EQ Compression technique

[Doobah]

I am interested to find out how people get the most out of sounds. I make alot of music, though i have great difficulty in taking the time to eq everything and compress before mixing. I often will eq or compress a sound quite randomly based on what it sounds like to me, though often when i listen back, or listen to professionally mastered pieces i see alot of problems with my techniques. Perhaps it's the standard of equipment i use, or an attention defecit. I want to be able to hear all of my sounds in amix, though they often tend to drown each other out. [organs clashing with bass clashing with drums etc]

Would it be a good idea to find a good eq/compression setting for a specific setting then kill the other frquencies and eq other sounds in reference to the killed sections of the other sounds, or is there a big difference between professional mastering eq's compressors and the amateur ones? I use reason, is this the problem?

I would love to hear how others go about eqing and compressing, with out having to read through a 10. 000 word document writen by an engineer based on the science of sounds rather than their musicality.

[ian-s]

[Uncle Krunkus]

I rarely compress while mixing, I usually only compress on the way down, (very gently). I EQ tracks of sonically "wide" sound by taking out a bit of what is hiding the "narrow" one.
And I agree with G2ian, when you pull something out a bit it frees up front space which can actually let you hear it better as part of the background.

[flipside]

What I'm writing is not necessarily -the- way of doing things, or the most "correct" way of doing them, but it is a way that I found working for me. So these are my experiences.. Feel free to read and try out what I tried, and maybe some of it can work for you too.. Alot of stuff is hard to explain too, some things gets intuitional after a while..

I work with hardware synthesizers, so everything I record goes through a TL Audio Ivory2 5052 (tube eq, compressor, limiter). However, that doesn't mean I boost this frequency and that frequency on every thing that goes through it. It's important that you know what your main elements are. And these may have a boost in their appropriate frequencies. Or rather, if you can, first cut the frequencies you don't need.

I would advice on reading some articles anyway though, find EQ- and compressor-articles on Sound on Sound's website, and also Future Music. Alot of the technical stuff there will probably be boring to read, but try to understand what's important for you and what's not in these articles, and try applying it to your music. When I started out with a few synths and a mixer, I sequenced my whole track via midi, and recorded everything in one go. Needless to say, a lot of sounds were battling with eachother. After reading a lot of boring articles, however, I started to understand the basics of how I should work with an EQ, and things started getting better..

Basically, what I learned was this:

1. lowcut everything that shouldn't have low-end frequencies to it. (Usually that's everything except your bassline and bassdrum, and maybe some percussive sounds). This alone will clean up your mix considerably (it cleaned up mine atleast)

2. 300hz is usually considered a "muddy" area.. If an element in your track seems "muddy", "wooly", and you don't like it, it may help to reduce around 300hz

3. every sound has atleast one "sweet spot" I think.. Find it by doing this: (First make sure your volume is turned slightly down).
Set up an EQ, using something between a bell and a spike shaped curve (you need an eq that features a "Q" parameter), start at the higher end of the frequency spectrum (16khz+), boost it a good deal (don't worry if it's too much), and sweep slowly down the frequencies and pay close attention to when the sound "jumps out", and gets noticeably more clear. Now reduce the boost so it's not too harsh. If this is, say, a lead sound, set up another EQ point with a lowcut, start at the bottom end, sweep upwards till it sounds a bit thin, and then back down a bit.

4. You can probably lowcut your mix at around 40hz. Some will argue and say either a higher or a lower value, but this works for my music. Find out what works for your music.

5. If you have more melody/lead elements to your track, and did what I described in step 3 on one sound, you should try -reducing- the boosted frequencies of the first sound on the second sound. It's all about allowing every piece of your track to interact over the frequency spectrum. You only have so much space, and if you put too much in, some things will cancel eachother out and sound messy.

There's more to it, and probably other/better ways of saying it, but I can only hope it helped a little Good luck!

Best,
Alex

[opg]

I do just about the same things - the lowcuts, and 40Hz for the whole mix. Sometimes I even choose my instruments by what frequencies haven't been "used" yet in the mix. This can be rather limiting, though (pun intended).

Where I get frustrated is mastering. Not necessarily the process, but deciding whether or not to think about it when I'm still making the song. What I mean is: I have goals as to how I want my songs to sound (tape saturated, playing from a cheap speaker, lots of vinyl crackle, etc) and I don't know whether I should make adjustments for that while I'm making the song or wait until it's done. There are a lot of effects plug-ins out there that can lead you down bad paths or habits.

The next time I finish a song, I've decided that I will record the whole thing onto cassette or reel-to-reel, and then back into the computer. Then I may tweak it with some mastering program like TRacks. If I do this, should I back off on any tape saturation effects plug-ins that I've been accustomed to using? Who knows? All I know is that there is going to be a lot of experimentation to get it the way I want it. I have a feeling though that if I'm trying to get a vintage hardware sound, I'm going to have to use vintage hardware.

This still doesn't make me comfortable about the samples/instruments I use and what EQ, compression, and limiting I should add to them. Say for example, I want a fat sawtooth synth bass and a kick drum from an old drum machine (Korg Minipops, Watford, etc - thanks Hollow Sun!) playing at the same time and they're hanging around that 65Hz area. I'm going to have to use some compression, I guess, or pull out a little with EQ on one or the other. I hate doing this sometimes because the instrument never sounds as big as it does by itself. Most of the time, I'll lowcut the kick drum and then watch a spectral analyzer to see what the synth is doing way down low.

I guess I'm saying that some of these processes that are easily available to us now on computer were not there to use on those songs that were recorded for vinyl, 8-track, or cassette. And therefore, should I be doing them if I'm planning on using some vintage hardware after the song is done?

Mother of God that was long-winded.

[paul e.]

[paul e.]

[paul e.]

[opg]

[flipside]

[paul e.]

'tis true..very subjective

any habit, that is done 'just because', can get you in to trouble sometimes

i.e. always rolling of 300hz, or always recording your bass lines thru a tweaked tube EQ ;]

it really depends on the mix, and the interaction of the frequencies of that mix

[elektro80]

[elektro80]

It must be said that some popular genres are clearly showing signs of having been tuned to what actually will work both live and on "tape". How you write the music will in many ways affect how easy your life will be in the studio. Certain genres, like big band brass stuff are all about getting the arrangements right. And some genres are also showing signs of spatiality being a very big part of the sound.

[elektro80]

[elektro80]

http://solomonsmusic.net/insrange.htm

Check out Larry Solomon´s "Instrument Range Chart"

Hot!

[opg]

[elektro80]

[elektro80]

[mosc]

I write mostly electronic music made with modular synths or some such stuff. The most important part is making the sound right before it is recorded. As a rule, I think if you have to do eq or compression during the mix or mastering process, you are in trouble.

Now, if you are doing pop music, this of course is not the case because there is a set model of what the music should sound like; heavy compression on the vocals, almost zero dynamics, etc. Like G2IAN says, professionally mastered pop productions make all the lites turn on all the time, but to me these recordings don't sound very good. I get fatigued listening to them. That's ok, Britney Spears won't probably be looking to me to mix and master her music.

My background is more appreciative of classical music. Here, if the performance is good, the recording is good unless it is botched somehow by poor mic placement or such. I used to do a lot of recording of acoustic folk groups and small rock bands. I paid very careful attention to levels and panning, but not too much with eq (mostly used to reduce wolf tones) and hardly ever did I use a compressor. People said they really liked my recordings because it was so transparent and made the music sound more musical. I think the recordings of Creme were made with this same keep it simple philosophy. When you get three guys that play that good, you don't start messing with EQ and compressors. I saw them play in Toronto in 1968. I talked to the sound man. That was pretty much what they were up to. I guess that made a lasting impression on me.

[seraph]

[opg]

Hmm. Indeed. White noise.

I've always tried to follow Howard's remark on "if you need to use compression or EQ, you're in trouble." I always thought that was one of the reasons some artists were drawn to minimalist electronica. If you were working with pure sine waves (or any other waveforms, for that matter), you could avoid some of those headaches.

I remember when I first discovered VST plugins, and like most people, I WAY overused them. That caused a lot of problems and essentially sucked the life out of all of the songs.

[paul e.]

[Kassen]

[elektro80]

Dynamics processors, filters and the lot are creative tools as well as problem solvers.

Unprocessed isn´t better than processed or whatever. It is easy to demonstrate how an untreated multitrack recording of an acoustic jazz quartet will sound more artificial and less natural than a dynamics processed recording. And it is possible to prove the opposite is true too. The main point is to understand what is going on and have a fairly sensible reason for doing whatever is called for.

There is a mythology related to EQ in the sense of there being some kind of secret code among pros as to what is to be tweaked and what is to be left alone. This is all true. If you need to tweak something you will of course do so. Imagine how the proximity effect of a certain microphone in some cases will be both unwanted and also very vital. How do you solve the various issues in order to get the sound you want?
Imagine running the mikes hot through coloured vintage preamps. The whole gain structure makes it important to sort out bleeding issues between the microphones at an early stage. Yes, you just might have to cut bass on some inputs. You might even reconsider the whole thing and choose to use very transparent pres and sensible levels, and then instead sort out spectral issues on each channel before you run the old vintage console hot as hell at line levels. And it might even be a lot smarter to emulate some of the analogue audio/vintage console shit in the digital domain using stuff like the UAD-1 anyway.

[opg]

Man! This is some good shit! It's making my brain smoke, though...

Follow this train of thought and see where I can improve:


I don't know if this is the same for others, but I love that feeling when you say to yourself "Alright, I'm going to start out with a clean slate." I'll think very carefully about what I'm adding; I'll watch a spectrum analyzer as I play whatever instrument I've chosen. Usually, I'll start with drums, as I am more particular about what kind of sound (vintage drum machines!) they should have as opposed to bells or electric piano. As I may have mentioned before, I like drums that are overdriven and bit-crushed (8 bits so you get that hiss and crunch - mmmm... that should be a cereal "Binary Crunch"). I've therefore added some plugins that will make EQ changes (and possibly some compression from the overdrive/distortion). I will continue to view the spectrum analyzer.

ANYWAY, at this point I feel that I have limited my choices for other instruments. Perhaps the synth bass sound I would like to hear clashes with the kick drum (65Hz, 129Hz, 150 Hz, whatever). So, what are my options, then? I could:

A. Make EQ adjustments on one or both clashing instruments
B. Run both through a compressor

If I choose "A," I usually run into the problem where my synth bass is no longer "warm." Perhaps this means I overdid the EQ. I would hesitate to make any more EQ changes because it could lead to more problems with other instruments later.

If I choose "B," I usually run into the problem where the life gets sucked out of my drums (BTW, I usually send all pieces of the drumkit into the same effects channel where they are receiving the same amount of overdrive and bit-reduction. I go with my ears, and I watch the analyzer, and I like what I hear and see so I am comfortable with this decision). What do I mean by "life gets sucked out?" Well, the bass and kick drum sound great, and neither is overpowering the other, but that "hiss and crunch" seems to be lessened on the drums, but more importantly - the drums loose their sense of "closeness." The high end, I guess? So what do I do? Do I boost the EQ on the drums to make up for this? It also should be noted that when I add in the drum samples, I will view each hit through the analyzer and determine if any HP filter/bass roll-off is necessary.

MAN! It's like I need some kind of flowchart or something...