Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: A Modest Proposal for Ending the Loudness War (Read 11360 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

A Modest Proposal for Ending the Loudness War

It's fairly well known in these parts (but less known everywhere else) that loudness equalization mostly solves the loudness war, in that attempts at increasing program loudness are immediately compensated on playback.

iTunes and the iPod, which arguably control a majority of the music playback market right now, have SoundCheck, which is imperfect, but works. Everybody else has ReplayGain, which rocks (Thanks David!). Beyond Apple, only a couple other music providers dominate the market. I'd put MySpace and Bleep on those lists.

If only two or three of the biggest music playback providers made loudness equalization required - ie, it can't be turned off except by advanced means - the loudness war would probably end fairly quickly. Instead of trying to argue from the standpoint of increased fidelity, one would be able to argue that hypercompression/limiting would do nothing to increase loudness, and would only reduce fidelity. Once producers realize this, the loudness war would probably be stopped dead in its tracks.

The problems with this scheme are as follows:
  • Convincing Apply/MySpace to do anything is kind of hard.
  • Like I said, SoundCheck kind of sucks, notably because of a lack of album gain.
  • Playback gain would need to be adjusted substantially. Volume levels on the iPod are sometimes considered (by crazies IMHO) to be quiet as they are.
  • Very few people actually understand what SoundCheck does. Apparantly, some audio engineers seriously believe it's a dynamic range compressor.
  • I'm not sure it is possible to read off RG tags from Flash, so MySpace might need Adobe's help to implement it.
Nevertheless I think this is far more feasible than the other usual alternatives proposed, which include: getting hardware manufacturers to add DRC to all their devices; informing the music listening and production public of higher fidelity mastering; moving to some intrinsically loudness-agnostic format such as vinyl or SACD; etc. No additional hardware is required, and it only takes convincing a few organizations to do something cheap for everybody else to likely follow suit.

So, opinions? Is this worth getting out my petition-writing paper?

A Modest Proposal for Ending the Loudness War

Reply #1
My guess is that 70% of people don't care, 29% don't understand it, and the other 1% are on HydrogenAudio.

A Modest Proposal for Ending the Loudness War

Reply #2
That's a major supporting argument for this proposal.

A Modest Proposal for Ending the Loudness War

Reply #3
Sheep dont want to have a voice and therefore shouldn't have a voice. If the majority doesn't care, then others should determine what happens. The problem is that with the music industry's growth in power going unnoticed for a long time(perhaps because of too many "didn't care"), it has a much larger influence than sensible audio enthusiasts and sensible music creators. Well, at least thats how it has been up until now. There are new players on the playfield now which have a rather high degree of influence, compared to their age.

So yes, i agree... it is naive to expect that large manufacturers of hifi-gear or major labels, etc will support this. But targeting players which focus on the portable and online market may be more efficient... it will still be difficult, but chances are much better than with the old "dinosaurs". And they are capable of provoking a change of mindset regarding loudness.

However, you will NOT convince them to support this, by telling them that it is "more correct" and that those "restrictions" are good for some idealistic reason. Even if they listen more closely to enthusiasts, they are still senseless and greedy short-term-profit machines. That means that to convince them, you will need to explain to them why RG would ADD value to their products - why it makes their products more attractive - that implementing it is cheap and giving them a finished solution to make this tech easily usable for joe average. Remember, greed-machines - if you can give them almost everything for free and convince them that it benefits their products value/attractivity - so, an "you can increase the value of your products for free" case - then they will start listening.
I am arrogant and I can afford it because I deliver.

A Modest Proposal for Ending the Loudness War

Reply #4
If you don't like the product, don't buy it.

The studios may not understand sound quality, but they do understand profit.

In addition to stop using DRC they also need to increase the mastering budget. No album will sound great with a $1500 budget - loud or not.

A Modest Proposal for Ending the Loudness War

Reply #5
The studios may not understand sound quality, but they do understand profit.

True statement at the base, but it fails to account for interpreting those lost profits. Dropping sales are just dropping sales, not more.... as long as they do not understand WHY those sales drop, the "message" does not arrive.
I am arrogant and I can afford it because I deliver.

A Modest Proposal for Ending the Loudness War

Reply #6
Sheep dont want to have a voice and therefore shouldn't have a voice.


Buy they do have a voice/vote -- every time they buy a song.

A Modest Proposal for Ending the Loudness War

Reply #7
Most people under the age of 30 don't understand how lots of DRC ruins the music.  They pretty much grew up with loud, compressed music, and haven't experienced the difference.

A lot of these 20-somethings listen to urban hip-hop music and care more that their music is loud enough to rattle windows and justify those big subwoofers they installed in their cars.  They'll have major, major, hearing loss by the age of 40.

All these know is that the music sounds great on the radio, in their cars, and on their iPods.  Very few of them actually listen to music on Hi-Fi equipment.

Giving them charts and graphs, some document that looks like it was made for a master's thesis, and lecturing them will only cause their eyes to glaze over and everything you tell them will quite literally go in one ear and out the other.  They'll nod yes and then forget what was said 10 minutes later.

So, to help end the loudness war, one needs to show them, via audio and listening examples, how today's music is being mastered in such a way that it often turns into an incomprehensible mush.

If possible, get a nice album that has been mastered to audiophile standards, so that there is very little to no DRC applied.  Then when you apply RG and peak computation via fb2k, it will come back as an excellent set of numbers like 0.5 RG and peak of about 0.75

Now, take that same music and make copies that have been processed like most CDs mastered today, with DRC up the wazoo, so that now the RG is around -10.0 and peak about 1.2

Have these people listen to the original "audiophile"-quality track and then listen to the "modern 200x" mastered version.  They should be able to easily understand how all this DRC is ruining music, if they haven't already damaged their hearing to the point of futility.

As they hear the same track, both in original audiophile format and the modern 200x version, show them the equivalent Audacity (or whatever) analysis so that they can see the wall of sound in the overly processed DRC version as it's playing.

Once they understand how all this DRC is ruining their precious music that they spent money on (in some form or the other), they can then start demanding that the music industry actually produce better quality mastered recordings without all the DRC and hopefully refuse to "buy" any music that is ruined this way.

A Modest Proposal for Ending the Loudness War

Reply #8
Moreover, if you follow the idea that reduced fidelity is reducing sales, the labels' hands are still tied. They can't increase dynamic range because a surprisingly large number of people will still complain that they can't hear their music anymore, the new CDs don't sound as good as the old, don't know how to use a volume control, etc.

Getting a few playback vendors to go along with this would eliminate that concern. Labels could do damn near whatever they want with dynamic range and they would not be punished in the marketplace for going "too quiet".

A Modest Proposal for Ending the Loudness War

Reply #9
So yes, i agree... it is naive to expect that large manufacturers of hifi-gear or major labels, etc will support this. But targeting players which focus on the portable and online market may be more efficient... it will still be difficult, but chances are much better than with the old "dinosaurs". And they are capable of provoking a change of mindset regarding loudness.
Maybe I wasn't quite clear in what I was targetting with my proposal. I propose changing the behavior of the iPod, and of iTunes, and of the MySpace player. That's it.

Quote
However, you will NOT convince them to support this, by telling them that it is "more correct" and that those "restrictions" are good for some idealistic reason. Even if they listen more closely to enthusiasts, they are still senseless and greedy short-term-profit machines. That means that to convince them, you will need to explain to them why RG would ADD value to their products - why it makes their products more attractive - that implementing it is cheap and giving them a finished solution to make this tech easily usable for joe average. Remember, greed-machines - if you can give them almost everything for free and convince them that it benefits their products value/attractivity - so, an "you can increase the value of your products for free" case - then they will start listening.
Quite true. I'm hoping that the lower the required cost, the less important the reasoning has to be, although it still needs to be ironclad.

Remember, the goal here isn't merely to get RG installed everywhere, it's to REQUIRE its use in music listening.

MySpace has a really good reason to do this already: amateur musicians. Getting the most loudness out of a song while preserving some sense of fidelity requires a lot of production/mastering experience. Amateur/indie musicians are at a disadvantage here. Loudness equalization will eliminate that and will lower the barrier for entry for new musicians to make a dynamic, competitive song.

Apple has much less of a justification to do anything like this except out of idealism. They already have SoundCheck, and everybody who wants it is already using it or using ReplayGain, so why would they want to remove the ability to run without it? One could argue that it would add some goodwill with certain regulators in that modern mastering would no longer be implicated in hearing loss issues. And that by taking a stand like this, Apple stands to give its users a higher quality listening experience, by recommending that records be produced with a higher dynamic range, or to go all Mudcrutch on everybody and release two versions of the same record.

Still, I think the biggest problem for Apple doing this is that iPods and computers are already stretched for SNR as they are - I'm sure most iTunes listeners are using crappy computer speakers that hiss like a mofo, most iPod listeners listen to them in low-dynamic-range environments or with low-sensitivity headphones, etc. Compelling Apple to compromise their listening volumes or noise levels is going to be a stretch.

A Modest Proposal for Ending the Loudness War

Reply #10
Moreover, if you follow the idea that reduced fidelity is reducing sales, the labels' hands are still tied. They can't increase dynamic range because a surprisingly large number of people will still complain that they can't hear their music anymore...


They can easily release two different versions and even charge a little more for the audiophile version.


Getting a few playback vendors to go along with this would eliminate that concern. Labels could do damn near whatever they want with dynamic range and they would not be punished in the marketplace for going "too quiet".


Then the users would compain to the playback vendors and not buy their products...

A Modest Proposal for Ending the Loudness War

Reply #11
They can easily release two different versions and even charge a little more for the audiophile version.
No, that's naked handwaving, and it is much less feasible than my proposal. The added cost of an audiophile mastering is a hell of a lot more than just $1500, it confuses the hell out of consumers, and on top of all that, I've heard anecdotally that labels rarely even make a profit out of them, even at a higher price. (Certainly that seems to be the case with vinyl releases.) Any solution that lets them produce a higher quality record while only requiring a single release is going to be a big plus for them.

Quote
Getting a few playback vendors to go along with this would eliminate that concern. Labels could do damn near whatever they want with dynamic range and they would not be punished in the marketplace for going "too quiet".

Then the users would compain to the playback vendors and not buy their products...
Only if they really became quieter, and only if customers aren't adequately informed.

A Modest Proposal for Ending the Loudness War

Reply #12
It's interesting that no one seems to think it's the artists responsibility.
Too often IMO there's a sense of: "oh those poor artists, controlled by profit hungry men in grey suits".
Artists choose to sign with labels (and become brands that function to sell products often made by parent companies i.e. Missy Elliott functions to sell Gap clothes).
Artists choose to allow their music to get squashed.
And the public choose to buy their squashed output.
Personally I choose not to.
Things will change when artists choose to take responsibilty for their work.

C.
PC = TAK + LossyWAV  ::  Portable = Opus (130)

A Modest Proposal for Ending the Loudness War

Reply #13
Still, I think the biggest problem for Apple doing this is that iPods and computers are already stretched for SNR as they are - I'm sure most iTunes listeners are using crappy computer speakers that hiss like a mofo, most iPod listeners listen to them in low-dynamic-range environments or with low-sensitivity headphones, etc. Compelling Apple to compromise their listening volumes or noise levels is going to be a stretch.

I cannot be bothered to explain this in more detail, but a practically efficient implementation of RG, would *require* optional automatic DRC! The idea is to:

- use RG automatically
- combine digital-amping, output-amping and DRC into ONE volume control

In short, the user would - besides of waiting for an automatic RG-scan with new music - not need to think about RG at all. All he would need to know, is that there is this volume knob, which strangely can be turned up above 100% (default RG target, full output-preamp) into a "red zone" which is louder but may decrease soundquality (digital amping above default RG-target, limiting active, full output-preamp).

That way, people can make their music as loud as it is now, if they wish, but informing them that it may decrease quality with dynamic music. So, they are free set the loudness/fidelity however they like, with just one simple volume knob - and everything will automatically be at equal loudness.
I am arrogant and I can afford it because I deliver.

A Modest Proposal for Ending the Loudness War

Reply #14

Moreover, if you follow the idea that reduced fidelity is reducing sales, the labels' hands are still tied. They can't increase dynamic range because a surprisingly large number of people will still complain that they can't hear their music anymore...


They can easily release two different versions and even charge a little more for the audiophile version.


They might easily make two different versions, but the economics are most likely not there for the distribution of two different versions.  The RIAA is not in the business of selling low volume, high margin items.  With the limited shelf-space per unique retail location available, there is not room for two versions of albums. 
On-line only small volume sales are boutique, and major labels own boutique labels expressly for that market.  The boutique labels, however, don't seem to be participating in the loudness wars to nearly the same degree as their mass-market siblings. 
In other words, I'm not holding my breath - you're a market segment they have long-since decided not to serve with their mass-market products.
Creature of habit.

A Modest Proposal for Ending the Loudness War

Reply #15
I cannot be bothered to explain this in more detail, but a practically efficient implementation of RG, would *require* optional automatic DRC! The idea is to:

- use RG automatically
- combine digital-amping, output-amping and DRC into ONE volume control

In short, the user would - besides of waiting for an automatic RG-scan with new music - not need to think about RG at all. All he would need to know, is that there is this volume knob, which strangely can be turned up above 100% (default RG target, full output-preamp) into a "red zone" which is louder but may decrease soundquality (digital amping above default RG-target, limiting active, full output-preamp).

That way, people can make their music as loud as it is now, if they wish, but informing them that it may decrease quality with dynamic music. So, they are free set the loudness/fidelity however they like, with just one simple volume knob - and everything will automatically be at equal loudness.


In even shorter, Lyx's volume knob goes all the way up to 11, for all the people who like their music loud and distorted.



(Us folks who enjoy audio fidelity, on the other hand, can have our nice quiet signal with proper dynamics.)

A Modest Proposal for Ending the Loudness War

Reply #16
If you don't like the product, don't buy it.

The studios may not understand sound quality, but they do understand profit.


They will only 'understand' this particular point, if the public has a clue as to *why* they aren't liking the music.

Do you think the first thing the industry thinks when it looks at dropping sales is, oh, goodness we'd better start increasing the dynamic range of our releases? Consumers are going to have to be taught to articulate their displeasure.

Here are some moves in the direction of changing the status quo.

http://www.turnmeup.org
http://www.metalliance.com

A Modest Proposal for Ending the Loudness War

Reply #17
In even shorter, Lyx's volume knob goes all the way up to 11, for all the people who like their music loud and distorted.



(Us folks who enjoy audio fidelity, on the other hand, can have our nice quiet signal with proper dynamics.)

Thus taking the loudness-decision completely away from the artist and label, and making it a user-decision. "Complete" in this case means, that users are also capable of choosing overcompression if they wish to - no argument left for oversaturation on the record itself, except of implementation costs.
I am arrogant and I can afford it because I deliver.

A Modest Proposal for Ending the Loudness War

Reply #18
The problem I see with your proposal though Lyx is that often the mastering techniques used to get albums as loud as possible don't translate well to automation. Perhaps there needs to be some procedural data stored as well to allow for more graceful volume maxing for the people who desire that...

A Modest Proposal for Ending the Loudness War

Reply #19
Good point, and artists and labels wont do that large-scale by themselves - which was one of the goals in the first place: not being dependent on label support.

Though, maybe that isn't even necessary? Just put a slightly larger amp on portable players, and then to suit those yurop decision makers put a matching cap on the digital amping. But wait, existing amps are already capped for yurop, so maybe bigger amps arent even necessary, just a higher cap?

I guess the question is: Is a slightly lower possible max loudness (at "mass-listenable" fidelity) a killer-downside of this idea? Does tech which implements this really need to be "exactly as loud as without the feature", or is "almost as loud" enough?
I am arrogant and I can afford it because I deliver.

A Modest Proposal for Ending the Loudness War

Reply #20
Personally, I think "almost as loud" should be loud enough, but some of these pop pushers are pretty persistent about dynamics compression... I'd love to see your idea in action. I'd have no problem with more distortion in loud masterings so long as I get nice clean dynamic recordings without all the dynamic compression.

A Modest Proposal for Ending the Loudness War

Reply #21
I agree that Lyx's approach is the technically superior approach in the long term. (And a wise old bird in a tree just reminded me that Vista/WMP basically already does this.)

But my original point is that this is not going to happen as the first step. If it really were that easy, all the playback vendors would have done this by now. As it stands, only Microsoft and Creative have tried. That should tell you something about the barrier to entry. Certainly it's easy to make any DRC integrated, but past experience shows that it is very expensive to make a good DRC.

Moreover, adding DRCs introduces a lot of complexity into the playback process that mastering engineers will wind up gaming for maximum loudness. It's possible that they will still employ hypercompression, merely to eliminate the need to use crappy DRC that might exist on some playback devices. It is simply not going to guarantee what you think it will.

Not only can we not rely on labels to make the first step, but we can't rely on playback vendors to make the first step, if it is anything beyond minuscule. Which is why I'm saying that DRC is simply not necessary to win this battle.

I think we've already seen the result of the education effort regarding the loudness war - most people roll their eyes. The fidelity losses, while significant, are simply not enough to change listening habits. However, loudness differences do have a very real impact on listening habits and purchase decisions, as evidenced by people buying louder remasters, tuning out of less-compressed radio stations, etc. By simply focusing on the loudness eq side, which is an extremely solved problem, and deferring the whole issue of DRC, I'm saying that the strong market bias against lower loudness will be eliminated, which will eventually allow the weaker market bias for higher fidelity to flourish.

 

A Modest Proposal for Ending the Loudness War

Reply #22
If I may weigh in on this subject...

The point of ReplayGain, as I understand it, is to reduce the perceived loudness of "excessively hot" tracks to a standard level, where the notional VU meter reads -20dBFS on the loudest sections.  Tracks (or albums) that are already recorded at sensible levels are not reduced (or, not as far), and their dynamic peaks are not truncated.  ReplayGain does not intrinsically do any compression of it's own.

To compensate, the analogue gain of the headphone amplifier must be set about 12dB higher, to allow tracks to be heard at the same level as the present worst offenders.  The trend recently has been to limit the maximum amplifier level on mobile devices, for health and safety reasons, precisely because people persisted in listening to these "hot" tracks at maximum volume.  And so people keep buying "hot level" tracks because anything else is too quiet to hear properly above the traffic.

With ReplayGain and a "normal" level headphone amplifier, the total effective gain on these hot tracks would remain the same as with newer devices, while the total effective gain on high-dynamics tracks would remain the same as on old, pre-H&S players.  Both of these are the "correct" listening level, where the VU meter sticks at about 80dB SPL.  The question is whether new portable hardware is still capable of producing the higher gains, if the firmware were upgraded with ReplayGain support.

I happen to have an old Panasonic portable CD player, from when CD players were just starting to get skip protection and remain affordable, and when the Health & Safety goons were just beginning to make worried noises.  If I put on my open Sennheisers (not the most sensitive of cans) and turn the volume to 10, I get a perfectly normal listening level with no obvious distortion - when I play my Berlioz CD, which is properly mastered.

Thus, the headphone amp in this old player is perfectly capable of handling the appropriate gain and level - on a bog-standard 3-volt battery supply (two AA cells).  A standard lithium-ion cell, as used in iPods, gives you 3.6V nominal.

Now, there *are* situations where DRC is appropriate.  Listening to a high-DR piece at night where you don't want to wake the neighbours, but you still want to hear the quiet bits.  Or, listening to music in the car (or worse, in a light aircraft) where the ambient noise level is very high.

For these situations, there should be a separate button - I remember we had a Kenwood car radio with a "Loud" button that simply cut in a mild compressor (though the manual was extremely vague about what it actually did and why you would want to use it), and my parents' new Yamaha AV-receiver has a "Night mode" button that does the same thing.  My old portable CD player doesn't have a "loud" button, and neither does anything else I have in my flat (and believe me, that's a lot of stuff).

It does have a "Super Extended Bass System" button - a simple bass-boost EQ - which causes horrible distortion if you turn it on at high gain levels, and is useless anyway if you have decent headphones or earbuds.  It also has a button to turn the anti-skip on and off, since the battery lasts longer with it off, but then it skips if you look at it sideways.  These are not useful buttons, to be honest - they smack of checklist-featureism.  A DRC button would have been more useful.

And I agree, it should be either impossible or tedious to turn ReplayGain *off*.  On simple or interface-constrained devices, it should be impossible.  On things like a full-fat iPod, it should merely be tedious (an option buried in the Settings panels).  On the plus side, the volume control should become much less important, because now the playback level is standardised - another thing that can disappear into the bowels of the Settings, with a default that works for standard earbuds and cans, instead of cluttering up the primary UI.

Apple would *love* that.

So would the "made in Taiwan" bunch, because a stereo pot is so many cents per unit!