EQ tips for common instruments

[Skol303]

In case anyone finds this useful...

I'm currently geeking away and trying to get a better understanding of using EQ as a mixing tool (something I've largely ignored till now!).

As a result of this, I've compiled various hints and tips from across the Interweb into a handy document, which I thought I'd share with y'all here, in case it helps any other EQ novices:

[attachment=87410:EQ_Instruments.doc]

^ None of this advice is my own: it's been cribbed from various websites and forums (including this one!), so please note I don't claim to hold any copyright on it.

I'd also take some of this advice with a proverbial 'pinch of salt': EQing is a very subjective art, as I'm finding out, so the document is meant to provide a general guide only; a starting point to help you get tweaking. Oh yeah, and it may well contain numerous mistakes! All part of trial and error...

[algmusic]

[quote name='Skol303' post='1345464' date='Aug 19 2011, 02:09 PM']In case anyone finds this useful...

I'm currently geeking away and trying to get a better understanding of using EQ as a mixing tool (something I've largely ignored till now!).

As a result of this, I've compiled various hints and tips from across the Interweb into a handy document, which I thought I'd share with y'all here, in case it helps any other EQ novices:

[attachment=87410:EQ_Instruments.doc]

^ None of this advice is my own: it's been cribbed from various websites and forums (including this one!), so please note I don't claim to hold any copyright on it.

I'd also take some of this advice with a proverbial 'pinch of salt': EQing is a very subjective art, as I'm finding out, so the document is meant to provide a general guide only; a starting point to help you get tweaking. Oh yeah, and it may well contain numerous mistakes! All part of trial and error... [/quote]
looks good

[cheddatom]

Interesting! I started writing the below when I thought it was a breift one page summary. I'll have a good read at home. Cheers!

Personally I would have thought the Kick Drum treble boost (if any) should be nearer 7-10KHz than 5KHz which I found could be too clicky. Also the increase in top end on the snare drum would only be acceptable on a top mic - if you did that on the bottom mic it would be all rattle and fizz surely?

[Mornats]

This is brilliant, thanks for posting it! I particularly like how you've explained what each cut or boost is doing - very helpful.

[51m0n]

Three words:-

[size=7][b]Use Your Ears!![/b][/size]


Nothing else matters, no guide lines, however well written, apply to mixes in the general sense all that well.

Some idea of where you may find various parts of the sound are useful, but using a sweep eq properly to find them [i]in the context of the mix[/i] is the real truth. Any other approach will not work nearly as well.

Some of those eq points I would take with a [b]huge[/b] pinch of salt, kind of like the size of Jupiter, but heavier....

The idea that you can compress something with various settings as a starting point is pretty bogus too to be honest. Instead learn how to set one up to do various things to a sound, practice on various sounds to change the transient in some way you predetermine (soften the attack, thicken the punch, add sustain, cause pumping or release artifacts) try and achieve these results deliberately, a couple of days of playing around with a decent comrpessor vst and you will been way way better off than if you try and apply the same cookie cutter approach to various source signals.

Edited August 19, 2011 by 51m0n

[Skol303]

[quote name='51m0n' post='1345683' date='Aug 19 2011, 05:48 PM']Three words:-

[size=7][b]Use Your Ears!![/b][/size][/quote]

Very good point well said.

But then I wasn't intending to provide the 'Gospel of EQ According to Skol', simply a rough guide for novices, such as myself, to at least have a basic idea of where to start tweaking all of those knobs and dials...

If you're skilled at EQ'ing, then using your ears is undoubtedly the best way to go about it. If you don't have a damn clue what you're doing (like me!), then simply being told to "use your ears" ain't quite as useful as having at least some basic notion of where to begin, bogus or not.

At least that's been my experience.

I'm sure there are probably books written on the subject of EQing, but if not then you might have a super-concise best seller on your hands right there... "Master EQ'ing in Three Simple Words"

[Rimskidog]

Sometimes I think 51mon and I are brothers or somethin'.

At skol303, it may feel like as a newbie you need something like this but, honestly, it's a false crutch. It will take you in the wrong direction as sure as eggs are eggs. What 51mon is saying is instead of following a guide like you've posted the way which will give you better results and will more quickly get improve your mixing is to

1. listen to the sound of each channel you have [b]in the context of the mix[/b], think about what you don't like about
2. in respect of each channel grab the knob that changes the frequency on your eq and sweep it up and down the frequency range until it hits the offending frequency
3. boost or cut that frequency to suit your sense of aesthetic (twiddling the q knob as required)

Following this process just 2 or 3 times will help you 'get it' more than reading anything about it. Really. Just try it.

Good luck!

[Skol303]

^ Gotcha, and I couldn't agree more with either you or S1m0n. More than happy to bow to your superior know-how

Like I said, I wasn't suggesting that a list of frequencies should ever replace using your ears - otherwise the best studios would be run be deaf mathematicians... all I'm saying is that for me personally, it's certainly helped to know, for instance, that I can lose some of the 'boominess' from a typical kick drum by starting to tweak its frequency somewhere around [i]here[/i]; or I can usually make a snare sound more crisp by boosting it somewhere around [i]there[/i], so that's where I'll start twiddling.

Of course, it depends on the song in question and the context of everything else in the mix. But for me, it has helped to have a rough idea of where each instrument [i]usually[/i] lies in the frequency spectrum. Not as a replacement for using my ears, but just as a starting point to help me understand where different sounds might be sitting in the mix.

You guys can probably reel off the typical frequency bands of instruments in your sleep, so for me it's just been a basic exercise of getting at least some of this understanding for myself. The rest, as you say, is going to be all about listening, listening and then listening some more!

Anyway, you can count on me pestering you guys for advice when I get stuck!

[flyfisher]

Thanks Skol, looks like a useful set of hints & tips for a recording novice like me. I get what 51m0n is saying - after all, music is such a subjective subject that ears are the ultimate judge - but anything that helps the novice to understand where to start and what the effect of various changes might be, must surely be a good thing?

Anyone who dials-in the suggested eq settings and then sticks with them even if they don't like the resulting sound is never going to have a happy outcome. I can't imagine many people are likely to do this, so Skol's 'cribsheet' seems like a good idea to me.

Seems like one of those 80/20 rules to me; the first 80% is fairly straightforward and where rules-of-thumb can be very useful, whereas the final 20% is where things get tricky and experience really counts. I don't think Skol was suggesting the document was a 100% solution.

[Dad3353]

Good evening, Skol...

Just thought I'd point out a 'typo'...

 SNARE: +4 dB @ 7 kHz +2 dB @ [b]100 kHz[/b]: increases snap and adds fullness to high-tuned snare

...my snare is not that high-tuned; this is for those with ears like bats, methinks..?

Otherwise, well done; as has been stated, a reasonable jump-off point, which will save time when mixing without previous experience. Nice.

[51m0n]

[quote name='Skol303' post='1347215' date='Aug 21 2011, 03:38 PM']^ Gotcha, and I couldn't agree more with either you or S1m0n. More than happy to bow to your superior know-how

Like I said, I wasn't suggesting that a list of frequencies should ever replace using your ears - otherwise the best studios would be run be deaf mathematicians... all I'm saying is that for me personally, it's certainly helped to know, for instance, that I can lose some of the 'boominess' from a typical kick drum by starting to tweak its frequency somewhere around [i]here[/i]; or I can usually make a snare sound more crisp by boosting it somewhere around [i]there[/i], so that's where I'll start twiddling.

Of course, it depends on the song in question and the context of everything else in the mix. But for me, it has helped to have a rough idea of where each instrument [i]usually[/i] lies in the frequency spectrum. Not as a replacement for using my ears, but just as a starting point to help me understand where different sounds might be sitting in the mix.

[b]You guys can probably reel off the typical frequency bands of instruments in your sleep, so for me it's just been a basic exercise of getting at least some of this understanding for myself. The rest, as you say, is going to be all about listening, listening and then listening some more![/b]

Anyway, you can count on me pestering you guys for advice when I get stuck! [/quote]

This is the crux of it, not really, well, kinda, maybe.

I could list a load of frequencies where stuff can happen, but I cant tell you where it really will happen on a given source.

I dont need to, neither do you.

You have to learn to use your tools to work it out.

Mixing is a jigsaw, you have a load of sounds and you want to ge tthem to sit together such that you can hear them all and there are no areas where one piece obscures another.

You have three ways to do it (off the top of my head).

1) Cut holes in the frequency spectrum to allow them to slot together
2) Change the envelope of the sound so you effectively move the energy in the time domain to a differnt point
3) Use the wonder that is stereo to make things sit alongside each other.

You are concentrating on 1.

In order to do this you need to compare two sounds and figure out a couple of things:-

1) where are the 'nasty' parts of the sound that [i]you[/i] dont want in this mix - use a paramteric eq, boost a band by 6 to 12 dB and then slide the frequency up and down until you hear something about the sound you dont like, then set the gain to 0dB, we will adjust the gain at this frequency later if necessary....

2) when they are both playing, does one overlaps the other in a detrimental way, if so choose which sound you want to hear in that area then go to the other sound and repeat 1 but listen for more build up in that area that is overcrowded, then cut the gain until the two sounds are no longer a mess there, and you can hear the one you want to.

3) when you have got riod of the build up then listen again in context and see if you can still hear that nasty part of the sound, if you can grab that gain control on the frequency in 1) and pull it down, dont bin it (the nature and timbre of a sound is so often actually a result of the nasty part of it - more than you will ever believe!)

The point of this is simple it doesnt matter what the freuquncies are!

Put this another way, you need know the following:-

20-40 Extreme sub bass, rarely actually detectable on hifis, so you can bin anything below 35Hz almost always and it will just make things sound better
40-80Hz Deep bass, that biiigg warm pillowy bass area, too much and things become undefined, too little and things will sound empty

80Hz Lowest area of a normal male voice

80-130Hz Bass, like old school bass, not super deep but you know, punchy and full of energy - too much will sound thuddy

130Hz to 500Hz Low mid, this is the killer, somewhere in here is all the mud and crud, somewhere in here is the warm and round, too much of this on a non-bass instrument (ie guitar) is going to cause you a world of hurt on the bass, too little and hese instrumetns will have no impact. B ottom end of the snare, and female vocals are in here somewhere.

500-1KHHz - there is presence in here, but too much is nasal

1KHz -3KHz Again presence, can sound scratchy and very very tiring, but is also where your ear is most effective (Fletcher Munsen old boy!)

3KHz-6KHz Lots of toppy sibilant presence on vocals is somewhere in here

6-10KHz de-essing territory (can go as low as 4KHz as it happens)

10-20KHz can we say bats? This is air, too much is tiring and distracting and will show up on hard tweeters as painful, too little and whaqtever else you do will sound muddy.

Right now you could have written out something along thiose lines with a little thought, play with some sound sources (say a drum kit recording on 5 or 6 tracks to start with) and play the game (my bro Rimskidog and I) have suggested with eq, see how these areas of the frequncies interact and how they sound when you boost and cut them.

Now forget it (its like jazz in this regard!!!!) because you can intuitively understand this stuff within a very very short time, it is not necessary or even helpful, to put numbers to frequencies to mix, it is only useful when talking about mixing, and that is the rub, you are talking about how to actually do it, and Rimskidog and I are telling the truth.

If I said to you, what kind of frequency range is upper mid, you could hazard a guess, so listen to the sounds and think "is the upper mid ok?" and then go there (mentally) and concentrate on it. It is that simple, you dont need to get tied up in the numbers for this. Use the other tool (eq) to make changes and really get to grips with the specifics of the area.

Same with compression!

As for stereo there needs to be the following in the center of 99.99% of mixes:-

Kick
Snare
Bass
Lead Vocal

Everything ( I mean EVERYTHING) else needs to get out of the way, nothing else matters.

By out of the way I really mean at least 70% panned, on a rock track 100% is fine.

Double up guitars to pann them left anf right (do multi track them, do not just duplicate the same track and pan it) - if that doesnt fit the situation then send a delay or reverb to the other side to widen the guitar.

I could bang on for hours about reverb vs delay for ambience, and other special effects, but the core skills to mix are really, frequencies (eq), envelopes (compression) and stereo space (panning). Get those right and you are 90% of the way there.

One last point, dont change it unless its a problem. Hardest thing in the world is knowing when a sound is right....

Edited August 22, 2011 by 51m0n

[51m0n]

I think a lot of this way of working eq comes from using real desks to mix back in the day, and on a real desk two things become apparent very soon:-

1) You never ever look at the frequncies when you turn a knob on a parametric eq, cos your hands are in the way

2) the frequencies printed on most desks are laughably innaccurate anyway, so they are only good to document a mix on that particular desk, they are useless for any other instance of that device even.

So I would put money that Rimskidog and I , having learnt to mix before the advent of computer DAW mixing are totally at home with the idea of listening to a mix as we mix rather than lloking at the screen - I still always turn the display off very regularly when mixing, I even keep the lights down low as I want as little visual info coming into my bonce as possible.

[Skol303]

^ Superb and detailed advice as always, 51m0n - and greatly appreciated!

So in essence: there's no point me getting too hung up on frequency ranges, because ultimately I'm going to have to treat each instrument in each mix in a different way... so I might as well start training my ear now, as that's what I'll end up using anyway. I think I get it now, cheers for bearing with me!

I actually had some proof of this over the weekend, when I was practicing with EQing. I did start by using some of the frequencies I'd posted in the document, but in the end I ended up resetting all the dials/faders and doing it all over again by ear!

I'm reasonably confident using things like stereo width and panning, but compression is another 'dark art' that I'll probably be asking for your advice on in due course...

Anyway, thanks again.

[51m0n]

[quote name='Skol303' post='1348067' date='Aug 22 2011, 03:18 PM']^ Superb and detailed advice as always, 51m0n - and greatly appreciated!

So in essence: there's no point me getting too hung up on frequency ranges, because ultimately I'm going to have to treat each instrument in each mix in a different way... so I might as well start training my ear now, as that's what I'll end up using anyway. I think I get it now, cheers for bearing with me!

I actually had some proof of this over the weekend, when I was practicing with EQing. I did start by using some of the frequencies I'd posted in the document, but in the end I ended up resetting all the dials/faders and doing it all over again by ear!

I'm reasonably confident using things like stereo width and panning, but compression is another 'dark art' that I'll probably be asking for your advice on in due course...

Anyway, thanks again.[/quote]

Yeah exactly!!!

Every single person on this forum could learn how to do this in a short time, it is not hard, and the best bit, the result is the right one, every time. It is very intuitive.

As soon as you start imposing someone elses findings from some other mix you are going to be going wrong.

Training your ears to do this is totally and utterly different from the ear training a musician needs. Rather than identifying intervals and chords you need to learn to identify regions of detrimental frequncy build up.

Of course you still ge to make all the wrong choices about which of any overlapping instrumetns is the dominant one in that part of the frequency range

One 'gotcha' that will get you time and again (still gets me), in learning to concentrate your attention on a very specific part of the mix in terms of frequency I find it becomes increasingly easy to put aside other areas of the mix, and sometimes you come back to your work only to hear all the bits that you never got round to fixing. Suddenly the mix you had a warm glow about has become an absolute mess.

Time to get the eqs out and cut some more crud!!!

Compression is a similar beast, a lot of people find the concept hard to grasp, and the tools hard to hear, and the available parameters too many and too varied with bizarre names and so decide that comnpression is the work of the devil and the best approach is to use either a cookie cutter set of parameters without really understanding what that should/could/will/did do. Its as bad to do this with compression as it is with eq, both will damage your mix equally!

Learn how to use a compressor, and then you dont worry about cookie cutter settings, you just tailor the sound to your needs for the mix.

Put it another way mixing is a bespoke activity, like tailoring, if its done right. Off the shelf never fits as well, never looks as good, but it is cheaper....

[Rimskidog]

[quote name='51m0n' post='1347823' date='Aug 22 2011, 10:25 AM']I think a lot of this way of working eq comes from using real desks to mix back in the day, and on a real desk two things become apparent very soon:-

1) You never ever look at the frequncies when you turn a knob on a parametric eq, cos your hands are in the way

2) the frequencies printed on most desks are laughably innaccurate anyway, so they are only good to document a mix on that particular desk, they are useless for any other instance of that device even.

So I would put money that Rimskidog and I , having learnt to mix before the advent of computer DAW mixing are totally at home with the idea of listening to a mix as we mix rather than lloking at the screen - I still always turn the display off very regularly when mixing, I even keep the lights down low as I want as little visual info coming into my bonce as possible.[/quote]

Now I just feel old

As ever though, great explanations from 51mon. More articulate than I could ever be.

[51m0n]

[quote name='Rimskidog' post='1348434' date='Aug 22 2011, 09:19 PM']Now I just feel old

As ever though, great explanations from 51mon. More articulate than I could ever be. [/quote]


Me too!

I feel better when I remind myself that 16 tracks (not 8) was the defacto standard in a decent small commercial studio when I started. It doesn't help a lot, but it kind of takes the edge off!

And thanks

[Skol303]

[quote name='51m0n' post='1348206' date='Aug 22 2011, 06:07 PM']One 'gotcha' that will get you time and again (still gets me), in learning to concentrate your attention on a very specific part of the mix in terms of frequency I find it becomes increasingly easy to put aside other areas of the mix, and sometimes you come back to your work only to hear all the bits that you never got round to fixing. Suddenly the mix you had a warm glow about has become an absolute mess.

Time to get the eqs out and cut some more crud!!![/quote]

^ Ha ha! Yeah, that's exactly where I'm at right now. I finish something and it sounds "ok", then I come back to it the next day and it sounds terrible. Practice makes perfect, and all that... so I guess I have a lot of practicing to do.

Cheers again for the top notch advice. You've definitely given me some 'clarity' in terms of how I should approach EQ'ing, which has been a great help.

Honestly, you and Rimskidog should a write a book on this stuff some day! "Never Mind the ****ocks: Here's the Basschat Guide to Mixing"

[quote name='51m0n' post='1347816' date='Aug 22 2011, 10:19 AM']Double up guitars to pann them left anf right (do multi track them, do not just duplicate the same track and pan it) - if that doesnt fit the situation then send a delay or reverb to the other side to widen the guitar.[/quote]

^ PS: I'm probably being very thick here... but what do you mean by "multi track" in this instance? I understand this in the context of working with two separate guitar tracks. But if I have only one guitar track, then the only option I can think of is to duplicate it and pan the original track left and the duplicate track right (both in mono).

Like I said, please explain further for the sake of my addled brain!

Edited August 23, 2011 by Skol303

[51m0n]

[quote name='Skol303' post='1348781' date='Aug 23 2011, 10:41 AM']^ Ha ha! Yeah, that's exactly where I'm at right now. I finish something and it sounds "ok", then I come back to it the next day and it sounds terrible. Practice makes perfect, and all that... so I guess I have a lot of practicing to do.

Cheers again for the top notch advice. You've definitely given me some 'clarity' in terms of how I should approach EQ'ing, which has been a great help.

Honestly, you and Rimskidog should a write a book on this stuff some day! "Never Mind the ****ocks: Here's the Basschat Guide to Mixing"



^ PS: I'm probably being very thick here... but what do you mean by "multi track" in this instance? I understand this in the context of working with two separate guitar tracks. But if I have only one guitar track, then the only option I can think of is to duplicate it and pan the original track left and the duplicate track right (both in mono).

Like I said, please explain further for the sake of my addled brain![/quote]

I really do mean multitrack, double the guitars, especially in rockier stuff, and hard pan them. Its so much the sound of modern rock that without it you are almost not in the same genre.

If you only have a single guitar source then you can do all sorts nasty tricks like send a slightly delayed signal to the other side, adding a spot of reverb or compression or extra dirt to it, or some kind of chorus/flange/modulation to it or the reverb or the delay, eq it differently over there, anything really just make it slightly different as well as slightly delayed. Or you can put some other instrument over there - Hammond is always a good one - to balance things out. Hell even an automatic double tracker vst is better than nothing!

If you get serious about this you can certainly use massive amounts of time copying the original guitar part and then editing it to be out ever so slightly with the original track, cut here, stretch there, chnage the levels here and there so the compression acts differently, make it sound real, and subsequently the two together will sound fatter. In this day and age nothing is unfair. This will work better than the simple delay trick, and give you ultimate control compared to an ADT effect.

If you just duplicate ithe two all you are doing is raising the volume of a track panned central. That is exactly what we are trying to avoid at all costs, its far too much competition with the lead vocal.

Get it right and you have: Wall of Guitar + Perfect Vocal = Mix WIN!

Other possiblities are to take this approach and in the heaviest choruses edit two more tracks up and place them at about 90% L and R and add them to the mix just for the biggest payoff choruses. Doesnt half add to the wall of sound if doen with care....

All from a single guita track.

Forget what the band line up is, forget what the band say go mad and produce a mix so devastatingly effective that they realise thats what they actually wanted ion the first place (HINT: this may only work sometimes!)

[Skol303]

[quote name='51m0n' post='1348967' date='Aug 23 2011, 01:20 PM']If you only have a single guitar source then you can do all sorts nasty tricks like send a slightly delayed signal to the other side, adding a spot of reverb or compression or extra dirt to it, or some kind of chorus/flange/modulation to it or the reverb or the delay, eq it differently over there, anything really just make it slightly different as well as slightly delayed. Or you can put some other instrument over there - Hammond is always a good one - to balance things out. Hell even an automatic double tracker vst is better than nothing!

If you get serious about this you can certainly use massive amounts of time copying the original guitar part and then editing it to be out ever so slightly with the original track, cut here, stretch there, chnage the levels here and there so the compression acts differently, make it sound real, and subsequently the two together will sound fatter. In this day and age nothing is unfair. This will work better than the simple delay trick, and give you ultimate control compared to an ADT effect.[/quote]

^ Gotcha, cheers 51m0n! (that's another beer I owe you...).

I have a track with a single guitar that I'm going to try this on. I think I'll test it out using delay/reverb/compression first (only because that seems like the easier solution), and then have a go at editing/cutting/stretching the original guitar part later on, once I've nailed the first.

[quote name='51m0n' post='1348967' date='Aug 23 2011, 01:20 PM']If you just duplicate ithe two all you are doing is raising the volume of a track panned central. That is exactly what we are trying to avoid at all costs, its far too much competition with the lead vocal.[/quote]

^ Cheers for pointing that out too. I'd have probably tried that otherwise and walked right into a new world of mixing pain!

[cheddatom]

I do the doubling thing quite a lot, just because we do a lot of "live" recordings so only one track of guitar available. I tend to just nudge one track slighty forward. I fnid that the one that comes in earlier tends to have more presence so I have to cut a bit of presence from the early track and boost it on the delayed one.

[Wil]

It's called the Haas effect I think, hearing the same guitar part with a very slight delay tricks the ear into hearing one big fat guitar part. I've been using it recently on some band demos too and it really does work a treat, using a 34ms delay with no feedback on the doubled guitar.

I've been experimenting with slow attack, fast release compression settings on the guitar too to give it a bit of a "pumping" sound - still refining the best way of achieving that but it seems to add a lot of whallop on our heavier sections.

[51m0n]

34ms is a long time in terms of this kind of processing...

Not wrong at all just long.

The human brain can detect time alingment differences between left and right ear down to around 3 or 4 ms

Edited August 23, 2011 by 51m0n

[BottomEndian]

[quote name='51m0n' post='1349058' date='Aug 23 2011, 03:10 PM']34ms is a long time in terms of this kind of processing...

Not wrong at all just long?[/quote]
Beat me to it. I believe the Haas effect ([i]per se[/i]... not to say you can't use a longer delay for a different effect) really only applies up to 30ms at the most. Beyond that, you may be pushing into the region where some people will hear it as a separate delay, while others may not.

Under 30ms, if you pan the delayed signal across the field, it should sound like a "widening" of the original source rather than a separate signal.

[cheddatom]

I tend to find phasing issues when doing this, so I play about with the delay between 3 and 20ms until I get a sound I like.

If I were to EQ or compress one of the tracks differently would that somehow stop the phasing issues?

[51m0n]

[quote name='cheddatom' post='1349073' date='Aug 23 2011, 03:27 PM']I tend to find phasing issues when doing this, so I play about with the delay between 3 and 20ms until I get a sound I like.

If I were to EQ or compress one of the tracks differently would that somehow stop the phasing issues?[/quote]


Ideally you wouldnt use a simple delay at all, because it will cause comb filtering and phase issues.

The editing the second track will significantly improve the situation. It is time consuming but it can certainly be done and really quickly in today's DAWs.

In place of that then anything else you can do to change the nature of that second track a bit (without turning into something pants in the mix of course) will really help.

Remember the ideal is a true multitrack. But even then you can treat the two differently to a certain extent.