Can we talk frequencies for a minute?

1. I'd say bass frequencies are below 200 Hz, mid is 200 Hz to 5 kHz, and highs are above 5 kHz. It gets easier if you split it 5 ways:

Lows: 20 Hz to 150 Hz.
Low-mids: 150 Hz to 700 Hz.
Mids: 700 Hz to 2.5 kHz.
Hi-mids: 2.5 kHz to 6 kHz.
Highs: 6 kHz to 20 kHz.

This is NOT universal, just my personal view.

2. I tend to boost cymbals with a high shelf, centered at about 16 kHz, where they shimmer the most. If there's a harsher cymbal frequency, it's probably in the 4 to 7 kHz range. "Harsh" frequencies are usually in that hi-mid range.

3. "Muddy" frequencies are usually in the low-mids... I tend to cut a lot of stuff across the board at about 400-500 Hz. I just did a project where I cut 440 Hz on half the tracks... Now that's mud!

 

Cuts between 300 and 600 hz reduces the mud in my experience. I usually won't cut the low mids below 300 hz on a stereo mix because this often weakens the fundamental frequencies for the snare, which are often around 250 hz or so. Unless you want to hollow out the snare tone, which I generally don't like to do. This is my experience on consumer grade gear (Roland VS 1880 and the like).

 

Another one. Gainy rhythm guitar tracks can often clash with gainy lead guitar tracks. While lowering volume will often move the rhythm guitars back in the mix to allow the lead to poke thru, try cutting eq at between 1.4 khz and 2.5 khz instead. This is another way to push the rhythm tracks back into the mix and often opens up the lead guitar. On bass guitar tracks, boost the eq at 2 to 3.5 khz for more clarity and cut. Sometimes another boost at 700 to 900 hz adds some interesting complexity to bass tracks. But sometimes this sounds like arse, gotta trust your ears.

 

I mean no offense, but why does it matter how frequencies are classified? It's one thing to refer to a specific section of the audio spectrum with an adjective like "bass" or "sub-bass" or "upper mids," etc. just so other people know generally what you're talking about, but you should NEVER think of these terms, much less USE them, when mixing. It's a bit like telling the chef that you want something "green" and expecting to get what you really have in mind. (Or dancing about architecture... Sorry, had to do it.)

I'd really love to be able to sum up a nice "how-to" about mixing a rock tune, but there's just no way to do that. What I wouldn't want to do is throw a bunch of adjectives your way and tell you that you understanding them will improve your mixing chops.

Even though I don't like "bass" and "treble", one adjective that's specific enough to refer to a pretty tight frequency is "mud." When I have a track that's too muddy (usually bass or kik) I almost always start with a cut at 250Hz, then move the freq around until it's right. Rarely do I go very far from 250. If I go beyond that it wasn't "mud" to begin with.

The Izotope thing sounds cool. I'd wished for that product 20 years ago, and if I knew enough about writing software code (as I did 20 years ago ) I'd have tried to write that myself. That software seems brilliant, and the concept certainly should be within the realm of human possibility. In a very simple form, it could take the sonic signature of the reference tune (the average amount of each frequency throughout the tune), then compare it to the sonic signature of your tune, then add or subtract eq as necessary to make your tune match the reference. I'm sure Ozone is more advanced, but that's a great idea.

What I'd rather see is a SUGGESTION from Ozone, not the actual eq'ing, that tells me that I have too much or too little happening and at which frequencies, then let me eq myself. I think that would result in an even better final product because I could lean towards the reference, but with control.

 

*Boosting the bass too much at 16, 31, or at 60Hz will make it sound muddy. Boosting the bass too little at 60Hz makes it sound thin.
*Boosting too much at 125Hz makes it boomy. Boosting it too little at 125Hz makes it sound thin. When set just right, it will sound full without boominess.
*Boosting it too much at 250Hz makes it sound like a telephone and adds listening fatigue. Too little and you lose fullness of sound.
*Boosting it too much at 500Hz makes it sound honky.
*Boosting 1 to 2KHz sounds tinny.
*Boost too much of the high mids over 2kHz and you lose speech recognition, add a lisp, and listening fatigue takes over.
*Boost too much at 6kHz and the vocals hiss.
*Boost too much at 8kHz and you add sibilance.
*Boosting the 8kHz to 22kHz range just enough adds air and spaciousness, boost too much and you get sibilance on vocals.

 

I didn't know quite how to say that in as positive a way as you did.

 

Straticus, I hear you loud and clear. I started here a couple of years back with no experience to fall back on. I began to build knowledge in part on the discussions here on this board. Start with that and begin to build your own inventory of experiences, take notes and reflect, then start over again. Its an iterative process. FWIW, after getting your feet wet, always remember to get your tracks right at the tracking stage. It's easy to forget this and fall in love with gear like eqs and compressors, but get it right on the front end and you will have more meaningful and effective options at mixdown and mastering stages.

 

Absolutely! All I meant to say was that I agreed with dt's thoughts about expecting a piece of software to work "magic" vs. using one's ears, and about factors beyond EQ that the software can't detect.

Anyway...

The only way I was able to get started with EQ at all was to read what other people had done and ask questions. I'm still asking questions, all the time! There are guys out there whose records have a sold a LOT more than mine, and any insight they care to offer is welcome.

EQ is a bitch! Miking and compression are (IMO) much easier to tackle.

 

Hi Straticus.

Early on you asked for some ballpark EQ frequency info. I had a really handy guide around here for live mixing, and it took a while to find it on the hard drive, but find it I have and here it is...

(One caveat: IMHO freq points are like recipes--good starting points. No substitute for good taste. These may just get you close. Sorry for the caps on the font.)

**********

FREQUENCY PRESENCE EQ POINTS FOR INSTRUMENTS​

DRUMS:
KICK BOTTOM 60-80HZ SLAP 2.5K
SNARE FAT 240HZ CRISP 5K
HI HAT CLANK 200 SIZZLE 7.5-12K
MOUNTED TOMS FULL 240 HZ ATTACK 5K
FLOOR TOM FULL 80-120HZ ATTACK 5K

GUITARS:
BASS:
BOTTOM 60-80HZ PLUCK 700HZ-1K POP 2.5K
ELECTRIC:
FULL 240HZ BITE 2.5K
ACCOUSTIC:
BOTTOM 80 120HZ BODY 240HZ PRESENCE 2.5-5K

ORGAN:
BOTTOM 80-120HZ FULL 240HZ BITE 2.5K

PIANO:
BOTTOM 80-120HZ ATTACK 2.5-5K CRISP 10K

VOCALS:
FULL 120HZ BOOMY 200-240HZ PRES. 5K SIBILANCE 7.5-10K

********

In addition to the above, the writer recommends sweeping the mids on the house curve to about 2500HZ and backing 'em off a notch or two for listening comfort. I've had good luck with that tip in live mixing, but I'd say Use Sparingly for recording.

Also, if it isn't a bass or a kick drum, or occassionally keyboards, using the hi-pass (centered at about 100HZ most of the time) will keep out the mud.

Hope this is sorta what you had in mind.

The original document can be found on the Yahoo! RigTips discussion group, under Files / PA Docs / "Gain Structure Basics and Beyond," by Humphrey Audio.

Happy Holidays!

 

Very useful post, epluribus.

 

The writer - and you - are correct.

The 3K range is a "punchy" area for a lot of mid-range instruments and very necessary, but it sometimes too much of it builds up in a live room. When a mix sounds irritating but you can't put your finger on why, it's a good place to start a little notch.

A trick with notching that works well but causes anyone else in the room to hate you:

Start with as narrow a bandwidth (Q) as your EQ will allow and boost it maybe 6dB or more. Sweep till you find the spot where the offending frequency is strongest, coming at you like a freight train (careful not to hurt your ears or damage your speakers!!). Then pull it down. Adjust the gain reduction and Q till it sounds right. Keep in mind that your first impulse after it has assaulted your senses is to pull it out completely, but that might not be best. If you can, A/B that band only, off and on.

There are other tricks you can do with 3K, such as selectively boosting it on an otherwise lifeless vocal track in emphatic spots to add character and "oomph." But it's very tricky. Too much or too often and the singer sounds nasal and annoying, like an unpleasant neighbor in Forest Hills. If and when I have the time I'll try to remember to post specifics.

 

I can see where that notching technique will be quite handy, thanks for posting it Michael. Every once in a while you get a room or a mix with a nasty edge to it that you can't put your finger on.

--Ray

 

Depending on what you're hearing, it might be a good place to start, anyway.

Some rooms are just ugly no matter what you do.