Topic: Advice on taming booming bass

[HarpDoc]

Hi guys,

When I record through my omni mics (homemade using Panasonic caps with Linkwitz mod>CA pre>H120) the bass is overpowering and boomy for many loud rock concerts. I'd like some advice on taming it post-recording. I usually use Soundforge 7. I also have Sonar 4.

I know the standard response is "use your ears." What I'm looking for is advice on where to start and what you usually do. For example, do you do a gradual roll off, going all the way down to 0, starting at a certain frequency? Do you begin the roll off at a given frequency, but flatten it out after a certain amount of reduction? Do you use a parametric EQ and try to tame specific frequencies with a narrow Q? Ect, ect.

I have decent monitors and do my best just using my own hearing, but I'm inexperienced at mixing/mastering and could use some help from people with experience. Thanks!

[aegert]

If you have a really good mastering EQ plug in you can use that...

I would adjust the HP (high Pass) filter and I would start at about 20hz for my q point....

I would start with 6db / octave and I would increase slowly to 15db per octave to see if I found a point where it rolled to the point I liked...

I would start moving the q point up to 40hz and repeat the exercise.. then 50hz
then 60 then 75 then 100. This is an exercise! Get familiar with the effects of these changes.

Get a feel for what it does in these types of sweeps...

Every show is different and every pull of the show is different so you will need to get comfortable with it.

Remember you are now editing the show. You will want to reclaim the gain lost as well as bring this to CD volume for listening. When you do this you will be changing the perception of your eq work as well... So when you do the eq work listen to it loud so you here the effects on the Mid and high tones from your roll offs of the lows!

At lower volumes your edits might sound crisp and tight but when you turn it up they can be fatiguing .. There is a balance and you need to find it for every recording .. There is no one answer for every recording you make

A

[HarpDoc]

Thanks for the detailed reply. Really helpful stuff! I love the info I get from TS.

[aegert]

Oh
If you are doing these edits you have to do them in 24bit.. This might start an argument but if you want to retaint he character and effective bit rate and do more non destructive edits record in 24 bit!

A

[BC]

there are often presets for bass rolloff that could be useful as a starting point.

[DSatz]

HarpDoc, just to make sure--when you look at the waveforms in your editing software, do you see clipping (overload)? If that's happening, the first order of business is to prevent that in the future.

Please forgive me if that was so obvious that it was an insult to have mentioned it.

Now, going on--if nothing is being overloaded and this is really just a question of setting the EQ--the first thing you might want to check out is that your main problem may be with low-frequency standing waves that are occurring at particular ("spot") frequencies. In other words the overall amount of bass your microphone pick up may well be perfectly OK in general, but due to the room acoustics where you were recording, one or two (or three or ten) particular frequencies have a tendency to "jump out" and if the band is stimulating those particular room resonances, they're dominating the sound of your recording. It would be a shame to kill the overall bass of your recording just because a few spot frequencies have this problem.

If this is what's happening in your recording, the ideal way to find out would be to take a peaking filter (+10 dB maybe, but narrow) and sweep it slowly through the range from maybe 30 to maybe 250 Hz. If you get an alarming increase in output when the filter is centered on a particular frequency, then you might want to try notching that frequency (maybe reducing it by 6 dB as a starting point) to see whether that cleans up your recording without, again, killing the whole bass region. (I mean, I assume that the main reason you use omni mikes is that you like full, deep bass energy.) If you find the two or three or four worst offenders this way, you might be able to get a real improvement in clarity while retaining most of the good stuff.

Otherwise for a general solution I would suggest a shelving reduction of the low frequencies rather than a cutoff filter. The turnover point could be anywhere from 80 to 200 Hz, and the depth (degree of attenuation) could be just 2 or 3 dB or it could be more, depending on the severity of the problem.

[mwilker]

Sometimes rolling off to me is the 'cheap fix' and I find that the 'bass' that I don't like lies somewhere in the area 80-150hz. When lowering this area I find that the bass below 80hz is just fine. Just my .02.
-Michael

[nedstruzz]

In post I typically do a gradual roll off starting around 50hz and then if additional fine tuning is needed I do a more specific dip typically between 80-120hz.  The dip between 80-120hz is really a sharp curve as appose to a roll off. 

[HarpDoc]

Guys, thanks again for the sharing of wisdom. To answer DSatz' question, there's no clipping, just a tendency to over-do the bass (depending, of course, on lots of factors such as venue, my location, and the FOH mix). The last show I went to was just mixed poorly, with way too much bass throughout the venue.

DSatz, I like your advice on doing a frequency sweep to find particular offending frequencies. What Q would you recommend while doing the sweep?

This may sound stupid, but can I get some clarification on some of the terms that have been used to describe reducing bass?

Roll off: does this mean starting at a given frequency and gradually reducing down to 0db (or some other amount of reduction)? On graphic plot, the reduction would look like a decending straight line?

Shelving reduction: is this where, starting at a given frequency, there is a flat reduction that stays the same (say 4db) all the way down to the lowest possible frequency? Graphically, this would look like a horizontal line 4db lower than the rest?

Cutoff filter: No idea.

High pass/Low cut filter: is this the same as one of the above terms?

Wow, the more time I spend on this board, the more I realize my level of audio cluelessness (but thanks to you guys, I'm getting a semblance of a clue). Thanks!

[DSatz]

HarpDoc, no--you are the opposite of clueless, whatever that is (clueful?).

I'm glad that you apparently have a way of sweeping a filter--I mentioned it somewhat wistfully, wishing that you could do so and not realizing that you actually can. Use a high Q filter when doing a sweep of this kind--you want an offending frequency to stand out as much as possible when you encounter it. The music won't sound much like music, but you'll know when you've found something.

Good on you (or as they say around here, "+t") for asking about the terms you weren't sure about. But it seems that you do understand them for the most part. Rolloff and shelving are exactly as you inferred (if by a "descending straight line" you mean diagonal rather than vertical). Rolloff and cutoff are essentially the same, except for degree: a rolloff is more gradual (say, 6 dB/octave or less) while a cutoff is steep (say, 12 dB/octave or more). A frequency that's significantly beyond the cutoff point of a filter is mostly not there any more. Also correct: High pass = low cut, and low-pass = high cut.

--best regards

[DSatz]

One more thing--I was just looking through the service manual for the Nagra IV-S analog tape recorder, and I came across the frequency response diagrams for its built-in filters. Over the years these have proved to be good choices for a very wide variety of recording situations. If you can emulate them in your sound editing software, you might want to give some of these curves a try.

For music recorded with full-range mikes in an "acoustically challenged" room, for example, the M + LFA curve might really help. It's not really as drastic as this graph makes it look--the dB scale here is less squished (that being the official technical term for it) than in many other published frequency response graphs.

--best regards

[notsofast]

Great thread and extemely helpful. I record in a venue that oftens has this issue and have been working through these strategies and playing to get the desired results but these explainations gave the visual I needed and it is making sense, I see it now. The other advice I have been playing with is using Hypers, just did that last night and that did help.

+T for starting thte thread and the great support of those who helped HarpDoc and others out.

Tim

[dactylus]

Thanks for the useful info gentlemen.

+T's all around