Hey all,
 
So I just recently gave parallel compression a whirl for the very first time and I liked it alot.  As common for me, if I tried it one one instrument and liked it, I'd try it on them all.  I liked it best on piano, drums and backing vocals.  The cymbals were included on the stereo drum track from Stylus, and I'll be avoiding those next time for PC.  From your experience, any guidelines for types of tracks that seem to benefit the least from this treatment?
 
Thanks!
 
Donny

The higher the overtones, the less I like parallel compression.  Or the way I squash stuff w/ it, anyway.
 
@

Anything that sounds anemic is usually a place where parallel compression can apply. The reason is that a thin sounding source is usually high in mid range and upper end transients. A parallel feed through a compressor will tend to flatten the dominant range and raise the lower level transients. Added back in the mix, it will fatten the sound of the source. This can apply to Kick Drums, Snares, Vocals.
 
Another use is for side chain control, where you want one source to be lowered in volume to let another source dominate the track. Specifically, when a bass compressor that is side chain keyed by the kick drum. This lowers the volume of the bass guitar when the kick drum plays. This keeps the track consistent from a rhythm perspective.
 
The above examples are the most often used areas but really, there are no rules. You can even parallel compress an entire mix when mastering. It can really fatten up your track and get it up to competitive levels of volume.

doncolga
Hey all,
 
So I just recently gave parallel compression a whirl for the very first time and I liked it alot.  As common for me, if I tried it one one instrument and liked it, I'd try it on them all.  I liked it best on piano, drums and backing vocals.  The cymbals were included on the stereo drum track from Stylus, and I'll be avoiding those next time for PC.  From your experience, any guidelines for types of tracks that seem to benefit the least from this treatment?
 
Thanks!
 
Donny

I get pounded for this advice all the time, but those who have either become students or have trusted me in what I say have agreed with me in the long run.
 
The advice is, be careful with it and try NOT to use it unless you feel a need, a reason or a cause and effect. Engineers (especially bedroom hobby guys) are in search of big, fat tones. In turn they think having this on all instruments makes things sound better when in reality, it does not. Some sounds need to be thinned out so they fit in the mix. Not every instrument is a focal point instrument so it shouldn't be treated that way. In other words....you don't "just do" something "for the sake of."
 
I actually don't like p-comp on anything but drums in certain situations. The added impact can be nice for certain things but you wind up losing some of your dynamics due to the additional resonance you pick up. I'd just about never put this on a piano or an acoustic guitar due to how dynamic those instruments can be. Too much p-comp across multiple instruments and you can totally ruin your song.
 
The next question I ask all my students is "how do your mixes sound without all these additional techniques?"
 
The fact to be concerned with here Donny is NOT to run before you can crawl or walk with authority. This is one of the things I hate about forums and reading books. These techniques are introduced to people (not talking about you here, just in general) that can't mix right just yet so they shouldn't be going here. It's like buying an old dirt bike used for trail riding....you get the basic idea on how to ride and have a little fun, then next week you come back with a race bike made for moto-x. You can get jacked real quick if you're not careful.
 
My reason for stating that is due to all the people I've heard abuse this technique to the point of me just shaking my head. Hearing previous work they had done made me shake my head some more. It's like this....we had incredible mixes without this technique for quite a few years. Don't depend/rely on it to work magic. You should be able to come up with great mixes without it. When you're at that point THEN experiment a little and see where it works and where it may skew your mixes....because it can, and it will.
 
Unfortunately, this isn't one of those effects/techniques that you can use "a little" of. Meaning, you have to use a decent amount of it to hear the effect of it. Subtle doesn't do it justice really. So for sure it's something to be super careful about in MY opinion only. I've used it on drums when the drums needed that sound.....I've used it on bass for a bit of the "now sound" for clients.
 
I've used it on vocals for "effect purposes" in parts of songs....but it's not something I would ever put on a bunch of instruments at once because not everything needs to be big, and not everything needs that kind of impact. So just be careful with it and try to come up with a reason as well as a cause and effect for why you feel you need to use ANY technique.
 
The reason being, sometimes less truly is more and will do the job even better. If the sounds you are working with are not sounding big enough or they don't have enough impact, find out why. Record new sounds if you need to and experiment there. 9 times out of 10 you win when you capture a killer sound from the beginning instead of trying to make something bigger or give it more impact via effect or technique. Other times, sure, effects and techniques are great for that. I'm not trying to talk you out of anything....I'm just trying to feed your head while making you think.
 
I had a few students recently that were sending me mixes with p-comp in spots. They were having issues with the mixes and weren't crazy about them. I made them kill the p-comp, they re-eq'd and compressed properly, applied some effects differently and replied back "now I know why I work with you...what a difference for the better!"
 
So the moral of the story here....just like in real life, you have to pick your battles. Know when to have impact, know when to allow dynamics and natural sounds to flow through while getting them right instead of hoping a technique or effect will make the difference. Learn to do anything and everything you can at all times...but always be sure that you can record good tracks first and foremost, know how and when to eq, how to compress and where to pan. If you do that much and it sounds great, you honestly won't need much more other than "producing" the material a bit. If that fails though....p-comp and everything else won't save the mix. Just stay focused. Good luck. :)
 
-Danny

Dude...thanks for the post.  Awesome read.

I concur with Danny regarding excessive use of parallel compression. It kills a mix quickly. The trick is to bring up the compressed track under the original so that it fattens but does not dominate the transient information of the original. More importantly, does it really sound better? A lot of times you can mix yourself into a wall of flat sounding compression. Of course if a wall of flat sounding compression is your goal, hey, go for it. There are some genres which depend on that but a lot of times its important to pull back and see how the original tracks balance with little compression or FXs.
 
The goal, and this is a tracking issue, is to record a balanced mix from the get go. When you push up the faders, if you can get very close to the desired results, you won't need a lot of excessive compression or EQ. There was a recent seminar that Al Schmidt gave here in LA. He tracked an entire musical group with no EQ or compression. Just mic to preamp to hardrive (there was a Neve board in there) . If he needed more midrange or bottom, he moved the mic to get the balance he wanted. When he pushed up the mix, it was very close to the final outcome. If you focus more on the actual sound being recorded, you won't need as many plug ins to pull things together.

Middleman
I concur with Danny regarding excessive use of parallel compression. It kills a mix quickly. The trick is to bring up the compressed track under the original so that it fattens but does not dominate the transient information of the original. More importantly, does it really sound better? A lot of times you can mix yourself into a wall of flat sounding compression. Of course if a wall of flat sounding compression is your goal, hey, go for it. There are some genres which depend on that but a lot of times its important to pull back and see how the original tracks balance with little compression or FXs.
 
The goal, and this is a tracking issue, is to record a balanced mix from the get go. When you push up the faders, if you can get very close to the desired results, you won't need a lot of excessive compression or EQ. There was a recent seminar that Al Schmidt gave here in LA. He tracked an entire musical group with no EQ or compression. Just mic to preamp to hardrive (there was a Neve board in there) . If he needed more midrange or bottom, he moved the mic to get the balance he wanted. When he pushed up the mix, it was very close to the final outcome. If you focus more on the actual sound being recorded. You won't need as many plug ins to pull things together.

+1000!!! Spot on Middleman!
 
One thing I got to thinking about while reading your post that I wanted to share for everyone messing with this...
 
Sonar X1/X2 allows us to p-comp in one spot due to us having the bus compressor. The bus comp (don't let the name fool you) is incredible as a compressor and one of the best comps out there in my opinion. Use it on everything, not just busses. It smokes more compressors out there than those that smoke it...trust me.
 
Anyway, the cool thing about this compressor is it has a mix control. Meaning, wet/dry. This is awesome for quick p-comping as you do not need to create a separate bus for it. For example, when I do *some* of my p-comps, I will create a drum bus, send all my drums to it and then put a compressor in the bin to lightly comp the kit. This keeps the kit tight and glues it so to speak. Then I add the bus comp in the pro channel....pump up the compression and then use the mix knob to mix in the comped sound with the sound I already have and boom..instant p-comp which is completely controllable. Hats off to Cake for including a mix knob on a compressor...brilliant!
 
Now the down side of that technique there is, you don't have eq control over the p-comp or the light comp because they are on one bus. If you were to create an additional bus and add the bus compressor there and then insert sends on all your drum tracks, you would then have control over eq'ing the p-comp which in reality, you would want that kind of control. Here's why...
 
When you p-comp, the resonance you pick up sort of simulates a reverberated type "decay" sound. With too much highs, your drum kit (in this example) might end up hissing like a snake depending on how you have your hats and cymbals eq'd. Also, the resonance from the compressor should be eq'd anyway because this is how you literally control your impact. But it all depends on what type of sound you are going for as well as how much impact.
 
I find that most times I can get away with just p-comping my drums on one bus...which is of course my main drum bus. I don't use enough p-comp to where the impact it is adding it crucial...but without it, I notice it's not there. So I'm using just enough of it to make a difference without going the extreme. If I DO want to go the extreme route, I'm going to put the p-comp compressor on its own bus so we can eq it, effect it and have full control over how the extreme p-comp is going to sound in the mix. :)
 
-Danny

The following are my thoughts in conjunction with that of Mike Stavrou a well respected engineer over here in Australia. He wrote an excellent article on this very subject in issue 93 Audio Technology Magazine. (February 2013) I think it is also highly over rated. If you have the nerve to suggest to a parallel compression zealot that you prefer a different approach they come out swinging! I have met a few.
 
The problem with it is that we need compressors to control the voltage but the moment you start introducing the dry signal component back in, what you are effectively doing is putting the out of control dynamics back into the signal again which is sort of useless. Some say yes but it adds level to the low level stuff. And it might but because the signal is out of control again (due the dry signal being brought back in) you will need a compressor again at some point down the track anyway to bring everything under control thereby reducing the effectiveness of the low level stuff anyway. The louder stuff will be pushing that away. Who says you need the low level sound of a smahed compressor (artifacts) added to a nice clean signal anyway.
 
A very very good approach is to use a dual in line approach. ie two dynamics processors in series. The first one should be a limiter set for a ratio of 4:1 or higher, fast attack and a high threshold so all it does is tame the loud peaks only. This creates the desirable effect of taming the loud bits down to match the overall signal level much more in line with the rest of the sound. This in itself brings the lower level detail up much better than parallel compression. (especially as you can now add some makeup gain at this point too)
 
The second processor after the first one is a compressor set for a low ratio 2:1 or less, slow attack 20 ms, and a lower threshold so it is working much more. The two things combined will give a nicer result IMO than any parallel compression setup. With vocals BTW I sometimes put the limiter on the track but the compressor on a sep vocal buss. That way you can ride the fader level and control the signal going into the second vocal (smoother) compressor that comes later.
 
The great thing about the dual in line approach is the huge amount of control you have over the sound due to the various settings that you can adjust from both processors.
 
The thing to remember is that parallel compression is not effective for predictable level control, it is more of a flavor.
 
Some say parallel compression is really only effective on one thing and that is drum overheads. It can be said it seems to do a slightly better job there than the dual in line approach. In nearly every other situation the dual in line approach works and sounds much better. (BUT! the drummer needs to also be very controlled in how hard he is hitting everything and it requires the ability to be able to create even sounding and more level sounding drum parts, something most drummers are not capable of! Even parallel processing has problems with drum OH if the hits are all over the place level wise)
 
With me BTW I tend to only use a dual in line setup on one thing and that is the vocals. It works very well and it is fantastic at keeping vocals just sitting perfectly in a mix, (never jumping out either, limiter takes care of that) it offers total level control and can sound due to the nature of the second processor being so relaxed that there is nothing on the vocals at all. (remember higher ratios make things sound small, lower ratios make things sound big and as a result you end up with maximum illusion, minimum voltage and you end up turning the vocals down even further in the mix yet thay are still loud and clear)
 
I don't use the dual in line approach with tracks on much else. I find single compressors on most things do an excellent job of level control and that after all is what we are after. I use it more now on buses and it works in mastering too by the way. The limiter first to catch a lot of silly peaks and do it's thing in a dynamic manner. Second into EQ > Compressor > Final limit for increased loudness (only if necessary) Editing software can replace the final limit if you just want to trim overs and maybe add another 1-2 dB of rms level or two. A limiter like Xenon is good too because the mix comes out still sounding sweet if you use it first. A lot of limters sound better just not allowing a level to rise above a threshold as opposed to adding large amonts of rms volume.
 

Great food for thought guys.  Very much appreciate your time replying back.

I use parallel compression mostly on vocals, but may at least audition it on just about anything. Think of parallel compression as upward compression, making the quiet bits louder. That'll help you decide which instruments are most likely to benefit from it.
 
Jeff's right: parallel compression partially defeats leveling efforts. However, if you use a compressor with a wet/dry mix control rather than the traditional bus method, you then have the option of preceding it with a conventional leveling compressor when the uncompressed levels are all over the place.