Correct me if I'm wrong about cutting frequencies...

BroKen_H

BroKen_H

Re-member
When I'm trying to use a limiter on the master, it seems that certain undesirable frequencies boost (not always the same ones, but unique to each song). If I try to cut them out after limiting, it seems like a nightmare trying to track down all the different frequencies that keep cropping up. On the other hand, if I cut those frequencies before the limiter, it's done. Easy. Not sure if this is sound, but through much experimentation, it's what seems to work.
 
I just want to share what I've found. I've not seen this information in print anywhere, so I thought I'd put it here. If it's not accurate, however, I'd like someone to say so and why...
 
I think you've stumbled upon a pretty common practice. This is what mastering engineers do every day to every track.
 
So what is it you're trying to say? That limiting can change the tone, and in a way that's hard to fix- unless you fix it.. before adding the limiter?
Partly don't quite get it as I've not seen it.
 
Massive Master said:
And watch what happens to your peak meters if you put the EQ after the limiter...
Click to expand...

I never figured that out either. I usually just tried to cut what was boosted, and it never made sense that it became a bird's nest of problem frequencies that had to be cut in little weird sections and it just started to sound worse and worse...nightmare. AND things started clipping even though I was only cutting...:confused:
With the EQ before the limiter, t was SO easy to kick the limiter up, listen, cut the frequencies I heard pump, DONE! Course, it may be my ears are getting better trained to the task...:)
 
BroKen_H said:
I never figured that out either. I usually just tried to cut what was boosted, and it never made sense that it became a bird's nest of problem frequencies that had to be cut in little weird sections and it just started to sound worse and worse...nightmare. AND things started clipping even though I was only cutting...:confused:
Click to expand...

Eq works by shifting phase, which moves some waves slightly in time relative to others. It's inevitable that there will be points in the waveform where one peak is moved onto another peak causing a higher peak, even though you cut a frequency.
 
When you cut a frequency, you will also boost the frequencies around it a little bit. The more you cut, the more it boost the frequencies on either side of the cut. This isn't a problem before the limiter, but after the limiter, you don't have the headroom to allow for the boosts.
 
Thanks, guys! That's exactly the kind of information I was needing. :)

Maybe, just maybe, it will help someone else as well. Maybe I'm the only one who couldn't see this, but I doubt it.
 
Farview said:
When you cut a frequency, you will also boost the frequencies around it a little bit. The more you cut, the more it boost the frequencies on either side of the cut. This isn't a problem before the limiter, but after the limiter, you don't have the headroom to allow for the boosts.
Click to expand...

Is this also the case with using a linear phase EQ after the limiter? Never really thought about it before :D
 
Jake_JW said:
Is this also the case with using a linear phase EQ after the limiter? Never really thought about it before :D
Click to expand...
That's because it's not really true. Really steep filters can resonate and have a bit of boost near the cutoff, but especially if you're using an EQ that shows you the curve, you would see it there.

bsg has the answer. A low pass after a real hard brickwall limiter can really fuck things up. The limiter wants to make sharp corners, but the low pass won't let it happen, so the signal will tend to overshoot as it slows to a stop rather than stopping on a dime the way the limiter would do on it's own.
 
A limiter shouldn't be actually turning the peaks into square waves. If that were the case, you could just clip the converters.
 
You must log in or register to reply here.
Back
Top