Limiting and compression in Classical orchestral releases

[mohthom]

It worked out pretty well, as the film was to be screened only at selected venues we were able to spec pretty much exactly what we wanted and had control over playback. From a whisper to ripping the cones!

MohThoM

[joelpatterson]

Originally Posted by mohthom

It worked out pretty well, as the film was to be screened only at selected venues we were able to spec pretty much exactly what we wanted and had control over playback. From a whisper to ripping the cones!

MohThoM

Ah, I see... you're lucky to have control over the playback and immerse the audience, I was thinking of the scenario where they're flying down the highway with the top down.

[mohthom]

. . . is this thing on??

[adjusts stereo]

WWWAAAAAAAAAAAAAHHHHH

[rubbing ears in pain]

[5flagsaudio]

Most of my large ensemble recordings get dynamic treatement. Usually I'll use a very small ratio and very large threshold on a compressor. This usually results in about 4-5 dB of reduction. On top of that, if I feel it needs it, I'll take another dB off the top with a limiter (really percussive things tend to get this treatment). Naturally, I'm not the major league engineer that many of you here are, but my clients are always exceedingly satisfied.

[aracu]

I honestly think that the equivalent of compression by machine can be achieved better manually by creating volume envelopes in a sequencer. The results are completely based on one's detailed perception of the music rather than an automatic process. I think it is worth the extra effort.

[desotoslo]

Originally Posted by aracu

I honestly think that the equivalent of compression by machine can be achieved better manually by creating volume envelopes in a sequencer. The results are completely based on one's detailed perception of the music rather than an automatic process. I think it is worth the extra effort.

That's what I do. I comb through each track slowly, 90% of the time. Rarely do I used computerized compression. I'm the computer

[MBishopSFX]

I've NEVER used limiting or compression on at least a couple hundred major classical releases. I would never think of it. All involved, i.e.: conductor, orchestra members, producer, etc. have never asked for it. Our consumer response indicates they WANT the dynamic range. Maybe once every two years we might get a letter of complaint that a CD isn't loud enough. That's certainly not enough to change a thing in how I work.

On the other hand, I'll occasionally apply some dynamic range control with judicious and subtle use of the master fader during the sessions. That's far more musical than hitting it with a limiter. Because my sessions are always direct-to-stereo and direct-to-surround mixes, these are the masters (other than the post-production editing). Max instantaneous peaks are ideally kept to +2 dBfs in DSD and no higher than -0.9 dBfs in pcm.

.

[Roland]

I'm afraid I'm firmly in Michael's camp here. I don't view myself as a purist and in an extreme situation I might be persuaded to apply a couple of db or limiting just to knock out the transients, but I just don't find that it is usually necessary. People rarely listed to CD's in cars these days, particularly classical ones. More often than not they listen to classical radio stations that apply the compression as standard, this is fine for in car listening. Few (if any) of the classical releases I have from any of the major labels are listenable to in a car situation, so even if they are compressing the hell out of it, it's not working effectively.

As Michael says above, a little bit of judicious gain riding is effective enough.

Regards


Roland

[5flagsaudio]

Originally Posted by MBishopSFX

I've NEVER used limiting or compression on at least a couple hundred major classical releases. I would never think of it. All involved, i.e.: conductor, orchestra members, producer, etc. have never asked for it. Our consumer response indicates they WANT the dynamic range. Maybe once every two years we might get a letter of complaint that a CD isn't loud enough. That's certainly not enough to change a thing in how I work.

On the other hand, I'll occasionally apply some dynamic range control with judicious and subtle use of the master fader during the sessions. That's far more musical than hitting it with a limiter. Because my sessions are always direct-to-stereo and direct-to-surround mixes, these are the masters (other than the post-production editing). Max instantaneous peaks are ideally kept to +2 dBfs in DSD and no higher than -0.9 dBfs in pcm.

.

I must admit, I've long been a fan of the Telarc recordings I own. Keep up the fantastic work. My situation is somewhat different. My recordings arent commercial releases. They're given to the performers and most of them are listening to them on their iPods, in their cars or loading them in myspace. The bulk of my clients are college groups, local public school groups or community groups. I'll certainly look at the judicious use of the master fader as an option in the future.

[andy_simpson]

Originally Posted by d_fu

Sure... Show me one CD release or recording which comes anywhere close to actually utilizing the theoretical 96 dB dynamic range of 16 bit audio.... Show me a recording with even 80 dB of actual musical dynamic range...

It really depends on how you define 'musical dynamic range'.

By musical dynamic range I would assume you mean from not playing ('silence') to the loudest peak.

For example with a drum kit, you might consider the threshold of hearing (0dB SPL) or the noise-floor of the room (might be anywhere from 30dB to 60dB SPL) and the peak might be around 130dB or even 140dB SPL.

If by musical dynamic range you mean crest-factor, 96dB of crest factor is unlikely unless you call explosives music!

Andy

[jamwerks]

A lot of the sonic information about how load or soft a given instrument sounds is in the tonal quality. For example, a violin playing FFF has a much different tonal quality than the same instrument playing PP. Even if you gain match the two (FFF & PP), we still "hear" dynamic range via the quality. IMO, bringing up the bottom levels adds to the listening experience.

[joelpatterson]

I'm one of these people who would say the "way it sounded" is a different thing from "a recording that's enticing to listen to." If you were only after the way it sounded, then you could set up a pair of mics from the third row, or hey the back row, set the levels so low that the crescendos with the timpanis blazing away would only graze -2, and be done with it.

And you'd end up with a recording that the mics and preamps would love to listen to.

When you think about your audience being human beings, far removed from the actual concert, you start to think about the medium of recording and its limitations and how it's utilized in the modern world.

If this sounds like a distinction without a difference, then I've lost you, I guess.

Maybe a little metaphor will serve to shed some light: I'm a big believer in listing the times of the CD tracks. I tell people if they have any expectation that the music on their CD will find its way into any brand of media, listing the times is essential. You alert your media person how long each track is, which ones are under three minutes, which group will fill up ten minutes... say the guy at the radio station has ten minutes to kill at the end of his show... in listing the times, you've made it easy from him to play your CD. By not listing the times, you've made it a big chore.

At the actual concert, who cares how long each piece is going to be? This information, in that setting, is entirely superfluous.

To summarize: the actual show is one thing, the recording is a different thing.

[Roland]

Originally Posted by jamwerks

A lot of the sonic information about how load or soft a given instrument sounds is in the tonal quality. For example, a violin playing FFF has a much different tonal quality than the same instrument playing PP. Even if you gain match the two (FFF & PP), we still "hear" dynamic range via the quality. IMO, bringing up the bottom levels adds to the listening experience.

I hear what you are saying and of course you are right that the colour of the sound is significantly (or should be) different between ff and pp and should range like a rainbow in between, however, having genuinely quite passages and genuine dynamics adds yet another colour to the palette. Quiteness or even silences can add poetry and a sense of stillness that much great music (and performances) exhibit. I never met a great performer that didn't only vary their colour, but also their genuine dynamic range to achieve great effect. I truly believe the more you can play with peoples senses the more effectively you can communicate.

Regards


Roland

[d_fu]

Originally Posted by Roland

Quiteness or even silences can add poetry and a sense of stillness that much great music (and performances) exhibit. I never met a great performer that didn't only vary their colour, but also their genuine dynamic range to achieve great effect.

It appears some seem to equate compression with the complete removal of dynamic range, which is of course not the idea.

A crescendo may not lose all of its musical effect by being reduced from a 30 dB rise to 25 dB or whatever, IMHO, esp. if mild compression makes the pianissimo passages better audible. But I just realize I've said that before here... It's an old thread.

But I would certainly refuse to master e.g. a harpsichord solo CD anywhere even near 0 dBFS, somewhat similar to Ben's piano trio experience..

And I really wonder whether anyone's ever measured and compared the dynamic range that comes out of a multi-mic setup with a lot of spot mics to the actual dynamic range in terms of SPL that reaches an average listener's position? I have a suspicion that such mic setups may actually create more dynamic range than said listener might experience. If this is so, then purism of the "no compression on classical music" variety (based only on the recorded material) would lose some of its justification, methinks.

Daniel

[desotoslo]

Originally Posted by Roland

I hear what you are saying and of course you are right that the colour of the sound is significantly (or should be) different between ff and pp and should range like a rainbow in between, however, having genuinely quite passages and genuine dynamics adds yet another colour to the palette. Quiteness or even silences can add poetry and a sense of stillness that much great music (and performances) exhibit. I never met a great performer that didn't only vary their colour, but also their genuine dynamic range to achieve great effect. I truly believe the more you can play with peoples senses the more effectively you can communicate.

Regards


Roland

I agree with you, Roland. It's the space in music that gives its identity. Not necessarily only silence (the most striking), but in harmonic/melodic progression, timbre, etc.


Listening to classical/concert music requires greater sensitivity, in general. It only follows to have more sensitive recording practices...

and by more sensitive, perhaps I mean, more hands off!

[David Spearritt]

Originally Posted by d_fu

And I really wonder whether anyone's ever measured and compared the dynamic range that comes out of a multi-mic setup with a lot of spot mics to the actual dynamic range in terms of SPL that reaches an average listener's position? I have a suspicion that such mic setups may actually create more dynamic range than said listener might experience. If this is so, then purism of the "no compression on classical music" (based only on the recorded material) would lose some of its justification, methinks.Daniel

Bingo, hit the nail on the head.

Purist axioms like "never use compression", "only record with two mics" are pretty silly. Recorded classical music can be hugely improved for the listener with a little high quality compression, and it is all about bringing the low level room reverb and pianissimi up rather than controlling the top.

[Chaellus]

PlushQuote:
Originally Posted by Chaellus
classical is the last type of music that needs crazy limiting or compresion...now te want to ruin that for us too?

Hello, Chaellus

You didn't read the thread, did you?

We are not advocating "crazy limiting or compresion." (sic)

We are advocating controlling the dynamic range of the program material.

Many programs don't need any limiting or compression. Some program material does indeed lend itself to using these tools.

Why you wanna do me this way?



yea i just realized that right now....thanks for the correction and it was also 2 am when i wrote that...so my mind was on mixing a song at the time and being tired as well.

[Roland]

Originally Posted by David Spearritt

Bingo, hit the nail on the head.

Purist axioms like "never use compression", "only record with two mics" are pretty silly. Recorded classical music can be hugely improved for the listener with a little high quality compression, and it is all about bringing the low level room reverb and pianissimi up rather than controlling the top.

The only problem with compression (or for that fact Limiting) is that most of these devices have a sonic signature, often less that subtle. A couple of db "knocked off" transients on a particularly dynamic orchestral CD, I can understand, wholesale compression of classical recordings becomes a nonsense, too high a price.

Of course "gain riding" produces a manual form of compression, but this doesn't (or at least shouldn't) change the sound, compressors do. It was this same argument used on rock music that led to a whole raft of CD's coming out with completely "smashed" levels. I've had people bring projects in for mastering with 2-3db dynamics (meters not moving) aimed at producing "louder" cd's. In classical music, historically small amounts of compression were needed to prevent distortion/control levels to tape and subsequently vinyl. With the introduction of digital and the compact disc this became unnecessary, now there are those interested in going that route again. I haven't seen any compelling post here giving any real value as to why this is a benefit. However much compression is used, enough to make a classical title "car friendly" would be way too much for domestic listening. Michaels comments are real world and come from a well respected (with a reputation for a high level of sound quality) moderately sizable, classical label and as he say's it is unnecessary. Tony Faulkner posted a couple of times recently in here, I'm pretty sure he also doesn't use compression. In fact I'd go as far to say that any of the respected engineers I know at most, stick to using the Junger or alike for a couple of db limiting. This has nothing to do about being purist, but reducing dynamic so it can be comfortably, listened at low volume as one can on presume "background" music isn't improving sound quality. If you want more reverb or detail, mic and record for it, compression is a poor way to achieve this.

Regards


Roland

[Plush]

Handwringers sometimes have a good point when they appear and denounce ANY processing of the classical program material. What they guard against is heavy handedness I suppose.

However, there is a new dimension of depth and ear tickling detail to be had when the right program is processed to bring up low level detail.

I suggest experimentation instead of an outright ban on complimiters.

Ear candy may be the result.

Of course for broadcast, some dynamic range treatment is de rigeur. (Your program is being heard by over 400,000 people in one week.)

[3rd&4thT]

In the analog period, major classical labels kept their dynamic range to about 45 db, and classical radio stations further compressed to a range of 18 db to cut over road noise.

When digital came in, a few labels experimented with wider dynamic range, but listener complaints in most cases caused them to return to a range of 45 db, even though tape hiss was no longer a problem. Some boutique labels, Telarc prominently among them, continued with a wider dynamic range, but they always sold more to the "tweak" audience than the major labels did.

It was fascinating in the 60's to contrast classical recordings from Columbia/CBS/Sony and Decca/London. Both had to work with the same tape hiss, but CBS tried to push everything as loud as possible, while Decca tried to keep general volume down close to the floor in order to make room for climaxes. CBS believed that a louder sound was perceived as a better sound, and Decca believed in the illusion of dynamic range even if the chain had trouble delivering it.

BTW, both spotlit shamelessly, with very different effect. Decca had sliders with finger cups, and the engineer did gain riding by rippling his fingers like an octopus, rebalancing every two bars or so. It made editing takes together really rough, but the sound was usually stunning.

3rd&4thT

[Roland]

I think that Decca is a good case in point, as pointed out they for the most part tried to preserve the dynamic, even as far back as the 60's. Sure there was obviously some gain riding and spotlighting, but there are a few releases particularly in the 80's where the general level is pretty low, quite obviously to encompass large dynamic ranges, the Planets (MSO Dutoit) is a good example of this. It is also fair to say that Decca were far from a "niche" label and probably, more than any other label, consistently produced high caliber recordings. Much of this was due to both their high quality artist roster and the huge efforts made to record in handful of great acoustic venue's. Contrast this with other labels such as CBS who were more happy to manipulate recordings for what they probably viewed as their consumer's demand. Being that I don't hear people looking to create or reproduce the "CBS sound" I think that it is a reasonable conclusion that these recordings are not looked upon with the same regard as those of Decca. Ultimately you make your choice and decide which side of the fence you reside on. Having worked with a number of "rock/pop" acts over the years I have used compressors extensively as I have in mastering work. I personally choose not to use them for classical work, however I have used a limiter on maybe 2 or 3 occasions ever on a classical project and of course I have done the odd fader ride/spot highlight when it helped for balance or musically. I have to say I am a little surprised at how many here do, I expected it to be in the minority.

Regards to all (irrespective of your stand on the matter)



Roland

[aracu]

Another advantage, or disadvantage, depending on how you want to look at it, to creating the equivalent of compression by manipulating the volume envelope manually in a sequencer, is that it does not percievably colour the sound.

[mattssons]

For a classical orchestra to record Original material, (Admit, most are just large cover bands.) Then have it recorded in a more contemporary pop (Read close miced + sections)
Then have the dynamic range down to level that could be broadcast. I try to listen to classical stations in my car but usually give up, it´s all noise & blurry Ambience.
Classical music sound fantastic i think but i understand that a project like this is nearly hopeless. An Orchestra is not a cheap wagon to carry and most rely on sponsorship and government help i suppose, so it´s easier to fill those seats replaying the classics. /Toby

[just.sounds]

I'm not trying to start a war here but if you listen to spot microphones the dynamics will sound harsh and not how one would like to perceive it. So the main pair is closer to the orchestra than the avarage listener where it does not get a lot of support from early reflections and reverb. And the spot microphones are even closer to the instruments.

How do we manage to mix this without compression?

[Roland]

Originally Posted by just.sounds

I'm not trying to start a war here but if you listen to spot microphones the dynamics will sound harsh and not how one would like to perceive it. So the main pair is closer to the orchestra than the avarage listener where it does not get a lot of support from early reflections and reverb. And the spot microphones are even closer to the instruments.

How do we manage to mix this without compression?

That's why we have conductors & producers! Even with rock bands, if you get real pro's they balance themselves on stage. I used to work with Kenny Craddock the keyboard player. He was Van Morrisons MD for 2 years and originally played in Ginger Bakers Airforce, once you had set a decent onstage monitor mix you barely had to touch the faders as he and the players he worked with were more than capable of controlling their dynamics musically. With orchestras it is often the same, the last thing you want to do is spend all your time overriding the balance. If you get it right little extra input is needed.

For a classical orchestra to record Original material, (Admit, most are just large cover bands.) Then have it recorded in a more contemporary pop (Read close miced + sections)
Then have the dynamic range down to level that could be broadcast. I try to listen to classical stations in my car but usually give up, it´s all noise & blurry Ambience.
Classical music sound fantastic i think but i understand that a project like this is nearly hopeless. An Orchestra is not a cheap wagon to carry and most rely on sponsorship and government help i suppose, so it´s easier to fill those seats replaying the classics. /Toby

As for multimiking orchestra's this has been done ad-nausium, in the uk you can hear examples of this on Radio 2 with the BBC concert orchestra, 1 mic per pair of players. It works ok for listening, however it doesn't sound like a classical orchestra does live, not even close, copious amounts of digital reverb are used to give it a bloom, ok if that's what you really like. Often multi-miking is used for TV, as directors like to highlight the sound of players being shown in the picture, sounds ok on a TV set with it's mediocre speakers, less convincing on Hi-Fi.

Regards to all


Roland