Splitting the signal

[Rasbah]

Hi all!
I would like to split my stereo mix into 3 band:
the low band from 20 Hz to 90 Hz
the middle band form 90Hz to 8 Khz
the high band from 8 Khz to 20 Khz

What I would like to have is a choice of compress, equalize or give more enhancer to the single band.
I know that there are more multiband compressor plug ins on the market, but I prefer to use some analog gear.
Well the only manner ( I think) to split the signal is a digital crossover, 30/36 db oct., so the cutting point and roll off can be very precise.
With sonnox eq you could use high and low pass with high roll off, so the phase problem between the band is minimized.
So I try to do this chain:

-I had create 3 aux and on each I have put my sonnox eq with the xover.
The first one have a low pass with a 90 hz cut, the second aux have a high pass that start to 90 hz and a low pass that start to 8000 Hz, the third aux have a high pass whit a cutting point of 8000 Hz.

-I send my original track to all aux at 0 db so I have a low band middle band and high band.

In this way I can work on different band.
I try to send my three signal into three neumann w492 and than I summed them in a passive summing mixer, the result is very good for me.

My question is this:

Is this a common practice to split the band in this way in a mastering session? I usually use this method to balance my final mix, but I'm not sure if this technique is good for mastering, because some sort of inaudible comb filter....

[lowland]

When bandsplitting in analogue or digital hardware (Prism MLA-3 and TC 6000 come to mind) manufacturers tend to use a much gentler filter slope than you propose, with consequent minimised phase issues at the crossover points - it's no good if the cure is worse than the disease.

[Rasbah]

Ok! thanks for the advice!
Well, is the signal splitting a good method for mastering, or is better to process the tracks whit a multiband comp/eq, so there aren't phase problems?

[t_d]

if you need to master a track this way it's better to go back to your mix and fix these things first. if it's *your* stereo mix (which i gather from your post) then you likely have access to the whole individual track mix. go back to that and ask yourself why you wanted to split the master into 3 different bands and fix those issues in the mix, then you don't need to multiband the master.

[Jim Williams]

No mastering fixes a bad mix. That's why 20~30 years ago record companies regularly rejected mixes for various reasons.

That is no longer done, and it shows.

[SASMastering]

OT edit

[dcollins]

Originally Posted by Jim Williams

No mastering fixes a bad mix. That's why 20~30 years ago record companies regularly rejected mixes for various reasons.

That is no longer done, and it shows.

I don't know what labels you work with, but it's done constantly.

To the OP, if you're using analog to split the signal bands, there's only one slope that is really blameless: 6dB/octave.


DC

[Rob Murray]

What DC says, mixes get turned down all the time, often impacting our schedules for the worse. For myself I'd say the rule is that mixes won't be approved before the mastering date, and the the refreshing exception is getting mixes a day or two before. Even with non-label clients musicians these days are pretty picky about what goes out the door, especially since they know that in this era, everything can be fixed, overdubbed, tuned etc. they're always striving for the last inch.

Back to the topic, can't say I've ever felt the need, but something like this...

Drawmer Three-Sum Multi Band Splitter | 3sum 3 Sum | Soundpure.com

I think there's other similar options out there, or you can always ask a tech what it'd cost to put one together if you have certain specs you need.

[Rasbah]

Ok, I only want to learn how to do the job at the best, so I don't want to create problems..
I had read on many websites , that the multiband comp or eq (Waves, ecc..) are often used during the masterin, so I try to do it in analog way, just this.
I don't want to complicate the work whit some sort of tricks, so I thought whit a separate band you can work on the frequencies range in a specific manner, and also you can process only the band that you want to process, like a sidechain.
But.. my question is:
there are a lot of multiband plugins on the market..why?

[dcollins]

Originally Posted by Rasbah

Ok, I only want to learn how to do the job at the best, so I don't want to create problems..
I had read on many websites , that the multiband comp or eq (Waves, ecc..) are often used during the masterin, so I try to do it in analog way, just this.
I don't want to complicate the work whit some sort of tricks, so I thought whit a separate band you can work on the frequencies range in a specific manner, and also you can process only the band that you want to process, like a sidechain.
But.. my question is:
there are a lot of multiband plugins on the market..why?

While you hear a lot about it on the internets, multi-band compression is actually the least used technique, at least in my studio.

De-essing doesn't count.


DC

[Rob Murray]

Multiband compressors can be a life-saver in mixing. Something home-tracked, or just sub optimally done, or in electronic music where a sample (ex. drum loops) may have multiple elements. In mastering, minus de-essing, or the occasional single band compressor keying in on a problematic frequency, it just really isn't an everyday tool. Even the times I reach for one I usually end up taking it off. Kind of like using an axe where a sander is needed, it's a great tool, but if you're reaching for it during finishing touches, somebody probably screwed up already.

So IMHO Multiband compression is great, just not at the mastering stage.

[Paul Gold]

Originally Posted by dcollins

While you hear a lot about it on the internets, multi-band compression is actually the least used technique, at least in my studio.

I use it lot and like it. I know what I want to accomplish before I turn the knobs. If you try to 'see if it sounds good' it won't.

The usual dig is that it changes mix balances too much. IME it only does this if you turn the knobs too far.

[wado1942]

Buy a DBX crossover and remove half the filters.

[dcollins]

Originally Posted by Paul Gold

I use it lot and like it. I know what I want to accomplish before I turn the knobs. If you try to 'see if it sounds good' it won't.

I think that must be my problem. Although when I demoed the Maselec, I could get it do some cool things. I did tend to go back to full range though.

With the tc, it just seems to be initially impressive, but the more I listen to it, the less I like it. There are so many parameters to juggle, I tend to lose interest after the first six hours of fiddling with it too.

I guess the question is what you find to be beneficial, and if it's really turning out to be a daily process, not just one for 'troublesome' mixes.


DC

[Paul Gold]

Originally Posted by dcollins

I guess the question is what you find to be beneficial, and if it's really turning out to be a daily process, not just one for 'troublesome' mixes.

If I'm looking to smooth out the sharpness of a track slightly, just using the high band will give more transparent results than a wide band compressor.

If the bass drum is poking out a little using the low band is more transparent than a wide band compressor.

A out of control vocal often will have most of the outofcontrolledness in a small area like low mids or high end. Using a MB can often control the vocal level more transparently than a wide band compressor.

If I don't think it's needed I don't use it. I don't think the mix has to be "broken" to use it any more than I think EQ should only be used when something is "broken", what ever that means.

[huejahfink]

Cheers Paul! Nice to see someone talk about Multiband without waving a crucifix and holy water around as if fighting vampires.

[dcollins]

Originally Posted by Paul Gold

If I'm looking to smooth out the sharpness of a track slightly, just using the high band will give more transparent results than a wide band compressor.

If the bass drum is poking out a little using the low band is more transparent than a wide band compressor.

A out of control vocal often will have most of the outofcontrolledness in a small area like low mids or high end. Using a MB can often control the vocal level more transparently than a wide band compressor.

If I don't think it's needed I don't use it. I don't think the mix has to be "broken" to use it any more than I think EQ should only be used when something is "broken", what ever that means.

But that's just one band, right?


DC

[Paul Gold]

The examples I gave were for applications where I use one band. I also use multiple bands quite often and think I get more transparent results than using a wide band compressor in the same situation. It's not as easy to describe that though. I sort of think of it as controlling concentrations of excess energies. Tis just a tool. I like them but have worked without them and still manage send out invoices.

[Jim Williams]

Originally Posted by dcollins

I don't know what labels you work with, but it's done constantly.
DC

Could have fooled me. Some of the stuff that is accepted today would have been outright rejected 30 years ago. Clipping, mass compression, excess THD all would have been disqualifiers.

Quality standards have changed, for the worse.

I havn't figured out who's responsible, the AE's, ME's, record companys or a dumbed down public. Probably all of them have some responsibility.

[Waltz Mastering]

Originally Posted by Paul Gold

The usual dig is that it changes mix balances too much. IME it only does this if you turn the knobs too far.

I find this to be true too. I use the SMC 2b quite a bit but not in the bastardizing way people think of when they think of mbc. I like the sound of the box.

[Bonati]

Originally Posted by Jim Williams

Could have fooled me. Some of the stuff that is accepted today would have been outright rejected 30 years ago. Clipping, mass compression, excess THD all would have been disqualifiers.

Quality standards have changed, for the worse.

I havn't figured out who's responsible, the AE's, ME's, record companys or a dumbed down public. Probably all of them have some responsibility.

Ok, my tongue hurts from biting it. Your posts are essentially all the same. We get it already. You're old and you don't like anything and all this new stuff is crap.

[dcollins]

Originally Posted by Bonati

Ok, my tongue hurts from biting it. Your posts are essentially all the same. We get it already. You're old and you don't like anything and all this new stuff is crap.

But with the right opamps...............



DC

[dcollins]

Originally Posted by Jim Williams

Could have fooled me. Some of the stuff that is accepted today would have been outright rejected 30 years ago. Clipping, mass compression, excess THD all would have been disqualifiers.

Quality standards have changed, for the worse.

I havn't figured out who's responsible, the AE's, ME's, record companys or a dumbed down public. Probably all of them have some responsibility.

I have tons of examples that contradict this, although the concept of "quality" is not so rigid as you might think.

Just last December, a band I mastered had the entire record rejected by the label. Why? Because the raw mixes were way over-compressed. They re-did the whole thing, and the final result is much better. Cost plenty, I'm sure.

To some, there is an "optimum" setting, and quality basically just comes shooting out.


DC

[Paul Gold]

Originally Posted by dcollins

But with the right opamps...............

I have always wondered how you use magic bulk foil resistors with any regularity when they are only available in about 10 values.

Also the examples I gave had one application for a low, a mid and a high band. You could use all of them at once.

I think your real gripe is that a MB is too fiddly and you can get where you want to go faster with other methods. That is a fair criticism but not an indictment of a MB.

Some of my favorite records of the past sound terrible.

[dcollins]

Originally Posted by Paul Gold

I think your real gripe is that a MB is too fiddly and you can get where you want to go faster with other methods. That is a fair criticism but not an indictment of a MB.

Could be. Although I just did a track with a de-esser and a sidechain filter on the compressor which counts as the evil multi-band.

Some of my favorite records of the past sound terrible.

Likely caused by inadequate slew-rate.


DC

[Colin Leonard]

. paul, you're talking about using the maselec for multi band? the ds1 can be really nice in Mb especially for de-essing but i usually don't take it much farther than that, not to say it can't be done tastefully. my problem with the tc stuff is that they sound so grainy and bad to me and everything sounds better in bypass.

[Paul Gold]

I loved the M5000 but I've never used a 6000. I'd assume I'd like the sound of it but it's too much box for me.

I started out with an M5000. That's all I had so I managed to squeeze a lot out of it.

Now I'm using the Maselec and the Flux Alchemist. I even use both at once on occasion. Shhh...

[Bonati]

Originally Posted by Paul Gold

Now I'm using the Maselec and the Flux Alchemist. I even use both at once on occasion. Shhh...

And I sometimes use the Maselec and the DS-1 (mkIII) together. The tag team approach. Mostly for lacquer cutting situations where it's time to take the white gloves off and deal with sibilance that stabs like an ice pick.

[Jim Williams]

Originally Posted by Bonati

Ok, my tongue hurts from biting it. Your posts are essentially all the same. We get it already. You're old and you don't like anything and all this new stuff is crap.

Hey, when my cat runs out of the room when playing some of that crap, it tells me it's not going to be enjoyable.

Maybe you ought to ask those former customers that helped the music sales drop 1/2 in the last 10 years. Ask them how much they like the "new sounds".

No one said but you that all this new stuff is crap. A lot is, but not all. Even a broken clock is right twice a day.

I just don't see you or anyone else doing much of anything to improve that in the mastering domain.

Mostly, it's all compressed beyond what the format needs as a production decision. That is a decision that could be re-thought. However, it's not done for technical reasons, it's audio fashion. It's in style to compress the snot out of every CD you can. Then add a few digital overs, then adding some tape hiss on top. Many don't, but too many do.

Doesn't mean we have to like it. Many have voted with their feet. And their pocketbook. Take your arguements to them.

[Greg Reierson]

Originally Posted by Jim Williams

I just don't see you or anyone else doing much of anything to improve that in the mastering domain.

Then you're not looking in the right places.

There are thousands upon thousands of CDs made every year from all around the world. There is an endless supply of great music that has not been through the pop fad ringer. Music isn't always a competition.

GR