Side Chain Compression

M

martinweeks

New member
Okay here goes this particular question has been a thorn in my side now for months. So pardon me if I sound a bit cranky here but I'm completely fed up with all the other responses I've gotten from outside of this forums...and it's not that big a deal it's just a pain in the butt for me due to my equipment in the box.

I'm very familiar now with using parallel compression in different ways, however, there's a specific thing I wish to do using side chain compression.

So we are all on the same page...I AM NOT CONCERNED AT THE MOMENT WITH PUMPING BASS AND DRUM STUFF. So please do not waste time with all that drum/bass side chain discussion, there's already a universe of that on the web. It is not what I need to know about.

We (as in my fiancee and myself) are writing songs. The lyrical punch is important ergo there are specific points in the songs where the vocals stop and start.

All I am trying to do is one single thing at this time. And that is:

Use side chain compression to squash digital delay in a Send track during the vocals and release the effect when the vocals pause or stop. NOTHING MORE SO AGAIN NO DRUM BASS TALK PLEASE.

My DAW does not support "Native" Side Chain compression however, I do have compressors that will do this. Specifically I am talking about "Ducking" nothing else. I wish to use side chain compression to "Duck" the delay during the vocals, and release the "tail" of the effect (slight delay nothing radical) during the pauses on the vocals. For dynamic effect only, nothing crazy. Think live concert sound where there's that real small slap back from the back of the auditorium or whatever. (See you next Tour....tour.) Got it? That is all I want to accomplish.

I've now seen dozens of videos and read dozens of articles and they all do the same thing. They talk about pumping drum and bass. They also all do the same thing and talk about either Pro Tools, or Logic, or Reaper or studio 1 which all have native abilities for this or come with side chain specific compressors that have the "KEY" icon to make this process simple.

I do not.

What I have is the reaper "Realcomp" compressor which will do this without the need of a "Key". It sort of works. I worked on this last night using a demo recording that came with the DAW just doing the drum/bass thing so I could work with threshold and ratio trying to figure this out. However,

When I placed a vocal loop in a new track (vocals had long phrases and long pauses so I could over drive stuff and clearly hear what was going on.) I couldn't tell if I was actually doing the squash or not.

Their is also a well known side chain compressor called sidekick 6 which does all this, but i''ve downloaded it and it will not load into my DAW (don't know why and don't care at the moment.) also tried another "Side chain" compressor download and it too also failed to load. But the Reaper compressor works fine in my DAW. So this is the one I will be using for the time being

Other aggravation: I've seen videos that tell me I need to have two compressors to do this...one for the track sending the audio and one for the track that is to be affected by the signal. This doesn't make sense to me as I don't need to do that when it comes to drum/bass stuff. the kick signal causes the compressor in the bass track to work and compressor stops when the kick signal is gone. So why I would need to do this for ducking a reverb or delay effect makes no sense to me. If I do have to do this fine, but I would like the clarification once and for all so I can get back to work.

Next Aggravation. "Language or Lack thereof." Too many times I hear/see these videos and they refer to Send Tracks as Aux buses. Excuse me but a Send Effect is essentially send/return (old school PA stuff) The effect is controlled by the return of the effect back to the source in other words blend of levels. Want just barely audible return...turn down the send track signal as it's going back to the source signal. an Aux bus is a straight line...i.e. multiple drum tracks being aux bused through the aux to the main out. Straight line. Send tracks do not do that. (well maybe in the digital world but I have always thought the reverb or delay was sent back to the original instrument track. then out to the main. If I'm wrong please correct me.)

So here's the outline of what I wish to do.

1. I always use send tracks for delay and/or reverb as it's CPU conservation. Yes I also will use send tracks for compression, as it is a means of doing parallel compression (nice effect too by the way) But for this post we are focusing on what the vocal track does to the delay effect in the send track. or send bus whichever is easier for you to wrap your head around.

I know some of you folks most likely do use the Reaper Software. I do not. So I need a basic 1,2,3,4 step by step instructions on how I can set up my vocal track to trigger the "Realcomp" compressor in the send track to squash the delay while there are vocals. Nothing more please just this one specific function.

I've also seen just recently a video where the narrator used a Limiter (huh and why?) and the limiter was after the delay effect. (????) He was of course using Studio 1 DAW and I am not. So that training video was a total waste of my time again.

One more detail as I'm trying to be real specific and give you a visualization of how I set up my vocals.

I record the vocals mono.
Then I copy the vocal track and pan them left and right about 10 and 2 for basic widening. (and of course their's EQ and level balancing and all that)

These two vocal tracks are identical in every way to avod "Phase Issues" and they are then "AUX Bused" to the main stereo out.
the Aux Bus is where I place Compression in the traditional manner used for mixing. It is for the sake of gluing and making the vocals sit in the pocket as the lingo says. Works very well for basic static vocals, but I need to step this up to a higher level hence the need for the dynamics. I do not want to mess with the aux bus effects chains as I like the vocal tone that is created...but again it's still static. Doesn't breath (well it does as I use automation but that's another issue) So again...do I need to create yet another copy of the vocal track to use as a trigger or can I just use one of the two identical vocal mono tracks to trigger the compressor reaction?

I've already done this with manual automation but it takes forever to get done...especially when the song vocals are bouncy and sophisticate (which is the norm not the exception) so I need to use the compression technique to do the automation of the ducking. And it simply sounds better plain and simple.

So can someone please just send me a 1,2,3,4... set of instructions so I can set this up and I can then work with the thresholds, and ratios and release times to fine tune the sound I am seeking to create?

Summary of questions:

How many compressors and where do I place them in the chain of effects before or after?
What is "seeing" what? Is the second compressor seeing the vocals or the configuration of the compressor?

Below is a link to a song we recently did where I do this effect using automation. It's just not as good and clean as it could be but you will get the point. the vocals start out in lower register and are essentially flat. (I do use the Softtube Saturation Knob in the VOX Aux bus to fatten up the vocals) and as the vocals pitch gets higher and builds up to the guitar instrumental interlude the delay gets to go a bit nuts. then is toned down but remains present through the remainder of the song. Like I said it's not as clean and precise as I'd like but it's what I'm aiming at. Control of delay for dynamics.

Please let me know...we have a lot of unfinished material that needs this specific style of dynamics.

Thanks and again my apologies for sounding cranky. Working two jobs and dealing with an assortment of disabilities issues leaves very little time to just sit and be as creative as I'd like to be so I need to be more efficient with my work flow and time management. This is a means to an end as they say. Here's the link to the song as an example.
A Place I'''ve Never Been by LinMar Productions
 
It would probably help people if you told us what DAW you are using, as opposed to what DAWs you are not ...
 
Go get izotope's DDLY from their website. They're offering it free until March 9, and it does exactly what you want to do.
 
Really? Cool. Actually did this last night...haven't checked the email yet for the download stuff. Thanks will let you know how this works.
 
There is no difference between what you want to do and what people typically do with the pumping drum and bass thing. Exactly the same thing. You said you got that working...

The delay should probably be 100% wet (0% dry) and must come before the compressor.

I would at least start by doing whatever else you're going to do to the vocal first. If that means automating volume or compressing or EQing or whatever. You need to make sure that what you're sending to the sidechain comes after all of that so that the compressor is reacting to what you're hearing.

Leveling especially is important. If one word or phrase in the vocal is quieter than another, it will trigger less compression in the side chain, and the delay will duck less = stay louder. I'd think this is almost the opposite of want you want. I suppose if the delay itself is also getting the quieter signal it comes out in the wash? This, though, is probably why people are recommending the second compressor. You don't need that with most combinations of synth bass and drums because they're usually already pretty flat dynamically.

All that said, I suggest you just switch to Reaper. We have a number of ways to do what you're trying to do. In fact, we don't even need a compressor to accomplish it. :)
 
my apologies...as I said I've been inundated with other folks's "their stuff" replies. Okay.

Currently using the "Mixcraft 7" DAW (not their upgrade "Pro" Version. Serious financial issues preempt any more upgrades of anything for a while.)

From the tutorials I've researched I keep seeing the same quote: "this process works on any DAW"...etc. I've asked Acoustica Tech Support about this and they say it's not "Native" to my version of the DAW. But it still seems to me that with the right compressor it wouldn't matter. The "Realcomp" compressor from Reaper works fine with my DAW in all other respects and no issues, so it's something in the instructions I'm missing or a settings. I know it's me missing a detail but can't figure out what I'm "Not" doing correctly.

For the most part MC works really well with just about everything I've downloaded through the past Year. Only time I have issues is when I get too much stuff going on at one time (CPU and memory issues) on my computer. So I do keep things as simple as possible limited actual tracks, correct routing and bussing. It's just this one specific task that's making me nuts. Sorry about the rant again.
 
thank you for your reply... running late for work so will respond in detail later tonight. You've been VERY helpful. More to come.
 
My replies keep going away... thanks for your help. i have some more questions but just spent an hour writing them and lost them as I timed out or somethingg in the post. Will reply in more detail tomorrow.
 
So you're wanting to use the vocal signal to compress the delay FX on the vocal channel? Am I reading that right?

Here's how I'd go about it:
1. send the delay effect to it's own channel, 100% wet
2. Apply the "bass pumping" sidechain to that channel and adjust to taste

Things to adjust:
ReaComp has a "pre-comp" slider - essentially lookahead. Your big difficulty here seems like it would be getting the compression to stay engaged between words but not stay active over the pauses. I would try using the pre-comp to get it to start compressing early rather than using hold to keep the compressor on.
e.g. Let's say you have the phrase "here's the pause <pause> Yeah!"
You want compress the delay in the space between all of the words but not after, "pause", right?
Normally, you'd use the hold feature to keep the compressor active from "here" to "the" and "the" to "pause", but that will also keep the compressor active after "pause". With pre-comp, you can get it to start compressing for "the" right after "here". Then with a quick release, the compressor will drop off right away after pause. It will kick in before "yeah", but presumably the natural decay of the delay will mask that?
 
VHS - You may have jumped a couple of steps ahead, but it is a good point. Just do understand the lookahead moves the whole envelope forward in time. In the example you've given, the compressor will start to release as some point during the word "pause". Depending on the actual length of the release (and depth of the squish, and probably all of the other parameters also) this may or may not be noticeable, but I would expect to hear the delays start to swell in a bit prematurely. Then again, if we really wanted exact precision, we'd draw in a volume envelope on the delay track. ;)

Edit cause...

Ooo! It just occurred to me that we can take my last hint about volume envelope and add it to the other hint about not needing a compressor. Use parameter modulation on the output volume (not mix) of the delay plug. Set the envelope for that parameter to Write mode. Play the song through one time. It makes an envelope for you. Switch it back to Trim, and turn off the PM, and then edit the envelope as necessary.
 
Last edited:
Thank you so much for taking the time to provide some explanation for me. It is MUCH appreciated.

Apparently one of the things I was doing incorrectly according to your instructions was where I was placing the compressor. I was under the impression that the compressor had to go before the delay in the same manner that we use dynamic compressors in individual tracks (i.e. EQ, compressor, then other stuff.) Hopefully this will help.

Just a couple of clarifications though so I don't repeat past mistakes.
When I record vocals the process I always use is:
1. Record the vocals mono.
2. Do necessary EQ etc in that track. Once satisfied with that I...
3. duplicate the track so they are identical to avoid phase issues from EQ.
4. Pan individually left and right about 10 and 2. (easy widening trick)
5. Set up Aux bus and make the two “Child Tracks” blend to the aux bus. This is where I usually place normal dynamic compression for the purpose of “Glueing” as they say and making vocals sit in the “Pocket” as it is usually referred to.
So as far as sending to the “Send Track” with the delay and compressor do I send the aux bus or the individual or original vocal track?

The DAW I use is Mixcraft 7. Before you dismiss it as a joke I need to let you know that this DAW is one of those “What's under the Hood” type of things. As I've explored it more thoroughly I have realized that for pretty much all of the essential necessary functions of a DAW this one is as sophisticated as Pro Tools or Reaper. (no criticism of Reaper here just a comment) It's routing capabilities and automation are top notch...so my question regarding “sending” the vocal track to the “Send Bus where the delay and compressor is, is this:
Am I sending the entire vocal sound from the aux bus track to the send? Or just volume, or compressor. (Still a bit fuzzy on whether to use one or two compressors.)
Where you refer to making signal totally wet and totally dry...are we talking about the effects control in the actual delay plugin? Or are we talking about what is “Sent” from the original track and what is “returned back” from the Send Track? (forgive me I'm a “gigger” and still used to working with hardware delays inserted into a PA. On those devices you control send and return manually with knobs directly from the unit and or the mixer PA. So working inside the box get a bit confusing sometimes visually.)

Apologies for not replying sooner. The forum keeps timing out on me. I've written this same post about five times, and every time I click to reply to thread it knocks me out and I lose the reply post.

Thanks again for the help. And as to Reaper. I have it, but for reasons I cannot figure out every time I attempt to save a mix project, the software eats the mix. All the tracks come up when I come back to them to work on as: “Offline”. When I attempt to find these tracks and/or project they are nowhere to be found.

I also get totally confused when attempting to set up all the sends and routing in Reaper. It does stuff radically different from 90% of other DAW's out there so the learning curve has absolutely frustrated the daylights out of me. I've under a time pressure deadline on several submission opportunities and working two jobs at the same time, so I do get a bit frazzled. Don't mean to be rude or snooty in the posts, just exhausted a lot of the time.

Thanks again for the help.
 
Did and played with it a bit last night. Where do I go to find the manual for this plugin? Saw a bunch of videos about how to increase and decrease the effect "while" the audio is playing...but I'm looking to duck the audio while vocals are active and bring in the tail when there's a pause or stop of vocal signal. Hence the side chain compression thread/post. Looks like it will work very well but I'm still figuring it out and on a very limited time schedule/frame. (two jobs and other stuff makes time management a real you know what. Thanks for the heads up though...great plugin.

Also got the "Bus Driver" compressor (another freebie I found in this forum.) Really sweet compressor. Easy to work, and the presets are really well configured.
 
Ashcat_IT. Sir you are a prince. Thank you again for taking the time to walk me through all this.
re: >>Ooo! It just occurred to me that we can take my last hint about volume envelope and add it to the other hint about not needing a compressor. Use parameter modulation on the output volume (not mix) of the delay plug. Set the envelope for that parameter to Write mode. <<
While the initial purpose of my theard/inquiry was to more or less automate just a slight slap back return for general dynamics purposes there will be opportunities where I might want to make the delay go radical for a minute. so are you saying that I can manually adjust the strength, speed and so on of the delay "On The Fly" without disrupting the "Ducking" effect of the compressor? (I'm still working the learning curve re: automation.)

Below is a simple example of the above mentioned. However, I had to literally copy/paste the very last word in the phrase to a new track...re edit and eq it and add an additional send track with effect to avoid conflict with the original send track effects used for the other music tracks. So are you saying i can set up the side chain send track and still adjust the intensity of the effect in real time?
Our Lord'''s Prayer by LinMar Productions
 
Ummm...

First of all: When you hit the Reply button on a specific post in the thread, the Quick Reply box pops up right under the post you're responding to, but when you hit the Post button, it's going to put your post at the end of the thread, so it gets a bit confusing. Reply with Quote can help because then we can see what you're responding to. I personally prefer if you edit the quoted text down to something short and relevant to your reply so that we don't have quite so much redundant information wasting space. Another way to do it just to adress your post to the specific poster to whom you're responding.

So what was the question? Oh, there were several. I'll just spew some random answers and probably not answer everything. ;)

You want the controls on the delay plug itself set so that there is none of the dry signal coming out of it. Sometimes that's a mix knob, sometimes that's separate controls for wet and dry output. This has nothing to do with sends of returns or anything. (We don't have returns in reaper, just tracks)

If you put the ducking compressor before the delay plug, you'll have the vocal sound going into the delay "ducking" itself (lol! Go duck yourself!), which is essentially the same as just putting a compressor before the delay - you don't need sidechain for that. That ain't what you want. Compressor after delay so the repeats from the word before are squashed while the new word is happening.

The compressor won't really have any affect on the function of the delay. Go crazy. Have fun. The compressor just turns whatever the delay (its final output) does down when there's something happening on the dry track.

Duplicating a track and panning the copies away from each other is not an "easy widening trick", it's an easy loudening trick. Just makes the whole thing louder. Period.

But since you've already got that whole redundancy set up, you'll probably want a post-FX send from the aux bus. The compression on that bus is making your vocal dynamics more consistent, and it would be nice for the delay to duck similarly consistently, rather than following the (relatively) more erratic dynamic of the unsquished vocal.

In can't tell you how to actually set up the routing in MC cause I've never touched it. You said you had the sidechain thing handled...
 
oops this reply went to the wrong place. sorry.
You kind of lost me when you said:
>>1. send the delay effect to it's own channel, 100% wet<< I always get befuddled with send, aux and so on. What exactly do you mean by "send" the delay to it's own channel??
 
VomitHatSteve said:
So you're wanting to use the vocal signal to compress the delay FX on the vocal channel? Am I reading that right?

Here's how I'd go about it:
1. send the delay effect to it's own channel, 100% wet
2. Apply the "bass pumping" sidechain to that channel and adjust to taste

Things to adjust:
ReaComp has a "pre-comp" slider - essentially lookahead. Your big difficulty here seems like it would be getting the compression to stay engaged between words but not stay active over the pauses. I would try using the pre-comp to get it to start compressing early rather than using hold to keep the compressor on.
e.g. Let's say you have the phrase "here's the pause <pause> Yeah!"
You want compress the delay in the space between all of the words but not after, "pause", right?
Normally, you'd use the hold feature to keep the compressor active from "here" to "the" and "the" to "pause", but that will also keep the compressor active after "pause". With pre-comp, you can get it to start compressing for "the" right after "here". Then with a quick release, the compressor will drop off right away after pause. It will kick in before "yeah", but presumably the natural decay of the delay will mask that?
Click to expand...

You kind of lost me where you said:
>>1. send the delay effect to it's own channel, 100% wet<< Can you explain exactly what you mean by "Send the delay" ?? I'm not using reaper so I have to figure out what you are actually saying for me to do. Reaper does sends stuff a lot different from the more traditional DAW's. Thanks.
 
Why not just print the delay to a separate track, chop it up and use only the portions you want? Then you can automate the delay tails to sound exactly like you want at every instance. Real time ducking is for broadcast or live applications.
 
martinweeks said:
Just a couple of clarifications though so I don't repeat past mistakes.
When I record vocals the process I always use is:
1. Record the vocals mono.
2. Do necessary EQ etc in that track. Once satisfied with that I...
3. duplicate the track so they are identical to avoid phase issues from EQ.
4. Pan individually left and right about 10 and 2. (easy widening trick)
5. Set up Aux bus and make the two “Child Tracks” blend to the aux bus. This is where I usually place normal dynamic compression for the purpose of “Glueing” as they say and making vocals sit in the “Pocket” as it is usually referred to.
Click to expand...

Not directly related to what we're talking about in this thread, but I wanted to note that duplicating a track and panning them out does nothing to make it sound wider. It's just making it louder. I'd recommend actually double tracking your vocals and doing this if you're wanting a wider vocal sound.
 
You must log in or register to reply here.
Back
Top