I hate to admit it: I don't understand sidechains.
-
- audio school graduate
- Posts: 22
- Joined: Thu Jun 12, 2003 11:32 pm
I hate to admit it: I don't understand sidechains.
What is the purpose? What are some interesting/unusual ways to take advantage of a sidechain on my compressor? Is there any application that you can only accomplish with the aid of a sidechain?
-
- zen recordist
- Posts: 8876
- Joined: Mon May 19, 2003 12:10 pm
- Location: NYC/Brooklyn
- Contact:
Re: I hate to admit it: I don't understand sidechains.
The sidechain is a way to make your compressor (with a sidechain I/O obviously) more frequency dependant in the detector. I am not a tech, but in practice:ScienceOne wrote:What is the purpose? What are some interesting/unusual ways to take advantage of a sidechain on my compressor? Is there any application that you can only accomplish with the aid of a sidechain?
If you put an EQ in the sidechain, and really crank the shit out of 1K, you will be forcing the compessors detector circuit (NOT THE AUDIO) to react to 1k much more drastically than any other frequency. You still hear all the audio you sent into it coming out the other side. A de-esser is a compressor with a fixed sensitivity (or sweepable) to high frequencies. That doesnt mean you dont hear the whole spectrum, it just means the COMPRESSOR "hears" mostly high end and reacts accordingly.
If you thought there was too much "honk" to a vocal, way below de-esser range, you could insert an EQ into the sidechain and go after the 800hz in a dynamic way, rather than an EQ notch so the voice continues to sound lifelike, and breathe rather than sound EQ'd to death....
I am not explaining this well... Hopefully someone will post about this that is awake...
- snuffinthepunk
- pushin' record
- Posts: 220
- Joined: Mon Dec 13, 2004 5:59 pm
- Location: Nashville, TN/Destin, FL
- Contact:
Re: I hate to admit it: I don't understand sidechains.
that's basically the gist of it man...it's just like a key input on a gate, but instead of gating, you're compressing. de-essing is the first thing that came to my mind...I suppose you could use it to compress a screaming guitar when you want to have some vocals come in and not get drowned out. run the guitar through the compressor and a copy of the vocal into the side chain. the compressor will then act on the guitar (with whatever parameters you set) when the vocal crosses the threshhold. maybe useful in some kind of metal track. I'm not real good at making it work like that yet, but, I do know that's how it works. instead of the detector circuit monitoring the signal running through the compressor, it'll monitor the signal you tell it to.
"no dream is worth being underachieved"
I love signal flow.
Imagine the possibilities!
www.primalgear.com
I love signal flow.
Imagine the possibilities!
www.primalgear.com
-
- zen recordist
- Posts: 8876
- Joined: Mon May 19, 2003 12:10 pm
- Location: NYC/Brooklyn
- Contact:
Re: I hate to admit it: I don't understand sidechains.
Oh yeah... For ducking as well as described above, like a kick going to a bass compressor (one of them) so it just nudges out of the way and both sound huge...
Not needed that much in my experience, but it can help in a really dense mix. There are a zillion creative ways to duck.
Not needed that much in my experience, but it can help in a really dense mix. There are a zillion creative ways to duck.
- joelpatterson
- carpal tunnel
- Posts: 1732
- Joined: Tue Jun 10, 2003 5:20 pm
- Location: Albany, New York
Re: I hate to admit it: I don't understand sidechains.
I'll do some very simple side-chain compressing to keep the mix in the very precious "maximum loudness" zone without going over the top, f'rinstance:
I use a send from the Vocal Aux as the input for a compressor on the Instrument Aux--in other words, if there's gonna be a showdown between vocals and instruments, I let the vocals win by letting them triumph and letting them nudge the instruments down, slightly.
I use a send from the Snare Aux as the input for a compressor on the Drum Overhead Aux, so that snazzy snare hit can rise and shine and dominate the whole kit, just for an instant.
Sometimes it's ridiculous, everything is cantilevered around everything else, it's like that old board game Mousetrap, anyone ever have that?
So it's a way to automatically keep a nice healthy level even though on their own everything together would tend to overload.
I use a send from the Vocal Aux as the input for a compressor on the Instrument Aux--in other words, if there's gonna be a showdown between vocals and instruments, I let the vocals win by letting them triumph and letting them nudge the instruments down, slightly.
I use a send from the Snare Aux as the input for a compressor on the Drum Overhead Aux, so that snazzy snare hit can rise and shine and dominate the whole kit, just for an instant.
Sometimes it's ridiculous, everything is cantilevered around everything else, it's like that old board game Mousetrap, anyone ever have that?
So it's a way to automatically keep a nice healthy level even though on their own everything together would tend to overload.
-
- zen recordist
- Posts: 10890
- Joined: Mon Jun 16, 2003 1:26 am
- Location: Charlotte, NC
- Contact:
Re: I hate to admit it: I don't understand sidechains.
Here's another practical example of useful sidehain compression:
Say you have a bass player with a bass that for some reason is just loaded with a loud low "A." For some reason, whenever the guy plays an "A," it's WAY louder than every other note. In this case, you can send the compressor sidechain out to an equalizer and boost 110Hz, which is the fundamental frequency of A2, then send the EQed bass signal back into the compressor's sidechain input. If you listen to the equalized signal before going back into the compressor's sidechain input, you'd hear the "A" sticking out even more. What this does is essentially lower the compressor's threshold for all the 110Hz signal, or the "As" that appear. The bass is still pulling down the entire bass signal, but moreso whenever the "A" appears.
A cooler version of this is frequency-dependent compression from a narrow-band unit (which a de-esser ACTUALLY is) like the Urei LA-22 or the BSS DPR901. In these units, there is an internal sidechain filter (much like on the Drawmer or Aphex 600-series gates) which allows you to find the offending frequency. When this filter is activated and the compressor is in narrow band mode, the compressor only pulls down the frequency accentuated in the filter. In the case of the previous bass example, the compressor would only reduce 110Hz by a certain amount and leave the rest of the frequency spectrum untouched. The typical compressor's sidechain circuit doesn't allow for this sort of thing, but works as described above where you can sort of "fool it" into working in a similar manner.
Is that convoluted enough for you?
Chris Garges
Charlotte, NC
Say you have a bass player with a bass that for some reason is just loaded with a loud low "A." For some reason, whenever the guy plays an "A," it's WAY louder than every other note. In this case, you can send the compressor sidechain out to an equalizer and boost 110Hz, which is the fundamental frequency of A2, then send the EQed bass signal back into the compressor's sidechain input. If you listen to the equalized signal before going back into the compressor's sidechain input, you'd hear the "A" sticking out even more. What this does is essentially lower the compressor's threshold for all the 110Hz signal, or the "As" that appear. The bass is still pulling down the entire bass signal, but moreso whenever the "A" appears.
A cooler version of this is frequency-dependent compression from a narrow-band unit (which a de-esser ACTUALLY is) like the Urei LA-22 or the BSS DPR901. In these units, there is an internal sidechain filter (much like on the Drawmer or Aphex 600-series gates) which allows you to find the offending frequency. When this filter is activated and the compressor is in narrow band mode, the compressor only pulls down the frequency accentuated in the filter. In the case of the previous bass example, the compressor would only reduce 110Hz by a certain amount and leave the rest of the frequency spectrum untouched. The typical compressor's sidechain circuit doesn't allow for this sort of thing, but works as described above where you can sort of "fool it" into working in a similar manner.
Is that convoluted enough for you?
Chris Garges
Charlotte, NC
-
- steve albini likes it
- Posts: 361
- Joined: Sun Sep 26, 2004 8:57 pm
- Location: Portland, OR
Re: I hate to admit it: I don't understand sidechains.
Hmm...just thinking here. By no means a master of compression but...When you EQ the signal that is triggering the compressor's detector circuit, via the side chain, you are not making the circuit any more selective for one frequency or another, you are simply forcing the circuit to compress the material in response to the over amplified signal in the sidechain, which just so happens to be 1K or any other frequency content.?? If you boost the hell out of 1K in the sidechain signal, all frequencies above the threshold in the compression circuit get affected. So when the signal in the compression circuit is your sibilant vocal, and the sidechain signal is a copy of that vocal EQ'd to accentuate the sibilance, you are merely providing more level at that point in time to the detector circuit, so that the threshold can be exceeded and compression occurs. All content above that threshold is compressed regardless of frequency content, but with quick attack and release settings, the apparent effect is reduced sibilance.
-
- zen recordist
- Posts: 10890
- Joined: Mon Jun 16, 2003 1:26 am
- Location: Charlotte, NC
- Contact:
Re: I hate to admit it: I don't understand sidechains.
Yes. You are entirely correct.cetanorak wrote:Hmm...just thinking here. By no means a master of compression but...When you EQ the signal that is triggering the compressor's detector circuit, via the side chain, you are not making the circuit any more selective for one frequency or another, you are simply forcing the circuit to compress the material in response to the over amplified signal in the sidechain, which just so happens to be 1K or any other frequency content.?? If you boost the hell out of 1K in the sidechain signal, all frequencies above the threshold in the compression circuit get affected. So when the signal in the compression circuit is your sibilant vocal, and the sidechain signal is a copy of that vocal EQ'd to accentuate the sibilance, you are merely providing more level at that point in time to the detector circuit, so that the threshold can be exceeded and compression occurs. All content above that threshold is compressed regardless of frequency content, but with quick attack and release settings, the apparent effect is reduced sibilance.
Chris Garges
Charlotte, NC
-
- audio school graduate
- Posts: 22
- Joined: Thu Jun 12, 2003 11:32 pm
Re: I hate to admit it: I don't understand sidechains.
Interesting, and thanks for the responses. Concerning the de-essing and related methods: Now, so as I understand it, this is not compression of only certain frequency bands that we're talking about. This is compression of the entire signal when certain frequency bands pass the threshold. Right?
-
- steve albini likes it
- Posts: 361
- Joined: Sun Sep 26, 2004 8:57 pm
- Location: Portland, OR
Re: I hate to admit it: I don't understand sidechains.
Right. You are triggering the compressor circuit with levels and levels only. However, you can amplify specific portions of the frequency spectrum to pinpoint locations in the material to be compressed.
- snuffinthepunk
- pushin' record
- Posts: 220
- Joined: Mon Dec 13, 2004 5:59 pm
- Location: Nashville, TN/Destin, FL
- Contact:
Re: I hate to admit it: I don't understand sidechains.
yeah dude, technically it compresses the entire signal...but think about attack and release time. we're talkin milliseconds. if you're talking about de-essing, you set the attack real fast and the release real fast...and no one notices. the average person won't notice a difference in milliseconds.
"no dream is worth being underachieved"
I love signal flow.
Imagine the possibilities!
www.primalgear.com
I love signal flow.
Imagine the possibilities!
www.primalgear.com
-
- steve albini likes it
- Posts: 361
- Joined: Sun Sep 26, 2004 8:57 pm
- Location: Portland, OR
Re: I hate to admit it: I don't understand sidechains.
Hey, ScienceOne. It say audio school graduate under your name, as it does mine. Did you go to audio school, or did that kind of show up there. When I signed up for the board I do not remember if I said anything about going to audio school or not, and I cannot seem to find out where to change that. But, if you did go, they did not get into sidechain usage there??? Thanks
-
- buyin' a studio
- Posts: 877
- Joined: Mon Apr 19, 2004 5:30 pm
- Location: NYC
- Contact:
Re: I hate to admit it: I don't understand sidechains.
I had moustrap. Loved it!joelpatterson wrote:Sometimes it's ridiculous, everything is cantilevered around everything else, it's like that old board game Mousetrap, anyone ever have that?
Side chains + key inputs is one of the greatest mixing tricks I know and yet it's the one I forget about most. Thanks for the reminder!
FWIW, the term side chain is also used to describe a compressors method of detection i.e. optical, vari-mu, or VCA side chain.
- psychicoctopus
- buyin' a studio
- Posts: 890
- Joined: Wed May 14, 2003 3:01 am
- Location: Austin, TX
Re: I hate to admit it: I don't understand sidechains.
Any Portishead fans? Their drum sounds on the albums are unique. I could be wrong, but it seems like they're using sidechain compression on the drum reverb. The inrush of 'verb after the drum hits sounds like the drum reverb is run through a fast attack, slow release compressor keyed off of the drums. Anybody know the details?
Armed with seven rounds of space doo-doo pistols
-
- mixes from purgatory
- Posts: 2908
- Joined: Tue Mar 23, 2004 12:34 am
- Location: London, England
- Contact:
Re: I hate to admit it: I don't understand sidechains.
The audio school thing relates to how many posts you've made, when you start it says audio school, after you make a few post it becomes graduate, then it goes on, buying gear, setting up mics, recapping neve, banning rappers from your studio, having a row over payment, going bankrupt, and on and on...cetanorak wrote:Hey, ScienceOne. It say audio school graduate under your name, as it does mine. Did you go to audio school, or did that kind of show up there. When I signed up for the board I do not remember if I said anything about going to audio school or not, and I cannot seem to find out where to change that. But, if you did go, they did not get into sidechain usage there??? Thanks
Who is online
Users browsing this forum: No registered users and 27 guests