limiter questions

[davecg321]

I have a few questions on using a limiter that need clearing up. I apologize now for my lack of understanding on the subject etc...

When one uses a limiter in mastering to increase the overall perceived volume how many Dbs is an ideal maximum to increase by, and likewise with gain reduction...? I realize this may be different for every given situation, for now lets just say a 4 piece rock band..

Why do we makeup the gain loss with a limiter in the mastering stage..? why not just make our mixes slightly "hotter" (assuming we have mixed below 0db) to attenuate the overall mix?

again sorry for my naivety haha


many thanks

dave

 

[Chili]

In a nutshell, the limiter prevents anything from going over a certain level. Contemporary music is loaded with a lot of transients and peaks; maybe from a snare hit, cymbal crash, etc. These peaks are prevented from going over the limiter's set level. In essence, this brings the peaks and transients down closer to the overall average of the rest of the music. So, it limits the ratio of peaks to average.

As an example, if the peaks are hitting at near 95% and the overall average signal level is around 50%, making the signal hotter will cause the peaks to clip and it won't sound good. By using a limiter, you can prevent the peaks from going over, say, 75% which gives you a lot more room before clipping. Then you can raise the overall volume so your peaks are back up to 95% and your overall average is now 70% and the perceived loudness is greater.

As for how many dB's to increase by, well that depends on where you mix is running at. No one can tell you that except you. I don't know about others, but I run the limiter so I get about -0.5db as a final output. That doesn't tell you how much I increase the signal by, because each song is different, but that's my ending goal. I increase the levels by whatever it takes to get that level.

 

[Massive Master]

I realize this may be different for every given situation, for now lets just say a 4 piece rock band..

No, you had it before. It's different for EVERY given situation. Every individual mix, on every individual project, number of musicians or genre has nothing to do with it. Some mixes fall apart immediately (especially those that were tracked too hot) some mixes can handle freakish amounts of abuse (almost universally the ones that were done with obscene amounts of headroom at every possible stage previously).

Why do we makeup the gain loss with a limiter in the mastering stage..?

You're not making up a gain loss -- You're adding gain and wrecking the mix's natural dynamics. We're not dealing with "normal" levels here -- We're wrecking the dynamics in order to participate in a pissing contest between other artists and labels. The listening public never asked for this (and personally, I think if they actually knew what they were missing, there would be an uproar -- Metallica's Death Magnetic release would've been the catalyst for change. Many of us were hoping and praying that those "demonstrations" by the public would've actually put the brakes on this nonsense).

why not just make our mixes slightly "hotter" (assuming we have mixed below 0db) to attenuate the overall mix?

ASSUMING by "0dB" you mean "-0.0dBFS" -- If people were dealing with 0dBVU and using *that* as a guide (which it always has been), then mixes would actually have the dynamics they were meant to have again.

A lot of people *do* make their mixes hotter. As mentioned before, a lot of people track way too hot right at the start -- Wrecks a lot of mixes from the get go and many times makes sure a mix will never be able to compete with the ridiculous levels some expect these days. Mixing too hot puts a shiv in the mastering phase (if you're going to do it, do it in the context of the entire project).

Assuming you DON'T "get it as hot as you can without clipping" when you're tracking (as that's an inherently terrible thing in many cases and I still can't believe it's actually in many instruction manuals and "how to" books), you have some amount of headroom at the early stages. If (and I can't believe I'm going there) you want mixes that can handle the abuse of "loud" then your point should be protecting and cherishing that headroom at every possible point in every possible stage post-tracking. Every track, every buss, every aux send, group, effects return, etc. Make a mix with dynamics that serve the mix and worry about volume at the very last step.

Sorry, I don't mean to "rant" this early on a Sunday. I mangled the hell out of a project yesterday (as I do most days it seems) because the client "loves it, but can you make it even louder like (this band's recording)?" So I did. And personally, I did a helluva job on it. It's loud as **** with very little "apparent damage" due in no small part using a chain of processors that again, has an absolutely obscene amount of usable headroom and a conversion chain calibrated specifically to work at those voltages equating to a specific digital reference level.

Now let's go back in time 15 years and ask me if I'd ever be calibrating my converters to a specific reference level just for the sake of volume...

Sorry - Ranting again... This isn't the job I wanted to do when I started this. I mean, it's exactly the job I wanted to do - I'm just not able to do it the way I used to.

 

[GONZO-X]

folks that listen to music for fun (that means, everyone except musicians) have gotten used to the sound of flattened dynamics.


who knows, if maybe they heard their favorite artists' mixes WITHOUT limiting, more or less just the way the mix engineer left it at final mixdown.


they might like it even MORE!


but until things change on that front, the best you can do, is to learn how to use limiters with THE LEAST AMOUNT OF DAMAGE DONE.


that means, you are not going to use the limiter to flatten the waveform...

only, to tame the worst peaks, and hopefully THOSE peaks, aren't the wonderful signature snare tone of a world class drummer.

bottom line is, if the waveform peaks are FLAT on top, you have already changed the mix from what the mix engineer wanted.

and it is HIS call, you know.
that's what he's paid for.

oh yea, and to listen to the producer tell him what to do, but that's another conversation.

 

[ashcat_lt]

who knows, if maybe they heard their favorite artists' mixes WITHOUT limiting, more or less just the way the mix engineer left it at final mixdown..

...they might just be frustrated that they can't hear half of what's going on over their air conditioners or road noise or the people around them on the bus bleeding through their earbuds. They just might get tired of reaching for the volume knob every time your delicate verse goes to the explosive chorus. Or else they will just miss (as in not hear, they probably wont "miss" it as in know or care) all of the subtleties that you worked so hard to preserve.

I agree to a certain extent that modern techniques have gone too far and probably for the wrong reasons, but let's not forget that real world listening environments don't really have a whole lot of dynamic range available. The noise floor is surprisingly high, and the maximum volume is usually not going to particularly high. You must take these things into account unless you're purposely aiming for the cork-sniffing audiophile set.

 

[davecg321]

thanks for all the info guys. my mix "unmastered" is sitting at about -5.00 db with plenty of headroom on my faders. Would it be safe to say that i can then increase the overall volume by 6db with a limiter thus getting 1db in gain reduction. or is 6db too much for a fairly busy song. I definitely do not want to destroy too much of the songs dynamics. Is there any way to tell what a particular songs dynamic range is i.e plugins, meters etcc...?


thanks again

 

[Massive Master]

It depends on how the mix reacts to it.