"EQ as little as you can" a good rule?

[rasputin1963]

From the discussions we've had of mics here, and from comments I've read all over the web and in SWEETWATER catalogs, I repeatedly hear sound people say, "I used (X) mic, and by the time I got to the mix, almost no EQ was required."

 

Now here's a question sure to brand me as the n00b I really am: Why is this a desired thing?

 

Or, to put it another way, nowadays, when digital toys can make anything sound like anything, why would you want to refrain from EQ'ing a signal?

 

Seems to me, in this day and age, you can make even a lousy mic sound darned good if you twiddle with the signal long enough and in just the right way...?

 

Comments?

[Dean Roddey]

It's not like it's a bad idea. It's kind of a purist ethos. The less processing the better, I'm sure. But it's not exactly a practical rule for those of us who don't have 10 pre-amps and 25 mics to play around with until we get just the right sound combination.

 

I try not to use any more than is necessary, but if it needs an EQ, it needs an EQ. Particularly for those of us DI'ing in instruments, all of the processing that might have been done outboard (the player's foot pedals, effects boxes, and amp, plus any outboard compression and EQ) in a traditional recording scenario never happened in that case, so in many ways are way ahead in terms of purity of the signal on the disc, having gone through nothing perhaps but a pre-amp. So the EQ and compression we'd do in that case is not any more than would have often otherwise been done anyway, and maybe less overall, we just do it after the fact.

[D Charles]

Well you can use the same argument about cheap mic pre's; yeah, you might get a good solo sound, but when you start stacking parts together you'll have a mess.

 

I remember a bud of mine went to a big Nashberg studio once and commented to the engineer that he was getting a great sound. The engineer replied; "I'm not doing anything, those are the tones in the room."

 

If EQ magic could make any mic sound good on any source, there wouldn't be expensive mics. No one would need them.

 

That said, I almost always eq a bit.

[blue2blue]

I'm with Dean and D. The less processing the better, as a general rule.

 

(And, of course, there's still more to a good mic than a favorable eq response. Also, let's not forget that EQ response changes with proximity, particularly with cardioid mics, changes with impedance, and changes with dyanmics. That's one reason why convolution type effects fall so flat when trying to make a given mic sound like another.)

 

As D alludes to, EQ stacking can be a real issue when you're mixing on a board with fixed point EQs, if you get aggro with the EQ across the board. That's why I'm a BIG fan of sweepable EQs (better yet, fully parametric).

 

But I use EQ when I need it. No question about it. I have an EQ I really like and I'm not afraid to use it in a mix.

[Lee Flier]

 

I repeatedly hear sound people say, "I used (X) mic, and by the time I got to the mix, almost no EQ was required." Now here's a question sure to brand me as the n00b I really am: Why is this a desired thing? Or, to put it another way, nowadays, when digital toys can make anything sound like anything, why would you want to refrain from EQ'ing a signal?

 

 

Because you really CAN'T make "anything sound like anything," no matter what the marketing guys say.

 

 

Seems to me, in this day and age, you can make even a lousy mic sound darned good if you twiddle with the signal long enough and in just the right way...?

 

 

If the mic and/or the technique you used to record it is truly lousy, no you can't. And even if you could, it still won't sound as good as if you'd used a decent mic and a decent recording technique in the first place. If the signal is distorted when you don't want it to be, you can't undo that. If you want to boost 80 Hz with an EQ in the mix and there's nothing in the source track at 80 Hz, all the toys in the world won't help you. You COULD do all kinds of plastic surgery with sound replacement and the like to get it to sound "decent" - but why would you want to? It causes hours of pain and suffering and still doesn't sound as good as a high fidelity source would have.

 

When an engineer says "I used this mic and no EQ'ing was required," they're not really saying that EQ is a bad thing or that they wish to "refrain" from using it. They are simply saying they followed one of the fundamental principles of audio engineering: get the sound right as close to the source as possible. The closer you can get it to the source, the better it will sound down the line - no matter what happens to it down the line. If you want to intentionally mangle a track later for effect, there's nothing at all the matter with that, and most of us like a little EQ or compression to add a little sizzle or punch in the mix here or there. But no one's ever been fired for getting a nice, high fidelity signal to tape or disk. Whereas if you work at a commercial studio where your work is likely to be mixed by someone else, and they bring up a vocal track expecting to hear something approximating a vocal, and instead it sounds like distorted poo that they have to EQ and tweak the crap out of just to get it to sound like anything... well, that IS likely to get your ass fired.

 

Now you might not think that is so relevant in a home studio, where you're both tracking and mixing your stuff. But even then, the fundamental rule applies - get it as close to the sound you want at the source as you can. It'll sound better - fuller, more detailed, more present, etc. - than if you try to polish a turd with a lot of digital toys. In a home studio you're likely to have some less than top quality plugins and if you use too many of them added across too many tracks, you'll end up with a mix that sounds like poo.

 

Even if you go on to apply EQ or anything else later on in the mix, you'll appreciate it if it's only a gentle curve, unless you're really going for some way-out effect, in which case anything goes. Otherwise, if you're trying to get a voice to basically sound like a voice (only maybe a little enhanced), a piano to sound like a piano, a guitar to sound like a guitar... just get it right at the mic, as closely as you can. There are inherent problems with small untreated or lightly treated rooms that make it so you'll have to use some EQ anyway in all likelihood - but the more of it you have to use the more it will sound like poo. A very steep EQ curve generally will sound really crappy unless you've REALLY got some top of the line EQ.

 

In short, effects are great, but use them because you want to (i.e. you like the sound of the effect and it enhances the power of the music), not because you have to (because you did a crap job of tracking).

[Ernest Buckley]

Close your eyes & tweak the sound until its right. Disregard everything else.

[Gus Lozada]

 

Close your eyes & tweak the sound until its right. Disregard everything else.

 

 

Amen

 

 

Edit: Tweak or not, until the sound is right.

[Dean Roddey]

I wouldn't go that far. I'd say that the wisest mixer uses all the tools available, and that includes frequency analyzers and meters as well. Unless you have a perfect room and numerous monitors and absolutely no hearing loss due to age/abuse, you need to be sure that what your ears are hearing is really what's coming out of the DAW, and generally we don't have perfect rooms and numerous monitors, and many of us are going to have limited sensitivity up high. I always use the frequency analyzer to make sure the lowest and higest ends of the mix are reasonable.

[Kerouac]

I'm mixing a project for one of the teachers I work with at our music academy. He frequently records "band practices" of his students at similar skill levels and it's a lot of fun for the kids as they get to play with their peers, and then the music and/or video goes up on our website and lobby TV. Generally it's between 4-10 students all being mic'd and recorded and the bleed can be absolutely horrible, much less the fact there isn't a lot of time to tweak and get really great tones. It's all about capturing the kids having fun and playing their instruments, but getting everything to sound great sonically can be a challenge, especially with bleed and tuning problems.

 

The track I'm mixing right now has some terrible drum sounds on it (a verrrrry bad V-Drum patch....) and I'm going to have to layer up to replace the kick, but almost everything needs some form of EQ. Having to EQ so much is playing hell on the master fader and I have to do a lot of back tracking to make it all sound pretty even, so I took a break and started over.

 

Vocals and bass, then strings, then acoustic, then electric guitars and keyboards, and then last kick/drums/percussion, in order of least EQ to most EQ.

 

The vocals and bass actually didn't require any EQ, but I did have to do some automation for vocal levels and touch up the bass with some compression. Acoustic I just cut some flub and high end than added a hint of mids to tame some piezo quack. Keyboards and guitar were occupying the same space, so I dipped out some mids on keys and added a univibe plug for some extra shimmer, and dropped some treble and low mids from the electric guitar.

 

The drums are still quite a challenge. The kit sound is all high-mid and rubbery tom fill content, the snare was all sidestick, and the kick makes the snare on St. Anger sound full and punchy.

 

This project would've been a lot easier to work on if the tracks had been able to be recorded better and not need as much EQ, but in the end you do what you need to towards the goal of the track sounding GOOD. If that calls for EQ go for it... hell sometimes I write parts that are just begging for extreme EQ shaping, but keeping things sounding natural is always a good goal.

 

My $0.03.

[UstadKhanAli]

I try and use EQ for aesthetics, not for fixing things.

[Geoff Grace]

 

If you want to boost 80 Hz with an EQ in the mix and there's nothing in the source track at 80 Hz, all the toys in the world won't help you.

 

 

That's it in a nutshell. EQ can't enhance something that's not there.

 

To use an analogy, if you decided to work with a picture in Photoshop, would you want to start with a blurry mess or a crisp, clear image?

 

Best,

 

Geoff

[Anderton]

It's not like there's a dividing line between "Use EQ!" and "Don't Use EQ!". And there are also different situations. You might get a perfect drum sound, then halfway through the song, the vibe changes and you want to change the sound that had been perfect into something more fitting.

 

For example, I don't use EQ on too many sources but always use it on my voice. For the classical music recording I do, EQ gets added only at the end, and only subtly for aesthetic reasons. And sometimes I EQ for reasons like getting rid of subsonics.

[Magpel]

I think it is a good rule of thumb, sure. I think another good rule of thumb is cut rather than boost if at all possible.

[Lee Knight]

 

I think it is a good rule of thumb, sure. I think another good rule of thumb is cut rather than boost if at all possible.

 

 

True. But... (there's that ugly "but" again)

 

Sometimes it's best to forget all that and eq the crap out of a sound. A close miked snare for example. It's rare that you don't have to do some serious cutting of resonance, and major boosting of the things you do want. Some of the snares we hear today are seriously boosted in the highs and lows. Then the kick. There isn't a kick today that hasn't been scooped at somewhere between 250 and 400. And boosted somewhere between 3k and 6k.

 

Unless the mic already has those qualities. An Audix D6 is down more than 10dB at 1k! I don't happen to like that. I like my scoop centered somewhere between 250 and 350 or so. So I use an RE20 that is pretty flat, and I eq to get to where I want. Some tunes, and some kicks, need a bump at 3k. Some at 6k. The D6 is going to give me 1 thing. Up at 3K or so and WAY UP at 10k. No thanks.

 

But overheads? I'd rather find a mic that sounds perfectly right or almost perfectly right from the get go. If I have to make my OHs work through eq, it sounds crap. I want that natural.

 

On and on. To eq or not to eq. It depends on what I'm going after. If it's natural I'm seeking, the less eq the better. And when needed, a cut sounds more natural than a boost. When natural isn't an issue... I grab and twist to my heart's content listening all the way.

[Ernest Buckley]

 

I think it is a good rule of thumb, sure. I think another good rule of thumb is cut rather than boost if at all possible.

 

 

Yeah, thats something I started doing about 5 years ago and it really changed my mixes for better. I use EQ like salt: sparingly just to accent the sound.

[Kendrix]

All I know is that folks like Swedein, Massenburg and Manning all subscribe to the approach that Lee F and others have described here.

Get the sound as close as possible to the final sound at the source.

Use mic selection and placment as the first line of attack then EQ only if necessary during mixdown.

 

Based on all I've read and been told it seems that the most open/ narural and spacious sounding mixes tend to be produced in this way.

 

From my own experience I find that too much processing can cause the soundstage to collapse/become one dimensional at some point.

[Jimbroni]

It depends. If you are trying to get pure sounds you are better off getting it right while tracking rather than eqing in post, but lets assume you captured everything in an adequate manner, well its now it time play. Maybe you wann dirty some things up, maybe you wanna automate a sweep, maybe you decide to copy a guitar track and spike certain part of the distortion in the extra track.

 

Just because you have a bunch of great colors recorded, doesn't mean the painting is finished.

[UstadKhanAli]

I think that's the best way to go: get the sound as close as possible to the final sound at the source.

 

That's essentially what I meant by "I try and use EQ for aesthetics, not for fixing things." Like Lee Knight, I'll EQ the snot out of something for aesthetics or if I need to, but always try and get it as close as possible to the final sound at the source first, generally speaking.

[Jimbroni]

Yeah I'm with ya. I hate having to fix things with EQ. Its the worst time to use it.

[Lee Flier]

 

It depends. If you are trying to get pure sounds you are better off getting it right while tracking rather than eqing in post, but lets assume you captured everything in an adequate manner, well its now it time play. Maybe you wann dirty some things up, maybe you wanna automate a sweep, maybe you decide to copy a guitar track and spike certain part of the distortion in the extra track. Just because you have a bunch of great colors recorded, doesn't mean the painting is finished.

 

 

Of course not. Like I said - use effects all you want because you like them. But you don't want to start with a crappy canvas.

 

The only exception there is if you already know that the final product is gonna be a mangled track, and maybe you can mangle it better at the source. Like if I want a "megaphone" style vocal I'd just as soon cut the track through a megaphone than use EQ or a plugin to make it sound that way later. Others prefer the reverse, though.

[UstadKhanAli]

With the "megaphone" or "telephone" style vocal, I usually end up mangling it after recording it, although sometimes I'll help it along by recording it with a crappy mic. I don't know that this is necessarily better sounding, but it does offer more control when trying to fit the voice into the mix. It's that time-honored thing, though...you're delaying a choice until later, and sometimes, it might just be better to commit. Less futzing for later. The few times I've done "megaphone" vocals at the source, I've effed 'em up a bit too much and had to re-do them.

 

Speaking of doing this, those funny toy megaphones are really fun to use for mangling vocals. No plug-in can even touch that. Robot voice, odd reverb voice, etc. Damn, those are fun.

[Instrospection]

It depends on the song, too--sometimes you need something to cut through and it doesn't do it with it's natural instrument characteristics. For example, I recorded a Heintzmann grand piano with a pair of AKG 414's, and it ended up being too mellow for the rest of the mix. The rest of the mix was pretty bright naturally (especially the acoustic guitars), so I actually had to increase the brightness of the piano, reduce the bass EQ on the piano, and reduce the brightness of the acoustic, in order to get it all to sound good together. Different pianos have different characteristics--some are brighter, some have a higher attack, etc--so getting them to balance correctly takes some skill at times. Especially in rock mixes--things often sound great soloed, but then don't cut through when they're in the big mix, and need to be tweaked in order to "get along" with the other instruments.

 

Sometimes you need to reduce EQ, though. Instead of just boosting everything else, try reducing things.

[halljams]

I find the reality of this all is that sometimes there are {censored}ty sounding tracks to deal with. {censored}ty in a bad way and they need major reconstruction with eq to make them tolerable.

 

Another thing too is the more creative you get the more you may have the need to totally demolish a perfectly clean, nicely recorded and balanced track for the good of the vision of the mix.

 

The more i do this stuff the more i despise clean. Then it becomes a matter of taste for distortion of what ever sort you seek. It also starts to become a subtler matter, meaning; where i hear distortion and distortion type, other people may not even perceive any distortion at all. But it is important to me.

Distortion from over EQing is a type of distortion.

[philbo]

My tried & tested rule of thumb, which applies any time I've been listening/mixing for more than 20 minutes: EQ till it sounds right, then reduce the dBs of boost or cut by one half. For example: If you boosted by 5 dB, then the right amount was 2.5. This prevents the problem of coming back the next day and going, "Good God, what in the world was I thinking?"

 

This applies primarily to 'sweetening EQ', where you are trying to shape the sound. OTOH, If you're cutting trash out of a track, like sub-sonics or whatever, being brutal enough to get the job done right the first time is crucial.

[Kerouac]

My tried & tested rule of thumb, which applies any time I've been listening/mixing for more than 20 minutes: EQ till it sounds right, then reduce the dBs of boost or cut by one half. For example: If you boosted by 5 dB, then the right amount was 2.5. This prevents the problem of coming back the next day and going, "Good God, what in the world was I thinking?" This applies primarily to 'sweetening EQ', where you are trying to shape the sound. OTOH, If you're cutting trash out of a track, like sub-sonics or whatever, being brutal enough to get the job done right the first time is crucial.

 

That's a pretty good idea. I'll try it out when i get back to this track I'm working on.