Questions about leveling and EQing

[benherron.rrr]

I have been reading a few books about mixing lately. And there are a couple of rather simple things that I could use clearing up.

Having very little experience of mixing. I have been reading up alot about eqing, at first I thourght it was just used to add that little extra sweet ness to the track. but now I hear alot about Eqing being alot more to do with geling the tracks together as it would be.
I was reading how everything should have its own space in the spectrum to help bring out each element of the song, for example: as bass guitar and the kick drum hold down the bottom end of the mix they should be eqed and leveled to hold that range better? But how? does that mean giving a little boost to their home frequencies and cutting off the rest. basically giving each instrument there own frequency range and keeping everything else out of that? Also using cubase should I be using the graphic equalizer with what I can presume is 4 band (which I much perfer the look of) or should I be using the more detailed 30 band EQ with the sliders?

The other thing is about leveling. I usually have alot of changes going on in my songs from a vocal and an acoustic guitar changing to a full blown out guitar solo, obviously there is a change from the mellow acoustic to the distorted guitar and the levels change. I have noticed alot now how music generally stays the same volume throughout. I have tried matching up the levels, so the acoustic is playing at the same level as the electric. but I still seem to get the lack of volume. should I be matching everything up with the master buss volume? I genrally find it hard to keep consistancy. Like with vocals, im thinking sould I be using automation/envalope to boost those quite lines or is it a compression or EQ issue.

just thinking about these simple things just turns my brain to mush, as soon as I think i understand somthing I put it into practice and I can't seem to work with it. Im really beggining to get the tracking part down. now its the mixing thats dragging me down.

 

[miroslav]

...does that mean giving a little boost to their home frequencies and cutting off the rest. basically giving each instrument there own frequency range and keeping everything else out of that?

Yes...kinda...not always...sometimes you cut instead of boost...
...your ears are what tell you when and how much, don't think of it as any formula.

Also using cubase should I be using the graphic equalizer with what I can presume is 4 band (which I much perfer the look of) or should I be using the more detailed 30 band EQ with the sliders?

One is a graphic EQ (30 band) the other is a parametric...which means you can also control the width (Q) of the cutting/boosting frequency band.

I have noticed alot now how music generally stays the same volume throughout.

Yeah...and a lot of that music is tedious to listen to. You really want to have dynamics, differences in levels. Of course, you can still do a little level boost during mastering to bring things up to "competitive levels"...but you certainly don't want to adjust all your tracks/instruments/vocals to be the same level.

 

[bouldersoundguy]

There's really no objective right and wrong. It's all a matter of taste. Different people have different preferences for the sound of a finished mix, and different engineers have different techniques for getting there.

Probably the best thing you can do is get songs that sound good to you and play them on your monitoring system before a mix session. Take a break once in a while to listen to them again. That way you'll have a more concrete vision of your target sound. Then you can compare your mixes to the references and make specific adjustments to get closer to the target.

I find that I mix with eq. That is, I get one frequency range of a track right in the mix then eq the other frequency ranges to balance properly. This works well with things like bass, which has parts in the upper range that give it definition, and parts in the lower range that give it "weight". I'll get the heavy low end right to support the mix, then dial in the definition against the other stuff in that range.

With vocals it depends on how it was tracked. If the vocal was close miced (not typical of studio recording) I will compensate for proximity effect to get the voice sounding natural. Later I'll fine tune the vocal to the mix.

 

[Stylaproduction]

I am no expert, but i have had a good few years of experience in home recording.

As miroslav said it is often overlooked that eqing can be used to cut as well...which in turn can act like a boost in th remaining frequencies.
I too use cubase and am really not a fan of the graphic track eq there, i much prefer the waves plugins although there are others. the amount of bands i use generally depends on wat it is i eq, but im generally using like a 6 or 8 band graphic wen i really want to isolate and boost specific frequencies on like a guitar or synth. but for bass and kicks hats etc i usually use less

quote: does that mean giving a little boost to their home frequencies and cutting off the rest. basically giving each instrument there own frequency range and keeping everything else out of that?

this is not something i generally do but i think mixing is really about giving each sound its own space to breath in the frequency and in stereo space. although many sounds will share a frequency range. Each individual sound has its own 'sweet' resonant frequency.

 

[SouthSIDE Glen]

I was reading how everything should have its own space in the spectrum to help bring out each element of the song, for example: as bass guitar and the kick drum hold down the bottom end of the mix they should be eqed and leveled to hold that range better? But how? does that mean giving a little boost to their home frequencies and cutting off the rest. basically giving each instrument there own frequency range and keeping everything else out of that?

When people talk about each instrument having it's own range, that does not mean that that will be the only instrument has will occupy that range, with all other instruments being quiet in that range. One look at a frequency chart will show you how close to impossible that is for much of the spectrum, because most instruments have resonances and harmonics and such all over the spectrum. What it usually instead means is that your musical arrangement should decide the roles of each instrument and which part of the spectrum they should *dominate*, and that - *when necessary* - you should EQ to support that arrangement plan.

For something like kick and bass, this may not always be completly decided by the arrangement; the size and type of kick drum and the bass setup and technique can confuse the roles of the two within the spectrum. Or, if yuo havem say, two acoustic guitar parts or two rhythm git parts, they may occupy the same general spectrum space but you need to EQ them separately to help define them.

In such cases, listen to what you have on your recorded tracks and "see" with your ears what the leanings or inclinations of each track are. If it sounds like the kick really wants to punch low and the bass seems more melodic or clicky, then EQ to help them out in those respects, emphasizing the low kick and and the melodic definition of the bass. Or vice versa if it sounds like they want to go the other way.

For the guitars, usually one will sounds like it has a slightly lower character to it and one slightly brighter than the other. Rarely willthey sound identical. EQ to emphasize those differences, working "with the grain" in that regard instead of trying to force them the other way.

When EQing like these examples, rarely should you need an excessive amount of EQ either way, just a bit of helpful nudging is usually all you should need - unless there are some real issues with one or more of the tracks that need actual EQ repair. Often, a "differential EQ" technique works well. That is, instead of booust or cutting one trck excessively, boost one by just half the amount or so, and then cut the other one in the same place aby another small amount. The cumulative effect is one of a greater amount of separation between the two while simultaneously changeing each individual sound by only a small amount.

Also using cubase should I be using the graphic equalizer with what I can presume is 4 band (which I much perfer the look of) or should I be using the more detailed 30 band EQ with the sliders?

Should you use a chisel or a planer to smooth that piece of wood? Pick the tool that works best for you and your piece of wood. The only way you can find a favorite is through practice and experimentation.

The other thing is about leveling....I have tried matching up the levels, so the acoustic is playing at the same level as the electric. but I still seem to get the lack of volume.

As far as relative levels between the tracks, follow your ears and to hell with the meter levles. As far as overall mix volume, don't worry about it...yet. As long as your levles stay below clipping, you're OK. As far as the mix's final volume, worry about that in mastering after your mix is done, it's not your concern while mixing.

just thinking about these simple things just turns my brain to mush, as soon as I think i understand somthing I put it into practice and I can't seem to work with it.

It's not an easy job, for sure. But save your brain a bit and try not to think about it too hard. There are really only three things to think about: 1) How does what you have now sound?, 2) How do you want it to sound? and 3)What do you think you can do to get things from 1) to 2)? It's that simple...and that hard .

G.

 

[Chibi Nappa]

I was reading how everything should have its own space in the spectrum to help bring out each element of the song, for example: as bass guitar and the kick drum hold down the bottom end of the mix

Yeah, everything should have a purpose for being in the mix and a place to occupy in the mix. Just don't go thinking it's a 1:1 type thing. If it is the producer's intent, 4 different things could all be living together in the same frequency "spot". Those 4 things usually act as one whole in that case all fulfilling the same purpose. It's when you've got elements tying to "say" different things happening at the same time at the same frequency that you might run into trouble. Except when that's what you want (isn't art fun? ).

But right now and for the rest of your career put numbers and charts out of your head. You listen to the whole song. If you hear an "empty space", you can put something there. If you don't...maybe you should dial up a new sound that fits in a different space.

they should be eqed and leveled to hold that range better? But how? does that mean giving a little boost to their home frequencies and cutting off the rest. basically giving each instrument there own frequency range and keeping everything else out of that?

No! When recording, you choose a sound that either fits or it doesn't. Keep in mind that the notes of a part or the technique of a musician could make the same instrument tone work or not work. The arrangement and the performer are where you start. Not the amp settings/mic choice/midi patch/plugin effect. Don't think "this sound worked before, it will work again". Listen to how the part is fitting in while it is being performed. Choose sounds accordingly.

Anyway, the point is that everything is "home" and "holding it's range" before you even hit the record button.

Where does mix EQ come in? Well... how close did you come to your goal of recording parts exactly as they should be? Great but not perfect? Well, use as small of an amount of EQ as possible to nudge it over to "perfect".

How do you do that? The same way you knew the sounds were correct when you selected them for recording: You listen. Everything about recording and mixing starts and ends with listening. Got yourself great monitors and a great listening room? If not, get on that. You're lost otherwise no matter who you are.

Also using cubase should I be using the graphic equalizer with what I can presume is 4 band (which I much perfer the look of) or should I be using the more detailed 30 band EQ with the sliders?

Prefer the look, eh? Serious suggestion: Somewhere down the line get yourself a control surface so you can work with your monitor turned completely off. You'd be shocked how many EQ curves get set, how many notes are nudged, how many vocals are tuned, and how many dynamics are squashed because people liked or disliked the look of what was shown on the screen. I got rid of my computer screen and I'm happier for it.

Anyway, most people like to use a parametric EQ. Try 'em all. It's not wrong to use one over another.

The other thing is about leveling. I usually have alot of changes going on in my songs from a vocal and an acoustic guitar changing to a full blown out guitar solo, obviously there is a change from the mellow acoustic to the distorted guitar and the levels change.

This is a performance thing. You have to play the song with the volume changes/consistency that you want. If the dynamics are not as you want them after recording, you will not have good results trying to fix it up with mixing or plugins.

I have noticed alot now how music generally stays the same volume throughout. I have tried matching up the levels, so the acoustic is playing at the same level as the electric. but I still seem to get the lack of volume.

Read up on the loudness wars. There would be a volume change if not for the loudness wars. Most songs are written with intended volume changes. After all, an electric guitar just IS louder than an acoustic. But if you want your song to sound as loud as some other modern song for the first 2 seconds before the listener adjusts the volume to where he wants it anyway, then you can't have any volume change at all in your song. It's a long story. Read up on it.

Personally I hate the whole mess. Volume change is an important part of music. When you write sheet music, you notate the volume just as you do the notes, tempo, and time signature. Music sounds wrong and boring at a constant volume. Especially when it is apparent that the performer actually did scream louder or bash the drum harder only to have the result unnaturally evened out after the fact. The dude just screamed his ass off. He wanted to be loud. Let him be loud.

Let your songs be at an overall level much lower than what is being made today. Use the headroom for creative and intentional volume variation.

Like with vocals, im thinking sould I be using automation/envalope to boost those quite lines or is it a compression or EQ issue.

When you do have to fix unintended uneven volume (happens in vocals all the time), I find that fader automation usually sounds much better than compression. I use compression when I want to change the density and feel of a track.

just thinking about these simple things just turns my brain to mush, as soon as I think i understand somthing I put it into practice and I can't seem to work with it. Im really beggining to get the tracking part down. now its the mixing thats dragging me down.

Just remember that it is all simple stuff like you said. But it takes years of work to learn how to hear it. Spend as much time as you can with your mics and monitors (assuming you have monitors. Like I said above, you're dead in the water otherwise). It is unnatural to listen as obsessively as we have to. It comes slowly. But it comes.

As for the mixing: That should be relatively easy once you have tracking down. Maybe you haven't nailed tracking as well as you thought you did?

*Are you listening to your monitors, and only your monitors, to test the sound of the instrument you are about to record in context with the existing tracks before you press record?

*Are you keeping your max peaks at about -15 or lower?

*Are you recording in such a way that without touching a single fader or plugin, the song sounds just about done after recording the final track?

 

[benherron.rrr]

Thanks for the help guys.

My current set up is very small and just in my bedroom. which as far as my knowlage goes, it sounds quite alright. Its never going to be perfect but its as good as I can get at the moment. I have some monitors, but they are more on the cheap end, which for me is fine. A massive difference from when I was using computer speakers.
I am generally a big fan of getting it right from the start I hate the phrase "fix in the mix" so generally I try to get the best sound possible from the start, Even if it takes me a day to record a single guitar part.
I supose im having to take a giant leap backwards from how I have been taught through crappy education. like many people say, I have tended to add compression, EQ etc. because i think I should add it. not because Im listening. I've got a hell of a lot to learn, even taking a degree in the subject we seem to focus more on bussiness rather than on how to "produce music"
Its not alot but Im using a beringer bcf2000 which to be honest has really helped take my eyes of the screen. however Im generally more bothered about the sound then the numbers. Week after week I hear this guy arrogently talking about all the figures to get the best result, I hate it. If im adding a reverb even if it has a 2000ms pre delay I dont care if it sounds great. Makes me think Im on the right track. My main downfall is myself, Im a perfectionist I often focus on being that rather than the work.
Like most of us I want to introduce my music in the world being perfect. makes me hold back if I dont think is. In fact it is impossible to be perfect and perfection would lead to a bad song.

One major thing I need to do Is learn to listen, It fustrates me when I hear the words muddy etc. As I cant seem to hear things like that. and I cant really tell what I need to do and what I need to change. Maybe Im putting far too much pressure on myself, after all the 3 songs Ive got/getting tracked are the first 'real' mixes I have ever done.

 

[Farview]

I was reading how everything should have its own space in the spectrum to help bring out each element of the song, for example: as bass guitar and the kick drum hold down the bottom end of the mix they should be eqed and leveled to hold that range better? But how? does that mean giving a little boost to their home frequencies and cutting off the rest. basically giving each instrument there own frequency range and keeping everything else out of that?

Kind of. Every instrument gives off sound all the way up the spectrum. In your bass guitar example, low E is 41hz. But with most stringed instruments, most of the power of that note will be an octave up at 82hz. It will have some 'growl' at around 300hz, definition at 800hz and clarity at 3khz. So you have to choose what you want to accentuate or de-emphasize about the sound. When you EQ all the other instruments, you try to work around the emphasis of the bass guitar. Lather, rise, repeat.

Also using cubase should I be using the graphic equalizer with what I can presume is 4 band (which I much perfer the look of) or should I be using the more detailed 30 band EQ with the sliders?

I'm assuming the 4 band is a parametric EQ and the 30 band with the sliders is a graphic EQ. The parametric is more usefull most of the time.

The other thing is about leveling. I usually have alot of changes going on in my songs from a vocal and an acoustic guitar changing to a full blown out guitar solo, obviously there is a change from the mellow acoustic to the distorted guitar and the levels change. I have noticed alot now how music generally stays the same volume throughout. I have tried matching up the levels, so the acoustic is playing at the same level as the electric. but I still seem to get the lack of volume. should I be matching everything up with the master buss volume? I genrally find it hard to keep consistancy. Like with vocals, im thinking sould I be using automation/envalope to boost those quite lines or is it a compression or EQ issue.

The acoustic part should be quieter than the electric part. You are comparing your raw mix to completed masters. The mixes started out with the acoustic part at a lower volume and mastering leveled it out. It's a shame when that happens, but it does.

 

[BLP]

I'm asking this question in turns of understanding what I'm doing fully.

When people say boost or cut frequencies, that is all done with the EQ right?


How exactly is that happening when EQ's can have a wide variety of bands?

For example if someone says add about 4 kHz to add more presence to the vocal? What exactly are you adding it too on the EQ



I'm trying to grasp the technical side of all this. I'm steering away from the move knobs around till it sounds good and leaning towards knowing everything that I'm doing as I'm doing it

 

[SouthSIDE Glen]

For example if someone says add about 4 kHz to add more presence to the vocal? What exactly are you adding it too on the EQ?

You'd be doing exactly what it says; adding (aka boosting) via EQ at or around 4kHz to the vocal track. Think of it as turning up the volume, but just turning it up for a very narrow band of frequencies, and not the whole signal.

BTW, I know you're just giving an example, but be careful what you boost around 4k; that is a very busy frequency for a whole busload of instruments and can be very easy to overload in the whole mix, giving you a very tinny, midrangey sound. It can help vocal recognition for sure, but sometimes it may help to cut around that frequency range for other instruments to make room for the vocals instead of boosting the vocals. It all depends upon exactly what material you have in front of you, of course, but it's probably a good idea to keep in mind that that's a good example of the difference between a rule and a rule of thumb.

G.

 

[PDP]

Low E is 41hz. But with most stringed instruments, most of the power of that note will be an octave up at 82hz. It will have some 'growl' at around 300hz, definition at 800hz and clarity at 3khz. So you have to choose what you want to accentuate or de-emphasize about the sound.

I've been waiting to bring this up. I use a chart of frequency in musical notes, to learn what notes correspond to what frequencies and at what octaves. I read producers do this so I started to do it, but I havent seen it mentioned here until now.

I believe that if you sharply boost odd frequencies (relative to the key) the mix can actually sound out of tune. Is that possible?

 

[mixsit]

..I believe that if you sharply boost odd frequencies (relative to the key) the mix can actually sound out of tune. Is that possible?

I would certianly jump on frequencies that stick out rather high sometimes, and it does 'clean things up. But I never thought of it in reverse.
Snare ring can come off being way out of place.

 

[BLP]

You'd be doing exactly what it says; adding (aka boosting) via EQ at or around 4kHz to the vocal track. Think of it as turning up the volume, but just turning it up for a very narrow band of frequencies, and not the whole signal.

BTW, I know you're just giving an example, but be careful what you boost around 4k; that is a very busy frequency for a whole busload of instruments and can be very easy to overload in the whole mix, giving you a very tinny, midrangey sound. It can help vocal recognition for sure, but sometimes it may help to cut around that frequency range for other instruments to make room for the vocals instead of boosting the vocals. It all depends upon exactly what material you have in front of you, of course, but it's probably a good idea to keep in mind that that's a good example of the difference between a rule and a rule of thumb.

G.

Ok, but if there is multiple bands which one am I actually going to move?

I can have 10, 20, or 30 bands at a time

I realise now the whole time I'm using EQ I was only moving the "gain" on my 3 band EQ and never the actual frequencies.


honest mistake

 

[RAMI]

Ok, but if there is multiple bands which one am I actually going to move?

For your example, you'd move the band that's marked "4k".

And, as a side note, EQ is usually more effective when it's used to cut un-wanted frequencies, as opposed to boosting.

 

[dintymoore]

Here's a shot from the StudioEQ on Cubase 4:

1. turn on the band - this is a 4 band parametric eq

2. set the gain - here it's set for 3, which means "3 db boost"

3. set the frequency - here it's set to 4000, which is 4k (the k is "thousands")

So that's a 3 db boost at 4k. The Q-factor is how "wide" the boost is.

Now you have some ammo to pick up chicks with at that society party on Saturday.

Cubase 4 also has the GEQ-10 and GEQ-30 which are graphic eq's.

My take on eq - it always fucks up the sound so it's a trade off. It makes the sound less pure and less like real life. I heard some say it's the work of Satan and there's something to that statement! I always aim to use none and my proudest recordings have no eq. I have lots and lots of tracks with no eq. If you don't hear it don't use it!

But I do use it (all the time) and can say that almost all the eq's I've heard are horrible, I like my $2000 Summit tube eq and I growing up liked a friend's 40 ch Neve eq. Hate to put it that way but in general I haven't liked too many eq's. The eq on Cubase is ok sometimes, sometimes not...

I saw Phil Collins on Oprah (no I'm not) in the 80's and he was asked about how he sang and he said "I just try to make a nice sound".

 

[BLP]

Here's a shot from the StudioEQ on Cubase 4:

1. turn on the band - this is a 4 band parametric eq

2. set the gain - here it's set for 3, which means "3 db boost"

3. set the frequency - here it's set to 4000, which is 4k (the k is "thousands")

So that's a 3 db boost at 4k. The Q-factor is how "wide" the boost is.

Now you have some ammo to pick up chicks with at that society party on Saturday.

Cubase 4 also has the GEQ-10 and GEQ-10 which are graphic eq's.

My take on eq - it always fucks up the sound so it's a trade off. It makes the sound less pure and less like real life. I heard some say it's the work of Satan and there's something to that statement! I always aim to use none and my proudest recordings have no eq. I have lots and lots of tracks with no eq. If you don't hear it don't use it!

But I do use it (all the time) and can say that almost all the eq's I've heard are horrible, I like my $2000 Summit tube eq and I growing up liked a friend's 40 ch Neve eq. Hate to put it that way but in general I haven't liked too many eq's. The eq on Cubase is ok sometimes, sometimes not...

I saw Phil Collins on Oprah (no I'm not) in the 80's and he was asked about how he sang and he said "I just try to make a nice sound".

Ok so when they say "boost around 4k" you are literally just increasing the "gain" of the 4k frequency or whatever band you dialed in a 4k frequency? And when they say cut 250hz you go to the band that is in the 250z frequency and bring the gain down a few?


I was under the assumption that the terminology meant something way different


sounds simple enough!

 

[miroslav]

Right...that's how boost/cut works for a given band.

Keep in mind that there are different shapes to EQ curves for a given EQ plug (or hardware EQ)....bell, shelving...

Some are fixed, which is what most graphic EQs have...but the ones that have a Q option (parametric EQ) allow you to narrow or expand the width of that bell curve...so you control how much OTHER frequncies are being touched by the curve of a chosen CENTER frequency....wide reaches out to more frequencies on either side of the chosen center frequency...and narrow is, well...you get it.

Usually the very far end (low and high) frequency band options are shelving (half a bell). The low end drops off with a left shelf and the high end drops off with a right shelf.

 

[bouldersoundguy]

So that's a 3 db boost at 4k. The Q-factor is how "wide" the boost is.

To be picky, "Q" is a measure of the narrowness of the filter. A higher Q is a narrower filter.

 

[Eric Peterson]

Some damn good advice in this thread...