bouldersoundguy
Well-known member
That's not an unreasonable way to look at it, but that's just not how the term is used.
sidechain compression
- Thread starter banjo71
- Start date
That's not an unreasonable way to look at it, but that's just not how the term is used.
GuitarLegend
New member
I don't see where you got that. Maybe I should have been clearer but I didn't say that, nor is that what I believe. The input signal enters the VCA and exits the output in every case. That input signal also enters the detector except in sidechain mode in which case the detector gets its input from another source. Better?
JamEZmusic
Active member
Getting side-chain compression mixed up with (and/or) sharing terminology with Parallel compression if you are an older engineer is just plain dumb. I'd except that confusion from a complete n00b instead.
2 completely different mixing techniques entirely!
I'm out.
2 completely different mixing techniques entirely!
I'm out.
bouldersoundguy
Well-known member
They understood the techniques but used different terminology. You're the confused one.
jonny deep
Disappointing Member
Were you ever in? This isn't Dragons' Den.
Nah. It's just that for most uses of the term parallel path doesn't have to mean 'it's mixed back in somewhere. An aux send, that doesn't go back into the primary mix bus for example.
GuitarLegend
New member
I'm sorry, but it really DOES mean that the signal is mixed back in somewhere, otherwise it would not be parallel. It doesn't refer to the physical or diagrammatical layout, it refers to the fact that the signal passes through two or more processes and converge at the output. Any other arrangement is just a network. Parallel means parallel... it doesn't mean something else
Hmm. I would not have thought that saying for example an aux path in a mixer for example is a parallel path in the mixer. If goes to the main mix it is, but if it's used to feed something else it's not?
Well, I guess you did say that already.
GuitarLegend
New member
If you have a signal passing through a compressor to the master bus, and the same signal also passes through another compressor to the master bus, then the two signals are in parallel. Yes, the signal in the second path could be routed somewhere else that ends up at the master bus. In this case, of course it is in parallel with the first. There can be any number of serial processes in the second path, as long as you have one input to the two paths and one output where the paths converge. If you have two different outputs, then it is not considered a parallel process. Its just a fork
GuitarLegend
New member
Just to clarify, I am not talking about multiple inputs ending at a common output. I am talking about a single input passing through multiple processes simultaneously and converging at a single output. That is parallel processing. One signal split into however many paths, each with its own process and converging back into one path again.
thehitman2375
New member
Hi, I wrote this up to hopefully give people a better understanding of a valuable tool that often gets ignored; dynamic sidechains. There’s many great practical uses for them. I apologize if I am repeating information that has already been presented on this site, but I think this is something worth sharing to those who aren’t familiar with what they do, and what uses they can have.
Firstly, let me first try to explain what a side chain is. Basic dynamics processing is based around adjusting the amplitude (volume) of a sound in a certain manner (up or down) by a certain amount once the amplitude reaches a certain level (threshold). Most dynamics processors allow for an external signal to be used as the trigger (to reach threshold). So in other words, you can use any sound you want to trigger when the dynamics processor changes the level. Since it’s only for triggering, the processing is not done to the signal being sent to the side chain. In fact we don’t actually hear the signal going to the side chain at all. It’s the signal that is going into the main input of the *FX that is adjusted in volume. Simple analogy: the guy behind you in line is pushing you into me, but I only see you bumping into me when he pushes you.
(*Anywhere I use the term FX, I'm talking the reverb or delay)
So how is this useful? I’m glad you kept reading because it’s very useful in a lot of ways.
A. Ducking FX:
We don’t want our FX to clutter the mix. Delays can run over everything. Reverb can crowd a part and make it sound far away. But we like FX, and we want FX. We can always automate them to respond in just the parts we want them. But another easy solution is by ducking the FX. What this means is, when the sound that is being sent to the fx is actually happening, the fx are lowered in level. When the sound stops happening, the level returns to normal and therefore is heard in the spaces between the parts. This can prevent FX from obscuring the performance. You don’t want a sea of FX ruining your good performance, but you do want FX. assuming we are using our FX on an aux return, add a compressor AFTER the FX on the aux fader (not your vocal track). use a prefader send (set to unity is fine) to the key input of the compressor from the vocal track and a post fader send to feed the input of the FX to adjust the amount that’s on the vocal when it’s not happening. So the vocal is triggering the compressor to turn down (duck) the FX when it’s happening and letting the FX trail off naturally at the ends of words when there is space that the vocal isn’t happening. It’s important to use a prefader send going to the key input so that you can ride the main vocal fader without affecting the ducking of the FX. This isn’t as complicated as it sounds, and it’s worth trying. you’ll probably fall in love with it.
B. Tightening Parts Together:
Lets say you have 10 singers. You record them all on separate tracks. 3 of the singers didn’t rehearse enough and they had poor timing with the others. Sometimes they came in too early, sometimes they ended too late. These things can be adjusted manually by shifting parts, and that would be the best way to line them up, however, lets say the client doesn’t have the money or the time to sit there while you shift around the beginnings and ends of some of the vocals. Besides that, the song is 15 minutes long. Well, it’s not the ideal fix, but by putting an expander or gate on the suspect vocal tracks and using a (prefader) send from a vocal track that has all the timings right to trigger the gate on the external side chain can remove all the early entrances, and cut short late endings in real time. It’s a quick fix, maybe not the best fix, but you might not even be doing final vocals, you might be doing reference vocals and there’s no reason to spend so much time on something that is just going to get redone later. So this can be a helpful way to tighten things up.
C. Accenting The Groove:
Ahh yes, bass and drums. You always want them to be tight together. Maybe you want to emphasize their tightness to make the groove come out a little more. Well, by putting an expander on the bass and triggering it from a kick and snare via the external side chain can elevate the hits where they hit together. so even if the performance is spot on, you can enhance it even more.
D. Volume Level Priority Of A Sound In The Mix:
Ok, so you got a really big sound on the vocals in the chorus of your song. It’s stacked with 16 vocals (4 different notes x 4 tracks). Wonderful, but for the verses you only have a single lead or a lead and 1 or 2 backgrounds. The verses overlap with the chorus in the beginning and end of the verses and the huge sound of the chorus just overwhelms the verse vocal to the point you can barely hear the words. There’s a limited amount of headroom in the mix, and you are very happy with the overall level of both the chorus and verse, so you don’t want to turn down the chorus and turning up the verse in those parts might not sound consistent with the rest of the verse. Well, (Captain Hindsight..Errr, I mean) side chain to the rescue! Just output all the chorus vocals to a stereo aux fader and put a compressor on it. Then use a prefader send to send the verse vocal to the external key input of the side chain so that the compressor only affects the chorus vocals when the verse overlaps the chorus. This will allow the verse vocal to still be in front of the chorus vocal without turning up the verse and without having to turn down the chorus vocals during the overlap. Problem solved and nobody got hurt.
E. Auto Talkback:
This is my personal favorite. It’s so elegant and easy and slick. Are you sick of having to hit the talkback switch every time you want to talk to the artist in the booth? Let’s make it easy with side chains! Before you start recording, set up an aux fader in the session and input your control room talkback microphone to that fader. (Any microphone in the control room will do). Put a limiter on the talkback channel and pull the threshold all the way down so that it turns the sound down all the way (when it’s triggered). Now, you can use 1 of either of these things. You can use a prefader send from the tracks you already have or make a new track and put a small audio clip of a sine wave or noise (from the signal generator) that outputs ONLY to the key input of the limiter. If you use an audio clip, (which will work best) make sure the audio clip runs the entire length of the song. (Loop it) If you use the backing tracks, then the only thing that could be a problem is in spots where the background tracks aren’t happening. Once everything is set up the result being, whenever you press play, the limiter mutes the talkback. Whenever you are not playing, the talkback mic is wide open and you no longer have to push the talkback button. Cool, huh?
F. Cleaner Snare/Kick/Toms Gate Triggering:
When we use gates on drums, there’s sometimes just enough bleed from other drums to trigger the gate. This can be a real annoying problem. You can use the side chain path through filters or an EQ to get more accurate triggering, but let me also give you another way. Duplicate the snare track. Output the duplicate track directly to the side chain key input on the gate. Then proceed to cut apart the duplicate snare track to remove the noise. Strip Silence in Pro Tools is a quick way to do this. Now, normally that would sound awful because of the track stopping and starting abruptly, but we don’t actually hear that track. We hear the track that wasn’t cut up. The gate is triggered by the cut up track where you removed the bleed. Its also good to nudge the duplicate a millisecond earlier in the song so that it triggers the gate just before the snare transient. That way you don’t lose any of the attack on the snare when the gate opens. So, the bleed doesn’t trigger the gate and the gate can be set to open and close at a speed that feels natural.
G. Fattening Up A Weak Kick:
This is a weapon of last resort. The kick drum that you are working with doesn’t have any bottom to it. You can’t boost the bottom on something that doesn’t exist. So, we have to add low frequency to the kick. Here’s a method how you can do that. Create an aux track and put a signal generator or oscillator on the track. Set it to sine wave and choose a low frequency like 50-60 hz. The sine wave is continuous so you need it to only happen during the kick hits. Put a gate after the signal generator and use a prefader send from the kick track to trigger the opening and closing of the gate. You will have to adjust the attack and release controls to get it to blend in more naturally, but it is possible to get it right on point with the kick. So, now you have added low frequency that is “in” the kick.
See how useful dynamic side chains can be? There’s also so much more that can be done. I hope these ideas help people in developing their own uses and inspire creativity.
Thanks for reading. I’m happy to answer questions
Firstly, let me first try to explain what a side chain is. Basic dynamics processing is based around adjusting the amplitude (volume) of a sound in a certain manner (up or down) by a certain amount once the amplitude reaches a certain level (threshold). Most dynamics processors allow for an external signal to be used as the trigger (to reach threshold). So in other words, you can use any sound you want to trigger when the dynamics processor changes the level. Since it’s only for triggering, the processing is not done to the signal being sent to the side chain. In fact we don’t actually hear the signal going to the side chain at all. It’s the signal that is going into the main input of the *FX that is adjusted in volume. Simple analogy: the guy behind you in line is pushing you into me, but I only see you bumping into me when he pushes you.
(*Anywhere I use the term FX, I'm talking the reverb or delay)
So how is this useful? I’m glad you kept reading because it’s very useful in a lot of ways.
A. Ducking FX:
We don’t want our FX to clutter the mix. Delays can run over everything. Reverb can crowd a part and make it sound far away. But we like FX, and we want FX. We can always automate them to respond in just the parts we want them. But another easy solution is by ducking the FX. What this means is, when the sound that is being sent to the fx is actually happening, the fx are lowered in level. When the sound stops happening, the level returns to normal and therefore is heard in the spaces between the parts. This can prevent FX from obscuring the performance. You don’t want a sea of FX ruining your good performance, but you do want FX. assuming we are using our FX on an aux return, add a compressor AFTER the FX on the aux fader (not your vocal track). use a prefader send (set to unity is fine) to the key input of the compressor from the vocal track and a post fader send to feed the input of the FX to adjust the amount that’s on the vocal when it’s not happening. So the vocal is triggering the compressor to turn down (duck) the FX when it’s happening and letting the FX trail off naturally at the ends of words when there is space that the vocal isn’t happening. It’s important to use a prefader send going to the key input so that you can ride the main vocal fader without affecting the ducking of the FX. This isn’t as complicated as it sounds, and it’s worth trying. you’ll probably fall in love with it.
B. Tightening Parts Together:
Lets say you have 10 singers. You record them all on separate tracks. 3 of the singers didn’t rehearse enough and they had poor timing with the others. Sometimes they came in too early, sometimes they ended too late. These things can be adjusted manually by shifting parts, and that would be the best way to line them up, however, lets say the client doesn’t have the money or the time to sit there while you shift around the beginnings and ends of some of the vocals. Besides that, the song is 15 minutes long. Well, it’s not the ideal fix, but by putting an expander or gate on the suspect vocal tracks and using a (prefader) send from a vocal track that has all the timings right to trigger the gate on the external side chain can remove all the early entrances, and cut short late endings in real time. It’s a quick fix, maybe not the best fix, but you might not even be doing final vocals, you might be doing reference vocals and there’s no reason to spend so much time on something that is just going to get redone later. So this can be a helpful way to tighten things up.
C. Accenting The Groove:
Ahh yes, bass and drums. You always want them to be tight together. Maybe you want to emphasize their tightness to make the groove come out a little more. Well, by putting an expander on the bass and triggering it from a kick and snare via the external side chain can elevate the hits where they hit together. so even if the performance is spot on, you can enhance it even more.
D. Volume Level Priority Of A Sound In The Mix:
Ok, so you got a really big sound on the vocals in the chorus of your song. It’s stacked with 16 vocals (4 different notes x 4 tracks). Wonderful, but for the verses you only have a single lead or a lead and 1 or 2 backgrounds. The verses overlap with the chorus in the beginning and end of the verses and the huge sound of the chorus just overwhelms the verse vocal to the point you can barely hear the words. There’s a limited amount of headroom in the mix, and you are very happy with the overall level of both the chorus and verse, so you don’t want to turn down the chorus and turning up the verse in those parts might not sound consistent with the rest of the verse. Well, (Captain Hindsight..Errr, I mean) side chain to the rescue! Just output all the chorus vocals to a stereo aux fader and put a compressor on it. Then use a prefader send to send the verse vocal to the external key input of the side chain so that the compressor only affects the chorus vocals when the verse overlaps the chorus. This will allow the verse vocal to still be in front of the chorus vocal without turning up the verse and without having to turn down the chorus vocals during the overlap. Problem solved and nobody got hurt.
E. Auto Talkback:
This is my personal favorite. It’s so elegant and easy and slick. Are you sick of having to hit the talkback switch every time you want to talk to the artist in the booth? Let’s make it easy with side chains! Before you start recording, set up an aux fader in the session and input your control room talkback microphone to that fader. (Any microphone in the control room will do). Put a limiter on the talkback channel and pull the threshold all the way down so that it turns the sound down all the way (when it’s triggered). Now, you can use 1 of either of these things. You can use a prefader send from the tracks you already have or make a new track and put a small audio clip of a sine wave or noise (from the signal generator) that outputs ONLY to the key input of the limiter. If you use an audio clip, (which will work best) make sure the audio clip runs the entire length of the song. (Loop it) If you use the backing tracks, then the only thing that could be a problem is in spots where the background tracks aren’t happening. Once everything is set up the result being, whenever you press play, the limiter mutes the talkback. Whenever you are not playing, the talkback mic is wide open and you no longer have to push the talkback button. Cool, huh?
F. Cleaner Snare/Kick/Toms Gate Triggering:
When we use gates on drums, there’s sometimes just enough bleed from other drums to trigger the gate. This can be a real annoying problem. You can use the side chain path through filters or an EQ to get more accurate triggering, but let me also give you another way. Duplicate the snare track. Output the duplicate track directly to the side chain key input on the gate. Then proceed to cut apart the duplicate snare track to remove the noise. Strip Silence in Pro Tools is a quick way to do this. Now, normally that would sound awful because of the track stopping and starting abruptly, but we don’t actually hear that track. We hear the track that wasn’t cut up. The gate is triggered by the cut up track where you removed the bleed. Its also good to nudge the duplicate a millisecond earlier in the song so that it triggers the gate just before the snare transient. That way you don’t lose any of the attack on the snare when the gate opens. So, the bleed doesn’t trigger the gate and the gate can be set to open and close at a speed that feels natural.
G. Fattening Up A Weak Kick:
This is a weapon of last resort. The kick drum that you are working with doesn’t have any bottom to it. You can’t boost the bottom on something that doesn’t exist. So, we have to add low frequency to the kick. Here’s a method how you can do that. Create an aux track and put a signal generator or oscillator on the track. Set it to sine wave and choose a low frequency like 50-60 hz. The sine wave is continuous so you need it to only happen during the kick hits. Put a gate after the signal generator and use a prefader send from the kick track to trigger the opening and closing of the gate. You will have to adjust the attack and release controls to get it to blend in more naturally, but it is possible to get it right on point with the kick. So, now you have added low frequency that is “in” the kick.
See how useful dynamic side chains can be? There’s also so much more that can be done. I hope these ideas help people in developing their own uses and inspire creativity.
Thanks for reading. I’m happy to answer questions
ashcat_lt
Well-known member
Honestly, that is all really good stuff and great examples of ways that sidechains on dynamic processors can be used. I guess I was hoping for a bit more creativity form the way it started out. You said "dynamic side chains", which read to me like "dynamic processing on the side chain signal". You also talked about how you can do whatever you want to the sidechain signal, but then didn't really do anything to it. Yes, you used one source to control the dynamics of another, but you didn't do anything to that source on its way to the detector.
So, I've never actually tried it, but it might be worth a try to actually put a dynamic processor on the sidechain signal itself, before the detector.
Say you want to duck one thing with another, but that other thing swings a lot further than you want the ducking to go. Compress (or limit) it on the way to the detector so that the gain reduction only ever goes so far.
If you split the signal, run one copy through a compressor/limiter and then to the detector, and the other through the compressor's audio input, you can emulate that thing that some vintage circuits do where they "open back up" and start to apply less gain reduction than expected as the signal goes further and further above the threshold on a more standard comp and with more precise control over how it does that.
Likewise, if you put an expander in the detector path, it could emulate a soft-knee response in a hard-knee compressor.
OR Put other things in that detector path.
Filters are pretty common, and many comps have them built in nowadays. Be careful with that, though. A filter in the side side-chain is not the same thing as a multi-band compressor. An MBC splits the signal into frequency bands and applies discrete compression to each band based on its own content. A side chain filter causes the compressor to compress the entire frequency spectrum based on the content of the filtered signal. That is, if you run the HPF up so that it's only responding to high frequencies (typical for de-essing), it will still turn down the low-end too when the high end gets loud. Likewise, if you use an LPF it will still turn down the highs when the lows get loud. (Course, that's almost redundant with the attack/release/RMS time "filters".)
I have done things in the past where I wanted to key an expander on one track from another, but I wanted it to kind of "smear", and there wasn't attack/release controls on the expander, so I put a reverb in the detector path. Worked out kind of cool.
A delay effect might create a sort of "bouncing gate" effect that could be cool.
I'm not sure how much use most modulation effects would be, but I guess a relatively resonant flanger or phase shifter could do...something...
Ooo... A very fast delay makes a strange sort of resonator which could be used to either accentuate or attenuate the signal when a given note comes into the detector.
Things can get pretty whacky fast, but it's all fun to play with.
Just one more quick thing - if you set all of the time constants (including RMS window) to 0, a compressor turns into a distortion. If you then run a second input into the detector path, it works almost like a ring modulator.
thehitman2375
New member
i listed things that i felt would be very practical. things that might come up on a regular basis. understanding how it works is only the first step, knowing when to use it is the next step, & being creative is just taking it further by finding your own unique uses.
i would encourage people to try any of the ideas you mentioned. i probably wouldn't see any use for putting compression on the sidechain signal because you wouldn't actually hear the sound. if you're suggesting using compression on the side chain signal to alter the dynamic range of the triggering signal, i'm not exactly sure what that what advantage that would have since altering the parameters on the dynamics processor does exactly the same thing. you are suggesting putting a compressor on a side chain to change what amount of dynamic range (width) that would affect the signal. that's the same thing the threshold and ratio do. if it's still too hot even after having the thresh all the way up, then just turn down the send and you will have more headroom.
also in example F, there could easily be processing on the signal being sent to the sidechain. the filters, eqs, and the other things you mentioned are all creative ideas worth experimenting with. my aim here was to help with understanding how it works (the basics) and what kind of practical applications it could have in an everyday mix. i guess i could rattle off some of the more obscure examples of things that people may never want or need to use, but what would be the point? in any of these examples i don't quite give specific settings, all i'm saying is: here's a way that this can be useful
if anyone who reads this decides to try any of theses ideas, i hope they do take the time to try different things. but sometimes just using the tool to do a specific job is enough work in itself. i wrote all of 1 page here, not a book. so i appreciate you adding some more thoughts to what i wrote because the bottom line is you are thinking about it and looking at it from the angles i did not. that's what my hope was from the start.
thank you for your reply.
i would encourage people to try any of the ideas you mentioned. i probably wouldn't see any use for putting compression on the sidechain signal because you wouldn't actually hear the sound. if you're suggesting using compression on the side chain signal to alter the dynamic range of the triggering signal, i'm not exactly sure what that what advantage that would have since altering the parameters on the dynamics processor does exactly the same thing. you are suggesting putting a compressor on a side chain to change what amount of dynamic range (width) that would affect the signal. that's the same thing the threshold and ratio do. if it's still too hot even after having the thresh all the way up, then just turn down the send and you will have more headroom.
also in example F, there could easily be processing on the signal being sent to the sidechain. the filters, eqs, and the other things you mentioned are all creative ideas worth experimenting with. my aim here was to help with understanding how it works (the basics) and what kind of practical applications it could have in an everyday mix. i guess i could rattle off some of the more obscure examples of things that people may never want or need to use, but what would be the point? in any of these examples i don't quite give specific settings, all i'm saying is: here's a way that this can be useful
if anyone who reads this decides to try any of theses ideas, i hope they do take the time to try different things. but sometimes just using the tool to do a specific job is enough work in itself. i wrote all of 1 page here, not a book. so i appreciate you adding some more thoughts to what i wrote because the bottom line is you are thinking about it and looking at it from the angles i did not. that's what my hope was from the start.
thank you for your reply.
ashcat_lt
Well-known member
I didn't really mean to criticize. Just that I was kind of excited to hear some more "out of the box" kind of thinking. That's probably my problem.
Take it to an extreme and lets say we put a limiter on the (split signal going to the) detector. Up to the threshold of that limiter, the compressor will act just as though you had just run the signal in with no side-chain at all. Once that limiter clamps down, though, we will not get any more gain reduction at all no matter what the signal does. Comp threshold = -18, ratio = 2:1, limiter threshold = -12. If the input hits -16 the output will hit -17 (1db attenuation), at -14 output is -16 (2db attenuation), at -12 output is -15 (3db attenuation). Beyond -12, though, the detector stops moving, and thinks that it is just sitting at -12, and continues to only apply 3db attenuation. If the input hits 0db, a ratio of 2:1 would leave the output at -9db (9db attenuation), but it will actually hit -3db. The compressor "opens back up" with higher inputs like a few very popular hardware units do. Use a compressor instead of a limiter, and it's more of a curvy thing, but works about the same. This can not be done using any of the typical controls on a compressor. If you dig that, then I think you can see how it might apply in a ducking situation such that the "ducked" music only gets so much quieter and no more, no matter how much louder the "ducking" signal gets.
Likewise, if you expand the detector input, the compressor will see a lower level than is actually there up to the point that the expander's threshold is exceeded. This will mean there is less compression than the ratio would suggest for quieter signals, and will act very much like a soft-knee control. This is somewhat redundant with your compressor's knee control. What? Your compressor doesn't have a variable knee control? Mine (ReaComp) does, but this still gives you more control.
Now start playing with the attack/release on that comp or expander in the side-chain and you end up with even more things you could never do with one compressor alone.
I guess I don't know how practical any of these things are, but...
Not as I see it.
Take it to an extreme and lets say we put a limiter on the (split signal going to the) detector. Up to the threshold of that limiter, the compressor will act just as though you had just run the signal in with no side-chain at all. Once that limiter clamps down, though, we will not get any more gain reduction at all no matter what the signal does. Comp threshold = -18, ratio = 2:1, limiter threshold = -12. If the input hits -16 the output will hit -17 (1db attenuation), at -14 output is -16 (2db attenuation), at -12 output is -15 (3db attenuation). Beyond -12, though, the detector stops moving, and thinks that it is just sitting at -12, and continues to only apply 3db attenuation. If the input hits 0db, a ratio of 2:1 would leave the output at -9db (9db attenuation), but it will actually hit -3db. The compressor "opens back up" with higher inputs like a few very popular hardware units do. Use a compressor instead of a limiter, and it's more of a curvy thing, but works about the same. This can not be done using any of the typical controls on a compressor. If you dig that, then I think you can see how it might apply in a ducking situation such that the "ducked" music only gets so much quieter and no more, no matter how much louder the "ducking" signal gets.
Likewise, if you expand the detector input, the compressor will see a lower level than is actually there up to the point that the expander's threshold is exceeded. This will mean there is less compression than the ratio would suggest for quieter signals, and will act very much like a soft-knee control. This is somewhat redundant with your compressor's knee control. What? Your compressor doesn't have a variable knee control? Mine (ReaComp) does, but this still gives you more control.
Now start playing with the attack/release on that comp or expander in the side-chain and you end up with even more things you could never do with one compressor alone.
I guess I don't know how practical any of these things are, but...
thehitman2375
New member
i see u are giving me settings and what they would do, but that's a theoretical exercise. if that's the case, there's an plenty of theoretical things that can be done. all the settings in the world are not telling me why you would ever do it. i gave very specific examples of where the ideas could be applied.
16, 8998, 78214778, 49949
those are numbers, i not gonna say what they are gonna be used for or when or why, but they are indeed numbers and of those numbers some are higher and some are lower. they exist and can be counted. that's kinda what you are pointing out to me. values are nice, but musical purpose is the whole point. if you aren't telling me why you are doing all the things you are doing, and what they are used for, then it's really not gonna help me to understand why u would want to do that. i get that you CAN do it.
i like that you are thinking outside the box. give me an application for it where it will sound good and then the numbers will mean something. the only thing i take from what u wrote is, i can add this and subtract that and here's the outcome. i won't say what you wrote is false, nor would i criticize the idea as not worth trying, it's just try it on what and when? u give me a what and when and then everyone who reads what you wrote will say: "hey that's a clever idea" including me. until then, all i can say is, it's a good theoretical exercise.
i wrote what i wrote with the hope that what i wrote will help some people achieve some good real world results. it's my hope that by not giving specific values, that people will tinker for themselves and find the settings that fit their specific situation. every situation is different. so by giving settings, it may not even be in the ballpark. i record everything very hot. not clipped, but as close to 0 as possible. not everyone does that and my thresholds will just be different than theirs. this sidechain thing can be a great tool, but it's still a tool for the music. let's not overcomplicate this by giving all the theoretical things we can do. that will confuse those who are just learning it for the first time. i prefer simple and practical. where you and other's take these ideas is entirely what you make of them and that's a great thing.
16, 8998, 78214778, 49949
those are numbers, i not gonna say what they are gonna be used for or when or why, but they are indeed numbers and of those numbers some are higher and some are lower. they exist and can be counted. that's kinda what you are pointing out to me. values are nice, but musical purpose is the whole point. if you aren't telling me why you are doing all the things you are doing, and what they are used for, then it's really not gonna help me to understand why u would want to do that. i get that you CAN do it.
i like that you are thinking outside the box. give me an application for it where it will sound good and then the numbers will mean something. the only thing i take from what u wrote is, i can add this and subtract that and here's the outcome. i won't say what you wrote is false, nor would i criticize the idea as not worth trying, it's just try it on what and when? u give me a what and when and then everyone who reads what you wrote will say: "hey that's a clever idea" including me. until then, all i can say is, it's a good theoretical exercise.
i wrote what i wrote with the hope that what i wrote will help some people achieve some good real world results. it's my hope that by not giving specific values, that people will tinker for themselves and find the settings that fit their specific situation. every situation is different. so by giving settings, it may not even be in the ballpark. i record everything very hot. not clipped, but as close to 0 as possible. not everyone does that and my thresholds will just be different than theirs. this sidechain thing can be a great tool, but it's still a tool for the music. let's not overcomplicate this by giving all the theoretical things we can do. that will confuse those who are just learning it for the first time. i prefer simple and practical. where you and other's take these ideas is entirely what you make of them and that's a great thing.
ashcat_lt
Well-known member
I did.
1) You want the compressor to open back up because you like the way those old compressors do it, but don't have one, and/or want more control of how it happens
B) You want one thing to duck another, but you only want that first thing to get so quiet no matter how loud the other thing gets
III) You want to fake a soft-knee compressor when you don't have one or want more control over the knee characteristics
My other post also included the thing about using reverb to get an attack/release characteristic on an expander that doesn't have knobs for it.
These are all pretty practical uses. Maybe they're not super typical, but they could all be really useful in real world situations.
1) You want the compressor to open back up because you like the way those old compressors do it, but don't have one, and/or want more control of how it happens
B) You want one thing to duck another, but you only want that first thing to get so quiet no matter how loud the other thing gets
III) You want to fake a soft-knee compressor when you don't have one or want more control over the knee characteristics
My other post also included the thing about using reverb to get an attack/release characteristic on an expander that doesn't have knobs for it.
These are all pretty practical uses. Maybe they're not super typical, but they could all be really useful in real world situations.
BroKen_H
Re-member
I just experimented until I figured out how to "de-ess" one of my songs. It was very enlightening. Finding the frequency of the SSS on a parallel channel, high and low passing everything else, and feeding that signal to the bus compressor to compress the S! I know there's probably lots of vids out there, etc. But I just like to experiment and find out if I can figure this stuff out on my own.
I should try some experiments with CVs and manipulating them as a side chain...Be interesting to see what kinds of results that would make (or if it's even possible).
Yeah, I know de-essing is not the most innovative thing you can do with a side chain, but it was fun to learn how.
I really liked the idea of kicking reverb/echo down or completely out of the vocal until the vocal stops. I am going to have to try that.
My last experiment was doing a "preverb" by using the parallel vocal back 300 ms and running a gate reverb that allowed the 100% wet signal to hit 300ms before the original note.
A lot of what I do is based on parallel channels, but I'm starting to wonder where side chaining can lead...
I should try some experiments with CVs and manipulating them as a side chain...Be interesting to see what kinds of results that would make (or if it's even possible).
Yeah, I know de-essing is not the most innovative thing you can do with a side chain, but it was fun to learn how.
I really liked the idea of kicking reverb/echo down or completely out of the vocal until the vocal stops. I am going to have to try that.
My last experiment was doing a "preverb" by using the parallel vocal back 300 ms and running a gate reverb that allowed the 100% wet signal to hit 300ms before the original note.
A lot of what I do is based on parallel channels, but I'm starting to wonder where side chaining can lead...
thehitman2375
New member
yeah.. maybe i'm not seeing something here, because i still don't see where you mentioned what instrument/voice/other you are affecting with this idea and how it would benefit the music in the end result. also, i don't see any reasoning other than "because u like it better". i guess that's good sound reasoning for what you want to do, even though nobody but you can figure out what that might be.
i'm sorry for giving you a hard time. the point you make about liking it, is all that matters. i definitely like that you want to get the best out of what you are doing. that's a characteristic we both share. it may not be easy to understand what application u have in mind, but it is a good trick for achieving the particular type of balance you are seeking. at the end of the day, it's what is gonna make you happy that matters most. it's not really up to me to decide that for others and this sounds like something you've given thought too. there's good value in that.
i have to admit though, if you did give me the details of what this would/could be used for in a practical application, it would benefit me because i will have learned something that i never even thought of doing.
as for having/not having a particular dynamics processor, i guess i'm blessed that i have all the tools i would want to use. so, for others that don't, some of these ideas might get you there, so experiment away.
Last edited:
Farview
Well-known member
I don't want to speak for him, but the uses are pretty pedestrian...
1) This would emulate an old opto-compressor. It's great for vocals, bass, or anything that needs to be compressed pretty hard without stepping on the transients too hard.
B) This is like the ducker on the radio. When you talk, the background music gets quiet. But you don't actually want the music to pump along to how loud the person is talking. You want it to go down to a certain level and stay there.
III) This one is kind of the opposite of #1. This gives the effect of starting off compressing at a lower ratio on lower signals and ramping up to a higher ratio as the signal gets hotter. This makes the compression seem more transparent.
1) This would emulate an old opto-compressor. It's great for vocals, bass, or anything that needs to be compressed pretty hard without stepping on the transients too hard.
B) This is like the ducker on the radio. When you talk, the background music gets quiet. But you don't actually want the music to pump along to how loud the person is talking. You want it to go down to a certain level and stay there.
III) This one is kind of the opposite of #1. This gives the effect of starting off compressing at a lower ratio on lower signals and ramping up to a higher ratio as the signal gets hotter. This makes the compression seem more transparent.