Splitting the Mix Bus into different frequency ranges

Hello, guys

So, today I wanna talk about something I've been thinking for a couple weeks.

The goal I wanna achieve is to be able to split my Mix Bus into 6 aux tracks each with its own frequency range (Sub Bass Range, Bass Range, Lower Midrange, Midrange, Upper Midrange and finally Presence.

I have a Mix Bus towards which all my Track Stacks and Effect Busses are routed, so that all my composition stuff goes there. Said Mix Bus is then routed to my Master Bus for the final mastering stage. I would like to be able to split my Mix Bus into those frequency-divided aux tracks, and inside each of those 6 auxes I will have a Waves C6 Multiband Compressor with a proper band soloed for each of the ranges.

Now, hoping my explanation was clear enough (haha), I would like to know how would the correct routing be for achieving this.

Splitting your mix into 6 frequency bands for individual processing is most likely overkill.

Still I'm all for experimenting. and there are various ways to split an audio signal into frequency bands for individual processing, discusses in the following topic:

Making a plugin ignore / undetect certain frequencies

just a thought, based on this, and your other thread: you're over-thinking everything. a little common sense, and your ears, and brain... and you can be working, instead of discussing working.

making things more complex doesn't mean you're making things better, just... more complex.

but as david (who's far more diplomatic than the rest of us), said, there's no harm in experimentation. personally, i'd rather spend the day making music, than thinking about making music.

First off, thanks as always for your responses.

I'll start by saying that I do not wish to make thing more complex than they already are, but I am getting pretty interested into experimentation in order to see all the possible outcomes.

So, what I had in mind is to split said mixbus into 6 different frequency bands in order to be able to apply precise processing to each one of them (you cannot do it by using just one C6 for the all mixbus as you cannot insert effects on each individual band within the plugin of course). This way, I would be able to EQ, Stereo Spread, and do any needed surgical thing just for that single band (Sub Bass, Midrange, etc.). Just wanted to know how would you guys do it (if you were to do it) and how should I route the mixbus and the single band auxes in order to work correctly by my idea.

I hope it's a little clearer now.

personally, i'd rather spend the day making music, than thinking about making music.

How about thinking about overthinking about making music? (From the course "how to use the word about as often as possible in really short sentences")

First off, thanks as always for your responses.

I'll start by saying that I do not wish to make thing more complex than they already are, but I am getting pretty interested into experimentation in order to see all the possible outcomes.

So, what I had in mind is to split said mixbus into 6 different frequency bands in order to be able to apply precise processing to each one of them (you cannot do it by using just one C6 for the all mixbus as you cannot insert effects on each individual band within the plugin of course). This way, I would be able to EQ, Stereo Spread, and do any needed surgical thing just for that single band (Sub Bass, Midrange, etc.). Just wanted to know how would you guys do it (if you were to do it) and how should I route the mixbus and the single band auxes in order to work correctly by my idea.

 

I hope it's a little clearer now.

It was clear the first time. Did you read the thread I linked to?

You don't need to split out different bands in order to EQ specific ranges - that's what parametric EQ is for. For widening, there are tools like Ozone which let you, within one plugin instance, create whatever bands you want and widen them as much as you want independently, without any special extraneous crossovers and mixer routing required. And there are many tools to add distortions in particular frequency ranges if you want to experiment with saturation - simply do it serially on each section of the frequency spectrum you want to process.

I've never heard of, or seen, anyone practically do this convoluting band splitting method across the whole mix, including pros, and if you want to experiment and learn stuff in the process, then go for it (and in this case, I'm not such why you're asking, just try it, as by investigating these things yourself, including what techniques to use, is all part of the learning process, successes and failures inclusive).

Note that the more processes you do to individual "bands", the most likely you're going to get nastiness happening at the crossover points when you remix them back together again.

Let us know the methods that ultimately you settled on, and the results of the experiment...

First off, thanks as always for your responses.

I'll start by saying that I do not wish to make thing more complex than they already are, but I am getting pretty interested into experimentation in order to see all the possible outcomes.

So, what I had in mind is to split said mixbus into 6 different frequency bands in order to be able to apply precise processing to each one of them (you cannot do it by using just one C6 for the all mixbus as you cannot insert effects on each individual band within the plugin of course). This way, I would be able to EQ, Stereo Spread, and do any needed surgical thing just for that single band (Sub Bass, Midrange, etc.). Just wanted to know how would you guys do it (if you were to do it) and how should I route the mixbus and the single band auxes in order to work correctly by my idea.

 

I hope it's a little clearer now.

Yeah, well, you could do that and I tried (something like) that some time ago, but it doesn't make anything easier or, more importantly, better. It makes everything worse, instead. Muddier. More desperation. Or...

Why not split it into twelve bands? You could do that also. Or twenty? Or fifty? It will not work, because it will overstretch your CPU, and then you have to go about fixing that; not to mention that you may have one setting in any of those six C6's wrong or interfering and... all I can see happening is 99 % troubleshooting and 1 % "hm, that is interesting" moments and 0% "Ah, now that is a clear and unambiguous advantage doing it like this versus doing is much much simpler!" moments combined with 0 % better mixes.

But by all means, try it.

How you could set up the C6's:

1st: 20-200 (Hz) (Bass)

2nd: 200-500 (Muds)

3rd: 500-1200 (singings)

4th: 1800-3000 (more singings)

5th: 3000-9000 (sizzlings)

6th: 9000-20000 (young peoples' supersizzlings)

Cutoffs of those bands set as steep as possible (48 dB/oct if available... or maybe not, because of the "bump" effect which such steep filters, maybe 12 or just 6 dB/oct - o no, but then you get overlaps which will cause a different kind of bump effect at the "border" frequencies... mmmm... maybe don't do it all? Or setup an EQ after each of the C6's to attenuate those border bumps...

I foresee that this will make your mix much worse, but in a convoluted sort of way. There so many easier methods to really screw up a mix. You can even do that without any plugins, in my experience.

First off, thanks as always for your responses.

I'll start by saying that I do not wish to make thing more complex than they already are, but I am getting pretty interested into experimentation in order to see all the possible outcomes.

So, what I had in mind is to split said mixbus into 6 different frequency bands in order to be able to apply precise processing to each one of them (you cannot do it by using just one C6 for the all mixbus as you cannot insert effects on each individual band within the plugin of course). This way, I would be able to EQ, Stereo Spread, and do any needed surgical thing just for that single band (Sub Bass, Midrange, etc.). Just wanted to know how would you guys do it (if you were to do it) and how should I route the mixbus and the single band auxes in order to work correctly by my idea.

 

I hope it's a little clearer now.

It was clear the first time. Did you read the thread I linked to?

I did, David, thank you. And Fuzz on that thread seems to be talking about the thing I want to achieve, which is using a Muliband Compressor and soloing the bands.

Could you please explain to me what "in parallel" means in that case and how whould you route the various inputs/outputs for that to work properly?

at the end of the day, no matter what you do, if you're mixing stereo, your final mix is... stereo. so breaking things into all those parts and then putting them back together seems to offer nothing but possible phase issues..

i mean, if you're eq'ing the upper midrange, for example... you're eq'ing the upper midrange, then still combining everything back to 2 full-eq-range channels. where's the benefit?

anyway if you do this, let us know what you discover (but am glad it's you doing this and not me)

at the end of the day, no matter what you do, if you're mixing stereo, your final mix is... stereo. so breaking things into all those parts and then putting them back together seems to offer nothing but possible phase issues..

 

i mean, if you're eq'ing the upper midrange, for example... you're eq'ing the upper midrange, then still combining everything back to 2 full-eq-range channels. where's the benefit?

 

anyway if you do this, let us know what you discover (but am glad it's you doing this and not me)

Let's say, if I were to use Waves Vitamin direclty on the Master Bus instead, would it be the same thing I was trying to achieve but far less trouble?

i can't answer, as i don't know that plugin. but my philosophy would be...to try the easy route first, and then explore more complex options if the need arises. i always look for the most direct route to a result, and amend that if & when......

EDIT: i just looked it up. it already IS a multiband processor, so it's already doing the work for you; it's designed to split up the signal into different frequency groups. explore it, and you should get what you want, without a more-convoluted effort.

If you are willing to be "all Waves"

Insert the Waves "Studio Rack" (stereo). Use the frequency splitter to make 5 bands (sorry, only 5)

Now take the first split and add a frequency splitter. There 6 frequency bands.

You can add Waves plugins to each of the split frequency bands.

If you want plug-in independence just put 6 sends on the channel (set the channel to no output) and place a linear phase EQ on each split to define the frequency. 48 dB/octave high and low cut filters.

Hello, guys

 

So, today I wanna talk about something I've been thinking for a couple weeks.

The goal I wanna achieve is to be able to split my Mix Bus into 6 aux tracks each with its own frequency range (Sub Bass Range, Bass Range, Lower Midrange, Midrange, Upper Midrange and finally Presence.

I have a Mix Bus towards which all my Track Stacks and Effect Busses are routed, so that all my composition stuff goes there. Said Mix Bus is then routed to my Master Bus for the final mastering stage. I would like to be able to split my Mix Bus into those frequency-divided aux tracks, and inside each of those 6 auxes I will have a Waves C6 Multiband Compressor with a proper band soloed for each of the ranges.

 

Now, hoping my explanation was clear enough (haha), I would like to know how would the correct routing be for achieving this.

Create as many busses that you want freq splits.

In this video I created 5 busses and then used Logic's Bandpass Filter (make it the first effect) and adjusted each to a freq range.

Doing this means that I can apply effects to each of the freq ranges.

if you want to do the same with one plug-in, get Blue Cat Audio MB-7 Mixer, it does the same thing.

...and that's why this is such a great forum (and why i am occasionally humbled); good people, willing to help... no matter what.

anyway, will be interesting to see what the OP does with all these options... let us know.

To pick up where Fished left off, it's truly amazing to see how knowledgeable and willing to help the people on this forum are. That's really something.

Now that I have couple days off work, I can sit, check your suggestions and try them in my project!

So, now that I have tried what you guys suggested, I can share my point with you.

I have found that the best way for me to achieve that so yearned frequency split is to use 6 sends from my mixbus to 6 auxes each with a linear phase eq set to isolate the desired ranges within the frequency spectrum.

After that, I am able to insert any plugins I want (Waves, Logic's native ones and so on) to tweak and fine adjust thee sounds to my liking.

I would say after all this is probably the easiest and most effective way to do it.

So, now that I have tried what you guys suggested, I can share my point with you.

 

I have found that the best way for me to achieve that so yearned frequency split is to use 6 sends from my mixbus to 6 auxes each with a linear phase eq set to isolate the desired ranges within the frequency spectrum.

After that, I am able to insert any plugins I want (Waves, Logic's native ones and so on) to tweak and fine adjust thee sounds to my liking.

 

I would say after all this is probably the easiest and most effective way to do it.

Consider that the EQ isn't going to give you a clear frequency cut. For example if you use a low-cut with the cutoff at 1,000 Hz with a 48 dB/Oct filter (that's really steep) then you're not just eliminating everything above 1,000 Hz, you're only lowering its volume, so that at 2,000 Hz you're applying a gain of -48 dB.

Then there's the distortion that the EQ itself adds.

Try to not add any effects at all, just separate the frequency spectrum into bands and sum those bands back together again: the result won't be exactly the same as the original.

i still don't get the point of this; wouldn't there also be phase issues? anyway, seems there are better ideas on this very thread to consider.

but still... whatever works (or works enough to make one 'happy').