The Case For Graphic Equalizers

[MetroCenter]

I've seen a lot of criticism of graphic equalizers, and "rules of thumb" that say "don't ever use them for recording". So I wanted to present my experience with my graphic eq. I'm more than happy to take constructive criticism of my technique, as long as it doesn't get personal.

I have a dual graphic eq (DBX 231s). I'm not using it for monitor adjustment or compensating for room issues. When I use it (which is not too often), it is during the editing/mixing phase to radically shape the sound of certain instruments.

See, my songs often have 10 instruments or more, and I sometimes need an instrument to provide just a tiny sound to the mix. Sometimes the purpose is to remove lots of bottom end. And sometimes the purpose is to remove several frequency slots, to allow for other instruments. When I'm cutting out so much of the sound, it just makes more sense to me, to use the graphic eq than the parametric eq.

As an example, I sometimes require an acoustic guitar with almost all of the low-end rolled off, so that the guitar functions more like a percussive instrument. It just doesn't seem feasible or logical to try and do something like this with a parametric eq.

My recordings sound good to me, so I would say it's a good solution, right?

Is there a better hardware solution? (I am a hardware recordist and do not plan to switch to software/DAW, so please don't recommend plugins. )

 

[bouldersoundguy]

If you find a graphic to be convenient then go ahead and use it, but parametric eq can do anything graphic can do, and do it better. They just take more skill and time to set. The main advantage of a graphic is that it's quick and intuitive to use. That's why you see them at FOH in live sound systems.

Note that just because the filter centers on a graphic are 1/3 octave apart it doesn't mean the filters are 1/3 octave wide. In most cases they are substantially wider than that, so they aren't exactly precise.

 

[crankyelbow]

One of the best additions to my guitar rig was an EQ. Now the recorded sound starts about where I want it to be.

One could argue, however, that recording as dry as possible with as little in the signal chain as possible will yield higher quality results - as digital processing is mathematically superior.... but in this world of sonic voodoo where words like "wooly" and "warm" describe waves of pressure I've come to realize - anything goes =)

 

[Bristol Posse]

As an example, I sometimes require an acoustic guitar with almost all of the low-end rolled off, so that the guitar functions more like a percussive instrument. It just doesn't seem feasible or logical to try and do something like this with a parametric eq.

So not a big fan of the High pass filter on a parametric EQ that would do exactly that with one control then huh?

 

[bouldersoundguy]

If you're going to use eq it really helps to actually know something about what they are doing to your signal. That's why I ran a bunch of tests on some I had laying around the house.

EQ Frequency Response

 

[MetroCenter]

Thanks for all the comments!

I do always record "dry", so I can always revisit the original as necessary.

I think the real problem is when people expect them to produce precise curves (which they don't). I can imagine people getting really upset when they're trying to tweak a key sound to perfection, for it to end up mangled and distorted.

Bouldersoundguy, your graphs are useful in showing the limitations of graphic EQs, thanks for the link. (Although I wonder, given the condition of your test boxes which were "laying around the house", if they are representative of graphic EQs in general.)

So not a big fan of the High pass filter on a parametric EQ that would do exactly that with one control then huh?

The high-pass filter on my parametric only goes up to 300 Hz.

Maybe the best answer to my own question is: use graphic eq only as a last resort, when a parametric eq or shelving filter will not do the trick.

 

[bouldersoundguy]

Bouldersoundguy, your graphs are useful in showing the limitations of graphic EQs, thanks for the link. (Although I wonder, given the condition of your test boxes which were "laying around the house", if they are representative of graphic EQs in general.)

The White is an ancient eq, but it was serviced and up to spec when I tested it. The Rane probably has a bad ground in the power supply that only affects the boost side of the filter. In proper working order the top curve would mirror the bottom one. All the others were in proper working order. There used to be a site with a comprehensive series of eq measurements done by "real" engineers, and my results are consistent with theirs. Unfortunately that site no longer hosts those test results.

 

[MetroCenter]

I'll be the first to admit that I am still learning. Eventually I may decide to part ways with my graphic eq, once I'm able to recognize its limitations by ear.

BTW, on the broader subject of equalization, I just found this excellent webpage describing best practices in the studio. I've seen some other discussions of eq on the web, but this one is probably the best I've seen so far.

 

[diggy_dude]

There's no rule that you can't use both. Use a parametric for precise surgical eq'ing of individual tracks, especially if the track has a specific problem that needs help. Use a graphic to color the overall mix.

 

[jaygee]

Maybe the best answer to my own question is: use graphic eq only as a last resort, when a parametric eq or shelving filter will not do the trick.

Keep it as a tool in your bag of tricks.
It doesn't have to be a last resort. If it works to arrive at the sound you wanted, then it is just as valid as any other method.

 

[Bobbsy]

There's no rule that you can't use both. Use a parametric for precise surgical eq'ing of individual tracks, especially if the track has a specific problem that needs help. Use a graphic to color the overall mix.

Exactly.

One of the best live sound guys in the world (Dave Rat--he does the sound for Red Hot Chilli Peppers, Blink 182, Soundgarden and lots of others) recently posted a picture of his FOH racks HERE on his 23 June entry.

I was interested to see that he uses parametric EQ for all his "set up" stuff--but keeps one BSS GEQ in circuit to "grab and go" for any mid-show changes he has to make.

If it's good enough for Dave Rat, it's good enough for me!

Quite seriously, I use a mix of the two when I'm doing mixing. I can't even say why I choose one over the other (though the GEQ tends to be where I go for gentle wide-band boosts and cuts and Parametric for more "surgical" changes--but that's more habit than any need).

However, they're both tools in the tool kit.

 

[bouldersoundguy]

The only real advantage of a graphic over a parametric is how fast and easy it is to use. When you have the time a parametric will always get you closer to the ideal result. Pretty much all serious PA systems have parametrics at the input and output of their speaker processors. The graphic at FOH is there for quick and dirty "emergency" eq and to keep guest mixers happy. In the studio there shouldn't be eq emergencies.

 

[witzendoz]

Graphic equalisers are fine, I sometimes use one on problem kick drums to cut the mud or to add attack. I also use graphics to set up side chain compression, where you compress curtain frequencies on a compressor, also to set up a compressor as a de-esser.

The main thing that they should not be used for is to eq the monitors to make up for room problems.

If you can get your head around parametric eq's they are even more useful for cutting out problem frequencies.

Alan.