Need advice on taming the bass on a stereo file

noisewreck

New member
OK. I am in that non-envious place where I am stuck with a mix with too much subbass. Unfortunately I cannot go back to the project, because it's not there anymore, so all I have is this 48kHz/32bit mixdown to work with.

The genre of the track is a crossover between Drum'n'Bass and Industrial/Powernoise.

Thanks to mixing with headphones, I have overdone the subbass. It sounds great on the headphones, decent in my car and like a 400lb lady's flabby belly on the stereo.

Now, since it has Drum'n'Bass elements in it, the subbass needs to be there, and for this particular tune needs to be bombastic. However, as it is, it's obliterating everything else, eats up gobs of headroom, which affects overall perceived volume and all kinds of other calamities :D

At first blush it shouldn't be too hard to fix, since it is specifically a separate subbass line, panned dead center, that was done on a Juno-106, so it's sitting in it's own place, down around 40Hz - 70Hz range.

However, rhythmically it is duplicated by a couple of kicks that are sitting just above it also panned dead center. The trouble is, when I try to shelve it out somewhat, the kicks start getting affected as well.

So, I am thinking of two other options:

1. Automate EQ to follow the subbass notes. This is gonna be tedious as hell as the notes change all the time, and the track is around 6:30 min long. Also, the EQs that I have aren't gonna really cut it. I got the plain vanilla EQ that comes with UAD, which doesn't take to automation too well, the EQ modules in Reaktor that... (hey, that might work!!! just realized about Reaktor as I am typing this at work), and the built-in EQ in Cubase that sounds... meh.

2. The other option I was thinking of was getting my hands on a linear EQ, make a copy of the track, Hi-cut the copy down to around 80Hz or so, flip the phase on it and bring it's level up enough to lower the overall bass energy down by somewhat cancelling out the main track. Although I am not sure how different this would be from simply shelving out the bottom end, which I've tried unsuccessfully so far, and also I am thinking that even linear EQs aren't gonna be so linear as to not cause issues in the passband.

So, anything else you guys can think of? So, the goal is:
1. Tame the subbass
2. Not lose the overall punch in the combined subbass/kick combo.
 
Sounds like goats head soup. Try some lower band compression. If your troubled between the kick and bass competing try changing the attack time. A faster attach should bring out more bass. A slower attack will bring out more kick. Balance between the threshold and gain to get the punch you want.
 
I hate to sound like a broken record because I give this answer so often, but IMHO this is an ideal candidate for the ol' parametric sweep to map narrow-Q cuts to surgically liposuck the flabby tones from the sub-bass wile leaving the meaty tones of the kick alone. Sometimes after making the cuts a little low shelf boost of just a couple of dB can sometimes help to restore the overall bass energy out of the good stuff after so much of the bad stuff has been notched out.

I normally would also recommend M/S decoding the signal before doing so, so that you're isolating the center "channel", but because we're talking low bass stuff, that might not be so necessary. Your discretion there, but it probably would not hurt - unless you have some reverberation on that sub-bass stuff that goes beyond the center channel.

G.
 
Good advice about low shelf boost there Glen. Hadn't thought of it. Yeah, targeting the offending frequency with narrow Q parametric is not a problem. I just need to automate it to follow the notes. I just don't have a good sounding EQ that also takes well to being automated. Any recommendatins there?
 
Good advice about low shelf boost there Glen. Hadn't thought of it. Yeah, targeting the offending frequency with narrow Q parametric is not a problem. I just need to automate it to follow the notes. I just don't have a good sounding EQ that also takes well to being automated. Any recommendations there?
To be honest, I rarely automate EQ, so I don't really have any specific recommendations in that regard. Not that I have anything against it, I just haven't found a whole lot of need for it in my orbit. If I need spot EQ, I'll just apply it to the spot(s), and I have found (IMHO and all that) that usually if something needs more than just a couple of spot EQs like that, it usually will benefit from the EQ curve across the whole waveform and not just the spots.

But outside of that, for this kind of task, I'd simply recommend your favorite multi-band/multi-shape parametric. My go-to is the one from Elemental Audio/RNDigital, but that's kinda hard to find these days if you don't already have one. Any similar one that lets you set up multiple bands of parametric cuts and a low shelf in one pass will work great, though you can just as easily do it in multiple passes or multiple instances of plug if need be. Voxengo's GlissEQ is kinda nice too, but it's not as neutrally surgical-sounding to me as the EA one (though some may prefer that).

G.
 
i can't add to the content of the above posts so i won't try.

but for a different approach, (keep in mind i haven't heard the track so i'm guessing here) a lot of times car stereos just lothe the 100-150 range, and the strength here can make the sub bass sound like shit.
so just try a cut at like 120hz. a healthy cut, this is an experiment, and see if that doesn't make the sub bass tolerable (in the car).

let me know if this helps, i'd be interested to know.
 
i can't add to the content of the above posts so i won't try.

but for a different approach, (keep in mind i haven't heard the track so i'm guessing here) a lot of times car stereos just lothe the 100-150 range, and the strength here can make the sub bass sound like shit.
so just try a cut at like 120hz. a healthy cut, this is an experiment, and see if that doesn't make the sub bass tolerable (in the car).

let me know if this helps, i'd be interested to know.

That's the thing though. It sound great in the car! :D
 
If the narrow Q parametric does not work, definitely experiment with adjusting the M/S separately.

I know you said it's panned center, but that does not mean it's mono. You may still benefit from trying to lower the sub bass in the side channel. Sometimes the bass problems can be from phase issues. Making the sub bass more mono has helped some of my bad mixes sound better.

Also, a great tool is Voxengo Soniformer. It's basically multiband comp, but it has many narrow bands that can limit the offending frequency without affecting too much else. You can really tweak the attack and release, and you can process the mid or side separately with one click. It's not super expensive either.
 
Thanks for the advice Leddy. I'll check out the demo of Soniformer. Yeah, I need targetted narrow band control on this monstrosity :D
 
i can take no credit for the idea; only the blame for probably explaining it wrong. i was reading an article about a year or so ago by this small timer named Bob Katz (hardy har-har). it was pretty brilliant, as most things by him that i have read have been.
basically, he was talking about how most of the problems he encounters are, of course, in the lower frequency and there were a couple of things, aside from multi-band compression etc that he used.
one is probably no big secret: linear phase eq with a steep highpass filter set around 45 if the low end was sub-y and bloated.
the one i found great was the very musical approach of figuring what key the song was in and trying to target specific notes in the progression that were causing the grief, then cutting the offending notes...sure wish i had a nice analog copy of that chart. all nice and laminated and whatnot. hope this gets you searching in the right direction because i don't know squat about mastering!
 
I really like my T. C. Electronics Finalizer. The automatic part sucks but if you are judicious and hardly use the thing it's great.

Mostly I use the multi band compression - compressing the low band and not so much the mids and hardly any on the highs.

I fought with bass for a long time until I got the Finalizer. I just got Cubase 4, and it's got multi band compression, so it could be that I could use that and sell my Finalizer.

So that's what I reach for when the bass isn't tight. Multi band compression is a whole 'nother animal than regular compression. It really has an effect that I liked right off the bat.

I avoid eq like the plague, and would always try compression unless of course something jumped out as incredibly wrong that begged for eq. I do use eq, but as a last resort. If you record stuff right the need for eq is minimal.
 
Back
Top