gain structure and mixing aka THE MONEYSHOT THREAD

[danoldboy]

Excellent thread as already stated, quick question, would anyone be able to recommend a good (and not too expensive) eq vst that is best suited for shaping the frequencies suggested in the sub bass groups..

[slyman]

i have a question about splitting a bass into 2 groups and then back into one to make room for the kick whats the difference between doing that and just doing a cut on the frequency where the kick would be without splitting the bass?

[macc]

i have a question about splitting a bass into 2 groups and then back into one to make room for the kick whats the difference between doing that and just doing a cut on the frequency where the kick would be without splitting the bass?

Sometimes it's better one way, sometimes it's better the other.

That's the best, most sarcastic, truest, most accurate and least helpful answer all rolled into one

[macc]

Excellent thread as already stated, quick question, would anyone be able to recommend a good (and not too expensive) eq vst that is best suited for shaping the frequencies suggested in the sub bass groups..

Buy (!) Voxengo GlissEQ.

[gravious]

i have a question about splitting a bass into 2 groups and then back into one to make room for the kick whats the difference between doing that and just doing a cut on the frequency where the kick would be without splitting the bass?

Technically, not much. It would give you a little more control of the precise levels and EQ of the 2 sections of bass though, I suppose.

Another (slightly cheating?) way to fit both kick and bassline in similar frequencies would be to use a sidechain compressor to squish the bassline when the kick hits, although this can make a bit of a 'pumping' sound.

[gravious]

Maxx bass = crap distortion = mud = (often) loss of true sub bass as the energy is spread throughout the spectrum = great for shitty laptop speakers = great for making your mix sound like it is wrapped in wool on anything else.



It's just distortion with some simple filtering. Nothing more to it than that.

I tend to agree with this - most of these 'Bass Boost' type plug-ins seem to make subs that sound phat through small speakers, but a bit shit through a big rig. You are better off starting of with a nice clean sub sound, and using these effects sparingly I reckon.

Generally, I find a bit of tape compression/distortion followed by some LP filtering can get it sounding pretty weighty without heavily boosting any particular frequencies

[dubz]

this is a fantastic thread - thanks to all the contributors, been learning a lot!

Now my question:
Can someone clearly articulate how headroom works?
On some of my tunes I've noticed that if I have my bass sounding phat and my kick banging then my mids and hi's get drowned out. Is that because the bass and kick is eating headroom and the mids can't push through?

cheers for the info

[Sinus Sawtooth]

Thanks for these awesome tips!!

I always struggled with the "cleanness" of my tracks, I tend to use a lot of different hihat/rim loops and various kick/bass in my steps and it always sounds muddy... I followed the general DB guidelines, compressed some samples here and there and the sound improved 100%! It's so much cleaner and clear now

[legend4ry]

I'm sorry i'm impatient and can't get through the 12 pages to read the answer of my question because theres so many "thanks for the info" in the thread, obviously macc deserves appreciation but anyways..


What about hats, when ive checked stuff like Ramadanman's tunes - the hats are about 1-2db below the kick and snare - though when I do this on mine - they're way to overpowering, now I ask - since my samples are high quality samples is there a special way I should be filtering them, any EQ peaks what are common in hats what makes them over power everything else?

EDIT: Ive tried low and high passing them - EQing where I can hear that they're peaking / making to much noise and alas no cookie - this ones a tricky one for me

[dignan]

I've got a question or you, Macc... I've read a lot of what you said in this thread and it's extremely helpful, thank you, man!

I read what you said about having your master stay around -3db, and I've got a tune that peaks at -2.4. all my channel faders are at unity also. I want to start mastering it because I'm done with the mix. How do I get it to sound large? I put an Adaptive Limiter on it to raise the level, but I can only raise a couple more db and it's still sounding low in volume, please help me out, man...

[morro_e]

all this information readed carefully and understanded clearly

just improved how my track sounds like ten thousand times

thanx a lot macc

[chilpillwill]

i have to be careful about mastering tracks - sometimes i can get a bit obsessive compulsive and end up spending ages doing, effectively, nothing. I have a reasonably good pair of headphones and a fairly decent 7.1 speaker setup. i usually master stuff with the headphones and then listen to it through the speakers with the sub to get an idea of what it would sound like on a bigger rig. I would recommend this type of setup to anyone who is struggling to get their tracks to sound good on a larger speaker setup.

[macc]

Sorry chaps - been on holiday. Bloody magic it was too

Now my question:
Can someone clearly articulate how headroom works?
On some of my tunes I've noticed that if I have my bass sounding phat and my kick banging then my mids and hi's get drowned out. Is that because the bass and kick is eating headroom and the mids can't push through?

Headroom just refers to how far below digital full scale (0dB) a signal is. There's only so much cake to go round. To me it sounds like your bass and kick are too loud. They're eating too much of the cake. Turn them down a bit and things will probably balance better.

What about hats, when ive checked stuff like Ramadanman's tunes - the hats are about 1-2db below the kick and snare - though when I do this on mine - they're way to overpowering, now I ask - since my samples are high quality samples is there a special way I should be filtering them, any EQ peaks what are common in hats what makes them over power everything else?

Just turn them down. Who cares what numbers they are compared to the kick and snare? This is not related to/missing the point of this thread imo (sorry)

I read what you said about having your master stay around -3db, and I've got a tune that peaks at -2.4. all my channel faders are at unity also. I want to start mastering it because I'm done with the mix. How do I get it to sound large? I put an Adaptive Limiter on it to raise the level, but I can only raise a couple more db and it's still sounding low in volume, please help me out, man...

You've started another thread about this, but I'll repeat this here as it may be getting missed elswhere in the thread.

DON'T OBSESS OVER THE NUMBERS TOO MUCH. I don't recall saying 'leave all faders at unity' and if that's what is being taken from it then I'll have to apologise. Get the sounds right at source as often and as much as possible - THAT is the point I want to get across. If you have all the sounds right, then as a consequence the faders will be at unity. Be careful not to put the cart before the horse.

With that said, I would expect that a tune peaking at -2.4 could get another 4 to 5 or so dB equivalent level with just a simple limiter without souning too bad. You get 2 dB just by turning it up to 0dB, and then add 2 to 3dB or so of limiting on top of that. Any more and things might start to get damaged. If it's still very soft then would you need better mastering but ideally before that you need a better mix.

[dignan]

Sorry chaps - been on holiday. Bloody magic it was too

Now my question:
Can someone clearly articulate how headroom works?
On some of my tunes I've noticed that if I have my bass sounding phat and my kick banging then my mids and hi's get drowned out. Is that because the bass and kick is eating headroom and the mids can't push through?

Headroom just refers to how far below digital full scale (0dB) a signal is. There's only so much cake to go round. To me it sounds like your bass and kick are too loud. They're eating too much of the cake. Turn them down a bit and things will probably balance better.

What about hats, when ive checked stuff like Ramadanman's tunes - the hats are about 1-2db below the kick and snare - though when I do this on mine - they're way to overpowering, now I ask - since my samples are high quality samples is there a special way I should be filtering them, any EQ peaks what are common in hats what makes them over power everything else?

Just turn them down. Who cares what numbers they are compared to the kick and snare? This is not related to/missing the point of this thread imo (sorry)

I read what you said about having your master stay around -3db, and I've got a tune that peaks at -2.4. all my channel faders are at unity also. I want to start mastering it because I'm done with the mix. How do I get it to sound large? I put an Adaptive Limiter on it to raise the level, but I can only raise a couple more db and it's still sounding low in volume, please help me out, man...

You've started another thread about this, but I'll repeat this here as it may be getting missed elswhere in the thread.

DON'T OBSESS OVER THE NUMBERS TOO MUCH. I don't recall saying 'leave all faders at unity' and if that's what is being taken from it then I'll have to apologise. Get the sounds right at source as often and as much as possible - THAT is the point I want to get across. If you have all the sounds right, then as a consequence the faders will be at unity. Be careful not to put the cart before the horse.

With that said, I would expect that a tune peaking at -2.4 could get another 4 to 5 or so dB equivalent level with just a simple limiter without souning too bad. You get 2 dB just by turning it up to 0dB, and then add 2 to 3dB or so of limiting on top of that. Any more and things might start to get damaged. If it's still very soft then would you need better mastering but ideally before that you need a better mix.

should I turn it up to 3dB during the mixing or mastering stage? Wouldn't this cause clipping though?

[macc]

You might need to read the thread again; it seems you've missed the point somewhere.

Sorry if this is blunt... I wrote a massive essay but I've deleted it as it missed the point of your post.

[dignan]

You might need to read the thread again; it seems you've missed the point somewhere.

Sorry if this is blunt... I wrote a massive essay but I've deleted it as it missed the point of your post.

so... manually turn up the master output to 0dB then limit another 2 or 3dB?

[macc]

so... manually turn up the master output to 0dB then limit another 2 or 3dB?

Most limiters will do all of that as a part of the same process, that is, you'll find yourself adjusting the threshold by 5 to 6dB or so in total, in this case.

[macc]

Nice one putting it out here man.... Didn't meant o be rude about it, just thought more people could benefit.

Wasn't really one for this thread but I'll get back here proper laterrzrzrzz

[macc]

Right then.

A key point here that helps demonstrate this with regard to plugin/digital eq is: The gain for a bell curve boost = unity at the Nyquist frequency, in a non oversampled system.

So if you have a wide-q eq band at 18kHz and you turn its gain up, you naturally expect that it extends up past 22050Hz, even if you can't hear what it is doing above Nyquist as the signal gets anti-alias filtered. That's what an analogue eq does, that's what you'd expect it to do. It's very common to see analogue eqs with apparently 'out of the audible band' centre freqs.

But now think about that for a non-oversampled digital eq. Its gain at Nyquist has to = unity. As it approaches Nyquist the curve is forced downwards to meet the unity gain condition. You will have an asymmetric or 'cramped' curve (as they are known). The 'air' you may have been trying to put in just isn't going to be there. It's not quite right.

Other funny business occurs for different filter types, but to cut a long story short, non-oversampled eqs suffer from odd behaviour like this as you get up towards Nyquist.

Page 22 and especially 23 of this:

http://www.soniqware.com/pdf/SoniqWare-PE-1.pdf

show it extremely well. Look at the graphs and it explains all of this better than I can in words.

The way to get around it is to increase the Nyquist frequency. There's two aspects here;

Working at higher sample rates is probably the simplest, without having any conversions involved. It's then very easy to have eqs that don't suffer from cramping in the audible band, as the Nyquist freq is now, say, 48kHz (for a 96kHz SR) and not 22.05kHz (for 44.1kHz). Thing is, if you're going to CD then there's going to be downsampling at some point, so you have to choose your poison.

Another way to do it is up/oversampling. To the user, it all stays at the lower sample rate, say 44.1kHz, in the sequencer. The sequencer passes the audio to the equaliser, which increases the sample rate internally (ie sticks a load of zeroes in), applies its curve to the signal, and then anti-alias (LP) filters it before passing audio back to the host at the original sample rate. That gets around the problem quite nicely, but has drawbacks. 1) it takes more CPU depending on what factor the oversampling is being done at (your 1x, 2x 4x options), and 2) it means applying filtering to the audio before it gets brought back down, which has to be very carefully implemented so as not to cause damage to the signal (ring/prering etc depending on how it is done).

EDIT: I should point out that people are coming up with crafty ways to decramp without resorting to oversampling, there's a few eqs like that now.

So when you're using an oversampling eq it depends on several factors as to whether you use it and to what degree. How much CPU you have to spare, how much HF work you'll be doing, how much you care, how well you can hear the effects of the filtering, etc etc. Personally I generally prefer to use oversampling when available as the more natural curve usually outweighs the other stuff, BUT I do check to see. Plus I am usually doing high freq work in analogue anyway as it avoids all this and generally just sounds better anyway, depending on what eq you are using of course.

Help?