Sign in to follow this  
Master Mi

End the loudness war - back to the roots of full range dynamics and the best music quality

Recommended Posts

Heya, guys - in the last time I've read a lot about topics like loudness war, compression, limiting and have made my own experiences with this at my DAW.

I think most people here know that compression and limiting can increase the loudness of soundtracks within the production process a lot - much more than all the single tracks and the master track could bear it by just turning up the volume (dB) - without causing clipping sounds.
Compressors and limiters make it possible to give a soundtrack much more loudness without clipping - I guess this is one of the main reasons for the ongoing loudness war, better compressors >>> higher loudness possible.
So, many radio stations, television broadcasting transmitters, commercial broadcasters and of course music producers started to increase the loudness of their audio stuff because they thought (and still think) that a bit louder audio stuff could attract much more listeners than the competitors at the global market.
And I guess they were right when it comes to the habits of the mass of listeners - they obviously liked the slightly louder (but more compressed) audio stuff often more than the quieter audio stuff of the competitors.

But the problem of this phenomenon is a continuously worsening of the audio quality of soundtracks - 'cause of the increasing compression or even just by cutting or emasculating the peaks with limiting.
Some fans of high quality hi-fi music or hardcore audiophiles criticise this - and I guess they're absolutely right.

I also think that it's pretty nonsensical to do nearly everything (even cutting down the full range of dynamics which is often really beautiful to listen to - without taking anything away from this) to increase the perceived loudness like a mad hatter who just wants to be the winner of the loudness war.

I think we should move away from the war of loudness back into our ambition to increase dynamics and music quality to the upper limits - yeah, without limiters.

Maybe we don't even need any compressors/limiters for creating the best soundtracks if we just edit and mix/master them right - or at least we would use compressors only then if we want to just give some pressure to drums/basses.


So, I guess we need some kind of loudness rules - what do you think of EBU R128?
https://tech.ebu.ch/docs/r/r128.pdf

I have a loudness metering tool in my DAW (Samplitude Pro X2 Suite) based on this standards.
What with your DAWs/loudness metering tools?


Would be cool if the remixers of OCRemix would support the idea: "End the loudness war - let's back to the roots of full range dynamics instead".


What's your opinion about this topic?

Share this post


Link to post
Share on other sites

I feel like you are jumping to one distinct conclusion.  There are different appropriate RMS levels for every genre of music.  Frankly different RMS levels work better for different listening devices as well.  There is no 1 level to rule them all.  it is all a balancing act.

 

Also there is a scientific reason why people prefer sounds with higher RMS levels.  When the loudness is over a certain threshold we release serotonin.  One of the scientific reasons people love live music.  

 

Compressors and Limiters are not always used to make a sound louder.  They can also be used to make a sound quieter.  Hence making other sounds appear louder and more dynamic.  They can also be user artistically to bring out different timbres in the sound.   

 

Also another note, most of our best compressors came out 50+ years ago.  So the argument of better compressors = greater loudness is null and void.  To be honest people should learn how to use compressors and limiters in a more appropriate manner.  

Share this post


Link to post
Share on other sites

EBU R128 is already well on its way to being standardised worldwide across broadcasting and streaming services. And even those who don't follow EBU guidelines specifically usually have their own equivalent in place, like music published through YouTube's new music service. You're not really going to do yourself any favors brickwalling music today, as most outlets will have algorithms in place to gain-adjust the track pre-emptively so the average loudness will not be perceived as louder than music with more dynamic range. I think people who do this and notice their music being played next to more dynamic music are eventually going to wise up and utilize more of that real estate in dynamics that they're not getting.

 

I feel like you are jumping to one distinct conclusion.  There are different appropriate RMS levels for every genre of music.  Frankly different RMS levels work better for different listening devices as well.  There is no 1 level to rule them all.  it is all a balancing act.

 

Also there is a scientific reason why people prefer sounds with higher RMS levels.  When the loudness is over a certain threshold we release serotonin.  One of the scientific reasons people love live music.  

 

Compressors and Limiters are not always used to make a sound louder.  They can also be used to make a sound quieter.  Hence making other sounds appear louder and more dynamic.  They can also be user artistically to bring out different timbres in the sound.   

 

Also another note, most of our best compressors came out 50+ years ago.  So the argument of better compressors = greater loudness is null and void.  To be honest people should learn how to use compressors and limiters in a more appropriate manner.  

 

I don't see anything in the OP hinting towards compression being an inherently bad and destructive tool. I think you're jumping to conclusions here. And I believe the whole "different RMS works better for different devices" argument is largely blown out of proportion as well. And, at the end of the day, it just seems completely backwards to me that you would mix/master for the absolute lowest common denominator output in mind and deprive everyone else of the option for better quality. The actual logical solution for this argument would have been to implement processing options in the player devices themselves (which some seem to be doing for movies/TV with "late night mode").

Share this post


Link to post
Share on other sites

I think we should move away from the war of loudness back into our ambition to increase dynamics and music quality to the upper limits - yeah, without limiters.

Maybe we don't even need any compressors/limiters for creating the best soundtracks if we just edit and mix/master them right - or at least we would use compressors only then if we want to just give some pressure to drums/basses.

Um... yes, we do need Limiters? The whole reason why we DON'T clip is that we DO have Limiters. That's the whole point of having them in the first place. We DO need Limiters.

Share this post


Link to post
Share on other sites

Um... yes, we do need Limiters? The whole reason why we DON'T clip is that we DO have Limiters. That's the whole point of having them in the first place. We DO need Limiters.

 

Well, no, we don't need limiters in general, we need them when we need to raise the volume of our signal and have the peaks taken care of. Limiters are a loudness tool, not a necessity. They're the norm nowadays, but don't confuse that for some kind of technical system design truth. Limiters aren't the only way to avoid clipping, and they're a fairly recent development in recording engineering history that a lot (not I) would argue is fueled by mastering inexperience.

 

To clarify, they have insane benefits and I agree with using them; I'm just pointing out that your reasoning here is a little weak for a statement that limiters were invented for the purpose of stopping clipping, and that until we had limiters, everyone clipped, which is kind of insulting to the industry. This is the implication when you say the whole reason we DON'T clip is that we DO have limiters; ergo, you are saying not having limiters means we always have/had clipping, which is... incredibly false.

Share this post


Link to post
Share on other sites

Well, no, we don't need limiters in general, we need them when we need to raise the volume of our signal and have the peaks taken care of. Limiters are a loudness tool, not a necessity. They're the norm nowadays, but don't confuse that for some kind of technical system design truth. Limiters aren't the only way to avoid clipping, and they're a fairly recent development in recording engineering history that a lot (not I) would argue is fueled by mastering inexperience.

 

To clarify, they have insane benefits and I agree with using them; I'm just pointing out that your reasoning here is a little weak for a statement that limiters were invented for the purpose of stopping clipping, and that until we had limiters, everyone clipped, which is kind of insulting to the industry. This is the implication when you say the whole reason we DON'T clip is that we DO have limiters; ergo, you are saying not having limiters means we always have/had clipping, which is... incredibly false.

Except I never actually intended to say (explicitly, on the page, literally) that people always clipped without limiters. You implied that. What I DO intend to say is that without limiters, it was easier to clip. If someone says that "hey, we should try mixing without limiters!", well... we'd have to be really careful of our peaks, then, because there's no ceiling above the song.

 

That said, I'm pretty sure we OC ReMixers joke about the loudness war being an "avoid winning it" type of deal. :razz: So, it's not as if we (or at least I) don't agree with the OP on at least that point.

Share this post


Link to post
Share on other sites

I don't see anything in the OP hinting towards compression being an inherently bad and destructive tool. I think you're jumping to conclusions here. And I believe the whole "different RMS works better for different devices" argument is largely blown out of proportion as well. And, at the end of the day, it just seems completely backwards to me that you would mix/master for the absolute lowest common denominator output in mind and deprive everyone else of the option for better quality. The actual logical solution for this argument would have been to implement processing options in the player devices themselves (which some seem to be doing for movies/TV with "late night mode").

 

From the OP:

"But the problem of this phenomenon is a continuously worsening of the audio quality of soundtracks - 'cause of the increasing compression or even just by cutting or emasculating the peaks with limiting."

AND

"Maybe we don't even need any compressors/limiters for creating the best soundtracks if we just edit and mix/master them right"

 

 

Where and how frequencies are distributed based upon different listening devices is crucial.  Not once did I remotely mention mastering/mixing to the lowest common denominator.  Any music or sound should be tested on multiple listening devices.  

 

Imposing Loudness standards for TV and/or Radio is way more plausible than other mediums.

 

BTW Limiters have been around a looong time:

http://www.recordingmag.com/resources/resourceDetail/109.html

Share this post


Link to post
Share on other sites

I like big sounds and I cannot lie. An EDM track that doesn't have a brickwall on it just doesn't fit into the rest of EDM, and that's not going to change anytime soon. Like Avaris said above, it's subjective and should be handled on a per-case basis.

 

Meanwhile, I'm gonna be over here with my limiters and my compressors making phat and tasty sounds. :D

Share this post


Link to post
Share on other sites

The actual logical solution for this argument would have been to implement processing options in the player devices themselves (which some seem to be doing for movies/TV with "late night mode").

I would kill for a portable MP3 player with built-in compression and leveling options. It would make listening to classical music in noisy environments such as buses and airplanes so much more pleasant.

Share this post


Link to post
Share on other sites

Personally, I've always tried to keep my tracks in the -18 to -24db range before doing anything with them, and I shoot for a peak of around -5db before any further EQing. I've never really worried about limiters because the peak db is rarely anywhere close to clipping. That being said, my tracks have always been quiet(moreso than they should be, maybe), so I've recently been using limiters and pushing the volume a bit higher. I haven't really messed with compression, though. I like to have a nice dynamic range, and compression often takes my music's natural soft-loud dynamic shifts and squashes them flat.

 

I'm not really against using limiters and compression. I think they're great tools for making audio sound as good as possible as long as you don't overuse them.

 

 

 

I would kill for a portable MP3 player with built-in compression and leveling options. It would make listening to classical music in noisy environments such as buses and airplanes so much more pleasant.

 

I use one of those cheap Android-based phones as a portable MP3 player. About $20 for one, use a large micro-SD card for storage and has a headphone port. Some of them have some OK leveling options by default, and there are quite a few equalizer apps available. Beats the hell out of most MP3 players in both cost and function.

Share this post


Link to post
Share on other sites

Well, it definitely depends on your purpose. If you want to write EDM-style music, sometimes high compression is helpful for the glue you want, and a limiter would help keep that in check. Sometimes, if it's really bassy music and you want some insane bass presence without the clipping, again, it helps to have that Master track compression to glue things together and a brickwall limiter just in case. And if it's orchestral, you usually don't need Master track compression. (with the occasional exception of controlled/careful multiband compression)

Share this post


Link to post
Share on other sites

And if it's orchestral, you usually don't need Master track compression.

 

Are you sure that song was appropriate to connect to that statement? The volume in that song is yikes.

Share this post


Link to post
Share on other sites

Yeah, I'm sure. I heard the dynamics still preserved in there. Granted, it's a hybrid orchestral track, but it's still orchestral, and I didn't hear overcompression that I would point out. But, for you, I'll use a more suitable track (I'm aware of the poor youtube compression on this, but the actual audio file is pretty dang loud =P).

Share this post


Link to post
Share on other sites

I see loudness as more of a taste thing, though it really started to get beyond reasonable limits past the early 2000s.  I mean I don't know, it's about what sounds bad.  Unfortunately a lot of bad sounding stuff is getting approved and even desired because of a multitude of factors, despite there being nowhere else to go in terms of headroom.

 

I'm extremely unimpressed by loud stuff.  It can never shake its lack of dynamics and tiny sound.  Wow, okay, amazing,  you got it loud.  Almost NEVER sounds great in addition to that, though.

Share this post


Link to post
Share on other sites

I've uploaded a version of my newest remix without any compressors/limiters (not in one single track, not even in the drums and of course not in the master track).
It may lack in loudness - but I like it much more and it's much clearer than in the version with limiters/compressors.

Here's the previous version 1.1 with compressors/limiters (not in the master track - but in the flute and drum track):
>>> https://soundcloud.com/master-mi/star-tropics-the-fire-of-the-southern-cross-master-mi-remix-version-10-uc

Here's the version 1.2 without any compressors/limiters:
>>> https://soundcloud.com/master-mi/star-tropics-the-fire-of-the-southern-cross-master-mi-remix-version-12

I created this version pretty close to the EBU R128 standard - had still about 3 dB unused headroom in my DAW/master track.
Just the LUFS in my track was a bit louder and had a far bigger range (about 4 dB or +/-2 dB instead of the recommened range of 2 dB or +/-1 dB) between the loudest and the quietest part.
In R128 the absolute LUFS has to be around - 23 dB +/- 1 dB.
The absolute LUFS of my newest track is between - 21,5 dB and - 17,5 dB, so a bit louder - I just wanted to use as much safe headroom and "natural, uncompressed loudness" in my DAW as possible.

I don't get it why R128 only allows a loudness range of 2 dB between the loudest und the quietest part - doesn't make sense at all because I also like classical tracks where you obviously have differences of much more dB between silent parts und the heavy swells, the louder parts sometimes.

Share this post


Link to post
Share on other sites

It's all great to talk dynamics in classical music and whatnot, but standards aren't devised with musical ideals in mind; they're devised with technical efficiency in mind.

 

The simple fact of the matter is that with TV and radio, you're rarely going to have circumstances where the quiet parts of widely dynamic music is going to heard in detail. This is especially true with TV, where stuff happens over the music. Doubly true when listening to stuff in cars; road noise renders piano and pianissimo pretty inaudible. This is why I can't play, like, the Lord of the Rings score or something in the car. My dad complains he can't really hear anything.

 

Life isn't a concert hall where everyone sits down and admires in silence, and recordings with dynamics need to be listened to in proper environments like silent rooms with speakers (lazy sunday afternoon in your living room, maybe) or headphones in order for the quieter parts to be appreciated. For the world at large, dynamic music doesn't really cut it for the way people listen to music/audio. Technical standards are designed with those kinds of considerations.

 

That's probably also a small factor in what affects mainstream popularity. The popular stuff is the loud, bassy and in your face stuff. Rap, pop, rock, etc. It can be listened to anywhere, while jogging, driving, in a gym, etc.

Share this post


Link to post
Share on other sites

Lower dynamic range isn't a problem, but the amount of limiting and hard clipping in this age of music is absurd.  It sounds (and looks) like crap.  Again, there's nowhere to go, so past a certain range you're just reducing quality and nothing more.

Share this post


Link to post
Share on other sites

As others have pointed out, the mediums and listening environment are important factors in this.  I'll just add that mp3 itself, even at highest bit rates have a significant impact on the experience, particularly on more complex sounds (bells, cymbals... etc).  Reducing the dynamic range helps a lot to address this but from an artistic point of very is a major limitation.

As data storage and transfer technology develops, it is inevitable/(hopeful?) that lossless formats will take over.  Saying this, mastering techniques are always going to be important but hopefully in time it will become more viable to use them in more subtle ways.

Share this post


Link to post
Share on other sites

I like big sounds and I cannot lie. An EDM track that doesn't have a brickwall on it just doesn't fit into the rest of EDM, and that's not going to change anytime soon. Like Avaris said above, it's subjective and should be handled on a per-case basis.

 

Meanwhile, I'm gonna be over here with my limiters and my compressors making phat and tasty sounds. :D

 

I started a kneejerk, then I looked at your name, and a wave of reason overcame me. I always advise producing according to the emotional needs of the music, and many talented engineers follow this philosophy. We can abuse the limiters in fantastic ways, and so long as the emotion is there, the listener will disregard the functional aspects of the sound.

Share this post


Link to post
Share on other sites

I'm perfectly fine with heavy compression, so long as it fits the style of what you're going for, it's all intentional, not overly loud volume-wise (regular-old volume, for example, but also in terms of how much clipping there is), and it doesn't accentuate the harshest frequencies of the most annoying (or least accessible, to be nice) sounds in the mix. i.e. Part of the reason why I hate certain dubstep is that I tend to hear quite resonant wobbles and repetitive sirens (it's not THAT interesting! GOSH).

Share this post


Link to post
Share on other sites

This is an example of compression on the Master track actually doing something helpful ("Smooth Surfing" Before->After). Instead of turning the track into overcompressed mush that clips, it glues the components in the track together, evens out loud peaks, and makes the drums punchier. That's the kind of heavy compression I could live with. Doesn't have to be super obvious.

Share this post


Link to post
Share on other sites

Right but what ratio?  How much GR?  I doubt either of those numbers are ten.

 

I haven't found a lot of heavy compression in mastering.  Mostly that would be done in parallel.  The mix usually has compression on it, which is the sum of compressed busses, which are splits and/or sums from compressed individual tracks.  So really by the time it's at this final stage there's been 2-4 layers of compression already.

 

With limiting we've been in the era of clipping converters for over 10 years now.  It's whatever clients want, we went way beyond reason long ago.  If you ask me, anything post-early 2000s is generally pushed beyond reason.  There are plenty of exceptions, it's just the standard overall has continually been lowered, not raised.

Share this post


Link to post
Share on other sites

Right but what ratio?  How much GR?  I doubt either of those numbers are ten.

I'm assuming GR means gain reduction. Unsure what that corresponds to (perhaps Threshold), but I have:

 

Threshold: -16.0 dB

Range: -63.0 dB

Makeup Gain: 4.0 dB

Attack: 3 ms

Ratio: "4" (options: 2, 4, 10, so approximately "mild", "medium", and "high" respectively) --- I rarely use 10, but I rarely feel the need to

Release: 0.1 seconds

Peak Clipping: On

High Pass: 50 Hz (affects 50 ~ 20000 Hz)

40% Wet Mix, 60% Dry Mix

 

(Note that I didn't even say what the values are for the individual drum busses, which have gains closer to 10 dB, thresholds closer to -15~-20 dB, and about 50/50 wet/dry)

 

For practical reasons, I would much rather get everything "right" before doing anything substantial on the Master track, like Master track compression. If it doesn't sound good without Master track edits, there's probably something off elsewhere.

 

Even with this compression though, clearly it was an example of not overdoing it, and so I don't see why I should not. I understand that some people might overdo it, but I'm paying close attention to that these days so that I can avoid writing music that is way too loud.

Share this post


Link to post
Share on other sites

Timaeus,

 

That really isn't much of a comparison as the elements are balanced differently. A proper comparison is against a track that is already mixed with compression added after the fact then A/Bing between the two. Any other way completely negates the point of the comparison as something done to a single instrument such as the lead can drastically influence how individuals will interpret (i.e. hear) the main keys in a track. That is just a stupid example, but the point is that by changing one instrument in the mix the way the mix is heard is also changed in more than one way. 

 

Quick example of just turning on & off some light Master Compression (~2dB GR: 30ms attack, 100ms release, 4:1 ratio)

Before

After

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Sign in to follow this