Hard Limiting will kill the dynamics

"Hard Limiting will kill the dynamics"

I was motivated to think about this from a post by Bag recently. I also Googled the phrase and it comes up a lot. So, does Hard Limiting (in AA 2.0) kill the dynamics, and what are the dynamics of an audio file, what does this mean exactly? Does Hard Limiting affect speech in the same way as music?

As luck would have it, SteveG recently posted a very interesting article relevant to this.  Have a look at:  http://audiomastersforum.net/amforum/viewtopic.php?t=5695

As for your questions, I'm sure Steve will have a far more scientific definition, but basically the dynamics of a piece of audio could be defined as the difference between the quietest parts and the loudest parts.  This applies whether we're talking about a recording or a live performance.  Indeed, one of the biggest tasks for the conductor of an orchestra is controlling the dynamics of the orchestra.

Yes, used incorrectly the Hard Limiting function can very definitely kill the dynamics of an audio file.  Hard Limiting is very much a manual function since it's up to the user to determine both the maximum level (the point where limiting occurs) and the amount of gain applied to the signal.  In theory, you could set the limit at 0dBFS and apply 100dB of boost, raising even the noise floor to the maximum level!

That said, used sparingly, Hard Limiting can also be a valuable tool in some cases.  A lot of my work is doing sound effects for use in theatres.  Theatres (nowadays) are noisy places, with ventilation noise, audiences coughing and rattling sweet wrappers and the lighting deparment's moving fixtures whirring away.  As a matter of course I eliminate some of the dynamic range of most of my effects to ensure the quiet bits stay above the background racket.

I imagine that, in your radio work, you face similar problems and apply similar solutions.  Certainly, when I was working in TV, we would apply a compressor/limiter to most voice as a matter of course.  The amount we'd apply depended on the programme style.  Certainly, even voice needs some dynamics left in, otherwise it becomes flat, uninteresting and tiring to listen to.  However, different radio styles tend to have different standards...we've all heard the super-loud, alway-yelling top 40 format.

It's worth saying at this point that the compressor limiter function does something similar, but allows you to vary the amount of gain applied depending on the initial level...you can raise the quiet bits by more than the louder parts, vary how extreme the limiting is, and so on.  Both functions have their places.

Bob

As for your questions, I'm sure Steve will have a far more scientific definition, but basically the dynamics of a piece of audio could be defined as the difference between the quietest parts and the loudest parts.  

I do have a more detailed one (for measurement purposes), but that one willl do just fine in this context...

I think radio is the bottom-line. Everyone in the music business and in radio has asked themselves at least once 'why does music sound so good on the radio?' And the reason is compression. It's addictive actually, so I can't really agree with the people who say things like the brain can't take it for very long, because otherwise radio people wouldn't be able to do their jobs (and yes some of us keep the headphones on all the time or for long stretches, and nothing sounds unbearable). I think that what's hapened is that people have cottoned on to that compressed radio sound and they are desperate to impress the radio people who choose what to playlist, knowing they are addicts of the compressed sound. You cannot have two standards, one that says the compressed radio sound sounds great, and a completely opposite view. And if all this is so, why haven't the radio people themselves complained that the records sound too compressed? the only time I have ever heard compression madness on recent recordings was with Prince records (and now Justin Timberlake, who seems to be copying him).

I will admit to noticing distortion on some CDs in recent years, and being somewhat preplexed at what could be causing that .. some vocals have seemed to "screech" and not be as clear as you'd expect, and this might be all about the make 'em louder approach, if it is, then I am disapppointed. The clipping the article mentions is all too clear to witness when you import any current Top 40 track into edit view.

The article does not mention the production revolution called "The Wall Of Sound" used by The Beach Boys, and other incredibly revolutionary make 'em louder techniques used by Motown producers. You can't in one breath appreciate all that, and in another, be totally opposed to it. Actually to me, many Motown records sounded like a distorted "screech" to me, and like most people I guess, put it down to the inferior recording equipment of the era .. not so, it was all down to those nasty producers trying to make records sound "loud"   How they didn't bring the recording industry down is a mystery.      

Back to my original question. This is all about my continuing quest to make sure that my voice-tracked files are correctly treated, and all I'm trying to achieve is to be heard clearly over music in my programmes. Because it's mostly loud pop music I have been enduring a battle to do this well. I have gone through compressing my files, over-compressing, not compressing at all, and a few months ago I thought I'd found the prefect thing, just do a Hard Limit. This seems to have a similar effect to the compressors, in that any quieter spoken parts can still compete well with the music while not sounding over-compressed. If I didn't do anything at all to my files, I am drowned out more often than not, so it's not easy. Now I learn that I might be removing the 'dynamics' of my spoken links by Hard Limiting, wow, so much for a good idea - so I panicked again. But if all Hard Limiting is, is another form of compression. In the context of my radio programme productions, I think I might be justified in making myself sound "louder", and if I'm going to start worrying about losing 'dynamics' (that word belongs in a car catalogue) then we should email all the radio stations in the world and warn them that all their transmission compressors are causing all their broadcasters to lose their 'dynamics'.

The article does not mention the production revolution called "The Wall Of Sound" used by The Beach Boys, and other incredibly revolutionary make 'em louder techniques used by Motown producers. You can't in one breath appreciate all that, and in another, be totally opposed to it. Actually to me, many Motown records sounded like a distorted "screech" to me, and like most people I guess, put it down to the inferior recording equipment of the era .. not so, it was all down to those nasty producers trying to make records sound "loud"   How they didn't bring the recording industry down is a mystery.      

I'm not surprised they didn't mention the Wall of Sound - well not like that, anyway. The Beach Boys never used the technique specifically; the Wall of Sound was a Phil Spector idea, although it's true that Brian Wilson was significantly influenced by Spector. If you want a good UK example of the technique, then listen to the Beatles (yes, the Fab Four) 'Let It Be' album. And the sound wasn't achieved by compression either; initially, it was based on sound reproduced in front of a real wall, in the echo chamber of the Gold Star Studios. And it was specifically a mono technique as well, all picked up with a single mic. It's certainly true that the Wall of Sound didn't have much in the way of dynamics, but that was because the sound never stopped. It wasn't specifically compressed at all, it was reverberation. McCartney said that Spector's use of a modification of the technique ruined Let It Be, and that's why the 'naked' version was later released.

The ultimate example though is generally considered to be 'Be My Baby' by the Ronettes - of course one of them was Spector's wife, so I suppose that this is hardly surprising. And this was also the track that supposedly influenced Wilson the most.

All of these techniques were very much of their time, and fortunately, for a while, the industry moved on. And as far as Berry Gordy was concerned, all Motown records had to sound good on his grot box in the office, otherwise they weren't released - simple as that. The production techniques were actually more sophisticated than Spector's, and not based on the same model at all. Most of it was down to using two, or sometimes three drummers, four guitar lines and a single line bass - check out the Funk Brothers for more detail.

If you are going to slag off production techniques, then please at least get the facts straight...

I think radio is the bottom-line. Everyone in the music business and in radio has asked themselves at least once 'why does music sound so good on the radio?' And the reason is compression.

I don't think I've ever heard anyone ask why music sounds so good on radio.  More often I've heard them ask why it sounds so bad.  However, the answer is still the same:  compression.  (Often coupled with things like an Optimod in the transmitter chain.)

Seriously, I think there are two things at work here.  First, is the radio station's desire to sound the "loudest" when a listener sweeps through the dial.  Loud is equated with good reception.  This has nothing to do with improving quality...it's just trying to hook listeners.

The second issue is one that affects all music these days:  cars, walkmen, discmen and MP3 players.

Gone are the days when people would listen to radio or records in the quiet of their own homes.  Nowadays, lots of music and certainly most radio, is listened to on the move in noisy environments.  The quiet portions of recordings with a reasonable dynamic range (or the quiet portions of BFM's voice for that matter) are going to get lost in the general hubbub of today's life.  So...people now think that a good recording is one that can drown out the background (and probably damage their hearing in the process).

So...for his world, BFM is probably right, just as I have to compromise my effects recordings for a noisy theatre.  However, you can count me amongst those who find most modern recordings tiring to listen to.  Even more, I find them flat and uninteresting.  At the risk of showing my age, if I want to feel INVOLVED in the music, I'll dig out my old half-speed mastered vinyl recording of Dark Side of the Moon (or maybe some of my older classical recordings) and throw them on the turntable.  Suddenly instead of a noise, I have music for which I can shut my eyes and "be there".

Bob

The article does not mention the production revolution called "The Wall Of Sound" used by The Beach Boys, and other incredibly revolutionary make 'em louder techniques used by Motown producers. You can't in one breath appreciate all that, and in another, be totally opposed to it. Actually to me, many Motown records sounded like a distorted "screech" to me, and like most people I guess, put it down to the inferior recording equipment of the era .. not so, it was all down to those nasty producers trying to make records sound "loud"   How they didn't bring the recording industry down is a mystery.      

I'm not surprised they didn't mention the Wall of Sound ...

If we're going to be fair about the LOUD effect, we should be surprised they didn't mention it in this new article. The truth is that record makers have been trying to do this for over 40 years, it's not new, the end result is the same. It's not implicit that I thought the wall-of-sound was acheived with compession, I knew full well that it was arrived at by mixing techniques. In fact I'm pretty sure the Motown article mentions that the Motown producers did interesting things with stereo panning, dubbing extra layers of vocals in the centre with certain EQ techniques and so on to get that louder sound .. which to me sounds like screeching and I have to cover my ears

I think radio is the bottom-line. Everyone in the music business and in radio has asked themselves at least once 'why does music sound so good on the radio?' And the reason is compression.

Gone are the days when people would listen to radio or records in the quiet of their own homes.  Nowadays, lots of music and certainly most radio, is listened to on the move in noisy environments.  The quiet portions of recordings with a reasonable dynamic range (or the quiet portions of BFM's voice for that matter) are going to get lost in the general hubbub of today's life.  So...people now think that a good recording is one that can drown out the background (and probably damage their hearing in the process).

I don't think today's world is noisier, there are noisier jungles. Maybe I was brought up in a particularly noisy home but I never experienced a quiet home .. the kids with their TVs/radios upstairs, Mum with her TV in the kitchen and Dad with his TV in the front room, all going at the same time, and somehow we could all listen to our own thing quite hapily without being disturbed by any of the other machines, which we would only hear on the stairs. Besides, and this is crucial, the way our hearing works is that we filter out the sounds we don't want to hear and we focus on those sounds we do want to hear. If we didn't have this filter we'd go mad trying to listen to it all simultaneously. Once we have our on-the-move-headphones on we don't hear anything else, so no our listening environment is not noisier today.

Aside from wanting to achieve the addictive compressed radio sound which, totally contrary to the new article, everybody actually loves, I suspect that what people are still trying to get past is the inadequacies of the playback equipment still. CD is still not the utpoian recording medium and digital playback sounds dull and lifeless even if a loud-producer has not gone near it. Vinyl still sounds better, putting aside the crackles. I don't know the reasons, maybe digital will start to sound good when much higher bitrates are possible. But all the CD has achived so far is a clean record, so records don't have scratches and crackles any more, good .. it's a start.

Maybe your home wasn't quiet...but I bet it was a heck of a lot less noisy that a typcial urban street or the interior of a moving car (unless we buy a Rolls of course).  A decade or two ago, it simply wasn't possible to carry any form of "quality" music with you, but nowadays my kids go nowhere without their MP3 players plugged into their ears.  Unfortunately, they want the output of their players to drown out the town noises around them, and these town noises are often around 80dB(SPL) with higher peaks as the bus or lorry goes by.

I'm sure it's this mobile listening lifestyle that responsible, at least in part, for the current desire for over-compressed music.  Alas, it's also why, at 54, my hearing recently measured better than my 15 year old daughter.

Bob

...all the CD has achived so far is a clean record, so records don't have scratches and crackles any more, good .. it's a start.

I can't go along with this - leaving aside the deliberate abuse of modern digital technology (and the frailty of the medium itself) - we have the best, by a very long way, recording and reproductivve quality that has ever been available to the public.

What about extended frequency response, greater dynamic range, increased channel separation, etc.?  Do none of things make a difference to your listening pleasure.

You can walk into any high street store and purchase, for a relatively miserly sum of money, an audio system which would put my first 'real' hi-fi system - Quad tuners and amps with Leak loudspeakers, this was not cheap! - to shame.

Aside from wanting to achieve the addictive compressed radio sound which, totally contrary to the new article, everybody actually loves, I suspect that what people are still trying to get past is the inadequacies of the playback equipment still. CD is still not the utpoian recording medium and digital playback sounds dull and lifeless even if a loud-producer has not gone near it. Vinyl still sounds better, putting aside the crackles.

Oh, come on Bernie, you must be winding us up.

I can occasionally listen to BBC Radio 3 or 4 in the UK, but the rest of what I used to listen to is now so tiring and distorted that I'm afraid I've virtually given up on radio. Speech, especially, is just appalling, with some voices hitting particular grating sounds that the whole experience is something I avoid.

I've just sold a decent FM tuner because I can't stand it any more. When you look at how some of these radio processors actually work - with the massive distortion in each of the narrow bands being filtered out before being re-combined into the final noise - it's not surprising how bad it is.

It's not implicit that I thought the wall-of-sound was acheived with compession, I knew full well that it was arrived at by mixing techniques.

Fine, except that it wasn't... there is a world of difference between achieving the effect temporally, and the effect of compression. This effect was achieved by means of natural diffusion, not mixing.

In fact I'm pretty sure the Motown article mentions that the Motown producers did interesting things with stereo panning, dubbing extra layers of vocals in the centre with certain EQ techniques and so on to get that louder sound .. which to me sounds like screeching and I have to cover my ears

Whatever it took to get Berry Gordy's grot box to sound okay, they did. But since the box very likely compressed everything anyway, they probably didn't have to resort to that at all. Once again, we are in temporal territory, not necessarily compression.

Aside from wanting to achieve the addictive compressed radio sound which, totally contrary to the new article, everybody actually loves...

There are several contributors to this thread who don't, for a start. And believe me, we are the tip of an iceberg.

Maybe there's something wrong with your hearing? Perhaps you need to have everything compressed in order to percieve it at all? If you spend enough time in this sort of environment, especially wearing headphones, there's ample evidence that TTS (temporary threshold shift) occurs, and if you do this for long enough, it turns into PTS. Now, if you advocate more compression, people listening on headphones are going to be exposed to a higher overall level of noise, and this will damage their hearing sooner. I'd call that thoroughly irresponsible, myself. Damage your own hearing by all means - but let's leave it at that, shall we?

And it's not just personal stereo listeners that suffer either - the amount of compressed bass that a lot of car systems produce, that has the effect of sounding louder outside, and in the surrounding vehicles (and I actually measured that, and know why, before you tell me it's not true) is also influenced by the overall level of compression - the bass has to stay louder to keep up with the rest of the mix. The effect of this is, to say the least, antisocial.

In the fullness of time, if there is any justice in the world at all (which I'm beginning to doubt   ), you will find yourself in a minority of one over this issue.

Steve, you keep assuming so many things that are just not so. Firstly that I am not supporting nor encouraging anything or anybody, I'm just telling it like it is, that there are many thousands of people who work in radio around the world are hooked on compression. Now Just because I say that does not automatically mean that I think that's a good thing, or that I agree with it and so forth. Take a chill pill my friend, you're very uptight

[I]

Steve, you keep assuming so many things that are just not so. Firstly that I am not supporting nor encouraging anything or anybody, I'm just telling it like it is, that there are many thousands of people who work in radio around the world are hooked on compression. Now Just because I say that does not automatically mean that I think that's a good thing, or that I agree with it and so forth. Take a chill pill my friend, you're very uptight

Doesn't read like that though, does it? You said it - why do I need to assume anything, or is this the royal 'I', etc?

I think radio is the bottom-line. Everyone in the music business and in radio has asked themselves at least once 'why does music sound so good on the radio?' And the reason is compression. It's addictive actually, so I can't really agree with the people who say things like the brain can't take it for very long, because otherwise radio people wouldn't be able to do their jobs (and yes some of us keep the headphones on all the time or for long stretches, and nothing sounds unbearable). I think that what's hapened is that people have cottoned on to that compressed radio sound and they are desperate to impress the radio people who choose what to playlist, knowing they are addicts of the compressed sound. You cannot have two standards, one that says the compressed radio sound sounds great, and a completely opposite view. And if all this is so, why haven't the radio people themselves complained that the records sound too compressed? the only time I have ever heard compression madness on recent recordings was with Prince records (and now Justin Timberlake, who seems to be copying him).

I don't think that it's me that's getting uptight here...  

Properly used, hard limiting should have only a transient effect on dynamics.  There should be no effect on the dynamic range other than at the precise and very brief points that the hard limiter is affecting the audio.  The way to demonstrate very clearly what's going on is to use Audition's hard limiter on a test file, save the result, then apply an equal amplitude increase to the original, save that, and invert one result against the other.  You should be left only with the tiny spitches at the points where the limiter cut in.  If more, then you've misused the tool, which should normally be used to curb out-of-the-normal transients, and turned it into an effect which is likely to sound quite unpleasant.

Other forms of compression may affect the dynamics through much of the audio, depending on the settings used.

I routinely use hard limiting when preparing concert material for sending on CD to the musicians, to avoid having the whole level of the CD being reduced  by what might be a quite chance peak in levels.  I also usually use the Classic master limiter VST on the main output of DAW software when live recording, to enable me to hear the detail of quiet passages without being deafened by the loud.  In that scenario I'll quite possibly be lopping 4 - 6dB off the loudest peaks, but that's in a monitoring situation only and the actual recording of course retains its full dynamic range.

Of course if you get the level wrong with 16 bit recording in particular, the audio will not only be hard limited at full scale because there's nowhere further to go, but it will be distored (clipped) too.  To my mind it's better to limit just a little rather than clip, if you have to choose one or the other.