EQ Frequency

[peaka]

Hi all,

The two things that always seems to come up when discussing a good mix is EQ and compression. I understand that this is not the be all and end all of a good mix but I'm sure all of you will agree they're considered in the mixing process.

In this thread I'd like to talk strictly about eq'ing and techniques to find the right frequency to cut or boost. Yes, I do agree that that the most reliable source is your ears, but when you have 30-40 tracks laid out, it's sometimes hard to see the forest from the trees. Any tips on how to find the right frequency to fit a track into a mix? Is there any specific sound that you guys look for when sweeping across the frequency spectrum on your eqs?

Or better yet, please share your methods of how you eq during the mixing process to make all your tracks sound like 1 cohesive song.

Thanks guys

[Simulant]

345hz is the correct frequency to boost.

[alphacat]

1) Key of the song are important. You would EQ a C2 differently than an F3, for instance.
2) Boost wide, cut narrow.
3) It is usually better to cut than boost.

That's about what I got.

[SunkLo]

Well the conundrum here is that in order to make intelligent choices about mixing, you need to hear a track within context. But having everything playing at once is overwhelming and compromises your ability to pinpoint problems.

In a mix I start with the most important elements and set their levels. If you've already got issues with frequency clashes with only 3-4 faders pulled up, you're in for a long haul if your project has 30-40 tracks. At this point subtractive EQ is the technique to use. Trying to fill in empty space without the rest of the project playing is just gonna cause you to chase your tail. Once you've got the foundation of the track solid, it's a matter of bringing up the next important parts. Again, get the levels near perfect first. You can get a great mix without EQ or compression. You can't get close without getting those faders in the right spot.

Keep doing levels and then subtractive EQ until you've got the whole project playing and there's nothing obviously clashing. At this point you're really the majority of the way there. Now is the time to listen to the frequency spectrum of the whole project and decide what needs to come up or down. Cuts and boosts need to be applied intelligently. If you're looking for more top end, boosting it in your bass is gonna be a bit dim. Obviously if the frequency balance of any one element is skewed within the context of the whole project, that will need to be adjusted.

Now the mix should be in pretty mint condition. Most likely some compression will be applied at some point and the EQ will need to be tweaked a bit to compensate. Your compression shouldn't affect the levels since you should be matching the apparent loudness with the pre-compressed signal. The only difference should be dynamics, not level. You can also replace some EQing with multiband compression if it's applicable, to get more control over a track. When in doubt, go without. Don't be one of those twats with a multiband over their master buss and mixes that fatigue your ears after 2 minutes of listening.

You can write down the levels and drop all the faders. Then do the mix again a few times. The fader levels you write down should float around an average point where they sound optimal.

If your mix sounds like garbage right now it's most certainly because your original track was shit to begin with.

[peaka]

1) Key of the song are important. You would EQ a C2 differently than an F3, for instance.
2) Boost wide, cut narrow.
3) It is usually better to cut than boost.

That's about what I got.

But what you said speaks volumes
point # 2 is something I wouldn't have thought of but is a VERY valid point.

[peaka]

Well the conundrum here is that in order to make intelligent choices about mixing, you need to hear a track within context. But having everything playing at once is overwhelming and compromises your ability to pinpoint problems.

In a mix I start with the most important elements and set their levels. If you've already got issues with frequency clashes with only 3-4 faders pulled up, you're in for a long haul if your project has 30-40 tracks. At this point subtractive EQ is the technique to use. Trying to fill in empty space without the rest of the project playing is just gonna cause you to chase your tail. Once you've got the foundation of the track solid, it's a matter of bringing up the next important parts. Again, get the levels near perfect first. You can get a great mix without EQ or compression. You can't get close without getting those faders in the right spot.

Keep doing levels and then subtractive EQ until you've got the whole project playing and there's nothing obviously clashing. At this point you're really the majority of the way there. Now is the time to listen to the frequency spectrum of the whole project and decide what needs to come up or down. Cuts and boosts need to be applied intelligently. If you're looking for more top end, boosting it in your bass is gonna be a bit dim. Obviously if the frequency balance of any one element is skewed within the context of the whole project, that will need to be adjusted.

Now the mix should be in pretty mint condition. Most likely some compression will be applied at some point and the EQ will need to be tweaked a bit to compensate. Your compression shouldn't affect the levels since you should be matching the apparent loudness with the pre-compressed signal. The only difference should be dynamics, not level. You can also replace some EQing with multiband compression if it's applicable, to get more control over a track. When in doubt, go without. Don't be one of those twats with a multiband over their master buss and mixes that fatigue your ears after 2 minutes of listening.

You can write down the levels and drop all the faders. Then do the mix again a few times. The fader levels you write down should float around an average point where they sound optimal.

If your mix sounds like garbage right now it's most certainly because your original track was shit to begin with.

Solid advice dude. Thanks

[claudedefaren]

Don't thunk in terms of "frequency areas" like "warmth" and "clarity" and "snap" and "mud" and all that. Use your ears. Be on the lookout for ringing freqs in the highmids (like 2000 to 8000ish) when soloing an instrument, and bell them down with a super narrow curve. As alphacat said, boost wide, cut narrow. Don't be afraid to boost. Use a spectrum analyzer on every track to help find ringing partials that make your ears hurt. They will be slightly obvious to your ears, but clear as day with an analyzer. But most importantly, as Simulant said, 345hz is the correct frequency to boost.

[claudedefaren]

Oh, and only do as much as you NEED to. Don't eq for the sakr of eqing. Try dialing back any dips you've made and see how little of a dip you truly need to make in order to fix the tone.

[Genevieve]

1) Key of the song are important. You would EQ a C2 differently than an F3, for instance.
2) Boost wide, cut narrow.
3) It is usually better to cut than boost.

That's about what I got.

Boosting wide, sure. And narrow boosts can be a creative tool (only use it when you know what you're doing).

But with cutting, I don't really agree completely. He said, while going overboard with the narrow cutting, that it can affect the overal tonal quality of the sound... but that's bad how? When I hear a kick and I think it's a little muddy, I'm not going to surgically remove one or two frequencies. I will gradually remove some frequencies between 200 - 500 hz to make it a little less dark and make the highs more sparkly.

I know he added a disclaimer that it all depends on what youré hearing, but I feel he generalized too much with that one.

[Artie_Fufkin]

1) Key of the song are important. You would EQ a C2 differently than an F3, for instance.

Could you be more specific? What, do you boost the root? Boost the frequencies that would be accidentals on percussion?

I haven't gotten super advanced with my mixdowns yet, but here's what I do.
I try to get each individual sound sounding how I want them. If it's supposed to be a huge sound, you try to start off with a full sound and then make cuts if it's too much and you need to fit other stuff in. Mostly I just use highpass and sometimes low pass filters on tracks to create room, focusing mostly on the low end. If you want your music to be strong and have an impact, you have to manage the low end carefully. I will put a low pass filter on the master and check how everything sounds together. I'll push the cutoff higher and higher, checking where the snare and other stuff comes in frequency-wise. Sometimes I'll try bandpass and highpass. Sometimes it helps me find things to work on.

Note that it might cause less headaches for you to just arrange the elements of your track better than to fiddle around with eq and compression to try and make things play nicely together. Consider getting rid of unnecessary instruments or notes if your mix is too crowded and busy and you're having trouble eq'ing.

And check out therecordingrevolution.com There's some great tutorials and tips.

[alphacat]

^ For example; the "center" freq. of A4 is 440Hz; the "center" freq. of D4 is 293Hz. If your riff progresses only in A4, then the EQ on that riff doesn't need to change. If, on the other hand, it alternates between 440 and 293, if you leave the EQ alone one of the notes (usually the higher one) will sound louder, but that's only because it's sonically 'centered' in the frequencies reaching your ears, while the same EQ position on the lower D will affect the sound more to one side of the spectrum [where it has frequencies in common with the A.] Kind of like a Venn diagram in a way...

I feel like this is a terrible explanation though. Someone else can probably explain it more clearly.

[SunkLo]

That's a pretty good way to put it. You can think of EQ boosts like a spotlight, whatever's right in the center of the spotlight will be the brightest. As instruments move between notes, their relation to the EQ changes. It's like walking between streetlamps. The parts that are highlighted change as you move around.

A boost on guitar in an arbitrary frequency range might sound good, but if you transpose the guitar up an octave it might sound out of place. Sometimes an EQ setting can be perfect for one section of a track, but fall apart when it switches to something else in another section. Thankfully most of us work in the box, so you can just automate the EQ accordingly as the song changes. It's just like automating levels between sections. Ride those faders and controls, and your mix will benefit.

[alphacat]

Yeah, and this is where decent spectral analysis can help you figure out what's really going on too.

[peaka]

A common practice used by many seem to be boosting high and sweeping across the freq ranges to see which timbres to remove.

My question is,is there a specific sound you're listening for when you're sweeping across the frequency spectrum with your eq to find freq to cut?

[Samuel_L_Damnson]

IF you do that you should hear a whistling resonant irritating noise at some point which you then can cut out if you fancy.

[claudedefaren]

I've always hated that method. you should be able to hear the problem spots without doing that resonant sweep. Often times if you do the sweep you'll notice wherever you sweep will resonate and ring.. Well, duh!

[alphacat]

A common practice used by many seem to be boosting high and sweeping across the freq ranges to see which timbres to remove.

My question is,is there a specific sound you're listening for when you're sweeping across the frequency spectrum with your eq to find freq to cut?

When certain areas of the spectrum have more going on than others - like the kick and bass[es] duking it out in the low end - and things get lost, muddy, or distorted, that's where you want to go in and carve out a little room for some things to co-exist or bring up other parts of the spectrum that lack presence or are just plain old too quiet. It's really about clarity and balance for the most part.

With most of our gear and software designed to out insanely massive low end, boosting the highs to offset this is an obvious response. I think proper gain staging, however, should come first; it's very tempting for many to work in a low db overhead mix, maybe only -5 or -8 db because things just aren't pumping/banging enough, but this is why it's better to mix at -12, -15, -24 (proper monitoring/signal chain is key here too).

[SunkLo]

^All about grabbing some nice meters and calibrating them.

Actually I'd love to have a monitor dedicated to meters. I want these above my main monitor:






Whoah every thread turns to the gain staging thread.


As for EQ,
Map a knob to two different EQs' frequency controls.
Map another knob to filter gain, with the second EQ's mapping inverted.
Twiddle like you're playing with an etch-a-sketch until it sits nice.
Tweak the gain a bit individually for each filter.
Profit

[alphacat]

Whoah every thread turns to the gain staging thread.

Pretty much. This also explains why there's a whole class of old school pro (Bruce Swedien, etc.) that never even touch compression and whose use of EQ tends toward the minimal: their signal is so perfectly amped and clean going into an environment with so much headroom (relative to what we deal with) that it doesn't even matter.

[SunkLo]

They're also recording with gorgeous mics into gorgeous pres into gorgeous convertors. A well-recorded track on high end gear pretty much mixes itself.

But yeah compression is more valuable as a creative tool to me. I'm rarely compressing a lot for dynamic range. Ideally you'd be most of the way there using only levels and a touch of EQ. Doesn't necessarily apply as much with electronic music, but it's a good principle to keep in mind.