Brandon Strader

So, who knows how to use a limiter? And who wants to?

Recommended Posts

Can't help but think about capacitors, diodes, opamps and resistors; a signal and time. It's "basic" electronics even though compressor/limiters are programmed nowdays, but really I think that to have an idea of how a signal is treated in an analog context, its real physical path with the influence of components, would help knowing the relation between the ceiling and thresholds, or set values, processed values, overshoot and all that many stuff we see in regulation. Won't make a tutorial, and won't claim to know it all, I'm a hobbyist musician ;) but as far as things may appear to go deep, they're not "that much" in a relative way.

Share this post


Link to post
Share on other sites

The physical components involved in the construction of the device absolutely have a great impact on the behavior of the device, for sure; and lots of software attempts to model that behavior--for better or worse--with more or less accuracy than we'd like.

Share this post


Link to post
Share on other sites

I still remember the time I opened a Dimmu Borgir song in Adobe Audition and looked at the waveform... there was a part with just the guitar by itself  and it filled up the entire audio spectrum, like it had been limited and was at max volume.. despite ONLY being heavy guitar by itself. Crazy stuff, to this date I've never done that with a rhythm guitar. Pretty excessive. I'd be interested in hearing, however, from people who have done that kind of thing, and what the experience was like (limiting heavy guitars for max volume)

Share this post


Link to post
Share on other sites

I'd think it had to do with a final comp/limiter on the master and that they probably had an automation to raise the guitar volume a little bit when it was by itself so that there wouldn't have an apparent drop in volume. But I don't know ?

Dan, yes and I love both sides of that; I like the innacuracy of the analog world, where you control current through a knob to get a result, and to "feel" it as you listen. But I also like the whole nerd aspect of plugins where you can break things to the microsecond, put in relation with the Bpm and frequencies. As an electrotechnician, it's like my left hand vs my right hand, we like to affect the analog world with the best precision possible... and a calculator!:)

When people think I go too deep, sometimes, it remains an art, as technical as it can get. 
And, I by no mean can master a song the way I picture all what it involves, that's for mastering engineers. They are artists before engineers in my book.

With that said, when it's hard for someone to grasp a concept, we shouldn't judge. Not everybody has the same toolbox,.
Someone might have a good general idea of a concept but miss some tiny details he would only know if he studied maths or something.

I have no idea about fourier analysis, for instance, so even though I can design guitar pedals, I'll have some tiny details I don't know about.
Do I feel bad about it ? No way. We always learn new things, wether we want or not (it helps to want to lol).

Share this post


Link to post
Share on other sites

Subtractive EQing and side-chain compression do wonders for achieving a "loud mix" pre-mastering, which as anyone who mixes with loudness as a factor will tell you is where you need to start thinking about it.  Even conceptual decisions about priority come in to play before you write the first note.

Using these techniques, depending on you mix, you should be able to have most of your elements near the ceiling if desired (I NEVER mix with a ceiling of louder than -6 db; loudness beyond that is achieved during mastering).  If not, ask yourself:

A ) are they adding anything to the mix? Is the mix too busy? (this was a big learning difficulty for me)

B )  do the individual elements in question need to be that loud?

 

Remember, all the mixing and mastering in the world wont amount to a hill of shit if your piece doesn't work composition wise. 

 

As for that Dimmu Borgir song, I'd have to look at which one you mean but I can tell you that they usually don't have a whole lot of weight in the sub bass frequencies due to the genre.  Most of the sub comes from blast pedaling and there is no dedicated sub bass (like in dance music or other "loudness" oriented genres), so it likely that the guitars would be the loudest element in the the mix next to the kick drum, which is probably subtly side-chained to them and they probably have a subtle "dip" the guitar eq where the punch of the kick comes through.  Dimmu also knows what they are doing compositionally.  Their song "Puritania" is a very good example of this.  Notice how the instrumental drops in complexity and intensity during the vocal parts to let them pop through.  So yes, it is possible that the peaks of their guitars could touch the ceiling, especially after limiting (which is the stage at which you are seeing the audio).  A 0db harmonic at 500 hz is not going to reduce the headroom at 1000hz; headroom is not one dimensional.

 

(Keep in mind this is coming from an electronica/industrial producer with no formal training so take it with a grain of salt :P)

Share this post


Link to post
Share on other sites

I use Fruity Limiter basically every time for side-chaining kicks and snares to other basses such as growls or other basses and also for saw melodies usually (only on some rare cases do i use Grossbeat but that usually applies to more electro house style side-chaining as i am focusing on orchestral dubstep nowadays). It's really cool because in it i can easily apply somewhat fitting 'volume goes down' on selected sound when kick or snare hits. Don't know about any other uses for it though.

Share this post


Link to post
Share on other sites
2 hours ago, Cyberdrive said:

Subtractive EQing and side-chain compression do wonders for achieving a "loud mix" pre-mastering, which as anyone who mixes with loudness as a factor will tell you is where you need to start thinking about it.  Even conceptual decisions about priority come in to play before you write the first note.

Using these techniques, depending on you mix, you should be able to have most of your elements near the ceiling if desired (I NEVER mix with a ceiling of louder than -6 db; loudness beyond that is achieved during mastering).  If not, ask yourself:

A ) are they adding anything to the mix? Is the mix too busy? (this was a big learning difficulty for me)

B )  do the individual elements in question need to be that loud?

 

Remember, all the mixing and mastering in the world wont amount to a hill of shit if your piece doesn't work composition wise. 

 

As for that Dimmu Borgir song, I'd have to look at which one you mean but I can tell you that they usually don't have a whole lot of weight in the sub bass frequencies due to the genre.  Most of the sub comes from blast pedaling and there is no dedicated sub bass (like in dance music or other "loudness" oriented genres), so it likely that the guitars would be the loudest element in the the mix next to the kick drum, which is probably subtly side-chained to them and they probably have a subtle "dip" the guitar eq where the punch of the kick comes through.  Dimmu also knows what they are doing compositionally.  Their song "Puritania" is a very good example of this.  Notice how the instrumental drops in complexity and intensity during the vocal parts to let them pop through.  So yes, it is possible that the peaks of their guitars could touch the ceiling, especially after limiting (which is the stage at which you are seeing the audio).  A 0db harmonic at 500 hz is not going to reduce the headroom at 1000hz; headroom is not one dimensional.

 

(Keep in mind this is coming from an electronica/industrial producer with no formal training so take it with a grain of salt :P)

Headroom is one dimensional.

A signal (represented as samples in time) is a linear combination of waves its component frequencies with their respective amplitudes.

An assortment of frequencies at certain amplitudes may not *necessarily* reduce headroom at other places, but they certainly can. If a 500 Hz sine and 1000 Hz sine start at the same phase and the 500 Hz sine hits the 0 dB ceiling, you can bet they will constructively add and thus clip at various points in the waveform. It's just math.

Share this post


Link to post
Share on other sites
41 minutes ago, Neblix said:

Headroom is one dimensional.

A signal (represented as samples in time) is a linear combination of waves its component frequencies with their respective amplitudes.

An assortment of frequencies at certain amplitudes may not *necessarily* reduce headroom at other places, but they certainly can. If a 500 Hz sine and 1000 Hz sine start at the same phase and the 500 Hz sine hits the 0 dB ceiling, you can bet they will constructively add and thus clip at various points in the waveform. It's just math.

I suppose that makes sense.  Learn something new every day :)

Share this post


Link to post
Share on other sites
2 hours ago, Cyberdrive said:

I suppose that makes sense.  Learn something new every day :)

Though it's worth mentioning the flipside; if the phase is right, a 0 dB wave at some frequency added with a wave at some other frequency may make a quieter overall signal, because they add destructively given certain phases.

The bottom line at least for me is never measure headroom in the frequency spectrum. The waveform is where headroom matters, because its the amount of space before a sample value goes beyond the digital limit, and those samples are transmitted as waveforms through soundcards/devices, not as frequency plots.

Share this post


Link to post
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.