EQ pre/post compression

  • Thread starter Thread starter 5c0tt
  • Start date Start date
5

5c0tt

New member
When tracking vocals, how does the eq position in the signal chain affect the sound?

I guess any eq boost or cut before the comp would make the comp react differently but I'd like to know more about this & how it affects the sound.

Anyone?

S
 
Try using moving the mic first to find the best sound, then record the vocals, maybe with some compression, but only a little to control small dynamic variations. then when you mixdown after everything else is recorded you might want to eq to help glue the vocals in the mix where it fits best. Eq before the compressor reduced the effectiveness of the eq. Eq post compressor removes some of the dynamics and also changes the effectiveness of the eq. Moving the mic around and modifying the singers technique will prove to be the most effective and dramatic changes. Try both methods to see which one you think sounds best to you, then youll really know.

SoMm
 
theres a third option here. EQ'ing the sidechain of the compressor
 
pipelineaudio said:
theres a third option here. EQ'ing the sidechain of the compressor
Click to expand...
Oh gee! But I already got a compressor side chained into the pre! Then What?

Peace,
SoMm
 
pipeline, i don't understand the sidechain comment.

how does that work?

if you want to EQ some frequencies out, how can that be accomplished with sidechain?

the compressor is still going to compress the whole sound, not just certain frequencies like a multiband.

i understand that by using the sidechain you'll only compress based on certain frequencies, but how does that change the frequencies represented in the sound?
 
If you enhance a certain frequency in the sidechain, everything will be compressed harder, if this freq appears in the signal. This leads to effect that every sound that contains this frequency will be less loud. On a voice that (most often) has a rather 'single instrument' character (unless the singer is using VERY different registers...), you'll be able to reduce sibilance by pushing it in the sidechain or to even out lower frequency and higher frequency parts if needed...

You'll not get the 'dynamic eq' features of a multiband though...

But I'm quite sure you already know...

aXel
 
So, have I got this right?
Say I were to boost some siblance freqs BEFORE the compressor, this would cause the compressor to act faster on the siblance and the result would be LESS siblance going to tape (or disc)?
Yes? No? What's the deal guys?
Thanks
 
5c0tt said:
So, have I got this right?
Say I were to boost some siblance freqs BEFORE the compressor, this would cause the compressor to act faster on the siblance and the result would be LESS siblance going to tape (or disc)?
Yes? No? What's the deal guys?
Thanks
Click to expand...

It's probably better to cut the lower freq on the eq instead of boosting the sibilant freq of the eq going into the sidechain. You leave whatever freq you want to compress at 0dB and cut the rest.
 
5c0tt said:
So, have I got this right?
Say I were to boost some siblance freqs BEFORE the compressor, this would cause the compressor to act faster on the siblance and the result would be LESS siblance going to tape (or disc)?
Yes? No? What's the deal guys?
Thanks
Click to expand...
Yes, exactly.:D As long as the source is single instrument, the side chain eq can cause some frequencies to get 'ducked' (or others to not pull down with cut-eq) while they are most prominant. If you happen to have 'sis' and 'boom' at the same time, they both get hit.
An example of cut-eq could be a single band compressor on a mix. You can cut in the low band if it it catching to much on the low. This is also one of the reasons one comp can sould different than another -they have different eq voicings in their detection circuits. Your just getting in and changing it.
Fun stuff!

And, if I may make the connection:rolleyes: , in the thread about 'comps and dullnes', the subject is how the 'Time' spent while your compresson does it's thing, has an 'Eq' effect.

Don't you love it?:)
 
Yeah, better to cut than boost, I see.

So if I leave the siblance freqs up there, by slightly cutting the rest, then the troublesome freqs will hit the comp threshold first and bring down the comparative level of the siblance versus the rest of the signal?

I think I'm with it now....thanks guys.

S
 
Also, there's a 'gotcha'.. If the compressor driving like mad taming wild jumps in volume, tricks like side-chain de-essing might not work so well. It's almost like two seperate setups and tasks for it to handle.
Yea?...
Wayne
 
Cool. Thanks mixsit.

I can now experiment and know exactly what is happening & why.

There's always a 'gotcha' eh!

But that's what makes it fun.
 
I'm not perfecly sure, but I would swear that the difference between enhancing frequencies (additive EQ) or cutting (subtractive) in a compressor sidechain is not really present. While in a 'real' signal path, an additive EQ might spoil everything due to the phase shifting, it should not matter in the sidechain. IMO, it is much easier to set up, but YMMV... In contrast, I assume that you might even get some positive effects if you push the EQ so that it gets into a distortion. The compression will not be any stronger if distortion occurs in the sidechain. This would lead to a rise of the overall transfer function of the compressor, which in fact would give you a transfer function very similar to the one an optical compressor gives you... (But in this case only for the sibilance...)

aXel
 
There's nothing saying you can't use cut and boost to get double the strength -15 on one band, +15 on the another.
There is that headroom limit though. How would you know when you've run into it?
Wayne
 
So if I leave the siblance freqs up there, by slightly cutting the rest, then the troublesome freqs will hit the comp threshold first and bring down the comparative level of the siblance versus the rest of the signal?
Click to expand...

comparative meaning that the whole of the sound at that moment when the sibilance goes over the threshold will be compressed... not just the sibilance.

the only way to get rid of the area of sibilance without changing the other frequencies (thereby changing the balance of the whole of the sound) is to use a multiband compressor.
 
Son of Mixerman said:
Try using moving the mic first to find the best sound, then record the vocals, maybe with some compression, but only a little to control small dynamic variations. then when you mixdown after everything else is recorded you might want to eq to help glue the vocals in the mix where it fits best. Eq before the compressor reduced the effectiveness of the eq. Eq post compressor removes some of the dynamics and also changes the effectiveness of the eq. Moving the mic around and modifying the singers technique will prove to be the most effective and dramatic changes. Try both methods to see which one you think sounds best to you, then youll really know.

SoMm
Click to expand...

Hi,

I though I'd jump in with a question: You in essence say to EQ after doing the recording to "help glue the vocals in the mix" BUT what about if I can't hear myself over the backing track ? I'm perhaps a rare case here that doesn't play an instrument and sings to pre-recorded tracks. How do you resolve the problem of only being able to hear yourself while singing loud passages but being drowned out when having to sing softer in the same song ? I have a VERY dynamic voice that goes from soft to very loud. I clip the meters but then when I sing softer I can't hear myself over the backing track. (Yes, I go away from the mic and closer depending on how "hard" I'm singing). I know I should perhaps use a compressor (RNC) but what about EQ ?? I have onboard mid sweep on my 424mkIII TASCAM and tend to reduce mid freq of the backing tracks while increasing mids on my voice but then it all, especially voice, sounds unnatural. I'm stumped for sure.

Thanks in advance for your help,

Daniel

PS Gear I presently have: SHURE BETA 58A, TASCAM 424mkIII, LEXICON MPX110
 
Its especially difficult to do with only 4 tracks. Typically on the vocals are done where the loud parts are recorded on seperate tracks and soft parts are done on separate tracks, then after every thing is recorded you ride the faders during mixdown as far as controlling changes in volume. The only think you can do is practice your singing more. You could use a compressor to squish the crap out of your loud vocals.
If the music your doing is like Papa Roach and Disturbed, your going to have to practice how to sound like your screaming when your really not that much louder. Pull the mic ultra close when doing the soft parts and pull it back to 1/2 of your arms reach (10 to 15 inches) for the loud parts. Id put the compression at 4:1 with fast attack and medium slow release for starters. If you must use eq, try to figure out what your voices center frequency is, see where it fits relative to the center frequency to the backing tracks, sometimes who have to pitch the song differently to find the niche.

Try printing this list out...Its from a Mastering Engineer who tried to help clarify eq starting points for most music.
Its just guide though.

400Hz
1. Increase to add clarity to bass lines especially when speakers are at low volume.
2. Reduce to decrease "cardboard" sound of lower drums (foot and toms).
3. Reduce to decrease ambiance on cymbals.

800Hz
1. Increase for clarity and "punch" of bass.
2. Reduce to remove "cheap" sound of guitars.

1.5KHz
1. Increase for "clarity" and "pluck" of bass.
2. Reduce to remove dullness of guitars.

3KHz
1. Increase for more "pluck" of bass.
2. Increase for more attack of electric / acoustic guitar.
3. Increase for more attack on low piano parts.
4. Increase for more clarity / hardness on voice.
5. Reduce to increase breathy, soft sound on background vocals.
6. Reduce to disguise out-of-tune vocals / guitars.

5KHz
1. Increase for vocal presence.
2. Increase low frequency drum attack ( foot / toms).
3. Increase for more "finger sound" on bass.
4. Increase attack of piano, acoustic guitar and brightness on guitars (especially rock guitars).
5. Reduce to make background parts more distant.
6. Reduce to soften "thin" guitar.

7KHz
1. Increase to add attack on low frequency drums ( more metallic sound ).
2. Increase to add attack to percussion instruments.
3. Increase on dull singer.
4. Increase for more "finger sound" on acoustic bass.
5. Reduce to decrease "s" sound on singers.
6. Increase to add sharpness to synthesizers, rock guitars, acoustic guitar and piano.

10KHz
1. Increase to brighten vocals.
2. Increase for "light brightness" in acoustic guitar and piano.
3. Increase for hardness on cymbals.
4. Reduce to decrease "s" sound on singers.

15KHz
1. Increase to brighten vocals (breath sound).
2. Increase to brighten cymbals, string instruments and flutes.
3. Increase to make sampled synthesizer sound more real.

Peace,
SoMm
 
crosstudio said:

(snip)
the only way to get rid of the area of sibilance without changing the other frequencies (thereby changing the balance of the whole of the sound) is to use a multiband compressor.
Click to expand...

I wonder if there is any practical difference between this and "Dynamic EQ"? I've always wanted to try one (is it the BSS?) but never been willing to shell out the bucks. The concept seems very appealing.
Wayne
 
mixsit said:
There's nothing saying you can't use cut and boost to get double the strength -15 on one band, +15 on the another.
There is that headroom limit though. How would you know when you've run into it?
Wayne
Click to expand...

I thought of listening to the key out or so...

But I think it might really be some kind of feature if you loose the headroom in the sidechain. From a certain strength of sibilance in your sound, the sibilance won't be reduced further... This in fact would be kind of a dynamic EQ:

Sibilance bleow thresh : no reduction
Sibilance between thresh and headroom margin: reduction in gain
Sibilance above no further reduction.

The thing is: I don't know how the (most prolly used) algorithms would deal with the distortion, but it should work...

aXel
 
cjacek said:
Hi,

I though I'd jump in with a question: You in essence say to EQ after doing the recording to "help glue the vocals in the mix" BUT what about if I can't hear myself over the backing track ? I'm perhaps a rare case here that doesn't play an instrument and sings to pre-recorded tracks. How do you resolve the problem of only being able to hear yourself while singing loud passages but being drowned out when having to sing softer in the same song ? I have a VERY dynamic voice that goes from soft to very loud. I clip the meters but then when I sing softer I can't hear myself over the backing track.
Click to expand...

You can always compress the cue mix. (Insert the compressor into the signal coming from the tape/DAW on the way to the headphones.
 
You must log in or register to reply here.
Back
Top