Subtractive EQ (Especially Mids)

Recording Techniques, People Skills, Gear, Recording Spaces, Computers, and DIY

Moderators: drumsound, tomb

User avatar
iamthecosmos
pushin' record
Posts: 205
Joined: Sun Sep 21, 2008 4:20 am

Subtractive EQ (Especially Mids)

Post by iamthecosmos » Mon Mar 12, 2012 3:44 am

Okay, I think I get taking out lows and highs where needed, but I could do with understanding midrange a bit more. At the moment I'm just sweeping through until I find something that sounds right, but I'm struggling with finding space for multiple guitar overdubs. Especially ones with thick distortion.

Also, does everyone typically EQ individual drum mics (I usually don't aside from cutting lows), or do you tend to try and EQ the whole drum bus to avoid odd phase issues?

Any other subtractive EQ suggestions welcome, I've been avoiding too much EQ due to lack of confidence in what I'm doing. Keeping it simple!

kslight
mixes from purgatory
Posts: 2968
Joined: Tue Oct 13, 2009 7:40 pm

Post by kslight » Mon Mar 12, 2012 4:56 am

Well my general practice for finding offensive frequencies is to boost and sweep until I find the honk or whatever I don't like and cut it out.

When dealing with guitars it is beneficial to EQ while listening to the whole mix, not just the individual tracks...as if you're trying to make room in a dense mix your guitars may not sound stellar individually but will sound the sum of their parts...if that makes any sense. For example its probably important that the guitars "cut" through so I'd be looking at high-end to make sure there is definition there, but low mids may be covered by bass and drums so you don't usually need lots of bottom on the guitars. I also find it very important when dealing with multiple guitar parts for them to be very tight in timing to each other, otherwise its going to sound like mud.

I normally EQ individual drum mics at some stage or another, either while recording or mixing. I do a lot of high pass filtering, but also on bass drum for example there is a bit of boxiness usually around 220-260hz that I'll suck out almost automatically, and toms have a similar type of sound at different frequencies, snare may have a ringing to it...

User avatar
farview
tinnitus
Posts: 1204
Joined: Tue Aug 31, 2004 1:42 pm
Location: St. Charles (chicago) IL
Contact:

Post by farview » Mon Mar 12, 2012 7:01 am

I do about the same thing as kslight. However, the sound you need from the instruments will depend a lot on the style of music and the sounds you start out with.

With layers of distorted guitars, I tend to send them all to a group buss and EQ the group buss. I will tend to record the guitars a little dark because I've found that you can add a bunch of high shelf to a dark guitar and get a really smooth brightness. If the guitar is bright to begin with, the speakers will (most of the time) sound pinched and annoying. This is why some people use ribbon mics on heavy guitars, it's an attempt to soften the effect of really bright sounds coming through a 4x12.

With drums, I EQ and compress all the individual drums and use the drum buss for overall compression. I will actually add low end (50hz for kick, higher for the toms) and suck out around 800-900hz on the kick and toms.

With snare, the weight of it will be somewhere around 200hz, the 'crack' will be around 2k and the sizzle will be 6k-8k or above.


for the most part, there are only a few steps to EQing anything:

1. imagine the sound you want out of the instrument.
2. listen to the sound you have recorded
3. figure out the difference between what you have and what you want
4. Set the EQ for that difference

kayagum
ghost haunting audio students
Posts: 3490
Joined: Wed May 07, 2003 11:11 pm
Location: Saint Paul, MN

Post by kayagum » Mon Mar 12, 2012 7:53 am

If you're looking for room in a crowded mix, slightly panning the different tracks in different positions can make a huge difference. Even between 11:30 and 12:30 is plenty.
Last edited by kayagum on Sat Mar 17, 2012 5:05 am, edited 1 time in total.

GooberNumber9
tinnitus
Posts: 1094
Joined: Fri Oct 20, 2006 7:52 am
Location: Washington, DC

Post by GooberNumber9 » Mon Mar 12, 2012 12:36 pm

On the thickest mixes I've done I was able to squeeze things in by almost planning where they would go in the frequency spectrum. So I EQ'ed drums first and sorta picked one important freqency for the kick and snare based on their recorded sound, and then gave them a boost there. Then I did a lot more cutting of each drum track and the whole drum bus to make some room. Mainly I put in high passes as high as possible on everything but the kick, and did some low passing on everything but overheads.

Then I did the same with the bass: I gave it as small a box to live in as possible without making a big impact on the sound, and then I gave a boost to one or two frequencies that it "liked".

Then on to guitars, with the same idea, but each guitar would get its own boost frequency (we're just talking a dB or three) and those would be different from the boosts used on snare, kick, and bass. When things got too crowded, I would look to making some cuts to the bass or drum bus at the same points where I wanted to boost a guitar, or at least where I felt a guitar track had a natural peak. When you have 4 or 6 or 8 very distorted guitar tracks, you can't be sentimental about them. Feel free to cut them up with the EQ and be brutal. This is a good case where listening to the whole mix while you're cutting can help you hear how the different parts interact.

Of course the vocals were top priority, so I bussed all the guitars together and made some cuts in all the busses where there was either a peak in vocals or I wanted to emphasize the vocals.

As others have said, it's important to let go of trying to have individual tracks sound good on their own, and instead try to make them work together - especially for the guitars, since they have a more continuous sound. You can get away with the snare, kick, or toms taking up a lot of space if the drum part is sparse enough. If you're like me, you'll also want the vocals to sound as good as possible and never get stepped on by anything else. Between those two, the guitars will sometimes lose out, but with a wall of guitars going on, no one will miss a guitar part getting pushed back by a snare hit or a vocal line.

For my style, I always carved up the frequencies before I did any panning, because that meant the panning was like icing on the cake. Also, one trick I used a lot was to put the mixer in mono mode and lower the volume almost to the limit of audibility. That made it easy to hear which frequencies were poking through and which instruments dropped out first. If you can make a mix fit together when it's mono and very quiet, then when you turn it back up and get the panning going again it just sounds much bigger and cleaner.

This involved a bit of memory about where I had EQed other parts, a bit of going back and checking and adjusting the EQs, and a bit of intuition and paying attention to the sound, not the numbers.

I also managed the client's expectations a bit by explaining that the more you put in the mix, the smaller each piece has to be.

clowrymusic
audio school
Posts: 2
Joined: Tue Mar 06, 2012 12:29 pm
Location: New York, NY
Contact:

If they must live together, they can't be the same.

Post by clowrymusic » Fri Mar 16, 2012 7:34 pm

I find that most instruments in my mix don't really sound spectacular in solo. If they did the mix would be a crowded room. The audibility and beauty of a instrument within the song is subjective to its environment within the song as a whole.

Subtractive EQ, keeping things in their own boxes, and panning gives you lots of room.
When you go to put up thick walls of guitar, I find that the old Hair bands short delay from right to left on your guitars, opens a nice whole for them to live comfortable in a pocket.

I know that doesn't work for everyone, but give it a try!
Making music sound awesome is what I do best! Respect your music and mix it right the first time.

dicko
audio school
Posts: 4
Joined: Mon Feb 06, 2012 5:28 pm
Location: Hobart, Australia

Post by dicko » Sun Mar 18, 2012 5:41 pm

GooberNumber9 wrote: I always carved up the frequencies before I did any panning, because that meant the panning was like icing on the cake.
I've found this approach very useful. I'll also check just the hard panned elements (guitar, overheads, stereo rooms) in mono together then rinse and repeat for the elements that stay centered.

drumsound
zen recordist
Posts: 7474
Joined: Tue Jun 01, 2004 10:30 pm
Location: Bloomington IL
Contact:

Post by drumsound » Sat Mar 24, 2012 11:50 pm

GooberNumber9 wrote:On the thickest mixes I've done I was able to squeeze things in by almost planning where they would go in the frequency spectrum. So I EQ'ed drums first and sorta picked one important freqency for the kick and snare based on their recorded sound, and then gave them a boost there. Then I did a lot more cutting of each drum track and the whole drum bus to make some room. Mainly I put in high passes as high as possible on everything but the kick, and did some low passing on everything but overheads.

Then I did the same with the bass: I gave it as small a box to live in as possible without making a big impact on the sound, and then I gave a boost to one or two frequencies that it "liked".

Then on to guitars, with the same idea, but each guitar would get its own boost frequency (we're just talking a dB or three) and those would be different from the boosts used on snare, kick, and bass. When things got too crowded, I would look to making some cuts to the bass or drum bus at the same points where I wanted to boost a guitar, or at least where I felt a guitar track had a natural peak. When you have 4 or 6 or 8 very distorted guitar tracks, you can't be sentimental about them. Feel free to cut them up with the EQ and be brutal. This is a good case where listening to the whole mix while you're cutting can help you hear how the different parts interact.

Of course the vocals were top priority, so I bussed all the guitars together and made some cuts in all the busses where there was either a peak in vocals or I wanted to emphasize the vocals.

As others have said, it's important to let go of trying to have individual tracks sound good on their own, and instead try to make them work together - especially for the guitars, since they have a more continuous sound. You can get away with the snare, kick, or toms taking up a lot of space if the drum part is sparse enough. If you're like me, you'll also want the vocals to sound as good as possible and never get stepped on by anything else. Between those two, the guitars will sometimes lose out, but with a wall of guitars going on, no one will miss a guitar part getting pushed back by a snare hit or a vocal line.

For my style, I always carved up the frequencies before I did any panning, because that meant the panning was like icing on the cake. Also, one trick I used a lot was to put the mixer in mono mode and lower the volume almost to the limit of audibility. That made it easy to hear which frequencies were poking through and which instruments dropped out first. If you can make a mix fit together when it's mono and very quiet, then when you turn it back up and get the panning going again it just sounds much bigger and cleaner.

This involved a bit of memory about where I had EQed other parts, a bit of going back and checking and adjusting the EQs, and a bit of intuition and paying attention to the sound, not the numbers.

I also managed the client's expectations a bit by explaining that the more you put in the mix, the smaller each piece has to be.
The way you describe this reminds me of what I hear when I listen to a few things JJP has mixed, especially the Grays record Ro Sham Bo.

GooberNumber9
tinnitus
Posts: 1094
Joined: Fri Oct 20, 2006 7:52 am
Location: Washington, DC

Post by GooberNumber9 » Mon Mar 26, 2012 9:49 am

drumsound wrote:The way you describe this reminds me of what I hear when I listen to a few things JJP has mixed, especially the Grays record Ro Sham Bo.
Just as long as no one here thinks I'm anywhere near that good, but if I were I could see myself going for an almost identical style of mixing.

User avatar
Red Rockets Glare
tinnitus
Posts: 1132
Joined: Wed May 07, 2003 8:36 am
Location: Los Angeles
Contact:

George shows you how to EQ

Post by Red Rockets Glare » Sun Apr 08, 2012 10:31 am

Why not let the master show you?

http://www.sweetwater.com/store/detail/GML8200

I found the videos on this page really helpful in terms of explaining subtractive EQ, which was previously a huge mystery to me.

I still use additive EQ in the mids BTW.

ricey
alignin' 24-trk
Posts: 71
Joined: Sat May 20, 2006 7:27 pm
Location: Sao Paulo/NYC
Contact:

Post by ricey » Mon Apr 09, 2012 9:06 am

i'm a mix engineer specifically, and lately dealing with a project w/ Tons of layered, distorted guitars...

... first thing i do is subtractive EQ - i find the room tone and remove it, i find the capsule resonance and either remove it or sometimes add a little, then just keep listening for continuous tones that don't change with the notes/chords and remove them... by then things are pretty cleaned up.

User avatar
frans_13
takin' a dinner break
Posts: 177
Joined: Mon Mar 07, 2011 4:46 am
Location: Bavaria, Germany
Contact:

Post by frans_13 » Mon Apr 23, 2012 11:21 am

For a shedload of guitars I try the following:
lowcut, maybe hicut - if it's a less dense mix then you can be gentle, if it's dense, be brutal
Find and notch out pickup and maybe speaker resonances. If you hit them right, taking the notch out will suddenly sound unnatural. Be careful to cut the right ammount as too much can easily castrate guitars... but if thats whats needed..
Does this or that track play a specific melody or range the other guitars don't? Find the root note or a harmony and apply a little bell boost here, maybe even automate the boost frequency if you feel it should follow.
After you are through, listen if your bandpassing and boosting made the guitars honky, if, group stuff and throw on a wide and gentle cut or simply re-think your earlier choices of eq.
Different ammounts of short delays may help the lines step forward or backwards, soundstage wise.

If things get ugly you could put a sidechain-able eq on the guitars that take up most of the "guitar real estate" and make the vocals duck said guitars a little around 3-4k every time they come up. Not ducking the whole guitar, just the range where their bite eats into the vocals.

For drums you should decide if they need attack or sustain to do their job. SPL transient designer to the rescue. Then again, why not have the snare duck the whole wall of guitars a little? Just like grandpa had the kickdrum duck the bass sometimes. You get the idea, if all the elements in a mix have used up all the available space, then you got to ride levels to direct the listeners attention to different elements one after another, according to the "drama" the song follows.
Why you're at it, similarly automate the whole song to get more/less dense at times, according to the emotional content in the song at the time.

That said I always wondered if you can't get a piece of music happening with less than 16 tracks, the song's fishy...clearly there are exceptions. Tracking guitars benefits from using more than one amp/guitar/tube types/register/pickup/speaker/mic/preamp

User avatar
Dakota
re-cappin' neve
Posts: 740
Joined: Sun Sep 02, 2007 10:14 am
Location: West of Boston
Contact:

Post by Dakota » Mon Apr 23, 2012 1:22 pm

frans_13 wrote:Just like grandpa had the kickdrum duck the bass sometimes.
! Love it !

User avatar
vvv
zen recordist
Posts: 10139
Joined: Tue May 13, 2003 8:08 am
Location: Chi
Contact:

Post by vvv » Mon Apr 23, 2012 3:45 pm

FWIW, when stacking guitars, I find that using a different mic for each track on a side helps a lot - either Mr. Patterson or mebbe Mr. Carges, or both, said never use the same mic on vocal tracks - I love that, and for guitars, also. Taking the idear further, a different guitar or just pick-up, changing the amps out, pre-amps, etc. for each guitar track per side helps.

Other things you can do is use a chorus on one guitar, do one clean, one dirty, etc. Or compression, or delays ... Or even mic distance and orientation (on-axis, off axis, on the floor) ...

And if you use the same stacked tracks on each side (ex., Tele/Fender/dynamic/SS pre; Paul/Marshall/LDC/tube pre; Strat/Vox/ribbon/hybrid pre) you can build symmetrically. But really, the difference between a SM57 and a old AKG or something can be enough, even with the same guitar and amp.

Then, when put together, they seem to be individually more distinct in the guitar sub-mix, which I then might EQ.

Me, I'm less about EQ (other than pass-filters) then source sounds - definitely a weakness, but I just don't like EQ as more than a fixit.
bandcamp;
blog.
I mix with olive juice.

studiodog
alignin' 24-trk
Posts: 71
Joined: Thu Aug 16, 2007 7:43 am

Post by studiodog » Tue Apr 24, 2012 11:15 am

great post - I use different mic's for vox and guitar for each take; and go so far as making sure any effects used on guitar are moved up or down a tad and if possible use a different guitar. If your stacking the exact same sound on sound it turns into a mess fast.
"...because I'm good enough, I'm smart enough, and gosh darnit, people like me!"

Post Reply

Who is online

Users browsing this forum: No registered users and 67 guests