Order of Effects

There's plenty of information out there about the order in which one should apply effects such as EQ and compression (to individual tracks). However, I have found conflicting information. Some say EQ should come first while others say the opposite. Although I've come to learn a fair bit about how to use individual effects, I'm still unsure about the order.  I'm curious about what your view is on this topic and your reasons for it.

I currently apply the following order:

1. Noise Reduction
2. Compression
3. EQ
4. Reverb
5. Limiter

Is this right?

I currently apply the following order:

1. Noise Reduction
2. Compression
3. EQ
4. Reverb
5. Limiter

Is this right?

Nearly... and I am assuming that you are talking about a normally recorded track, not some sort of strangely effected one for a special purpose.

On a properly recorded track, Noise Reduction doesn't enter into this - you simply shouldn't need it. In general you would apply compression before EQ, certainly,  because if you apply it afterwards, it will screw with the EQ anyway! The way to remember the order is to think of what happens in a mixer - in order, there's a mic pre, an insert where compression can be applied, then the EQ and after that, any other temporal effects like reverb. Specifically what happens with compression after EQ is that if you have put any form of boost onto a signal, the compressor will react more to this than the rest of the signal, and what this means is that the EQ you thought you needed, you haven't got any more because it's been compressed.

Yes you might find alternative points of view, but these would be the ones who haven't considered what actually happens... 

The one thing you generally don't  do though is to apply limiting, which is a very severe form of compression, after the reverb. This would completely wreck any form of natural sound from it at all. If you want to severely limit an individual track, you'd normally do this after EQ and before adding reverb, not aferwards - compressed reverb isn't pleasant to listen to at all.

The argument is slightly different if you are considering the overall result - you might want to apply a little overall broadband soft-knee compression, depending entirely on the content and the result you desire, but limiting? No way! Not on a natural-sounding track, anyway. If you want to make one of those awful pumped-up noises that are supposed to sound deafeningly loud regardless, then yes, you could compress the life out of the track, but I  certainly won't be listening to it through choice, I'm afraid.

The order in which track processing is generally done has been agreed by the majority of people who mix for a long time now - and actually, there's less dissent about this than you might think.

I think a lot of it depends on the result you're looking for.  I won't argue with anything Steve has said regarding the most natural sounding track.

But beyond that, it's going to be a creative decision, just like everything else.

I think Steve will agree that sometimes individual tracks don't sound very natural at all...But when mixed, they fit together like a puzzle.  It's your job to make them fit.  If something will fit better by using the "wrong" order, do it.

If you've never done this, try mixing with NO effects.  Just doing this with one song should be enough to help you forever.  Once it's as good as it can possibly get, listen critically and add compression only where it's needed, and only as much as you need.  Then go back and apply subtractive EQ, again, only where it's needed, and only SUBTRACTIVE EQ.  Listen critically again and apply reverb...Just enough to add space and depth.  Finally, revisit your EQ and apply any additive EQ you may need, and other "creative" effects.

Someone told me to try this and it immediately made me better.  It really made me appreciate each tool, and I learned better how to effectively use each one in the context of a mix.  I now do a lot less "abusing" of tracks.  I use my ears a lot more now, as well.  My mixes breathe better, punch harder and shimmer more.  They sound a lot less "plastic" than they used to.

Emmett

I think Steve will agree that sometimes individual tracks don't sound very natural at all...But when mixed, they fit together like a puzzle.  It's your job to make them fit.  If something will fit better by using the "wrong" order, do it.

Usually tracks don't sound very natural after you've added FX to them! But in general, I'd agree. Almost just from the names of some effects you can tell that they are implied to be corrective, and I'd agree entirely about the principle that you mix dry and only put things right when you need to...

The trouble with that as far as mix novices are concerned is that if they do this, they can't always tell what's not working, or just as importantly what could be working better. This inevitably puts them into a conundrum; they you can't take a track from within a mix and treat it in isolation, but you can't always tell whether it's the track or everything else around it that needs fixing. And this is generally why people tend to take sections of a mix at a time, and generally start with getting things like rhythm tracks (specifically percussion/bass) sounding the way they want before fitting everything else around them.

Mixing in stereo is a giant artificial acoustic con anyway - there isn't really any absolute right or wrong about the results, because they invariably sound different wherever you play them. If you can get a good result from a grot box, then it invariably sounds okay on better systems, so really you have to say that this should be your reference, in all honesty - you have to mix for intolerant systems, not tolerant ones. And that's the problem - trying to do an un-effected mix that sounds good on an intolerant system is actually a very difficult thing to achieve. So most people starting out end up at something of a disadvantage, because it's not a gentle slope they are trying to climb, but a mountain.

Most specifically, your average grot box has no bass or treble extension, and is going to distort the midrange in as many different, unpleasant ways as I can think of. So if you mix on it, you invariably end up with a result that, if played on a more respectable system, has a scooped-out midrange, because you turned up the bass and treble, and tried to avoid as much of the nasty noise in between as possible. The perverse thing about this is that this is the sort of sound that people actually like - if you want to play it quietly, it still sounds okay, because you don't need so much loudness compensation, and invariably listeners don't know what the original sounds were like anyway. So, they don't think that there's anything 'wrong' with the results. And who's to say that this is actually wrong?

Well I can, I suppose, but nobody's going to listen to me saying it's wrong if they actually like the results, are they? That's why sometimes I have to prove that the result they like isn't as good as it could be, because there's one they might like even more, only they didn't realise it. But it's all these things that have led to the usual recommendations for doing mixes that we've made in the past, and I don't suppose that it would hurt to reiterate one or two of the most important ones (the order is not significant). Most of them are about trying to retain some degree of impartiality about what you are doing:

• Don't mix immediately after laying tracks - leave it a while.
• Check mixes on several systems.
• Get familiar with the sound of similar commercially successful tracks on the monitoring system you are using - that's really your reference, so refer back to it often until you are really used to the way it is - your ears can easily deceive you about what you are doing, even in the short term.
• Play mixes really quietly (I mean down at background noise level) and see what you can actually hear - this is a good way to set vocals at a sensible level, incidentally - you should still be able to hear that there is a vocal when it's really quiet.
• An alternative take on the above, that can often help with bass levels, is to listen to the mix from behind a closed door.
• Don't mix too loud - this is invariably more tiring, and gives you a very false idea of the result at normal levels.
• Take frequent rests - this invariably means that you can go on longer without tiring. And always make the final decision about a mix first thing in a new session the following day, not last thing in the previous one.

These are just off the top of my head - I'm sure that I've left things out, and if I think of any more in this category, I'll add them. And of course additional comments are welcome.

[...]
But it's all these things that have led to the usual recommendations for doing mixes that we've made in the past, and I don't suppose that it would hurt to reiterate one or two of the most important ones (the order is not significant). Most of them are about trying to retain some degree of impartiality about what you are doing:
[...]

Can I quote this advice on another forum, appropriately credited and linked back to the source?  No, it's not KVR, but a small community site based around a band you'd probably abhor.  But a lot of us on that site make music with varying degrees of professionalism, and do discuss production techniques.

Well I can, I suppose, but nobody's going to listen to me saying it's wrong if they actually like the results, are they? That's why sometimes I have to prove that the result they like isn't as good as it could be, because there's one they might like even more, only they didn't realise it.

This is exactly what I learned from going through all of those steps.  It got me away from presets.  One must admit that acceptable results can be obtained using entirely presets, and it's hard to let go of them.  But by taking the time to really critically listen to everything, you find a realization that presets mostly get in the way.  I have said that it made my mixes better, but I should specify that they became better because I really learned to listen better, make better judgements and problem-solve.  It's both a blessing and a curse.  My life used to be much easier, just using presets and being content with the result...But now that I can hear how much better it can be, I spend a considerable amount more time making everything fit.  So it is certainly a lot more work, but when I go back and listen to past stuff, the difference is night and day.

The ears are the most valuable tools you have...But they're  a lot harder to use properly than most people realize.

Can I quote this advice on another forum, appropriately credited and linked back to the source?  No, it's not KVR, but a small community site based around a band you'd probably abhor.  But a lot of us on that site make music with varying degrees of professionalism, and do discuss production techniques.

Really I'd prefer just a link - it's when stuff gets quoted elsewhere that others think that they can just pick it up from a site and use it, even when it's attributed. I know that I can't exactly stop anybody, and I don't doubt your integrity in the slightest (you asked, after all, and that's quite rare) - it's those other buggers...

The whole business of intellectual property is quite tricky when it comes to sites like this, and it puts people like me in a bit of a quandary sometimes. Individual contributors to the site do retain rights over the exact words they use, but the ideas are almost unarguably in the public domain, and there's nothing to stop anybody rewriting things in their own words and claiming the work as their own - even though it isn't. And it's in that light that you have to consider the above - most of those ideas are universal, some are self-evident, and only one of them really came from me in that form - that's the one about listening to almost silent mixes, and even that I can't guarantee that nobody else has ever thought of. But whilst the ideas are somewhat universal, the particular form of words, and grouping of those ideas is unique. And generally people are lazy, so they don't rewrite stuff or attribute when they really should, and that's when all the accusations start flying about.

So you can either rewrite it so I wouldn't recognise it, and just acknowledge the source, or provide a link directly to it, but I'd rather it wasn't just quoted verbatim, even with an attribution.

Really I'd prefer just a link - it's when stuff gets quoted elsewhere that others think that they can just pick it up from a site and use it, even when it's attributed.

No problem, I can quite understand that.

Your response has been quite comprehensive and helpful, as I've come to expect of this forum.  I appreciate your thorough input.

The difference in the opinions you have expressed on this matter can be reconciled. The order of effects should be viewed as a guideline, expressed as follows:

1.   Compression
2.   EQ
3.   Reverb;chorus;etc

Deviations from the basic order are appropriate, but only in specific and narrow circumstances, which are to be dictated by the overall goal and result achieved.

Yes, that's basically about it. Breaking the basic rules is fine, as long as you know why you are doing it, and what the consequences are likely to be.

[as long as you know why you are doing it]

Yes, that's basically about it. Breaking the basic rules is fine, as long as you know why you are doing it, and what the consequences are likely to be.

That's exactly what I was going for...I guess I could've saved some words typed, huh...

Emmett

Yeah - after the event, I realised that I should have removed a redundant 'basic' from it - and then it would be even shorter! 

There's plenty of information out there about the order in which one should apply effects such as EQ and compression (to individual tracks). However, I have found conflicting information. Some say EQ should come first while others say the opposite. Although I've come to learn a fair bit about how to use individual effects, I'm still unsure about the order.  I'm curious about what your view is on this topic and your reasons for it.

I currently apply the following order:

1. Noise Reduction
2. Compression
3. EQ
4. Reverb
5. Limiter

Is this right?

Leave out the noise reduction, limiter, and compressor and you'll be alright.  Why people continue to think you need to apply this stuff in mastering is beyond me!

Leave out the noise reduction, limiter, and compressor and you'll be alright.  Why people continue to think you need to apply this stuff in mastering is beyond me!

I got the impression from the OP that this wasn't about a mastering issue. If it comes to the crunch, I can't see why you shouldn't use a small amount of NR to make a not-too-good track into a usable one if it's not possible to re-record it if you can do it without introducing artefacts, even though this shouldn't really enter into the process at all. But as far as compression goes, it's not at all unusual to remove or reduce peaks in a track to make it sit better in a mix. Personally I wouldn't use hard limiting at all, certainly, but it's by no means unusual to apply soft-knee compression to a track during mastering as a rescue process - although I'd personally do a remix if I found that this improved things, I must admit. But in a very realistic sense, I don't think that one can be totally prescriptive about this - it's almost as bad as just using presets!

But, as a friend of mine said years ago - If it works, it's neat! (I don't entirely agree with the philosophy, but he has got a basic point, I think)

I should have clarified at the outset that I am not referring to mastering.  My question was with regards to mixing only.

Leave out the noise reduction, limiter, and compressor and you'll be alright.  Why people continue to think you need to apply this stuff in mastering is beyond me!

Now that we're on the mastering issue though...I'm surprised that you included "compressor" in that list. I find that sometimes a final mix can benefit from some subtle compression.  I'm not exactly sure why - I just know it sounds better.  I suspect that this has to do with the obvious effect of reducing dynamic range, but I think also because it makes all the tracks gel together.


NR is not part of my normal process at all, but I included it in my list anyway so that I could figure out in what order it would come in should the need arise.  Limiting, however, is something that I use to do often...it explains some of the problems I've had, especially when limiting after reverb.  I'll be avoiding it in the future.

I was using limiting to amplify sounds without effecting their dynamic range too much and ensuring that they don't clip.  How can I achieve these same results without a limiter?  I suppose that a compressor, when applied correctly, can be used to achieve the same results.  Or can it? Should I be using a normalizer for that purpose? or is the alternative to simply re-record the track at higher levels?