resonance and mixing

[joninc]

i am doing a lot of mixing these days and one thing that I fighting with is my own instincts.

for example - when a vocal track has a low end resonance - like say 200 hz - it catches my ear which makes me think "this vocal is too woofy in this particular freq, it's distracting". So i pull out a few dbs in the zone around that area - say 2-3 db between 180-220.

i know that this vocal was recorded on a very expensive vintage mic in a classy studio but it feels like it was recorded too close to the mouth and has a build of low mud (proximity effect).

or acoustic guitars with lots of low woofiness - I want to keep some of the body
of it and definitely don't want to hipass it really aggressively BUT it's like most of the tone is in this particular zone and when I cut a bunch of that zone down - it feels too thin.

ARRGGGG.

how do you address resonances like this? do you just cut a little eq?

do you use a multi band compressor to make it somewhat dynamic in it's response?

do you boost below where you cut to try and compensate a bit, to keep the illusion of fullness/warmth?

i am looking for general tips to keep fullness and thickness but without the muddiness.

[tjcasey1]

I usually cut the EQ like you do, but only by 1 to 1.5 db. I listen really carefully to the trade-off until I feel OK with it.

I've tried the multi-band compressor with it, but it always sounded weird to me.

[wren]

https://www.plugin-alliance.com/en/prod ... eq_v2.html

I hate to just offer a piece of gear/software as a solution to such a significant, nuanced problem (and one I often still struggle mightily with), but the dynEQ has solved probably 80% of that problem for me.

Also, I've found that sometimes thinking in harmonics helps. If cutting at 200Hz makes it too thin, I back the cut off some and try cutting a tiny bit at 100Hz and 400Hz as well. Doesn't always work, but sometimes does.

Also also: if you're particularly worried about the low end here, parallel processing can really be your friend. If I'm trying to get something to sound thicker without sounding muddy, I'll create a parallel bus and distort the hell out of it (thick, fuzzy, smooth distortion - think Big Muff), and then bring that in underneath the un-distorted version. Just a tiny bit of that really helps the lows and low-mids sound rounder and more present without being woof-y or muddy or obnoxious or overbearing. (You will probably need to EQ the hell out of that distorted bus, as well as probably compress it quite a bit before it hits the distortion device/plugin - you want a consistent, smooth distortion and not one that changes radically in character when, say, the singer really starts belting.)

[MoreSpaceEcho]

+1 on the dyneq. very useful plugin. they have sales all the time, if you're patient you can get that for 100 bucks.

DMG's essence is also really good for this stuff. in fact i think it's better at this sort of thing than it is at regular old fashioned de-essing.

if you feel like the eq is making it too thin, try shelving the low end up after you cut out the resonances. and/or try narrower q on the cuts.

in general though i think dynamic eq is the way to go for these sort of problems.

[roscoenyc]

DMG's essence is also really good for this stuff. in fact i think it's better at this sort of thing than it is at regular old fashioned de-essing.

.

I never worry about a de-esser but....
I'm mixing something that I didn't record right now.
The files cam w a Waves De-esser on all the vocals and it was life sucking.
When I got the vocals sounding good the sibilance was pretty pronounced.
A friend suggested the DMG Essence plug.
Downloaded the demo then immediately bought it.
Holy moly that thing is smooth.

[MoreSpaceEcho]

and just wait till you try some other crazy shit with it and see what it can do!

tame hihats/cymbals without touching the snare at all.
take some low mid tub out of the snare without touching the rest of the mix.
likewise boomy subs on a kick....edgy guitars...the list goes on.

i still like the venerable old spitfish for plain ol' dessing, but the DMG is a seriously deep machine. for what they charge it's a steal.

[vvv]

Me, I feel like the Q control is an especially useful parameter on a parametric EQ, whether it's a colour EQ or relatively transparent.

I read somewhere that the Pulteq uses a Q of 1.3, on at least some of the settings (possibley as default?). I don't recall the exact reason why that is said to be "musically pleasing" (something about percentage of an octave?), but it usually sounds like a good setting to me, with a cut of 1dB often sufficient to make a real difference, even ITB.

As well, consider the Pultec-oriented trick of simultaneous cut and boost with one or more parametrics at or near the same frequency; see here where it's described for a kick drum. I find that I can sometimes address specific problems a little better that way, even if it can also be used more radically (try playing with a kick around 80H, or 30 Hz, for example ...)

[Magnetic Services]

For acoustic guitar woofiness I'd reach for a multiband compressor, pretty slow with mild ratio and threshold set to engage just before the resonance starts to poke out. That way hopefully it will be pretty transparent and only do the job you want it to. And then some EQ!

[drumsound]

Dynamic EQ can be useful as expressed above.

Another thing to think of is, are you EQing in solo or in context? I find that working while I hear the whole track, it gives me a better idea of the effect things are taking.

[ott0bot]

One trick I've used with overly boomy acoustic guitars is to double the track and duck it alongside with the original with the volume quite low to use the phase cancellation and thin out the track. Usually I'll eq out some of the frequencies to prevent too much phase cancellation. Little labs make a rack unit and a plug in for phase that can work for this as well.

[losthighway]

There is a point in all of our learning where we start to see how much easier it is to get away with cutting than boosting and we stop making skinny spikes of high boost, but I think there's still a purpose for some good boosting when done tastefully.

So you cut a fair amount of muck out, but your worried there's still too much woof. I figure if you boost a couple sweet spots a little bit so you sweep around and see if you can find something (maybe a little in some upper harmonics- not too harsh, maybe a little pick attack if you're into getting more rhythm on the thing, still a fairly wide q), by the nature of a boost you just made the track a little louder, so you turn it down a couple db.

If you think ratio/math wise about your pre-boost track contrasted with the post-eq/volume adjustment the lower frequencies have all dropped in volume collectively with the frequencies you boosted staying up around the same level.

This doesn't happen to work in extremes on something broad like an acoustic and it kind of relies on being successful in mining for some frequencies that actually sweeten the track when boosted a little.

... sorry was that a longwinded way of saying something simple, and only kind of helpful

[SpencerMartin]

how do you address resonances like this? do you just cut a little eq?

do you use a multi band compressor to make it somewhat dynamic in it's response?

do you boost below where you cut to try and compensate a bit, to keep the illusion of fullness/warmth?

i am looking for general tips to keep fullness and thickness but without the muddiness.

The problem with the specific type of resonance you've described (guitar, vocal) is that it moves. Resonance is often more stationary in a drum for example (it's essentially playing the same note over and over) and can be addressed really effectively with EQ. Guitar/vocal resonance comes and goes based on what chords/notes are being played at any specific moment.

For example, an open E minor chord will have a greater likelihood of being too resonant in the low mids compared to a D major chord (assuming the guitar is in standard tuning) because the lowest reach of the notes will be nearly one octave lower. If you cut some low mids with EQ so that the E minor chord sounds perfect, the D major chord will probably sound too thin.

My approach is to first use some gentle EQ to get a suitable, general tone that I want before addressing shifts in tone that are time-based or change along with the changing of notes. To address that (here's your answer, finally!) I'll use either some form of side-chained, frequency specific wideband compression, or the same thing but split band (multiband). For low mids, I like slow attack and slow release so that it just rides along smoothly with the flowing shift of tonality in a very transparent, musical way. It works wonderfully on vocals too. Consistently balanced, perfect tone that is always full but never boomy.

+1 for parallel compression. It almost always makes a good sound sound even better.

[drumsound]

There is a point in all of our learning where we start to see how much easier it is to get away with cutting than boosting and we stop making skinny spikes of high boost, but I think there's still a purpose for some good boosting when done tastefully.

So you cut a fair amount of muck out, but your worried there's still too much woof. I figure if you boost a couple sweet spots a little bit so you sweep around and see if you can find something (maybe a little in some upper harmonics- not too harsh, maybe a little pick attack if you're into getting more rhythm on the thing, still a fairly wide q), by the nature of a boost you just made the track a little louder, so you turn it down a couple db.

If you think ratio/math wise about your pre-boost track contrasted with the post-eq/volume adjustment the lower frequencies have all dropped in volume collectively with the frequencies you boosted staying up around the same level.

This doesn't happen to work in extremes on something broad like an acoustic and it kind of relies on being successful in mining for some frequencies that actually sweeten the track when boosted a little.

... sorry was that a longwinded way of saying something simple, and only kind of helpful

This reminds me of something I was taught very early on from a friend who mixes FOH. He always said as a default to cut and re-gain. The reason many of us (early on) don't like cutting frequencies, is because things get a little softer, so we think "that's not good." If you cut a frequency and also raise the overall gain of the track/channel you get a better idea of the change in context by keeping the level similar.

Conversely, lowering the gain when boosting does the same thing. You get less of the "louder is better" phenomenon.

[JWL]

There is a point in all of our learning where we start to see how much easier it is to get away with cutting than boosting and we stop making skinny spikes of high boost, but I think there's still a purpose for some good boosting when done tastefully.

So you cut a fair amount of muck out, but your worried there's still too much woof. I figure if you boost a couple sweet spots a little bit so you sweep around and see if you can find something (maybe a little in some upper harmonics- not too harsh, maybe a little pick attack if you're into getting more rhythm on the thing, still a fairly wide q), by the nature of a boost you just made the track a little louder, so you turn it down a couple db.

If you think ratio/math wise about your pre-boost track contrasted with the post-eq/volume adjustment the lower frequencies have all dropped in volume collectively with the frequencies you boosted staying up around the same level.

This doesn't happen to work in extremes on something broad like an acoustic and it kind of relies on being successful in mining for some frequencies that actually sweeten the track when boosted a little.

... sorry was that a longwinded way of saying something simple, and only kind of helpful

This reminds me of something I was taught very early on from a friend who mixes FOH. He always said as a default to cut and re-gain. The reason many of us (early on) don't like cutting frequencies, is because things get a little softer, so we think "that's not good." If you cut a frequency and also raise the overall gain of the track/channel you get a better idea of the change in context by keeping the level similar.

Conversely, lowering the gain when boosting does the same thing. You get less of the "louder is better" phenomenon.

Great stuff and I agree. It's vital to preserve overall gain structure through a mix. I notice that it's when I stop paying attention to gain structure that things start to get out of hand.

[SpencerMartin]

Oddly enough I'm addressing some resonance in an acoustic guitar track at this very moment and realized that I forgot to mention a tactic that has been HUGELY helpful to me. When I hear the offending low, resonant tone, I'll reach over to my piano and figure out what note it is. Then I'll refer to my frequency chart that I keep handy. In this case, the problematic resonance is specifically a G at exactly 98 Hz. I'll then cut there with a really narrow Q before moving on to the aforementioned process (if it's still necessary). It's always worked super well and has been a huge time saver for me.