Mixing and Tracking for LOUDness

[faeflora]

How do you mix and track for maximum loudness?

Aside from tracking for clarity, what special tricks do you use when tracking or during a mix to make sure that each track is very compressed, free of large transients, and still clear?

If you are looking for a specific type of track, mostly kicks seem to eat up most of the bandwidth of finished mixes I work on. If I smash the attack, the kick loses the impact. How do you work around that? How about on snare? I don't multiband compress. Do you? Is that a good way to bring up the mids and highs to a level of the peak waveforms?

It definetly seems like a special skill to get super loud tracks that are still punchy.

Examples could be any modern pop, r&b, new metal i.e. slipknot/deftones, modern rock etc etc etc.

[RainbowStorm]

Originally Posted by faeflora

How do you mix and track for maximum loudness?

Aside from tracking for clarity, what special tricks do you use when tracking or during a mix to make sure that each track is very compressed, free of large transients, and still clear?

If you are looking for a specific type of track, mostly kicks seem to eat up most of the bandwidth of finished mixes I work on. If I smash the attack, the kick loses the impact. How do you work around that? How about on snare? I don't multiband compress. Do you? Is that a good way to bring up the mids and highs to a level of the peak waveforms?

It definetly seems like a special skill to get super loud tracks that are still punchy.

Examples could be any modern pop, r&b, new metal i.e. slipknot/deftones, modern rock etc etc etc.

I think the trick is buying the right compressors. No but seriously, you need great gear for that in combination with knowledge about setting up the gear right. Start from the beginning: maximize the signal-noise ratio and lower the noise floor on each instrument. Target the reference input level of your converter, for instance -9dB on 24-bit recording. Cutoff as much "dead" frequencies that you can, both in the low end and in the high end, on different tracks depending on their frequency range. Here is a quick reference on where the initial cuts should be:

Bass cut

80Hz Cut

Electric guitar
Accoustic guitar
Snare drum
Trombone

200Hz cut

Cymbals
Triangel
Harmonica
Violin
Vocals (female)


Treble cut

8KHz cut

Electric guitar
Kick drum

10KHz cut

Vocals (male)
Trumpet
Trombone

15KHz

Vocals (female)


From here, adjust according to taste. Sometimes piano/synth/organ can be cut at 80Hz or higher as well to leave space for the kick drum and the bass guitar and high cut at 15KHz or lower too, for making room for the vocals, snare and cymbals. When these cuts are done you have setup the basic environment for creating a loud mix. The drumkit should be left panned center or close to that area. The same with bass guitar and vocals.

Now to the effects.

Think stereo: Stereo reverb, Stereo compressor, Stereo delay.

Keep reverbs set for low reverb time but rather wet or you can add a lot of different reverbs set for low reverb time and rather dry. Use the "New York Compression Trick" to make everything louder and the rhythm section to really rock:

1) Buss the drums, and maybe even the bass to a stereo compressor.

2) Hit the compressor fairly hard, at least 10dB or more if it sounds good.

3) Return the output of the compressor to a pair(because of the stereo compression) of fader inputs on the mixer (digitally to a stereo track with the input set to the compressor bus).

4) Add a pretty good amount of high end (6-10dB at 10kHz or so) and low end (6-10dB at 100Hz or so) to the compressed signal.

5) Bring the fader levels of the compressor up until it's tucked just under the present rhythm section mix to where you just can hear it.

The drums will now sound bigger and more controlled without sounding overly compressed.

You can combine this with a multiband compressor on the drums element if you want it even louder from here. Compress each track a little to start with, adjust according to taste. Will you skip the compression on the stereo output and leave it to the mastering engineer? Add a stereo compressor on the mix! It will be recompressed during the mastering phase as well and yet again when played on a radio or television. But at some point you will realise the mix is loud enough but too compressed, then you lower the stereo compression on the mix bus.

When used for loudness add compression in the range 15 to 20 dB for electric guitars, room mics, drums, vocals and adjust down from there depending on the song. Gate the snare and maybe also the kick drum, depending on the kind of sound you want.

Unfortunately I don't have time to go through the settings on the different compressors. But maybe some other time...

You get the idea... Also, use a good dithering algorithm like for instance the uv22hr...

[Drumsound]

Try using parallel compression in groups and on individual tracks.

[cdog]

The kick and the snare SHOULD BE the loudest things on the track IMO.

The lead vocal should also be loud but it won't eat your headroom like a kick.

Try shelving off the kick below 40-50 hz, it can give your mix some more headroom.

[Jay Kahrs]

Maybe I keep missing the point of this...

Why is loudness important at all? A record shouldn’t be smashed flatter then Kansas.

By limiting the dynamic range you're reducing the soft parts of the music and by comparison the loud parts won't be AS loud.

Granted, the volume wars are out of control but why should we encourage artists to put out CD's with a dynamic range of 4dB? You can still compress and mix and have the mastering engineer do his bit of limiting and whatever, and yeah...the mix has to be built to survive the brutal processing but still...

If I want the mix to be louder I'll turn up the volume pot. At least the loud parts will have more impact and the recording will still have some tone and you won't hate the way the record sounds in five years.

[faeflora]

Originally Posted by Jay Kahrs

Maybe I keep missing the point of this...

Why is loudness important at all? A record shouldn’t be smashed flatter then Kansas.

By limiting the dynamic range you're reducing the soft parts of the music and by comparison the loud parts won't be AS loud.

Granted, the volume wars are out of control but why should we encourage artists to put out CD's with a dynamic range of 4dB? You can still compress and mix and have the mastering engineer do his bit of limiting and whatever, and yeah...the mix has to be built to survive the brutal processing but still...

If I want the mix to be louder I'll turn up the volume pot. At least the loud parts will have more impact and the recording will still have some tone and you won't hate the way the record sounds in five years.

I'm not so much asking about limited dynamic range where everything is horribly in your face. I'm trying to ask how you go about limiting and balancing the waveform level dynamics for each track but still make things sound good.

For instance if when mastering, someone hand edits the transient of a kick (which are far faster than a speaker can respond) to gain 4-5 db of headroom. Or for a another example compressing a loud vocal passage and then automating it to still be loud but less peaky.

BTW regarding loudness wars, I've had songs I've mixed called "weak" and "lacking balls" because they were quieter than smashed mixes. I told the listener "TURN UP THE VOLUME KNOB" but they shrugged and didn't understand.

[Lagerfeldt]

I appreciate that remark about loudness wars.

However, here goes..

Apart from the things mentioned above (not all of which I agree with), it's important that the original source sound has potential to be loud. All sounds are not born equal.

Multiband compression sometimes has a place, but mostly not. I mainly use multiband compression when doing vocals for broadcasting and fixing problems in mastering (usually just 1 or 2 bands max.).

Sum compression can help a bit but done wrongly everything will be wimpy loud, usually because of too fast attack times.

Limiting to a certain point can help on loudness but actually clipping (a bit controversial, I know) will sometimes be much better IMO. Clipping can introduce some artifacts (as can limiting) but clipping seemingly preserves hi freqs better than limiting (e.g. the L1 limiter by Waves) in some cases. Limiting or clipping should almost always be the last process.

Of course using high quality hardware compressors will help a whole lot. I use the Gyraf Gyratec X tube compressor and an SSL Type 4000 clone compressor. I'd say they can get stuff at least 1/3 louder than any software compressor like Waves Renaissance, etc. And they sound better too.

One thing not mentioned here is volume automation. You should turn down stuff not needed or not important (or remove it altogether), and turn up stuff that's important. Depending on what kind of music you're doing, I'd say loud snare and medium kick volume is the trick for rock/pop, while you definitely need a loud kick for hip hop and dance music.

Since the low end takes up a disproportionate amount of headroom a few dBs overall can sometimes be saved by sidechain compression on the bass, using the kick as trigger (this can save 2-3 dB if done right). Also the reverb predelay can help save a few dBs. Set the predelay so the reverb begins just as the original sound fades out. You can actually measure it on the master outpu (this can save around 1 dB).

You can also you sidechain compression on long delays, etc. to save headroom and clear up some of the mess :-)

[faeflora]

Originally Posted by cdog

The kick and the snare SHOULD BE the loudest things on the track IMO.

The lead vocal should also be loud but it won't eat your headroom like a kick.

Try shelving off the kick below 40-50 hz, it can give your mix some more headroom.

RE kicks usualy I do a lot of work on kicks from sample replacement/triggering and usually rely on subsynth or triggered sines heavily to get a clean hard hitting subkick.

[RainbowStorm]

Originally Posted by faeflora

RE kicks usualy I do a lot of work on kicks from sample replacement/triggering and usually rely on subsynth or triggered sines heavily to get a clean hard hitting subkick.

In this context I have to mention a trick that Guy Snider (mixing engineer for Tupac, Snoop) does a lot. To make the low end defined he takes the kick and the snare and buss them to either a stereo compressor or two mono compressors linked together and bring that back into two channels of the mixer. On these two channels he cuts all mids and adds tons of low end with lots of compression. Sometimes he mutes them, just to have them available if needed. You can do this digitally too by copying the original track, send the signal to a stereo compressor and return it to the copied stereo track. Actually this is a custom implementation of the "New York Compression Trick"...

The compressor effect is really one of the secrets behind successful mixes. Mixing engineers actually use a lot of compressors to create different colors and effects, much because they want to crank up the loudness of different instruments and color the sound of different instruments to improve the sound. Many describe the effect like it has the potential of making instruments come more alive.

Professional engineers know when to use what kind of compressor on what instrument. I would say focus more on that and less on "loudness", because loudness will actually automatically be there when you get it right...! But if you struggle with not getting the mixes loud enough or not the way you want it to sound you are probably using wrong compressors or wrong compressor settings. The key here is to mess with a lot of different compressors and compressor settings on different instruments, to learn when to use what kind of compressor how... Remember that you want compression but enough dynamics too.

You will laugh now, but you might soon agree... Have you ever admired the amount of dynamics found in a mix? Have you gone: "This sounds awesome, the mix is deep, yet very airy and lush with great dense, expression and touch!" This feeling of great dynamics is due to compressors!

[defjamm]

Originally Posted by Lagerfeldt

Limiting to a certain point can help on loudness but actually clipping (a bit controversial, I know) will sometimes be much better IMO. Clipping can introduce some artifacts (as can limiting) but clipping seemingly preserves hi freqs better than limiting (e.g. the L1 limiter by Waves) in some cases. Limiting or clipping should almost always be the last process.

or you combine both(limiting and clipping)

[Micgiver]

Originally Posted by TonyCrazyMan

In this context I have to mention a trick that Guy Snider (mixing engineer for Tupac, Snoop) does a lot. To make the low end defined he takes the kick and the snare and b..

Listen to real Jazz. there should be no compression there ever!!! thats deep music

[jdunn]

If you mix with your highest peaks -3 dB down, that's 3 dB the program level can be raised during mastering before compression/limiting is necessary. If you leave 6 dB, then the mastering engineer can increase the level by 6 dB before limiting.

I wish everyone would do this. If you slam the levels on your master, it's got nowhere to go but squashed!

Assuming you're using a DAW to mix, get an external summing box and some multichannel D/A's. It makes things sound a lot bigger to me.

[Jay Kahrs]

Originally Posted by faeflora

I'm not so much asking about limited dynamic range where everything is horribly in your face. I'm trying to ask how you go about limiting and balancing the waveform level dynamics for each track but still make things sound good.

I don't have a stock answer. Every mix is different because every song is different. Sure, I have tricks that I've gathered with experience but there's nothing that works all the time. Well...I never use brickwall limiting on anything and that includes the 2-buss. Actually I'm printing mixes that peak at -3 or -4dBfs and leave all the loudness stuff for the ME.

For instance if when mastering, someone hand edits the transient of a kick (which are far faster than a speaker can respond) to gain 4-5 db of headroom. Or for a another example compressing a loud vocal passage and then automating it to still be loud but less peaky.

Huh. I never hand edit anything at that level...transients of bass drums or whatever. I'd rather focus and spend time on the important things like making building the mix so it kicks ass rather then kickin' shins. Volume automation plays a part in that but not to the point of being obsessive over .5dB rides.

[joeq]

to me the stuff that sounds the loudest is the stuff that was arranged and performed that way. When each instrument is either "on" or "off" and not mushing around in between. When the spaces or holes in the arrangement go all the way down digital black, and when the ensemble hits together and its _really together- THAT sounds loud.

I am not a big fan of synth based music, but it often sounds louder to me. I love a bass player or drummer who can deliver that tight consistent envelope like a synth or drum machine.

[jdunn]

Originally Posted by Jay Kahrs

Why is loudness important at all? A record shouldn’t be smashed flatter then Kansas.

It's crazy, really. I guess people don't realize that their volume control is an attenuator, and it sounds considerably better when not at the extreme lowest possible setting. You're barely getting a trickle of the power your stereo can deliver when the attenuator is almost all the way off.

The thing is, any major label project will have the piss squeezed out of it. And most (all?) smaller labels don't want to take the chance of turning off listeners with CD's that have far lower level than today's standard.

I plan to sell music online, and educate people who visit the site. Explain to them that music played on their stereo system will sound much better if you can raise the volume knob above 1 without it ripping your ears off.

Most music today is just loud. I've said it before, I'll say it again. There's a whole generation of people out there who have never heard dynamics.

[Bob Olhsson]

Originally Posted by Lagerfeldt

clipping (a bit controversial, I know) will sometimes be much better IMO. Clipping can introduce some artifacts (as can limiting) but clipping seemingly preserves hi freqs better than limiting

Digital clipping only sounds better if NO further processing is ever done to the audio including things like a broadcast limiter or an MP-3. When digitally clipped audio gets reprocessed, it can shatter sonically and can even reduce your volume significantly on the air.

[jjblair]

This thread has me speechless. I hate when people tell you compressor and EQ settings without having heard a single note of the material. I mean, it's really a disservice to the original poster.

How about this: Listen to records that have what you consider to be "maximum loudness". Then examine the sounds both individually and how they work together. Then try to figure out how to get those sounds and do a lot of experimenting, recording and mixing to get your recordings to sound like theirs. Use different instruments, different mics, different pres, EQ and compressors so that you learn to recognize those sounds when you hear them, etc. Don't expect somebody to give you a recipe. For all you know, half the posters don't even know what they are talking about, myself included.

There is no simple solution. What Guy Snider does might work for Snoop, but it my not work for a different type of music. You have to learn the way that all good engineers have: through trial and terror. Good luck.

[MASSIVE Master]

Agreed - I had a client in today with a "halluca-metal" project. While I tried to keep the last line of dynamics & limiting similar for the whole project, every tune, while very similar sounding generally, required completely different and independent EQ and dynamics settings. Some of them required low cuts (all at different frequencies). Most were a bit muddy in the low mids and required adjustments (all different again). The guitars had a bit of a "grain" to them in the 1.5-3.5kHz range and needed a little notch here and there - Again, different for every tune.

Granted - A lot of this could have been worked on at the mixing stage. Still, all of these adjustments would have been different for every mix even if it were handled during mixing.

Presets anyone?

And as mentioned, careful listening and to a point, "trial and error" are key... With enough of that, the careful listening will simply reveal what the sound is asking for.

[RainbowStorm]

This thread has me speechless. I hate when people tell you compressor and EQ settings without having heard a single note of the material. I mean, it's really a disservice to the original poster.

How about this: Listen to records that have what you consider to be "maximum loudness". Then examine the sounds both individually and how they work together. Then try to figure out how to get those sounds and do a lot of experimenting, recording and mixing to get your recordings to sound like theirs. Use different instruments, different mics, different pres, EQ and compressors so that you learn to recognize those sounds when you hear them, etc. Don't expect somebody to give you a recipe. For all you know, half the posters don't even know what they are talking about, myself included.

There is no simple solution. What Guy Snider does might work for Snoop, but it my not work for a different type of music. You have to learn the way that all good engineers have: through trial and terror. Good luck.
__________________
"Eventually your experience catches up with your opinion." - David Palmer

"normally i brush off compliments as passing thoughts from people, but it is fuel for the fire of people like me everywhere and they are taking over whether you like it or not... and whipping your ass in the process.

word to the not so wise, dont growl at me because i bite back... and bite like a pitbull and do not let go until my prey is dead." - OmegaJerk (in PM)

Your ears will always be the best tool for everything, but as soon as your ears don't tell you what you want to know(which is, is this the best possible sound of this instrument in this mix?) it's time to start doing something OR leave it until you get the information from the rest of the mix. Trying out different engineers tricks can be very ear opening sometimes, because you get new perspectives and from that also new ideas to play with. Soon the ears tell you what you want to know and you know where you are heading. These tricks can also make it easier to confirm that you are on the right path by using only your ears.

I hate when people can't see useful things in options. So really, I don't agree and I don't think my post were disservice for the original post.

BTW, I personally think the New York compression trick is great! It makes perfect sense and I find that and similar tricks very useful!

[RainbowStorm]

Originally Posted by Jay Kahrs

Maybe I keep missing the point of this...

Why is loudness important at all? A record shouldn’t be smashed flatter then Kansas.

It's important because the combination of loudness and dynamic balance creates the perception of great dynamic range. This creates a mix that is very dense with lots of expressions and it's really just that you like in a song. So compression is a commersial trick for making the listener stuck with the song through its touch. I think it works, it works great! You really want that loudness, but without dynamic balance it's all dead flat! You don't want that! That's not how you make the listener stuck with the song!

There seems to be scary feelings connected to compression in the mixing world, yet everybody mess with a lot of different compressors on a lot of tracks and beautiful sounding mixes are born!

Something I don't like is when you ask compressor questions on different message boards and you always end up with the same "compression is bad" discussion. Open any mixing book and you will notice they go thorugh the compression area thoroughly (they only warn you once that using compression in the wrong way can make the mix dead flat). There's a reason for all this. Actually I think this is really a good thread, the person that posted it really touched something very important! Actually mixers have a lot to learn when it comes to compression. There are very important questions linked to this issue like what type of compressor can be used on what instrument to create what color, how to setup the compressors for creating the best result, how to automate compression successfully, how to EQ compressed layers, how to preserve tone in velocity, what kind of stereo compressors should be used to create a certain touch on the mix and so on, what limiter should be used in the mastering process to increase the loudness of the mix without ruining the dynamic balance. There are all kinds of important questions related to this effect.

Also remember that compression can actually make the sound in itself better sounding. Much of the beautiful color found in the velocity curve of an instrument can come alive by using compression. A good example is compression on lead guitars. Some weak tones on some parts of the guitar can be extremely beautiful when it is compressed! (the picking tone is not there which adds warmth to the undertone that shines beautifully through in a powerful way) It's the same with many instruments that hold much tone color in different areas of the velocity but are hard to generate by playing the instrument!

[RainbowStorm]

Originally Posted by MASSIVE Master

Agreed - I had a client in today with a "halluca-metal" project. While I tried to keep the last line of dynamics & limiting similar for the whole project, every tune, while very similar sounding generally, required completely different and independent EQ and dynamics settings. Some of them required low cuts (all at different frequencies). Most were a bit muddy in the low mids and required adjustments (all different again). The guitars had a bit of a "grain" to them in the 1.5-3.5kHz range and needed a little notch here and there - Again, different for every tune.

Granted - A lot of this could have been worked on at the mixing stage. Still, all of these adjustments would have been different for every mix even if it were handled during mixing.

Presets anyone?

And as mentioned, careful listening and to a point, "trial and error" are key... With enough of that, the careful listening will simply reveal what the sound is asking for.

BTW, what do you think about the Lavry converters compared to the Apogee converters? Do you think that either one is generally "better sounding" to you?

[RainbowStorm]

Hi!

I am sitting here with a cup of coffee thinking about an interesting trick that mixing engineer David Pensado (Christina Aguilera, Brandy, Jessica Simpson) uses a lot with effects:

"A lot of times what I'll do is put the effects only on the compressed sound. In other words, an effect I use a lot would be "Locker Room" or "Tile Room" on a PCM70 and I'll add that effect only to the compressed sound. As a result, the reverb actually has a snap and aggressiveness to it. Every once in a while I'll make it stereo where I'll take two 160s and I'll set them up identically, but on the insert of one I'll put like anywhere from a 9 to 15 ms delay so the tight compressed sound is out on the edges of my stereo spectrum, but the original sound is in the center. That creates an incredibly nice image. That setup works great for snares, kicks and hi-hat. Every once in awhile it'll make a guitar come alive too. So what you are doing is you're controlling the dynamics but you're actually increasing the dynamics. It's the strangest thing because psychoacoustically, it's not getting louder but your mind is thinking it is. On the radio it just jumps out of the speakers."

What do you think about this technique? Have any of you guys tried anything similar? Do you think it works?

[GRiFF]

I browsed this useful and interesting thread, but I dont think anyone really mentioned the actual tracking portion of the question. I've been frustrated by my AMEK CIBs when trying to capture vocals particularly but also on any musical parts - kicks, snares etc with large changes in dynamics.

What I want to do is get the hottest signal I can, benifitting from the equipment running at optimum. I just cant seem to crank up the input gain to the level I want without introducing unwanted clips.

Like the original poster, I dont want to kill the peformances, but I do want to find the secret to keeping the vocal or whatever tamed if runs into the red.
Compressing is a funny one, as soon as I think I understand it, I learn something new that makes me think I didnt know a thing.

An engineer friend always limits vocal takes gently and describes just what I'm after,- the compressor kicks in and gently pulls back the signal level if its over, but when I apply his suggestions which are a 2:1 ratio, with the gain reduction set to kick in at just below zero.

In theory I understand exactly why this should work. But it doesnt.
Please can someone explain how I'm getting this wrong?

I would have thought that setting the gain adjustment after compressing to zero should do the trick.

Hmmmm...am I the only one with compression dyslexia?

[thethrillfactor]

Originally Posted by GRiFF

I browsed this useful and interesting thread, but I dont think anyone really mentioned the actual tracking portion of the question. I've been frustrated by my AMEK CIBs when trying to capture vocals particularly but also on any musical parts - kicks, snares etc with large changes in dynamics.

You lost me here?

Originally Posted by GRiFF

What I want to do is get the hottest signal I can, benifitting from the equipment running at optimum. I just cant seem to crank up the input gain to the level I want without introducing unwanted clips.

Its possible your equipment is working at optimum level.

Sounds like the problem is you have everything working at optimum(in terms of loudness) and you have no where to go.

Originally Posted by GRiFF

Like the original poster, I dont want to kill the peformances, but I do want to find the secret to keeping the vocal or whatever tamed if runs into the red.

Is this during tracking or mixing?

When tracking you need to leave some headroom space in case some notes jump out.

When mixing with some creative tricks you can keep the vocal pretty steady and still be present.

Originally Posted by GRiFF

Compressing is a funny one, as soon as I think I understand it, I learn something new that makes me think I didnt know a thing.

Totally understand.

Originally Posted by GRiFF

An engineer friend always limits vocal takes gently and describes just what I'm after,- the compressor kicks in and gently pulls back the signal level if its over, but when I apply his suggestions which are a 2:1 ratio, with the gain reduction set to kick in at just below zero.

In theory I understand exactly why this should work. But it doesnt.
Please can someone explain how I'm getting this wrong?

I would have thought that setting the gain adjustment after compressing to zero should do the trick.

Hmmmm...am I the only one with compression dyslexia?

No sounds like your friend is working with a singer who knows how to control their dynamics.

I would suggest if you want a limiter try a 10:1 ratio with a very slow attack and med to fast release.

If it doesn't clamp down fast enough for you or its too fast, try putting the output on a fader an ride the vocal manually to tape.

Believe it or not some of us still work this way.

[jbuntz]

Keeping kicks and snares very tight (lacking length) can help. It makes me wonder if the whole dry drum sound trend has something to do with the loudness race. Clipping the mix does work well before L2.

[Jay Kahrs]

Originally Posted by TonyCrazyMan

It's important because the combination of loudness and dynamic balance creates the perception of great dynamic range. This creates a mix that is very dense with lots of expressions and it's really just that you like in a song. So compression is a commersial trick for making the listener stuck with the song through its touch. I think it works, it works great! You really want that loudness, but without dynamic balance it's all dead flat! You don't want that! That's not how you make the listener stuck with the song!

There's some truth in there. Adding some degree of limiting to the final mix can actually stretch the dynamics a bit, but after 3-4dB the mixes seem to fall apart and actually sound worse. The top gets grainy and harsh, the high mids start to leap out like nails on a chalkboard, the low end gets smaller...and with too much limiting you'll get that crackly distortion which is WAY worse then anything vinyl ever did.

How is any of that a good thing?

Besides, extreme limiting doesn't make the song sound better on the radio. Add too much and you'll actually invert the dynamics. That quiet banjo & vocal breakdown in the middle of your death metal march will be about 6dB louder then the rest of the song.

[Curve Dominant]

I'm always glad to see Dave Pensado and his methodology mentioned here. One can learn an awful lot from that cat.

I am a BIG fan of compression, so don't get me wrong here, but overall EQ balance in a mix will enable it to be mixed loud first. After that spectral management is achieved, compression can really "do its thing."

Disproportional buildup of low and low-mid frequencies carry a lot of energy, and will prevent a loud mix. Parallel compression is a great tool, but it makes a mess if the EQ balance of the compressed subgroup is not balanced, so I would go back to resolving that issue, then apply the parallel compression.

Unfortunately EQ settings cannot be "phoned-in," you really have to take it song by song, track by track. And making EQ fixes after compression is applied can create some very weird, unnatural sounding artifacts, so it's really important to get EQ balances happening before compression is applied for that reason. That's why HPF's on mic pres are so handy: Who needs -150Hz on a rock/pop vocal track in the first place?...not to mention those frequencies will create some unwarranted reactions from the compressor used in tracking, especially if you're close-micing a vocalist with a LDC where proximity effect comes into play.

Parallel compression of the drum submix adds a very nice punch when it's panned center/mono...a little trick Fletcher posted here once, which I've been using ever since and my clients are VERY pleased with.

This is a very big subject and all together we could easily compile a large book on the subject.

[GRiFF]

Always good to hear from you Thrill, your responses are consistently spot on.
Just to clarify, I was talking purely about tracking and trying to get solid takes to tape or hard disc without then having to normalise or boost the signal so much in the mix.

With better singers this is without doubt less of a problem, but recently one singer who had such a large change in dynamics ment that the only way to avoid clipping was to have such a low input signal that I felt the Amek was running less effectively. I wanted to know if there was something I could do to keep it as natural as possible, but still get a satisfying level.

The singer in question sings in an operatic style and I've heared other recordings she's done that appear far more controlled - the vocal sits better and yet still sounds natural.

Is there anything you can really say to such a singer, asking her to temper her style doesnt seem appropriate unless I just want a tame, un-emotional take.

[Riccardo]

Two words here.

Arrangement (lost art?)

Tracking (definitively lost art)

[BackHand]

Originally Posted by Lagerfeldt

Sum compression can help a bit but done wrongly everything will be wimpy loud, usually because of too fast attack times.


Of course using high quality hardware compressors will help a whole lot. I use the Gyraf Gyratec X tube compressor and an SSL Type 4000 clone compressor. I'd say they can get stuff at least 1/3 louder than any software compressor like Waves Renaissance, etc. And they sound better too.

I've have also the SSL Type 4000 clone compressor, just built it..all best components and everything...I feel that I sometime get that "wimpy" loud smashed feeling...generally I have release on auto, attack on 2 or so and the threshold compressing like 10 dB....doing it wrongly?
What's you're general settings?