gain structure and mixing aka THE MONEYSHOT THREAD

[Electric_Head]

Glad it's all helping boss

Simply limit the tune as much as you wish to, without causing distortion.

Make sure that you take advantage of the auto-gain matching function that most limiters have.

As you limit the tune more, it gets louder, right? Most limiters have a function that keeps the sound at the same level, allowing you to hear exactly what damage you are doing without being tricked by the different level. Once you are happy with the sound, you can turn the gain match off and the tune will get louder (mind your ears when you do this by the way!), and you know it isn't mashed up.

If people took advantage of this feature I'm fairly sure we wouldn't hear so many badly mashed up tunes!

It`s a pity so few people use their ears when limiting.
They just slam the hell out of there tune and ignore the distortion.

[nowaysj]

In monitoring my own habits over the years, I'm constantly learning to add more and more compression and limiting, at every level of the production. Haha, not an endorsement to go out and squash your sound but to those contemplative young producers that might be a bit squeamish about compression and limiting, I say go for it, give it a try, drop the hammer a bit. You might like what you find. Then again, you might not, but you'll never know until you try.

[Not Dead Still Naked]

i do most of my channels at, or as close to, -20, but not over.. and my sub a db to 3 under -20..

am i loosing quality from cracking the gain?

[Ldizzy]

isnt that a little low...

but well.. that always depends... do u stack a hundred channels?

[Not Dead Still Naked]

low.. yeah thats what i feel like.. i mean it works but thats how ive been taught.

i stack about 15, plus or minus.

[Ldizzy]

i guess the best would be to compare with a few reference tracks at unity gain... thatd tell u if it works

ive read tho that even the digital realm has something analog to the noise floor in the analog domain ( i dont know if there is a noise floor in the digital domain but id have a hard time conceptualizing it)...

basically if u have all these bits at ur disposal, u should use them... its extra resolution...

anyone wants to infirm, confirm!?

[vertx]

[macc]

i guess the best would be to compare with a few reference tracks at unity gain... thatd tell u if it works

ive read tho that even the digital realm has something analog to the noise floor in the analog domain ( i dont know if there is a noise floor in the digital domain but id have a hard time conceptualizing it)...

basically if u have all these bits at ur disposal, u should use them... its extra resolution...

anyone wants to infirm, confirm!?

Simply put:

Higher resolution is exactly equal to a lower noise floor. They are the same thing. Think of it this way: 'lower error'.

The greater precision you have, the more accurate things are, the smaller details you are able to capture. There's less inherent error which is exactly the same as having lower noise floor.

Does this help?


As regards 'using all the bits because they are there', well, yes, it's a point. But the REAL beauty of a higher resolution system (ie 24-bit) is that you don't have to worry about getting everything as close to 0dBFS as possible. If your tune is peaking at -12dB you are still using 22 bits, leaving you 132dB above the noise floor. Big deal!

The point: It's always better to keep things safely away from 0dB. 24-bit allows you to do that without worrying about noise floor concerns.

[Not Dead Still Naked]

thats really good stuff. thanks, helps a lot.

[Perfecture]

I always hear about cutting out lows on things like snares, cymbals, fx etc etc.
and cutting out 20hz and below on Sub bass to create more headroom for the sub bass etc.

But my question is, where do people stand with cutting out high frequencies? What do you usally cut out the highs on, synths? fx? cymbals? midrange? etc etc.

Since we apparently can't hear anything in the high range of 20khz, is it good to cut out the really high frequencies to create more headroom, create a cleaner mix and does it make for a better sounding tune overall? If so where abouts are you guys cutting out OR even boosting frequencies?

Any advice, opinions or tips would be appreciated.

[Ldizzy]

Thx Macc

i swear, one day ill finish a tune and make business with u (hopefully)!!!!!

[Not Dead Still Naked]

I always hear about cutting out lows on things like snares, cymbals, fx etc etc.
and cutting out 20hz and below on Sub bass to create more headroom for the sub bass etc.

But my question is, where do people stand with cutting out high frequencies? What do you usally cut out the highs on, synths? fx? cymbals? midrange? etc etc.

Since we apparently can't hear anything in the high range of 20khz, is it good to cut out the really high frequencies to create more headroom, create a cleaner mix and does it make for a better sounding tune overall? If so where abouts are you guys cutting out OR even boosting frequencies?

Any advice, opinions or tips would be appreciated.

Yeah ive been taught that its a good idea to drop a 18khz cut on the master track and 20-30hz for the lows...

i do 30hz + 18khz

[nowaysj]

In my personal opinion, if you're purely digital/in the box everything has way to much highs. But whatevs.

[skimpi]

i know your supposed to leave headroom and shit for mastering and like -3 to -6db is the norm right? is this RMS or peak level? cos i can have mixes that have like an average of like -7db but then there are peaks the go up to like -0.5db, nothing goes over zero, but for the split seconds theres things near zero, are they alright or should something be done to get rid of the peaks?

[symmetricalsounds]

But my question is, where do people stand with cutting out high frequencies? What do you usally cut out the highs on, synths? fx? cymbals? midrange? etc etc.

decisions like that shape the overall sound of what you're making, adjusting the filter cutoff whilst listening to the mix is going to tell you where to make the cut. use your own taste and vision of what you're trying to do to guide you.

[atticuh]

i know your supposed to leave headroom and shit for mastering and like -3 to -6db is the norm right? is this RMS or peak level? cos i can have mixes that have like an average of like -7db but then there are peaks the go up to like -0.5db, nothing goes over zero, but for the split seconds theres things near zero, are they alright or should something be done to get rid of the peaks?

You should be mixing with your peaks hitting -6dB on the master bus to have enough sufficient headroom for post-process. RMS levels are not the same as your peak level. RMS levels are the method in which you quantify perceived loudness or strength of the signal (root mean square), but even then, it is very loosely and ambiguously related to the overall perceived loudness of the track as there are many other factors which affect it.

[skimpi]

i know your supposed to leave headroom and shit for mastering and like -3 to -6db is the norm right? is this RMS or peak level? cos i can have mixes that have like an average of like -7db but then there are peaks the go up to like -0.5db, nothing goes over zero, but for the split seconds theres things near zero, are they alright or should something be done to get rid of the peaks?

You should be mixing with your peaks hitting -6dB on the master bus to have enough sufficient headroom for post-process. RMS levels are not the same as your peak level. RMS levels are the method in which you quantify perceived loudness or strength of the signal (root mean square), but even then, it is very loosely and ambiguously related to the overall perceived loudness of the track as there are many other factors which affect it.

ahh right, cos like i was kind of thinking of the rms as an average level, dunno why, but aswell its only like certain peaks that goin up near the top, the rest of the mix is averaging lower, its just annoying when the mix as a whole is alright, except for those nasty little peaks that pop up.

[nowaysj]

Well skimpi, this is a big part of mixing...

to catch some of those stray unnecessary peaks, trace them back to their origin, consider adjusting the volume of just those hits that add up into big peaks. Probably can reduce a db to three or so without even noticing the difference in impact or weight on those hits. If they do sound weak when dropping the volume, consider some light transparent limiting, there on the source, to control those peaks if they really are making your mix too quiet.

[macc]

Don't be afraid of peaks if it sounds good.

Impact is an important part of loudness perception as well as raw RMS. Don't shave off everything - of course, controlling things is good but don't get too trigger happy

[-[2]DAY_-]

Impact is an important part of loudness perception as well as raw RMS.

striking the right balance is what i've been aiming for ... really can be difficult, definitely varies widely from track to track as far as whats appropriate.
thanks very much for continuing to help us along. bigups, macc