parallel compression: when?

[NathanC]

A few months ago I "discovered" parallel compression (yes, I still have a lot to learn). I read about it on some forum (might have been here) and tried it on a drums group immediately. And I loved it! I've tried it on more things since and must say it's very tempting for me to use it on almost everything, but I especially like it on drums groups and acoustic guitars.

My question: when and how do you use parallel compression?

[Vank]

If it sounds good it is good! No rules apply...

[waxx]

i use it mainly on drums to make them more punchy without loosing the dynamics, but sometims it can work on other stuff (like piano/keyboard/organ or guitar or bas) also.

but experiment, and you will hear it if it works.

[morpheuzrecords]

whenever appropiate if it sounds good than use it.

[Mike Caffrey]

When I'm feeling lazy and I don't want to take the time to set my compressors right, I'll compress too much and then add the uncompressed signal to minimize the damage I've done to the dynamics and the transients.

I don't think there's truly no place for parallel compression, but I believe it's very misleading. Kind of like eating sugary food - it's great on the first bite, but after a little bit you realize your first impression was wrong.

It's not that hard to get the same sound without being in parallel.


I could see parallel a a good use for an excessively dynamic signal, like a vocal. Especially if you could sum the compressed and uncompressed and then run then in to a compressor.

Then the parallel signal could be used to boost the quiet parts, but it would limit the loud parts. The uncompressed woud then get compressed by the compressor they were being summed into. It would kind of be a pain in the ass unless you had a way to let some of the uncompressed signal through the first comrpessor because otherwise it would be a routing nightmage with extra parts of the signal path.

That would be a very cool use for parallel.

[Drumsound]

I like mixing with a drum parallel, a guitar parallel and a bass parallel.

[Mike Caffrey]

No vocal paralell?

Why stop there and not do everything parallel?

Don't you have gainstaging issues? You must have to put the faders way down since you're making everything "twice" as loud by doubling the signal. Maybe it's not exactly twice as loud, but when you blend in the parallel, you're going to hit the stereo buss harder.

[mattssons]

The reason i LOVE paralell compression is that you can get the really low level sounds right up there. If you would create that with serial compression you would have to compress really hard and the high level sounds would be severly squashed. I would like more compressors to have a "blend" control so i would´nt have to patch as much! /Toby

[redddog]

Could someone explain what it is as well as how to do it and when?

I've got way more to learn than you!

[BrettB]

Originally Posted by redddog

Could someone explain what it is as well as how to do it and when?

I've got way more to learn than you!

parralell compression is when you double a track and compress one copy very hard, and mix it under the original. It preserves the dynamics of the instrument but makes it sound more solid.

Somtimes both tracks are sent to an aux to be grouped and recompressed or limited. Some mastering engineers do it when the low end is missing or unpredictable. Rather than draggin the audio over to a second track, It can also be accomplished on a mixer, when you "mult" or use a send to a bus that has the compressor. In mixing, it is often used on bass tracks as well as kick and snare drums to add pressence without very much gain.

To do this you use a "send" routed to a bus that is set to input of an aux channel that has a compressor inserted. Hit the signal hard and blend that fx channel sound back under the original until it sounds fatter but does not get louder. When you mute the signal on the compressed track it should sound thinner but not quieter. The goal is to have the isnturment cut through the mix but still not be squashed. So you don't want to hear more of the squashed track than the unprocessed. Of course, you can do what ever you want to do with it until it fits the song you're working on. But, this is how I use it.

Also, what I do on Vox is similar but not the same. Record a first take, and a double take. The double track gets hit with a compressor very hard and limited, and mixed under the first vox take. They are singing the same things but two diferent takes. They get mixed to a group bus and compressed back together. For an effect you can have other 3rd and 4rth takes that switch around on the cadences of the chorus to change it up or send to delay. You just have to automate the mutes.

[Drumsound]

Originally Posted by Mike Caffrey

No vocal paralell?

Why stop there and not do everything parallel?

Don't you have gainstaging issues? You must have to put the faders way down since you're making everything "twice" as loud by doubling the signal. Maybe it's not exactly twice as loud, but when you blend in the parallel, you're going to hit the stereo buss harder.

I have done a vocal parallel, and I have chosen insert comps instead of parallel. The group thing is where I start. Sometimes the comp return is quite low in the mix, but it's adding a tonal presence that I don't get other ways.

It's a bit of a game with gain staging, and I will admit I tend to hit things hard. I guess I've listened to too much RTB stuff over the years. I like how things sound when they're really workin'!

I'll do a lot of experimenting with how hard the comps are hit, how much return, the send levels, and what goes to each comp. Maybe the snare and toms but not the BD, OH or room. Sometimes just the bass DI, but not the mic. Maybe the acoustic guitars, but not electric. Sometimes the comp alone sounds better and the direct signals come out of the stereo bus.

Am I making sense or rambling here?

[Jonkan]

Hey! what happened to my initial reply?

Must have something to do with the boards being down yesterday?

Anyway,

I like using parallel compression both doing live FOH mixing, and also when recording.

Works wonders for ITB mixing.
/J

When?

When I need something to have the impression that it is louder than it is, and when you look that the meters it really isn't.

Usually on kick/snr/Bass and sometimes vox (and other things).

I don't do a drum bus parallel per se (usually), but I do mults of my drums that usually have a copy that has 15-20 db knocked off with a dbx 160xt. I smash the living shit out of it and add it to the original as if it were a reverb return.


For vox, I do it, but not exactly for the same purpose. For vox, its usually a case of different chains/compressors sounding good on different parts of the vox. So I automate between faders.

[Tommylicious]

I like to use compression while tracking, recording both the compressed and the uncompressed signal. I do this with kick, snare, bass and vocals usually.

Then in the mix i blend the two signals, wich basicly is parallel compresion, sort of. You can always throw on an xtra compressor on the blend.

[redddog]

Cool.

[Mike Caffrey]

Originally Posted by Drumsound

I have done a vocal parallel, and I have chosen insert comps instead of parallel. The group thing is where I start. Sometimes the comp return is quite low in the mix, but it's adding a tonal presence that I don't get other ways.

It's a bit of a game with gain staging, and I will admit I tend to hit things hard. I guess I've listened to too much RTB stuff over the years. I like how things sound when they're really workin'!

I'll do a lot of experimenting with how hard the comps are hit, how much return, the send levels, and what goes to each comp. Maybe the snare and toms but not the BD, OH or room. Sometimes just the bass DI, but not the mic. Maybe the acoustic guitars, but not electric. Sometimes the comp alone sounds better and the direct signals come out of the stereo bus.

Am I making sense or rambling here?

No, it makes sense.

It's a little different than how I interpreted it.

I though you were putting all the drums to a stere compressor, blending that in, then the same with the bass and the same with a guitar group as a matter of course rather than selective.

I also always imagin one particular engineer who limited everything in parallel and he simply wasn't doing what he though he was and he had no way to A/B.

The snare sound wetn "uop" instead of "pop". All he heard was the excitment of a quick release time and didn't really hear what he was doing over all.

The net effect of what he was doing was blending in the uncompressed signal to fix his bad buss compression and he did it across the board.

[NathanC]

Originally Posted by Tommylicious

I like to use compression while tracking, recording both the compressed and the uncompressed signal. I do this with kick, snare, bass and vocals usually.

Why do you do that (recording a compressed signal)? There are many engineers doing that and I've never understood why.

I hardly ever record with compression. I want full controll when mixing and don't want to be stuck with wrong choices (wrong compression and EQ) that occured when tracking. When tracking, all my attention goes to mic choices, mic pre choices, mic placement, room, and performance. Not processing.

I do record with compression when I'm recording voice overs and there are clients sitting in the controll room. It's a psychological thing: it sounds fatter right away

[Tommylicious]

Originally Posted by NathanC

Why do you do that (recording a compressed signal)? There are many engineers doing that and I've never understood why.

Well for me it's because i know i'm going to use (parralel) compression anyway. Note; i split the signal after the preamp so i record the same signal uncompressed. My focus goes down to the clean signal regarding mic placement etc. When compressing parralel you're settings are pretty extreme anyway so not so much of a hassle to dial in.

So for example; when i know i'll be mixing a track ITB it's just handy to allready have a well compressed copy of the signal there. This saves me dealing with plugins or using hardware inserts. Makes it's sort of a convinience thing too, now i can mix it down anywhere.

In another case this could save you sacrificing your one and only great hardware compressor in the mix for the parralel thing.

Also i believe a compressor affects the signal some what different before it hits the converters, resolution wise, but that's just a thought and not the reason i do it.

oh and sometimes i might not be using the compressed signal in the mix anyway, it's like recording with room mics, when you're mixing you decide by the ear what fits.

[Sinewave]

Originally Posted by Mike Caffrey

No vocal paralell?

Why stop there and not do everything parallel?

Don't you have gainstaging issues? You must have to put the faders way down since you're making everything "twice" as loud by doubling the signal. Maybe it's not exactly twice as loud, but when you blend in the parallel, you're going to hit the stereo buss harder.

Why is that ? I've experienced the same thing with stereo bus overloading , does the same thing happen in an Analog console ?.

[NathanC]

Originally Posted by Tommylicious

In another case this could save you sacrificing your one and only great hardware compressor in the mix for the parralel thing.

This seems a valid reason. However, you could also use that great compressor when mixing, adjust it carefully and then record the processed track to a new track. Then your compressor will be free for use on other tracks as well.

But I understand it's just a convenience thing.

[castle]

Originally Posted by NathanC

This seems a valid reason. However, you could also use that great compressor when mixing, adjust it carefully and then record the processed track to a new track. Then your compressor will be free for use on other tracks as well.

But I understand it's just a convenience thing.

I also do this, and it's not just for conveinence. Recording both compressed and un-compressed tracks at the same time also avoids the converter latency nightmare of re-recording audio through an analog processor and then attempting to play it back PERFECTLY in sync with the original.

[NathanC]

Originally Posted by castle

I also do this, and it's not just for conveinence. Recording both compressed and un-compressed tracks at the same time also avoids the converter latency nightmare of re-recording audio through an analog processor and then attempting to play it back PERFECTLY in sync with the original.

Ok, you convinced this ITB guy

[thethrillfactor]

Originally Posted by NathanC

My question: when and how do you use parallel compression?

http://gearslutz.com/board/showthrea...sient+Designer


http://gearslutz.com/board/showthrea...sient+Designer


http://gearslutz.com/board/showthrea...sient+Designer


These days thank god the chains are much simpler. I don't do much vocal parallel compression(i've never been the biggest fan of it only for a last resort).

Basically whatever is needed...whatever works.

[Mike Caffrey]

Originally Posted by Sinewave

Why is that ? I've experienced the same thing with stereo bus overloading , does the same thing happen in an Analog console ?.

Which is louder, 1 track or 2 tracks? (assuming you diddn't change the volume of the first track when you added the second)

Yes it happens on an analog console.

Originally Posted by thethrillfactor

http://gearslutz.com/board/showthrea...sient+Designer


http://gearslutz.com/board/showthrea...sient+Designer


http://gearslutz.com/board/showthrea...sient+Designer


These days thank god the chains are much simpler. I don't do much vocal parallel compression(i've never been the biggest fan of it only for a last resort).

Basically whatever is needed...whatever works.

Thrill, What would you say is the reason your chains are simpler today?

Reading those old threads, kinda made me breakdown, remembering when my mom was living and battling Cancer. Man it's still hard to believe/realize/accept, that she won't be there to answer when I dial her number.

[thethrillfactor]

Originally Posted by no ssl yet

Thrill, What would you say is the reason your chains are simpler today?

3 main reasons:

1) I am tracking some of the productions i get to work on these days so early on i can get a handle on how things should sound later on.

2) A lot of the sounds i get to mix are much better now. Also more and more producers want me to stick to exactly to their vision so i deviate less and less and stick to the status quo. Which is fine by me. Its more a mastering thing than a mix thing but hey as long as i get paid...

3) I hear and think differently about stuff so you change with the times.

These days i am not as focused as much on the direct/upfront sound(which parallel, multing and such brings out and in droves) but more of the implied stuff especially when working digitally. It used to be cool to make drum sounds "pop out" but since everyone does it now you need to do something different to stick out from the crowd.

Originally Posted by no ssl yet

Reading those old threads, kinda made me breakdown, remembering when my mom was living and battling Cancer. Man it's still hard to believe/realize/accept, that she won't be there to answer when I dial her number.

Sorry to bring up those feelings bro.
(My condolescenses again).

Originally Posted by thethrillfactor

3 main reasons:

1) I am tracking some of the productions i get to work on these days so early on i can get a handle on how things should sound later on.

2) A lot of the sounds i get to mix are much better now. Also more and more producers want me to stick to exactly to their vision so i deviate less and less and stick to the status quo. Which is fine by me. Its more a mastering thing than a mix thing but hey as long as i get paid...

3) I hear and think differently about stuff so you change with the times.

These days i am not as focused as much on the direct/upfront sound(which parallel, multing and such brings out and in droves) but more of the implied stuff especially when working digitally. It used to be cool to make drum sounds "pop out" but since everyone does it now you need to do something different to stick out from the crowd.






Sorry to bring up those feelings bro.
(My condolescenses again).

It's cool Thrill, I get that way from time to time anyway when thinking about deceased family. (Especially MOM).

I asked because I wasnted to know if your reasons for simpler chains were the same as mine.

I find that eventhough sources are digital samples (rompler/MPC etc...) Hitting it before it goes ITB is a bit better than hitting it after it's been recorded to protools.

As often as possible, I try to feed "the box" with stuff that "sounds like a record" (Whatever the hell that is these days )

On occassion I have to wrestle with stuff, but like you said, a bunch of guys want their vision, and we aren't paid to be visionaries in that case.


(Thrill, It's actually kinda cool to be able to relive that part of life, and know where I was at the time. During the Hurricane, I lost alot of family pics etc... So it somewhat brought me back in a good way as I began to think of good times before the Cancer days.)

[Sinewave]

Originally Posted by Mike Caffrey

Which is louder, 1 track or 2 tracks? (assuming you diddn't change the volume of the first track when you added the second)

Yes it happens on an analog console.

But isn't doubling of tracks a normal part of modern music production ? how do some engineers control the volume ,Limiters, compressers ?
thank you for your time.

[Mike Caffrey]

Originally Posted by Tommylicious

Well for me it's because i know i'm going to use (parralel) compression anyway. Note; i split the signal after the preamp so i record the same signal uncompressed. My focus goes down to the clean signal regarding mic placement etc. When compressing parralel you're settings are pretty extreme anyway so not so much of a hassle to dial in.

So for example; when i know i'll be mixing a track ITB it's just handy to allready have a well compressed copy of the signal there. This saves me dealing with plugins or using hardware inserts. Makes it's sort of a convinience thing too, now i can mix it down anywhere.

In another case this could save you sacrificing your one and only great hardware compressor in the mix for the parralel thing.

Also i believe a compressor affects the signal some what different before it hits the converters, resolution wise, but that's just a thought and not the reason i do it.

oh and sometimes i might not be using the compressed signal in the mix anyway, it's like recording with room mics, when you're mixing you decide by the ear what fits.

Also, your compressed signal is converted once instead of 3 times as it would if you went in and then out and back in.

[andychamp]

Originally Posted by NathanC

My question: when and how do you use parallel compression?

Usually kick, snare (both with mucho gating and EQ) and bass.
For voices I recently started to develop a preference for a combination of serial compression and automation.