Pro Audio Files

Free Mix Workshop Premium Courses

How to Fix Clipped Drums in a Mix

Transcript
Hey folks, Matthew Weiss here — weiss-sound.com, theproaudiofiles.com. I want to play you a drum loop real quick.

[loop]

This is a cool loop. It’s got a good groove to it, and it’s got good sounds, except it’s really clipped, and if you’re an audio engineer, it is inevitable at some point in your career, you’re going to have to deal with clipped drums, particularly grungy kicks. They tend to be the main culprit.

[drums]

Maybe the producer, musically brilliant, but not so technically savvy. They just want to make it more thumpier, so they just keep turning it up, they don’t realize that the distortion that they’re hearing is not coming from their speakers, it’s actually coming from the sound itself.

Now, obviously the first line of defense is to call the producer and say, “Hey, could I get this stem but maybe turned down?”

In this particular case, that’s what I did for this record, but let’s say that’s not a possibility. Maybe they lost the samples, or whatever. How do we deal with it? Well, the easiest way to do it is simply highlight the sound and use something called iZotope’s RX. We go here to noise reduction, we can see something RX 6 de-clip.

This is literally an algorithm that’s meant to handle this exact situation. I’m going to turn the quality up to high, since we’re in destructive editing mode, and now I am going to play back the sample and I am going to remove some of that distortion.

[drums, adjusting de-clip]

That’s pretty impressive. Now, when setting this thing, you want to reduce as much of the distortion that’s there, as much of the clipping distortion, but at the same time, you want to impart as little of the artifacts that this produces as a result. So what’s happening is the distortion is getting removed, but in the process, we’re generating something called comb filtering. Now, if I exaggerate this, you’ll hear what I’m talking about.

[drums, adjusting de-clip]

So what we want to do when we use RX is find the point where we remove the distortion, but cause as little comb filtering as possible, and there’s usually a little graphic display here where we can see this line. That’s where our amplitude is getting hard clipped, and if we set our threshold just below that line…

[drums, adjusting threshold]

That’s the sweet spot. So what I’m going to do is I’m going to do this on a couple more kicks here.

[kicks, de-clipping]

There we go. So now we have a little loop going here.

[drum loop]

And if we listen to that a little bit later…

[loop]

Verse… You’ll notice that’s a pretty big improvement, and that’s cleaned up a lot of the clipping. We can do the last little bit with just some basic EQ where we can emphasize the cleaner tones, right? Because we still have a little bit of grunge in there.

[loop]

But you know, if we played the fundamental of the kick, and maybe some of the higher tones where the snare is mainly living, then we can end up getting something that sounds like maybe it really wasn’t all that distorted at all, maybe it has a little bit of saturation quality to it, but we can even impart our own saturation to it, and just sort of make that a characteristic of the sound.

Now, let’s say that we did not have RX, and we’re really left to our own devices here. What do we do?

[loop, distorted]

Well, we need to sort of analyze where the distortion is coming in. What’s happening is that the modulation that’s occurring when it hits the digital ceiling is lasting too long. You can see that not only do we have the distortion at the front, we also have prolonged distortion throughout the waveform.

That’s what we’re really hearing. When clipping occurs over a very short period of time, we don’t really hear too much change in terms of the tone, what we really hear is sort of the pancake effect where the speaker kind of stops pushing air. It kind of just creates sound rather than actual push.

That’s not terrible. What’s terrible is when it’s really prolonged, and we’re getting distortion that’s lasting. So what we need to do is focus the sound away from the sustain and clean up the sustain as much as possible, while at the same time, emphasizing the transient. That’ll also help to remove that pancake effect.

So I’m going to start with Transify. This is a little plugin that’s going to help us get more of the attack back in, and by getting more of the attack in, we can turn the overall signal down and ultimately clean up some of the sustain.

[loop]

ADVERTISEMENT

Now, it’s not that we’ve completely cleaned it up, but you can hear how it’s a much cleaner, much punchier sound when I do it this way.

[drums]

Then comes the tough part, where I sort of have to work with it, and I have to sculpt around it. Basically, what’s happening is that the low end in particular is clipping, and that distortion is showing up in the lower mids. So the stuff that’s between about 200 to about 800Hz is where it’s pretty grimey.

So I’m going to grab Ozone here and I’m just going to sculpt with it. The first thing I’m going to do is emphasize those clean frequencies — the ones I was talking about before — the fundamental of the click, and the cleaner stuff that’s up top, and what I’m going to do ultimately is de-emphasize this mid-range. That stuff I was talking about. The 200Hz to 800Hz right in there.

So let’s turn our output down by about a dB to make sure it’s level matched, and let’s go from there.

[loop with EQ adjustments]

Now right away, that sounds a lot cleaner.

[loop]

Not perfect, but we’re getting to the better stuff. So can I keep working at it? Well, maybe I can start affecting things dynamically. Maybe I can make it so that when the sustain is in, it cleans up and attenuates the low mids, but when it’s out, it kind of gets out of the way. I mean, clipping is amplitude based, so as we get to the very far tail of these clicks, it’s pretty clean, it’s really just right in that initial sustain.

[drum loop]

Now one of the problems here is that when we’re clipping something, we’re pushing all of our signal to the ceiling, and by doing that, we’re creating something that sounds louder and more impactful. Especially with the distortion, because that ends up playing to our ear. We’re very sensitive to harmonic distortion, so in reversing this process, we do lose a lot of apparent level, and that’s one of the ironic things about clipping is that people do it to make the sound louder, but actually, if we want to make it clean, we have to take away from that, and ultimately, at the end of the day, the overall playback level is not going to be quite as loud as if we had just had the clean samples to begin with. Counterintuitive, but that’s the case.

Anyway, what I’m doing here is I’ve selected that band between about 200, and in this case, it’s like, 1.4kHz, but it’s that lower mid-range band that I was talking about, and I’ve set the threshold pretty darn low, I’ve set the ratio pretty darn high, and I’ve set the attack very long, and the release pretty long as well, and the idea here that I’m trying to really target the sustain. I’m trying to let the initial attack through, and then try to attenuate the sustain here of the mid-range.

[drum loop]

And here we’re getting cleaner again, right? We’re always step-by-step working in that direction. Yeah, it takes work to clean this stuff up, we’re trying to clean distortion out of something that is inherently now imprinted with distortion. It’s fundamentally changing what the sound has become.

Now, the last thing on here, and I’m going to be really honest with you guys, I don’t know what this thing is doing, I just know that it works. They’re calling it spectral shaper, it’s for attenuating signals that are problematic. I guess it does it kind of in a multiband sort of way, but it sounds different than multiband.

So this spectral shaper here I’m using it in the same sort of range, 200 to 600, where that problem frequencies are — those are right there. So I’ve set it so this is going to attenuate that even a bit more.

[loop with spectral shaper]

If I just go between everything that I just did in the chain here on Ozone, you’ll hear the difference pretty clearly.

[loop, with and without Ozone]

Here is completely without anything.

[loop, no processing and with processing]

Now, is it as good as just using RX right off the bat? I don’t think so, but is it helpful? Absolutely, and really, if we want to really, really work it in reverse, we can do some kind of combination of both where we use a little bit of RX, we use a little bit of this stuff, we probably wouldn’t have to go as far with it, and then we end up getting a drum that’s still not quite what it could’ve been if we’d just had it clean to begin with, but pretty darn close, and we can really get that punch, and that knock, and that heaviness, and those clean tones that can really shine through into the sound.

Now, there is one more option, I know this video is getting kind of long, but I just want to put it out there. I don’t have a good example for this one, but one thing you can do if all else fails, you can take the distorted drums, you can filter them, and you can dub your own drums in. That’s not off the table, that I consider the last resort, that’s only if there’s really no way to work out some pretty bad distortion, and as you can see, where there’s a will, there’s a way. We did just manage to get this reasonably clean, and it started pretty grungy.

So I don’t recommend that as a default, but it is an option that’s out there.

Alright guys, if you’re into this stuff, and you want to see me go in-depth, I mean, a lot of times, people end up clipping their sounds because they’re trying to make the overall playback louder, and the ineffective way to do that is to just shove everything into the digital ceiling or into a limiter.

The right way to do it is to get the mixing right to begin with — to understand the various techniques that can go in there, and it’s not just one thing. It’s a combination of dynamic shaping, of EQ, of using clever distortion, of seeing things in the right way balance-wise to make everything sound big, and that’s a whole mix process. That’s the way that you get it right the real way.

So if you’re interested in that, and you want to learn about your genre in particular, I have a couple of tutorials out. One is called Mixing Hip Hop, one is called Mixing EDM, you can go check those out. There will be a link here, and depending on what you’re doing, you can check those out, and you’ll learn how to really get your playback level up, and do it in a way where it’s stylistically effective, and musically effective, so it’s not just about loudness, but it’s also about the mix and the music itself.

Alright guys, I hope you liked this video. I know it was a little lengthy, but I think it was important and it was good, so if you dig what I was doing, hit that like button and hit that subscribe button so you can check out more that will be on the way.

You know what I was trying to say. Alright guys, until next time.

Expand

Matthew Weiss

Matthew Weiss

Matthew Weiss is the recordist and mixer for multi-platinum artist Akon, and boasts a Grammy nomination for Jazz & Spellemann Award for Best Rock album. Matthew has mixed for a host of star musicians including Akon, SisQo, Ozuna, Sonny Digital, Uri Caine, Dizzee Rascal, Arrested Development and 9th Wonder. Get in touch: Weiss-Sound.com

Free Video on Mixing Low End

Download a FREE 40-minute tutorial from Matthew Weiss on mixing low end.

Powered by ConvertKit
/> /> /> /> /> /> /> /> /> />