Loudness!!!!!!!!

seannash

dont want to get into the whole loudness war thing but i am curious to now how people get there tracks sounding loud without cutting the heads off the wave forms.


i get to a point with limiters and compressors(i know there pretty much the same) where the only way i can get it up to compareable levels with other tracks is to cut the head off the waveform.


so its a bit of a holy grail question but how do you get tracks loud without killing it,sound wise

be specific if you can

Neurojazz

Compressors all the way...

The perceived loudness i've found in dance music and the thump/wack/punch i've emulated with a compressor letting the 3 main components have a fair amount of attack ... so bass drum / bass / lead would be forced through a compressor with a low ratio 1:4 lets say... and a large attack of that to give it some bite... then another comp or buss comp used to lift it up even louder in the mix with a 1:20 / infinity ratio fast/zero attack... i use it as a limiter, probably frowned upon etc.... but it's the duende and it kicks ass imo.

If you focus on getting those main components sorted then the other sounds can be just crushed to death with a 1:20 ratio compressor with a fast attack so they are lifted (even over compressed) and will sit in the mix fine... so essentially you are flattening most the elements and leaving the main ones with the dynamics.

Sidechains are then used to keep the sounds you want loudest occupying the frequencies you want 'reserved' - a common example would be the bassdrum sidechaining the bass drum... this can be strong or used in a subtle way so it doesn't pump to much.

Basically i use the initial compressor to give the sound an artificial envelope and the final one deal with the spikes.

I've tried hardware/software comps.... hardware really doing it for me.... and i have a friend with software SSL and it's not standing up in the mix... go figure!

seannash

thanks man.

so you think its very much a part of the creative process(ie not bouncing out your master file and then having a mastering chain to bring up the level)

Neurojazz

Well, only my opinion - and you'd need to speak to someone like Paul Brewer to get the actual science - but i take each sound as find it... get them to match and then start the compression and sidechaining fine tuning towards the end.

Sometimes i need to change thing radically, sometimes a light touch to get it all balanced...

What i do sometimes is push the mix hard against a final buss compressor / fast attack/fast release and push it a little to hard to see what sound ****s it all up and then back that one off a few db - just to try and balance things even more... then the final buss compressor would be taking about 4db off peaks to be gentle... i've had more comments and praise for depth in the mixes when doing this (even though the music isn't up to par!)

seannash

thanks man,definitely a few things for me to try out.

alot of people seem to be mentioning saturation as a way of gaining percieved volume without raising the volume

jtsuited

just do what everyone else does....cut everything below 60hz and smash the fook out of it with the L3.

seannash

jtsuited wrote: »

just do what everyone else does....cut everything below 60hz and smash the fook out of it with the L3.

really,below 60hz.is that not a bit high?(genuine question)

Neurojazz

seannash wrote: »

really,below 60hz.is that not a bit high?(genuine question)

Can depend on the shelving used (like the db per octave) - the steeper the cut would result in losing bass yes, but a gentle one or certain models i suppose would result in different results) - if you have an eq that can do different levels of low cut then you'd hear the difference between the db per octaves.

seannash

Neurojazz wrote: »

Can depend on the shelving used (like the db per octave) - the steeper the cut would result in losing bass yes, but a gentle one or certain models i suppose would result in different results) - if you have an eq that can do different levels of low cut then you'd hear the difference between the db per octaves.

yeah i guess.logic can do 6,12,18,24,36 and 48 db per octave shelving.id imagine anything above 24 would be too severe

TelePaul

I mostly 'master' rock and pop type recordings - by 'master', I mean I apply a few plug-ins. The mix really starts to suffer noticeably when the soft-clipper makes an appearance...compressors don't bring the hurt so much.

Neurojazz

TelePaul wrote: »

I mostly 'master' rock and pop type recordings - by 'master', I mean I apply a few plug-ins. The mix really starts to suffer noticeably when the soft-clipper makes an appearance...compressors don't bring the hurt so much.

Yep, found that out the hard way... since just using very light 'up-lifting' with a limiter/l3/insert loudness tool here, i've found mixes really getting peoples attention... and for dance music where the bass/lead/bass drum need to really be thick and warm it does really seem to be the way to go.

I do find though that using the 'soft clippers' to crush the life out of incidentals/pads/fx and leave them flat in the back of the mix seems to be a new use for them - it seems to even out all the nasty frequencies/spikes without too much trouble... don't really have to be delicate with those items in the mix as large PA tend to drown any subtle effects that you might make.

I remember hearing mixes through older turbosound rigs (60k i think) and any subtle work on reverbs was totally wasted - strong signals made it out... the subtle didn't.

seannash

i read also that you should try and focus the main elements of the track(kick,snare,bass,lead/vox) upfront in the mix and then push everything else to the back with reverb.

this would be classed as a mixdown technique alright but again people maintain it gives more percieved volume.

i havent tried it yet but intend to

t4cell

Compressors, expanders and nice limiters MD 3 or Sonnox Limiter and decapitator lovely plugin gives your mix that extra bit of bite, don't really like L3 on the buss or in the mastering chain, not so bad on individual sounds in the mix but does very nasty things to your high end if your not careful.

krd

Sean, I've been listening some of your sound cloud stuff. You're pretty loud already.


A lot of stuff that's loud for the sake of being loud, just doesn't sound that good.

seannash

krd wrote: »

Sean, I've been listening some of your sound cloud stuff. You're pretty loud already.


A lot of stuff that's loud for the sake of being loud, just doesn't sound that good.

ah yeah my stuff isnt exactly quiet but theres a point that i reacj where i cant get it any louder or as loud as some other tracks out there and im just curious why.

i know what you mean by stuff being loud for the sake of it and sound quality being sacrificed for it but some of them manage to achieve great sounding tracks that are really loud

krd

seannash wrote: »

ah yeah my stuff isnt exactly quiet but theres a point that i reacj where i cant get it any louder or as loud as some other tracks out there and im just curious why.

I know what you mean by stuff being loud for the sake of it and sound quality being sacrificed for it but some of them manage to achieve great sounding tracks that are really loud

I'm nowhere near your standard. But I do know some of the techniques - though I can't really do them that well.

A certain amount of the loudness is purely perceptual - a quite sounding muddy track, isn't pushing any more air than loud clear sounding one. If frequencies push over each other - the overall effect is to make it muddier and it sounds quieter.

One of the techniques, I've heard is to ruthlessly gut frequencies from your sounds. If a pad has some bassiness to it - strip it out. It will sound really thin by itself - but in the mix it can sound like it's all there.

It's a bit like fluorescent light bulbs and ordinary light bulbs. Fluorescent light bulbs only put out light at certain frequency peaks - ordinary light bulbs put out light across a the whole band. All your eyes need to see to think there is light is a few of the frequencies. Ordinary light bulbs have to push more power to get the same perceptual light level as fluorescent bulb. The eye and the ear perceive things as been louder and quieter. You'd be surprised. If you had a sound level meter - a pub that doesn't sound that noisy, may have the same level of sound energy as a jackhammers breaking up concrete. The noise in the pub is broader band of frequencies so none are really cutting through. A Jack hammer is in a few frequencies. You're speakers are ultimately limited in how much energy they can push. Cut out frequencies you don't need and you can go perceptually louder

I think it's a whole bunch of different techniques - I haven't had the chance to practice them all. There's one I've seen, where you start by removing problems with a peaks that's are not doing anything - you find them - audition them, then remove them if they're not needed. Once they're removed you can raise your volume again and go louder - your speakers won't rumble.

Something like those huge sounding 808 kicks you hear. When they hit - there is absolutely nothing else going on in their main frequency range.

I'm really only coming to grips with Eq's and compressors now - I've noticed the Eq's I've been using don't give a absolute savage frequency cut when I want one. I can see they haven't on the spectral analyser. I think it's something deliberate in the Eqs to make them sound more "natural" - or smoother.

Mastering engineers have their own techniques for making things louder - a lot of it is rolling off unnecessary frequencies and then just upping the volume. Not just in the bass. A mastering engineer - a proper one, not a chancer with a limiter - could probably take one of your tracks and make it sound much much louder. Just through the experience of knowing what they can get rid of.

If you're handing over your mix to a mastering engineer, I've also heard you don't try and max it yourself first. They take it at a lower volume and work it up. If they have nowhere to go, they have to lower the entire track then bring it up again.

hubiedubie

I find if a snare/clap coincide on a beat, it forces the limiter to clamp down hard. So I use track delay to move the snare forward / backwards so that the transients aren't occurring at exactly the same time. This gives me a little more headroom to push things a bit harder on the limiter.

Neurojazz

krd wrote: »

I've noticed the Eq's I've been using don't give a absolute savage frequency cut when I want one. I can see they haven't on the spectral analyser. I think it's something deliberate in the Eqs to make them sound more "natural" - or smoother.

There are only a handful of software EQ with deep/steep cuts.

Sonnox EQ (Sony Oxford) - is one software EQ offer good control over the bottom end.

If you don't already know, even when you have a steep cut in the very low frequencies that the analyzers will only a show a smoothing of the extreme lows - this is what you need to happen.

Lets say there's a busy passage of bass drum, bass etc... and take the shelve off you should see interference/bouncy/jerky/spiky/lumps on those lower frequencies from about 60htz to lower (depending on analyzer) - and when you bring the shelve in they will even out - technical speak, you've removed subharmonic rumble that interferes with the program material above it.

Or, it could be that the shelve you are using is deep, but a different algorithm would be even deeper: - Butterworth/Gaussian/etc... some of these are extreme and change the tone of the bass above it if you drag it up to close.

So basically your analyzer won't show those lows completely disappear until you sweep much higher than you think - and not a problem that they may still be present.

krd

Thanks NeuroJazz,

I've just switched to using Ableton recently. I'm having real trouble coming to grips with it. The main thing I was playing around with before was a Roland MC 505. It has many limitations and doesn't really work as well with ableton as I'd like it too. The MC 505's sequencer is not that great. But it's good for letting you tweak a pattern on the fly - crap for letting you build a whole tune that isn't very repetitive.

It could be lots of these I'm doing wrong. But the hardware nearly seem to have a circuit in it that makes everything sit well. Though you have nowhere near the control you have with a Daw.

I'm confused.

Neurojazz

krd wrote: »

Thanks NeuroJazz,

I've just switched to using Ableton recently. I'm having real trouble coming to grips with it. The main thing I was playing around with before was a Roland MC 505. It has many limitations and doesn't really work as well with ableton as I'd like it too. The MC 505's sequencer is not that great. But it's good for letting you tweak a pattern on the fly - crap for letting you build a whole tune that isn't very repetitive.

It could be lots of these I'm doing wrong. But the hardware nearly seem to have a circuit in it that makes everything sit well. Though you have nowhere near the control you have with a Daw.

I'm confused.

Sample the drum sounds out of it, get a little VST synth and just use the 505 for the knobs to control filters etc..., but it may fail there as it's a known bitch for being friendly.

electrogrimey

Found this, it's not the most detailed thing, but it might help some people, and it was better than expected.

http://audio.tutsplus.com/tutorials/mixing-mastering/how-to-master-a-track-in-15-minutes-or-less/

Has a bit about perceived loudness, basically throw in a brick wall limiter at the very end of the mixdown, then crank up the gain, and it makes the track denser and fuller, and sound louder, even if it isn't.

Also, is there any way to find out the actual sound range of a sound so you can properly EQ it? Like if you could find out the exact frequency of a kick, you could take a lot less out of the rest of the track and still let the kick go through.

splitrmx

Is sticking a compressor/limiter on the stereo buss and smashing the track common practice nowadays? Almost every interview I've read with a mastering engineer advises against doing this. Or am I just old fashioned? Do people not just turn the volume up on their stereo anymore?

Neurojazz

splitrmx wrote: »

Is sticking a compressor/limiter on the stereo buss and smashing the track common practice nowadays? Almost every interview I've read with a mastering engineer advises against doing this. Or am I just old fashioned? Do people not just turn the volume up on their stereo anymore?

I think good practice (and if you want the volume to compete with people flicking through tracks on portals etc), would be to pre-treat all the channels that have harsh attacks, or the tracks you want to be at the forefront before just hammering the master buss.

This way you'd get even more apparent volume without crushing the life out of the whole mix.

splitrmx

Neurojazz wrote: »

I think good practice (and if you want the volume to compete with people flicking through tracks on portals etc), would be to pre-treat all the channels that have harsh attacks, or the tracks you want to be at the forefront before just hammering the master buss.

By treating, you mean compressing them to smooth down the attacks a bit? I'm using a hardware mixer and just recording the stereo out into Logic, so I don't have the luxury of sticking various compressors on every channel.

Neurojazz

splitrmx wrote: »

By treating, you mean compressing them to smooth down the attacks a bit? I'm using a hardware mixer and just recording the stereo out into Logic, so I don't have the luxury of sticking various compressors on every channel.

Ouch, then it's expensive!

Anima

electrogrimey wrote: »

Also, is there any way to find out the actual sound range of a sound so you can properly EQ it? Like if you could find out the exact frequency of a kick, you could take a lot less out of the rest of the track and still let the kick go through.

A spectrum analyser plugin is what you're looking for.