Jump to content

Solution for clipping?


Recommended Posts

Clipping in short is a form of "disortion" if you get over a specific dB/dynamic range of your production setup.

How does this occour?

- Mostly the instrument too boosted, too loud

- wrong leveling while recording

- seldom but possible: if the CPU can't take it anymore (but here are latency issues and CPU limits more influenced)

How does it sound:

It sounds cracled/disortet.

What can I do against it:

"Compression" is the most used answer, but trust me. Nothing has to be at the +0dB range while producing, that's just ported into your head that you "have" to do it. But that's simply wrong, because you kill a lot of the dynamic of a song with it. You can always make it physically louder with specific tools in the end (after the song is finished). This is what we audio engineers call CD Mastering, but this is another way more complicated topic to cover.

To avoid the most well known issues of clipping, just take the loudest instrument, level it so that the loudest peak reaches +0dB max and use this one as guideline for all other instruments. "Bah! Then the outcome is too quiet!!!", you might say. True but it's way more dynamic, not overcompressed and dull.

Try it! Listen carefully and you're way more satisfied with your productions. Trust me, I'm doing this for years now.

Link to comment
Share on other sites

To avoid the most well known issues of clipping, just take the loudest instrument, level it so that the loudest peak reaches +0dB max and use this one as guideline for all other instruments.

To find out what the instrument reaches, do I just look at the peak meter in FL Studio's mixer? If so, there are two columns that show two different dB for the instrument. Why are there two? I take it this method will also make my tracks just about the same volume as each other?

Link to comment
Share on other sites

I guess we have to clear some issues about "clipping", because it seems like that a lot of you don't understand what "clipping" really is.

First of all, like I said earlier, it's a certain amount of "overusing" the dB range. Everything that get's over it and your audiocard or software can't handle it anymore, this is called "clipping". The endeffect is a total disorted sound that can even have crackles in it. In old recording techniques (analog) this "overusing" was often used for adding a warmth feeling to a recording. A so called "band saduration" or "soft disortion" of the sound. If you still have an original tape left and you already bought the CD, compare both and you know what I mean.

Digital recordings have it's good and bad sides. Good on it - clear recordings. Bad on it - always fear the clipping. But how do you "define" clipping? When do you get it? Good question I might try to answer.

--------------------------------------------------

Threntian or Yoshi for example asked me to listen to their stuff to give some suggestions. The difference between those two: Threntian uses a 5.1 speaker system in stereo mode and FL, and I might suggest that Yoshi uses a normal PC speaker system and FL. Okay, set rules for now. As I listened to one of Threntians "Wrath" remix versions (the Sephiroth one at the RemixWIP section) I was like "whoa dude, the bass is too heavy and some instruments blow the dB range. Okay, back to Yoshies "Under the Mysterious Blue" remix. But here, besides of some volume issues - no clipping at all.

Analysis:

So why did it happen? First of all I'd suggest it counts to the use of the speaker systems and volume of the amp. If the amp is quiet, you push up the volume of your sequencer automatically and each system responds different. That's one point. Another point are the frequency ranges of the speaker itself. A studio monitoring system simply has a better frequency range than a PC speaker. Let's take a simple example (example, might not be true!):

Studio Monitors:

20Hz - 30000Hz

PC Speakers:

200Hz - 18000Hz

You see the difference? Now combine, Sherloc Holmes and what do we get if we listen on different systems a - let's say - bass heavy track? Right, the monitor speakers don't give a f*ck about the frequency range, while the PC speakers are barfing like hell. A form of clipping occours!

Next example:

I just fooled around with Cubase VST5 a bit. This tool has a dB volume range from unlimited/-70dB (lowest volume) up to +0dB (balanced) and still up to +6dB as possible headroom where I can use "soft clipping".

Okay... so I loaded up a sampler and made a simple OHMP-ZZZZ track with 140bpm. No effect, no EQ, no panning, just two simple mono tracks and a later added compressor.

These are the standard settings. Volume at +0dB - the samples aren't normalized so I got a regular -3dB signal, which is normal in Cubase and gives me a huge headroom to play around with:

:arrow:BD-SNR -3dB

Okay, now I simply raised both samples with 2dB. The outcome:

:arrow:BD-SNR -1dB

Still no clipping occours. Now I'm using up my full headroom of Cubase VST5.

:arrow:BD-SNR +6dB (loudness)

Still, the clipping is almost unnoticable. Well at least to me and my speakers. Not to mention that the Terratec Producer card has a huge headroom. You might discover something else now.

Okay... now I'm overdoing it the first time, I add a compressor without limiting and boost it 12dB upwards. That's +18dB now. Be warned with your speakers, they might get damadged, but you hear the difference.

:arrow:BD-SNR +6dB (loudness) +12dB Overcompression

And now I'm using the maximum possible setting from the internal compressor of Cubase just to show you what happens. +24dB boost, added to the +6dB that's +30dB! Be warned again, and this time... now we talk about clipping.

:arrow:BD-SNR +6dB (loudness) +24dB Overcompression

You see... "Clipping" depends on a lot of things. Speakers, AudioCards, Sequencer Software, own fault...

The next time before you scream again, think about if that situation really occours. And if you're serious with music anyways, better equipment helps a lot - especially with mixing. So stop with the "I have to compress" mumbo jumbo, because this often kills the dynamic of your song for sure.

cheers.

Fox

Link to comment
Share on other sites

"Compression" is the most used answer, but trust me. Nothing has to be at the +0dB range while producing, that's just ported into your head that you "have" to do it. But that's simply wrong, because you kill a lot of the dynamic of a song with it.

Thanks for the feedback. In the past, I've just tried to lower the volumes of specific instruments until I don't hear clipping (when switching to a new format, or using a different audio player); but it sounds like there's alot more advanced ways to do it without sacrifising anything in the music (like playing around with peak ranges), and that there's many ways to get around this without having to use Compression right away, or at all (which I'd hardly ever use anyways - lol).

There are people who like their music, and then there are people who like their music A LOT. You're obviously just somebody who just likes his music.

You might be right in some ways. This "Lo-Fi" thing you refer to must just be something I don't have a good ear for, but when programs automatically have stuff like Compression and Anti-Alias on as the default (say, when converting your song to a wave file or something); I immediatly notice the difference... nomatter how minimal (same with when I try to convert songs from 22khz to 44khz to put on a CD) - because I have a good vision of what my song was supposed to sound like; and 'advanced features' that change the peaks and frequencies that I original made my sound with usually do nothing but piss me off (even if they're just trying to make the song sound better in the end) - haha.

Even when I listen to one of my songs on a different audio player, it never sounds completely right... as every stereo / headphones have a different default for how they project the treble/bass (like, turning the "bass boost" button on makes it more bassy than I intended; but turning it off makes the bass not as apparant as it was on my original computer speakers - which can sometimes annoy the hell out of me! lol). Yeah, so, I may not notice the whole "lo-fi" thing when stuff goes over the red light (when there's no apparent clipping [distortion or popping] to be heard)... but I'm REALLY damn picky when it comes to how I want my music to sound in other ways. :)

Studio Monitors: 20Hz - 30000Hz. PC Speakers: 200Hz - 18000Hz. You see the difference?

Thanks again for the information. I was pretty convinced that clipping depended alot on the frequency range of different systems (since some people comment about clipping when others never notice it), but it's nice to know how big the range can get to as well. Sometimes I just go into Cooledit and knock the frequencies down to about 28000 when creating songs (or press the "Decrease Volume by 25%" button in Sound Recorder); just so that the song will be compatible with most speakers - but like you said; there's obvious more technical ways to get rid of the problem. The headphones I have right now claim to go from 10Hz to 25000Hz, which seems to be good for whatever I do (never really get much static and distortion, unless it's on purpose - haha).

I'm gonna keep reading your bigger messages in detail, as it's obvious that alot of knowledge has gone into everything you've said CompyFox. Thanks again for all the interesting examples and tips man; they are greatly appreciated! ;)

Link to comment
Share on other sites

If you mean the compressor...

For this testrange I simply used the built in Cubase VST5 Dynamics plugin, that is routable from AutoGate->Compressor->Limiter.

Activity: Compressor only

Threshold: 0dB

Ratio: 2.0:1 (doesn't matter here, because the threshold is deactivated anyways)

Attack: 10ms

Release: Auto

Make Up Gain: 12.0dB, 24.0dB (maximum)

The endeffect is a brutally boosted sound without limiting as you can hear in the last two sound examples.

Link to comment
Share on other sites

  • 3 weeks later...
  • 2 weeks later...
  • 3 weeks later...

not sure how relevant this might be, but here's a small little

guide with pictures and stuff that I threw together some time ago.. I

can't recall when.

Doesn't really say anything more than whats been said before now,

but the illustration might be a bit eaiser to understand for some people.

http://24.211.112.207/shared/remixing/clipping.htm

Link to comment
Share on other sites

The next time before you scream again, think about if that situation really occours. And if you're serious with music anyways, better equipment helps a lot - especially with mixing. So stop with the "I have to compress" mumbo jumbo, because this often kills the dynamic of your song for sure.

I know you're talking about using compressors to stop clipping in a mix, but still, compressors do have their uses other than that. You can use a compressor to give your drums some punch, for instance. If you're willing to shell out the money, you can buy hardware compressors that will stop clipping without (noticeably) dropping any dynamic quality, although a decent hardware compressor is going to cost several thousand dollars (there's still some good commercial software compressors, though).

Link to comment
Share on other sites

Oh sure they have their uses, they wouldn't be used so much otherwise. We're just saying that it's easy to use it inapropriately. It's pointless reaching for a compressor when all you need to do is lower the level of your fader. If you want to apply a characteristic to a sound, and you think a compressor will help you achieve this, then by all means use one. But just for the sake of it? This is bad producing.

It's also worth remembering that some compressors automatically apply make-up gain which will keep the level of the output equal to the input. So you'll still get clipping anyway if you don't turn down the fader.

Link to comment
Share on other sites

Oh sure they have their uses, they wouldn't be used so much otherwise. We're just saying that it's easy to use it inapropriately.

I'm aware of that. I only said what I said because some newbie might read what CompyFox said and interperate it as "COMPRESSORS = EVIL".

Anyway, you couldn't go past Inspector to aid you with monitoring your mix and helping to eliminate clipping. Use it in combination with your program's standard db meter (it doesn't display anything over 0 db), and tweak your instrument's volumes.

Link to comment
Share on other sites

  • 3 months later...

In general, i use compressors on all of my effects tracks anyway, i dont necissarily do anything with them i just put them there for control. either that or the parametric equalizer....

I keep forgetting that not everybody uses fruityloops... :oops:

of course sometimes you just gotta let the bastard track clip...especially in hardcore tracks

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

×
×
  • Create New...