+- Subvert Central (http://subvertcentral.com/forum)
+-- Forum: Main Forums (http://subvertcentral.com/forum/forum-4.html)
+--- Forum: Audio Production and Engineering (http://subvertcentral.com/forum/forum-17.html)
+--- Thread: Order of applying processing on tracks. (/thread-38711.html)
Order of applying processing on tracks. - choronzon - 1st October 2007
Ok excuse me for the terminology ( or lack of) . I hope that my question is clear as possible, can be tricky sometimes talking about this stuff when not used to it. So please try and bear with me.
The question I have is that recently I was told to put my compressor before the channel eq on my inserts (using logic) on my track for my main break, now my question is why??, and does this hold true for tracks with other insts on them or does it vary depending on what your trying to do to the sound, for instance if it was a sub would it really make a difference compressing before or after equing? Also Does it matter weather or not the compressor comes before or after other effects?
I am still new at really thoroughly understanding the science of the processing of sound sometimes, I can only really listen but if anyone could help me clarify this idea of order and the effect of it a little in my head espically when it comes to equing ( cause I used to always eq first!!!!)
that would be much appreciated. 
While on topic too,, Is it usually a good idea to do proccesing to a sub sample before bringing it in a sampler? Or does it not much make of a difference?
Much thanks in advance
Order of applying processing on tracks. - Macc - 1st October 2007
choronzon Wrote:ok excuse me for the terminology ( or lack of) . i hope that my question is clear as possible, can be tricky sometimes talking about this stuff when not used to it.
so please try and bear with me.
hello cat - don't worry about all that mate, you should see some of the luddites we get here
Quote:the question i have is that recently i was told to put my compressor before the channel eq on my inserts (using logic) on my track for my main break, now my question is why??,
and does this hold true for tracks with other insts on them or does it vary depending on what your trying to do to the sound, for instance if it was a sub would it really make a
difference compressing before or after equing? also does it matter weather or not the compressor comes before or after other effects?
i am still new at really thoroughly understanding the science of the processing of sound sometimes, i can only really listen but if anyone could help me clarify this idea of
order and the effect of it a little in my head espically when it comes to equing ( cause i used to always eq first!!!!)
firstly, don't do anything just cos some dude tells you, be it me or anyone else :d if he couldn't tell you exactly and specifically the reason why you should do this, i'd suggest he read it somewhere on an internet board, off someone who read it on an internet board
secondly, order is very important! you're right in that it's all a question of what you're trying to
do to the sound. 'creative' compression (ie spanking things as an effect, rather than for mix purposes) can often be the first in the chain, after which you can tailor the 'new' sound
to perfection using eq and so on. however, for general mix purposes, there are the two common approaches - compression then eq, or eq then compression. i don't have any rules exactly
but 90% of the time i use eq then compression, and i'll try to make it clear why i favour that.
the fundamental thing behind all this is;
compression can affect your tonal balance, while removing or adding with eq affects your level.
to demonstrate this, get a raw (preferably a bit shit-sounding) drum break and compress
the living shit out of it, short attack (as short as you can), short release (say 50ms). set the make up gain so that the perceived level is the same when the compressor is on or off.
listen to how certain things - the most prominient things at a given moment - get further accentuated by the compression. the tonal balance 'jumps' around according to whatever is boss
at the time. the kick makes the hi hats disappear, maybe if the kick rings on a long time, it gets 'cut off' by the snare, which pushes the kick to the back when it hits. maybe the big fat kick makes the cymbals 'duck' (as it is called), and you can hear them pumping/breathing/sucking (all manner of terms get used here). so, compression affects tonal balance (although for 'normal' mix compression it wouldn't be as severe as this example).
eq's effect on level is rather more obvious - for example, turn up the bass and the overall signal gets louder :d
sticking with drums (a good example as they use virtually the entire frequency range);
the 'compression then eq' school see it as advantageous as you are able to tailor the sound in the mix once its level has been evened out somewhat by the compression. that is, you
'regulate' the sound a little, bring it under control dynamically, and then make it work in your mix tonally. this way the level is sorted out, and the tonal balance won't be getting
messed with afterwards. the problem here is that if you need to add lots of low, the kick will be eating up a lot of headroom, being somewhat less 'under control' than it was after the compression.
the 'eq then compression' school see it as advantageous as you can tailor the sound to the mix, and then regulate the dynamics, based upon the 'new' sound. this is good because if you decide to add lots of kick, you'll still have that and it will be under control. however, as we have seen, this might make your tonal balance jump around as the kick could well duck everything else.
as i have said, i prefer the latter approach in most cases, and there's a number of reasons for this. 99.9% of signals - even if perfectly recorded - will have some region/s that will need cutting in order to make it work in the mix. this part of the sound contributes to its overall level, and this overall level is what dictates how much a given compressor will work. by removing the 'shit bits' from a signal before running it into the compressor, the compressor has less work to do, and so will do less 'damage' to the signal (again we are talking 'mix' compression rather than 'creative' compression). another way of looking at this is that if a 'bad' portion of the signal is very prominent, compressing first will accentuate it somewhat as well as making the comp work harder (see example above). then you'll have to do even more cutting to remove it and balance the sound. i suppose my point of view here is tied into my general approach - get rid of the shit as soon as possible, so that you're only dealing with the stuff you actually want to be worrying about.
bear in mind this ^ is all relatively small scales we are talking about. as always in audio, imho the best way is to do these things to extremes to understand them, and how they work. once you understand what happens on the heavyweight scale, you get to better understand what happens - albeit less obviously/audibly - when you're not working things so hard.
one thing that has to be mentioned is that i am always going backwards and forwards between the eq and the comp. i'll use a surgical eq on a sound to get rid of the unwanted bits, then use a fat eq to accentuate the good bits. when this is sounding right, i'll put the compressor in and set it about right. however, this - as we have seen - is bound to affect the eq work we just did. so that will need a little adjustment. but adjusting that will affect the level going to the compressor, right? so now we need to go back to the compressor to adjust that. but that can affect the tonal balance........... and so on.
with a careful, iterative approach like this you can get the best sound, maximum gain reduction and minimal negative compression effects on the sound 
.a quick word about other effects - generally i'd place them before the compressor. things like flangers can cause big volume bumps as they modulate (especially with feedback). filters too - when you're lowpassing a beat you can catch a region that really booms. these virtually always need taming.
Quote:while on topic too,, is it usually a good idea to do proccesing to a sub sample before bringing it in a sampler? or does it not much make of a difference?
it depends on the sound and what you're intending to do with it - but general things like removing sub-30hz rumble are always best done as soon as possible.
hope this helps
Order of applying processing on tracks. - Lephrenic - 1st October 2007
Depends if you're cutting or boosting.
I'm inclined to boost after compression rather than before. I'm no expert, but it seems to me that boosting before compression would create more work for the compressor, which otherwise could work more delicately on the material. And what's more, your boost can really leap out if it's not being held back by the compressor. Though that might not always be what you want.
I'm still unsure about cutting. Perhaps it depends on your compressor, but you could find either that cutting the nasties out first makes the compressor work cleaner or you might find the compressor ends up emphasising what you just took out. Try both. See which works best.
Order of applying processing on tracks. - Macc - 1st October 2007
lephrenic Wrote:i'm inclined to boost after compression rather than before. i'm no expert, but it seems to me that boosting before compression would create more work for the compressor, which otherwise could work more delicately on the material. and what's more, your boost can really leap out if it's not being held back by the compressor. though that might not always be what you want.
this is why i was on about the iterative approach, you will have to adjust one as a result of adjusting the other. however, you can accentuate things into the compressor without getting excess peak volume, if you get me - you can accentuate things by virtue of the fact they are causing more compression.
but anyway, yes - horses for courses and try it all out!
Order of applying processing on tracks. - tyler - 1st October 2007
i'd always go for eq then compression, makes more sense to sort out the tone then risk messing up the balance over time slightly with the compression, than sort out the dynamics first then mess it up with your eq.
Order of applying processing on tracks. - ALPHA OMEGA - 1st October 2007
I put eq before the compression...
not one to go into detail but I like to 'create a nice frequency range' for whatever sound then have the compressor hold the reigns and keep everything under control.
Order of applying processing on tracks. - ALPHA OMEGA - 1st October 2007
and like Macc said,effects before compression.
every now and then I'll use an eq after the comp just to bring back in certain frequencies the comp may have tamed a little too much.
some guys use multiband compression to handle this but I'm not too comfortable with it to swear by it.
Order of applying processing on tracks. - djfada - 1st October 2007
i usually always finish a chain with compression then a limiter, but just because i read it in computer music magazine
Order of applying processing on tracks. - choronzon - 2nd October 2007
Thanks guys for the help!! Its 3 am and my eyes are shutting will read in depth tomorrow. Just finished babbling in another thread about being creativley blocked
hahahsrsly appreciate the response!!!!!!!!!!!!
Order of applying processing on tracks. - choronzon - 4th October 2007
alpha omega Wrote:and like macc said,effects before compression.
every now and then i'll use an eq after the comp just to bring back in certain frequencies the comp may have tamed a little too much.
some guys use multiband compression to handle this but i'm not too comfortable with it to swear by it.
yeah im using a multiband on this main break in my song, i dunno it seems cool like wow i can compres diff frequeny levels at same time sweet, its just like i really dont feel too comfortable with it yet either, just started using it and yeah not so sure i really fully have complete control/understanding of compressing even tho i do theoritcally its hard to tell sometimes espically with breaks. im still really getting the hang of it.
i have been liking this young w1 limiter which ive been throwing at the end of every chain on any track with drums tho, its simple and i can really tell what its doing to the sound.
btw
thank you macc for this exercise im sure it will help me with the feel for compression more so.
macc Wrote:to demonstrate this, get a raw (preferably a bit shit-sounding) drum break and compress
the living shit out of it, short attack (as short as you can), short release (say 50ms). set the make up gain so that the perceived level is the same when the compressor is on or off.
listen to how certain things - the most prominient things at a given moment - get further accentuated by the compression. the tonal balance 'jumps' around according to whatever is boss
at the time. the kick makes the hi hats disappear, maybe if the kick rings on a long time, it gets 'cut off' by the snare, which pushes the kick to the back when it hits. maybe the big fat kick makes the cymbals 'duck' (as it is called), and you can hear them pumping/breathing/sucking (all manner of terms get used here). so, compression affects tonal balance (although for 'normal' mix compression it wouldn't be as severe as this example).
btw if you wrote tutorials/ manuals id read em (and i hate manuals),,,,hahahha...
learn japanese and work for yamaha..
Order of applying processing on tracks. - Macc - 4th October 2007
choronzon Wrote:Yeah im using a multiband on this main break in my song, i dunno it seems cool like wow i can compres diff frequeny levels at same time sweet, its just like i really dont feel too comfortable with it yet either, just started using it and yeah not so sure I really fully have complete control/understanding of compressing even tho i do theoritcally its hard to tell sometimes espically with breaks. Im still really getting the hang of it.
Why are you using multiband compression?
What I mean is, why do you need to? What is it about the sound that specifically means it needs multiband compression? Ithere a specific overall dynamic problem caused by one area that needs adjustment? Can it be remedied by less convoluted methods? Or are you just doing it 'cos it's cool'?
Multiband compression is potentially a very powerful tool. However in the wrong hands - and even in the right hands - it makes things sound worse rather than better in 95% of cases.
See below, regarding level and quality...
Quote:I have been liking this young w1 limiter which ive been throwing at the end of every chain on any track with drums tho, Its simple and i can really tell what its doing to the sound.
If you really could, you wouldn't use it
Just half-joking The same here applies as above with MB compression. With any limiter or compressor (well, anything at all), while adjusting the controls be SURE to bypass the plugin regularly AT THE SAME SUBJECTIVE LEVEL. I cannot stress this strongly enough.
It's well known that if you have two identical sounds and play one at a dB louder, people will say how much better it sounds. It is essential for the quality of your productions that you check the damage you are doing to the signal before you turn it up. The dangers of things like L1 (of which the W1 is a clone) is that the minute you touch something, it sounds louder. And louder = better, right? WRONG
Same goes for your multiband compressor. Turn down the output to match the bypassed level (most likely each band's gain control is turned up) and I'll place money the bypassed one sounds better overall. The MB will most likely be killing punch, especially in the midrange - check the snare out. It's possible that the spectral balance might be better due to the equalisation side of MB comping.... but then why not just improve the spectral balance with eq rather than mess with all manner of other things in the process?
Sorry..... This turned into a MB compression rant. It's alright in its place but its use MUST be jusified IMHO.
Order of applying processing on tracks. - Slothrop - 5th October 2007
Macc Wrote:The same here applies as above with MB compression. With any limiter or compressor (well, anything at all), while adjusting the controls be SURE to bypass the plugin regularly AT THE SAME SUBJECTIVE LEVEL. I cannot stress this strongly enough.What do you mean by the same subjective level? Is that the absolute peak level of the signal ie so the signal is taking up the same amount of headroom as it would uncompressed?
My main issue with compression is the way it takes all the things I've carefully made quiet and then makes them loud.
But I'm kind of n00bish at compression. I'm going to have to sit and play with a break, a comp and a set of headphones until I get my head around what all the theory mean in practice...
Order of applying processing on tracks. - ALPHA OMEGA - 5th October 2007
Slothrop Wrote:What do you mean by the same subjective level?
http://people.seas.harvard.edu/~jones/cscie129/nu_lectures/lecture8/lecture_8.html
Order of applying processing on tracks. - Macc - 5th October 2007
Indeed - what I mean is, it is the same loudness to your ears. It sounds the same level. Whatever you want to call it
Forget peak, RMS and all that, when processing - before setting levels etc - make sure you hear what you are doing to your signal BEFORE you turn it up to level and 'make it all loud'.
Good is more important than loud, but loud can fool you into thinking it is good.
Bypass bypass bypass - it's what it is there for!
Order of applying processing on tracks. - SETI - 5th October 2007
hey chorazon, fellowe Frisco head!!!! :P
ALPHA OMEGA Wrote:every now and then I'll use an eq after the comp just to bring back in certain frequencies the comp may have tamed a little too much.
this kinda my approach, if i compress at all and the sound changes as a result.. my recordings nowadays, unlike the dnb i used to make, is about dynamics and fitting tracks sound levels with each other and watching levels rather than have a comp whack the dynamics.. gotta have nominal, uncompressed volume levels bfore anything else gets done to it, like sending it to my brother for the mastering treatment.
Order of applying processing on tracks. - Slothrop - 5th October 2007
Macc Wrote:Indeed - what I mean is, it is the same loudness to your ears. It sounds the same level. Whatever you want to call itStill a bit confused - I thought one of the reasons for using compression / limiting was getting things to 'sound louder' (and therefore hopefully better) while still taking up the same amount of headroom.
I can see how you'd end up using the compressor as a gain knob thereby making yourself think your channel sounds 'well phat' when in fact you've just turned it up. But that'd show up in a difference in the peak level rather than the subjective 'loudness', no?
Order of applying processing on tracks. - Macc - 5th October 2007
Slothrop Wrote:I thought one of the reasons for using compression / limiting was getting things to 'sound louder' (and therefore hopefully better) while still taking up the same amount of headroom.
DAGGGGGGGGGGGGGHHHHHHHHHHHH
The idea of compression/limiting is to control the dynamics of a track. The whole loudness thing stems from that, but anyway....
Yes, that is one idea - increase of perceived loudness. Not necessarily improved quality!!
Quote:I can see how you'd end up using the compressor as a gain knob thereby making yourself think your channel sounds 'well phat' when in fact you've just turned it up.
No, that's a gain change, not compression. That's a different thing. But then you (obviously) know that.
Quote:But that'd show up in a difference in the peak level rather than the subjective 'loudness', no?
BOTH.
If you turn it up, it gets louder, right? So the subjective or perceived loudness has increased, right? The peak level also increases, cos you're just turning everything up. Pleaseeeeee don't tell me I have to explain volume knobs
Right, here's what I am getting at. Quickly, cos I am supposed to be doing it, not talking about it.
You have a clean track. You put some compression on it, JUST COMPRESSION, NOT MAKE UP GAIN. Say it is doing 6dB gain reduction max. That means that (because this is what compression does - TURN THINGS DOWN!) the sound will be quieter.
What I am saying is that at this point, if you bypass the compressor, the uncompressed version will be louder, cos the compression will turn it down. Okay so far?
To take the simple view, the loudest peak will cause the most compression, right? So now, stay with me
, you have 6dB 'extra' headroom (if the max gain reduction is 6dB), right?Now if you turn up - with the compressor's make up gain - by 6dB and bypass, the compressed version will sound louder. Why? Because only the very peak is getting 6dB gain reduction. Some of the rest might only be getting 3dB GR, or none. At most points the sound is gaining (by make up gain) more than it is losing (by compression), right? So it sounds louder when bypass, right?
So everything else aside, if louder = better, then in the first case (no make up gain) the uncompressed version is better, in the second (6dB make up gain) is better. But the compression part is still the same right? So which one actually SOUNDS better, quality wise?
The answer is to set the make up gain to a value where the sound changes but not the volume you hear it at whe you bypass. To pull a number out of the air in this case, it might be +2.3dB. Then you can hear what the compressor is doing to the sound without any 'level trickery' by your ears. For example, you'll probably hear that at that peak where it does 6dB it sounds shite when compressed. Then you know you're doing too much compression.
After all this, once you are happy with the SOUND, you can set the LEVEL in your mix, happy and safe in the knowledge that you haven't shat all over your sound and fucked it up thinking it was good when it isn't.
PHEW
Order of applying processing on tracks. - Macc - 5th October 2007
BTW all that - the whole 'bypass everything at the same subjective level ESPECIALLY applies to mastering.
'MAH CHOONS BARE PHAT'
'Yes but it sounds shit'
Order of applying processing on tracks. - choronzon - 6th October 2007
Macc Wrote:Indeed - what I mean is, it is the same loudness to your ears. It sounds the same level. Whatever you want to call it
Forget peak, RMS and all that, when processing - before setting levels etc - make sure you hear what you are doing to your signal BEFORE you turn it up to level and 'make it all loud'.
Good is more important than loud, but loud can fool you into thinking it is good.
Bypass bypass bypass - it's what it is there for!
Gotcha..Thanks...Noted.
Order of applying processing on tracks. - Slothrop - 8th October 2007
Macc Wrote:BTW all that - the whole 'bypass everything at the same subjective level ESPECIALLY applies to mastering.
'MAH CHOONS BARE PHAT'
'Yes but it sounds shit'
I think that was where I was going wrong...
ie thinking of compression as a tradeoff between the amount that it sounds worse becuase it's been (over)compressed and the amount that it sounds better because it's louder.
*goes off to make things actually sound good*