EQ & Compression Discussion

It seems that if you can get a firm grasp on eq and compression techniques that you've overcome most of the battle in mixing. I've become a little obsessed over this topic recently and there are a lot of resources for tips online but too often they are focused towards recording live drums. I would be very interested to read how you approach eq and compression for electronic styles of music, especially on drums.
How do you determine when compression is needed?, Do you compress every hit separately or parallel compress the drum mix bus?, Do you eq boost a lot or go more subtractive?, What are the average ranges that you high pass for kicks, snares, hats?. There are so many schools of thought with all of those processes that it would be good to read some opinions from people creating similar styles.

My reading and experimentation so far has taught me that it's usually a good idea to nail down the low end of your mix before moving on. This often involves high passing unwanted low frequencies, sweeping to find the sweet spot and boosting a little, cutting out the mud in the low mids and carving out narrow ranges in your bass that you have boosted in the low end of your drums. I've also experimented with compression on the bass side chained with the kick. The fine tuning of all these techniques are of course dependent on the source material and the kind of sound you are aiming for but some general guidelines as a starting point would be helpful to a lot of people I'm sure. The way that the subtle changes to frequencies of one track affects the relationship between the other tracks and the perceived loudness is a fascinating subject to read about.

I think it's really helpful to distinguish between creative and non-creative uses of eq and compression. For instance, compression can be used to either make your drums fatter and punchier but it can also be used to even out the dynamics, trim peaks, tighten up low-end, or to thicken up anemic sounds. I always put compression after reverb, too. Friggin' huge.

I use EQ to clean house. I HPF pretty much everything (even bass and kick). I try to make my mixes 30hz-15khz. Like you said there is always mud in the 300-400hz region that needs to get cut (although I've been listening to alot of early 90's electronic stuff lately and I actually like how muddy stuff was back then). The bottom line for me is that there is always way too much frequency information so if you want clarity you gotta hack away at everything, otherwise you get dull indistinct mushy 'amateur'-sounding mixes (which is not necessarily a bad thing).

  On 6/28/2013 at 9:04 PM, LimpyLoo said:

I think it's really helpful to distinguish between creative and non-creative uses of eq and compression. For instance, compression can be used to either make your drums fatter and punchier but it can also be used to even out the dynamics, trim peaks, tighten up low-end, or to thicken up anemic sounds. I always put compression after reverb, too. Friggin' huge.

 

 

I use EQ to clean house. I HPF pretty much everything (even bass and kick). I try to make my mixes 30hz-15khz. Like you said there is always mud in the 300-400hz region that needs to get cut (although I've been listening to alot of early 90's electronic stuff lately and I actually like how muddy stuff was back then). The bottom line for me is that there is always way too much frequency information so if you want clarity you gotta hack away at everything, otherwise you get dull indistinct mushy 'amateur'-sounding mixes (which is not necessarily a bad thing).

What kind of db level would you say you are typically cutting? I know in the mastering stage it should be very subtle at maybe only 0.5-1db but in the mixing stage I've seen people cut 10+db in the low midrange. I've heard it said more than once that you should never cut or boost more than 4db. I'm with you on going more subtractive with eq. I've been hacking the shit out of my tracks and it always makes me feel good when i see a lot of information in a range then sweep over it with the hpf and hear no change at all. It's a good feeling to know that you just cleaned up a nice bit of space in the mix with no detriment to the sound. The more I think about it, the more it makes sense to to go subtractive eq to compressor and then back into a 2nd eq to bring a little back in again.

If i'm gonna do something like cut out 300-400hz then it's usually something reasonable like 2-3db with a broad-ish Q.

The 4db rule seems very arbitrary to me. If there is frequency information buried in a sound that I want to bring out then I'll boost until it's at the level I want. Of course, that is in a perfect world where we all have beautiful outboard eq's and don't have to worry about phase distortion from making drastic moves in plug-in eq's (although phase distortion doesn't always sound bad). But phase artifacts aside, I wouldn't (and don't) hesitate to boost 10db if need be.

And yeah you're right mastering is totally a .05-1db game. It's all about a bunch of tiny little well-placed moves.

everybody uses spectrum analysers these days, even drum and bass people like noisia.

in your daw you can analyse whats there then use high/band/lowpass filters if you really need it. if you're having problems with phase then an oscilloscope might be useful.

i wouldnt generally EQ some arbitrary area like 400-500Hz on every single track, i would check with the analyser and listen and if there was a need for filtering, then i might add some.

where did you get the idea about 30Hz-15KHz? it's interesting, i think digital sounds too harsh at the high end, when i record my 808 then the hats and shaker sounds arent coming out very nicely. i guess analogue recording would solve that problem, along with keeping the levels down at the high end.

did you know about the STC coles 4038 BBC microphones? the upper frequency range was about 15KHz too. and it's difficult to prepare vinyl records when there is a lot of high frequency informatin above 15KHz(?).

apart from EQ, it's better to choose the right sounds in the first place, not exactly sure what style of music youre making? if i use compression it's probably hard limiting to make the track 3-6 dB louder but thats rare. i see compression as a special effect that people apply to get a hip hop or drum and bass type of sound so i try to avoid it

30hz-15khz is (roughly) the frequency response of a reel-to-reel at 7.5ips. That's where I got the idea. Plus I have sensitive ears.

Either way I like the high-end shelved off.

do you think things like sp1200 sampler get a good reputation for their '12 bit crunchy samples' because the dynamic and frequency range is being reduced naturally. i.e. the 12bit sound compresses the sample and reduces the high frequencies, making the sound more thick and pleasant

no rules

  On 6/29/2013 at 1:24 PM, chunky said:

do you think things like sp1200 sampler get a good reputation for their '12 bit crunchy samples' because the dynamic and frequency range is being reduced naturally. i.e. the 12bit sound compresses the sample and reduces the high frequencies, making the sound more thick and pleasant

I would say the samplerate has more impact on the character of the sound than the bitrate. When you hear a kick drum from an SP-1200 the bulk of the character comes from the aliasing (which is related to the samplerate). And judging by the amount of aliasing I would guess that particular Bibio kick drum was pitched down, which is a common practice on account of the SP-1200's 10-second sample time.

The filters are also a huge part of the sound.