrikT]

Originally Posted by UnderTow

(And no it is not classical, jazz or any form of accoustic music. It is electronica).

I've always considered electronica to be the perfect update regarding "classic music". It is built around methods which are very related to classical orchestra compositions.

Patrik

[UnderTow]

Originally Posted by PatrikT

I've always considered electronica to be the perfect update regarding "classic music". It is built around methods which are very related to classical orchestra compositions.

Patrik

I'm not sure we are talking about the same Electronica. I use the term very widely to describe anything electronic from Stockhausen to Gabber passing by new age ambient and hardcore tekno.

The track in question was "Shpongle - And the day turned to night" from the first album "Are you sphongled?".

Alistair