Mixing down using EQs

Hey,

So i'm aware it's an obvious/necessary practice to use Eqs (I use Ableton) on each sound to sort of sculpt the frequencies so that things don't overlap/you can make cuts/boosts to get things to complement and sit well together.

My problem is that I EQ each channel which tends to have one sound on each channel. So I sort of sculpt one sound via EQ whilst listening to the other channel playing audio...BUT...I'm not obviously seeing the EQ frequencies from both sounds together at the same time when referencing two/multiple sounds. I'd like to be able to do this. I'd like to be able to visually be able to say - ''yes, I can see how the frequencies overlap or clash and I will cut here/here to make the change required''.

How do you guys do this?

I realise I can just put an EQ on the master channel and see how both sounds look together but I want to edit each sound individually whilst seeing the impact on surrounding sounds.

Also, more of a general question, when looking at the frequencies of a sound, let's say a snare...there are multiple dominant peaks right. All of these are required to some extent I think but if you were to pick one of those dominant peaks when deciding where the 'core' of a sound is or where you really can't cut away at, where is that? How can you tell visually what is the essential elements of a sound?

I hope any of this makes sense.

Poly

Only do minor EQ to taste while building tracks, use Pan more to find where things should sit. Only use EQ when you're nearly done and moving onto proper mixing stage as you can usually tell at this stage when tracks are starting to step on each others toes frequency wise.

Try and use your ears more than you eyes, it's too easy to get bogged down with numbers and what should be done especially in the box. In the end, it's only ears that will witness the finished product so it does not matter how you got it to that stage as long as it sounds good. If some tracks do not work on paper but are sounding great then fuck it, no engineer can tell you it's wrong.

This is not fact by the way, just my opinion innit ;)

Thank you! Yeah, I think it's tempting to almost write down all the frequencies in each sound to then sort of sculpt that way but I trust my ears more than the numbers really.

I have this weird habbit of sort of mixing down during the creative process as it means I have a bed of complementary sounds which I'm ready to keep building upon.

That way I can open the session and keep adding...if that makes sense.

I think I need to use transient designers more too...rather than savagely boosting lows or highs with EQ to make a good sound great. Think are some free Max ones in Ableton.

Thanks again dude. Appreciate the wisdom!

  Quote

So i'm aware it's an obvious/necessary practice to use Eqs (I use Ableton) on each sound to sort of sculpt the frequencies so that things don't overlap/you can make cuts/boosts to get things to complement and sit well together.

 

Not really obvious or necessary if the right sounds are chosen to mix together. Minimizing the amount of needed EQing will generally give better results.

  Quote

My problem is that I EQ each channel which tends to have one sound on each channel. So I sort of sculpt one sound via EQ whilst listening to the other channel playing audio...BUT...I'm not obviously seeing the EQ frequencies from both sounds together at the same time when referencing two/multiple sounds. I'd like to be able to do this. I'd like to be able to visually be able to say - ''yes, I can see how the frequencies overlap or clash and I will cut here/here to make the change required''.

 

How do you guys do this?

 

One way to do it is to bus two or more tracks to an aux with a frequency analyzer on it to visually see what you're EQing is doing to the overall mix of these two or more tracks.

  Quote

I realise I can just put an EQ on the master channel and see how both sounds look together but I want to edit each sound individually whilst seeing the impact on surrounding sounds.

 

I don't understand your statement, but it might be because I'm not familiar with Ableton. If you have a frequency analyzer on your master, you can see the frequency changes happening to your whole mix when you are changing the EQ of certain tracks. In my view, that would show you the impact on the surrounding sounds.

  Quote

Also, more of a general question, when looking at the frequencies of a sound, let's say a snare...there are multiple dominant peaks right. All of these are required to some extent I think but if you were to pick one of those dominant peaks when deciding where the 'core' of a sound is or where you really can't cut away at, where is that? How can you tell visually what is the essential elements of a sound?

I would choose a snare-type sound that fits in the mix instead of shoehorning a snare that doesn't fit in the mix with EQ.

However, to answer your question, there's no clear indication of what are the core frequencies of a snare since there are so many different snare sounds. I'm sure there's a ballpark area of frequencies and transients that gives that snare-type sound, but I don't know the nitty gritty details of it.

  On 10/24/2016 at 2:49 PM, paranerd said:

 

  Quote

So i'm aware it's an obvious/necessary practice to use Eqs (I use Ableton) on each sound to sort of sculpt the frequencies so that things don't overlap/you can make cuts/boosts to get things to complement and sit well together.

 

Not really obvious or necessary if the right sounds are chosen to mix together. Minimizing the amount of needed EQing will generally give better results.

 

  Quote

My problem is that I EQ each channel which tends to have one sound on each channel. So I sort of sculpt one sound via EQ whilst listening to the other channel playing audio...BUT...I'm not obviously seeing the EQ frequencies from both sounds together at the same time when referencing two/multiple sounds. I'd like to be able to do this. I'd like to be able to visually be able to say - ''yes, I can see how the frequencies overlap or clash and I will cut here/here to make the change required''.

 

How do you guys do this?

 

One way to do it is to bus two or more tracks to an aux with a frequency analyzer on it to visually see what you're EQing is doing to the overall mix of these two or more tracks.

 

  Quote

I realise I can just put an EQ on the master channel and see how both sounds look together but I want to edit each sound individually whilst seeing the impact on surrounding sounds.

 

I don't understand your statement, but it might be because I'm not familiar with Ableton. If you have a frequency analyzer on your master, you can see the frequency changes happening to your whole mix when you are changing the EQ of certain tracks. In my view, that would show you the impact on the surrounding sounds.

 

 

  Quote

Also, more of a general question, when looking at the frequencies of a sound, let's say a snare...there are multiple dominant peaks right. All of these are required to some extent I think but if you were to pick one of those dominant peaks when deciding where the 'core' of a sound is or where you really can't cut away at, where is that? How can you tell visually what is the essential elements of a sound?

I would choose a snare-type sound that fits in the mix instead of shoehorning a snare that doesn't fit in the mix with EQ.

 

However, to answer your question, there's no clear indication of what are the core frequencies of a snare since there are so many different snare sounds. I'm sure there's a ballpark area of frequencies and transients that gives that snare-type sound, but I don't know the nitty gritty details of it.

 

Thanks man.

I think my point about not being to visually see things was Ableton related yes.

If I was to stick an analyser on a group bus to see, for instance, the freq content of a kick and snare together, I would have to actually click that channel to see the spectrum but when I then click back to an individual track, for instance, the kick in isolation, and change EQ there, I can nolonger see the bus channel spectrum at the same time...so I can't sort of live edit it and see the overall impact in real time...I have to keep clicking back and forth. 

This can maybe be solved by using a dual screen? 

You could download voxengo's span: http://www.voxengo.com/product/span/

and keep the plugin window open while you select different tracks.

It's free!

Izotope's new masking meter looks pretty swell:

Don't use it myself, but in Renoise you can set-up/choose multiple tracks to a/b in the scopes;

Dunno if span would allow multiple inputs, to compare similarly otherwise having multiple instances open could work?

If you're feeling flash Blue Cat have a plugin that does it (for $99):

or for $66 is MMultiAnalyzer (often for sale at half price)

of which you can actually identify collisions in the frequency spectrum:

Cheers. Went for Voxengo Span.

As posted above, there are EQ's that help you do this. But a surgical approach to managing frequency content is generally a bad idea. For bass and static, short sounds like drums it's helpful to make sharp cuts and mold things into the mix based on what you hear, but as soon as you go into more complex sounds you're going to run into problems. It's very easy to kill a mix with drastic EQ adjustments in an endless pursuit to make things "fit". Melodical instruments move around over a wider spectrum, and you'll generally find that, besides simple treble adjustment, it's hard to treat the instrument in a way that gets you a consistent and convincing sound out of the whole sequence. Complex sounds have frequency content all across the spectrum and can be quite disorienting with their absence of obvious peaks. Unless you've got a good ear you're never quite sure what you're actually affecting with your cuts. It's a well known fact that we tend to underestimate our adjustments when following visual cues so you generally make more extreme cuts and boosts when looking at a spectrum analyzer or paragraphic equalizer. It's important to use your ears and try to understand how things should fit together and not just go for the "big offenders" on the spectrum analyzer. You don't need to do a lot to balance the energy of the spectrum out. I really recommend using EQ's with as few visual cues as possible, preferably with classic style knobs that have a spectrum range in text and then you have to do it by ear. This forces you to listen more closely to what you're actually doing to the sound, and you're very likely to stumble into the tried and true practice of wider, gentler cuts or boosts that are more sensible and sound better than the paragraphic scalpel.

I don't make a ton of music, but generally the only thing that needs to be cleaned up is the low end to make space for the bass/kick/snare. otherwise it's a lot about getting the levels right for each channel, possibly altering balance.

I was listening to syro recently and I think it's good for ideas for mixing. He does a lot of little tricks on that record.

You are all amazingly articulate wonderful people. 

Thank you!! 

The thing with me is Kick/bass - I just hate it when I realise I've lost the kick and then I end up doing the surgical EQ thing to save the kick and then I realise I've fucked it.

I also have pretty much always done side chain compression on bass to get the kick to poke through but I'd prefer not to have to do that at all but it seems to work best for me. Just depressing when it's just like...ah i'll just stick another compressor on this bass to get the kick to come through. Then when you a/b it it's like....''oh I wish I didn't have to do that''

I'm a noob.

It's kinda long but I really enjoyed this series with Steve Duda where among many other things he discusses mixing. He generally takes a simple approach and works with the old cliche of first turning everything way down before even thinking about eq and compression.

For a long time I did the same thing of putting an eq on every channel but now I try to work the other way around and first decide if it even needs anything.

And especially if you're starting out I'd consider just using a normal sidechained compressor to bring out the kick through the bass rather than multiband but of course ymmv.

  On 10/24/2016 at 8:03 PM, Polytrix said:

You are all amazingly articulate wonderful people. 

 

Thank you!! 

 

The thing with me is Kick/bass - I just hate it when I realise I've lost the kick and then I end up doing the surgical EQ thing to save the kick and then I realise I've fucked it.

 

I also have pretty much always done side chain compression on bass to get the kick to poke through but I'd prefer not to have to do that at all but it seems to work best for me. Just depressing when it's just like...ah i'll just stick another compressor on this bass to get the kick to come through. Then when you a/b it it's like....''oh I wish I didn't have to do that''

 

I'm a noob.

You can probably tune the kick higher and leave the bass lower or vice versa. Can even add a mid-tone to the bass line if you want it to standout more. Depending on what you're using to listen to the tunes it's likely your bass stands out a lot more than you think, and doesn't need to be as loud. These are at least thing I've encountered. My monitors are very flat and do not accurately express the strength of my lower end sounds. something for which I have to compensate when I'm mixing a track.

Yeah Chim. 

It's easy to take away what was interesting about a sound without realising it. Compartmentalising every sound can sterilise a mix. The more interesting things happen when you meld sound together and you can't tell when one starts and one ends. As long as it does not sound mush of course. Character is being lost by endless ways to examine music on the screen. Of course, these things are valuable in moderation but electronic music is not about being a good "producer" it's still music and music is the best when it has feel.

Imagine it's like furnishing your room, you don't want to just go to Ikea and fill it full of souless square things, you want to put your stamp on it and add things and get the room feeling like home with your personality. Just don't get like a messy hoarder! It's a balance between fairly tidy but some worn in grime and life.

(what a load of shit^^^ ha)

Mixing 101:

1. Decide what "job" each instrument is going to do (ie. is the kick the bassiest thing or the bass synth? Filter accordingly. This changes according to style/genre)

2. Cut low frequencies if they're not bass instruments.

3. If two instruments share a frequency range, then pan them on opposite sides.

4. Filter out resonant frequencies.

a lot of good points here. one thing i like to do is try to find the one or two areas in the spectrum that really represent the instrument and push those just slightly, maybe 2 to at most 4 db and then bring the whole instrument down a bit. this way the loudest part of the sound fills its own narrow range. 

i'd agree with the people that say it starts in the arrangement. if you have several sounds that are really clashing and panning/light eq isn't helping, you might want to question if you should try rewriting the part. it could be as easy as just doing some kind of A-B back and forth between the instruments i.e. one instrument plays for one or two bars and then the next one. 

you spent how much on the multianalyzer??? I could've coded the same sort of thing in JS for you

  On 10/24/2016 at 4:09 PM, mcbpete said:

If you're feeling flash Blue Cat have a plugin that does it (for $99):

 

 

or for $66 is MMultiAnalyzer (often for sale at half price)

 

 

of which you can actually identify collisions in the frequency spectrum:

 

That looks pretty nifty

  On 10/26/2016 at 4:08 AM, Mesh Gear Fox said:

i figured i'd google this so i have the answer instead of just asking itt..

 

if you're using ableton and you want to have the plug windows stay open, untick 'auto hide plugins' under look / feel in ableton preferences. so you have to close each one manually all the time but i think i prefer it like that anyhow.

Yeah but this only works on vst/au plugins and not ableton's native stuff, right?

never use eqs on anything unless you are making frequencies louder. use a limiter on each track, follwed by an eq boosting the frequencies you want to be loud, followed by another limiter.

and a compressor, eq, compressor, and limiter on the master.

Yeah, depending on the melody and FX you're using you can end up with some horribly loud resonance at certain frequencies. Being able to roll those back is one of the main functions of EQ. No point in losing your dynamic range on the rest of it, unless you're going for a hyper-compressed sound in the first place.

Seriously though, visual aids are great, but your best friend in EQing is your ears. The spectrum on one eq plus listening to the other track is usually enough to get isolation.