| Author
|
Maximize in Mixdown?
|
br0d
IsraTrance Junior Member
Started Topics :
12
Posts :
355
Posted : Jul 4, 2006 16:02
|
Just sayin, a master fader limiter is not the best way to contain peaks which are too large. It being a readahead limiter (like L2/L3) makes it better than a compressor IMO, unless the compressor is being used for its specific schematic/color, but the gain of a brick wall limiter is based on 0dBFS, C1 is not. If you are mastering the track and you want your levels to be maximized without overages, you want a limiter as the last in the chain. Or maybe you don't. Whatever works for you. Me and my loud mixes will be over here enjoying a Guinness. 
|
|
|
D-Alien
Oxidelic
Started Topics :
51
Posts :
619
Posted : Jul 4, 2006 23:23
|
I also agree with that. After I've been trying a lot of "tecniques" the simplest and most geniuos worked best. NOTHING ON THE MASTER CHAIN!!! lower the fader to -4 or -5 or -6 db. and try to do this "very hard manual" work of automating the volumes, or maybe even some eq settings on Every Individual CHANNEL. Try to hear and equalize every single sound, better lower volumes than boosting them... Evry compression u make or maximization or maybe limiters.. you're procesing a digital (grained, pixelated, bitrated) material. it has bounderies and if u "process" it too much u just loose and loose and loose dynamics, color, warmth.. maybe u win some gainpower but.. flat sound. So if u stay as clean as u can, if u left the dynamics of your track Intact, (no limiters, compressors whatever), if u leave a room for later mastering (-4db -6db lower the master volume) you will have JUST GREAT SOUND! bounce this kiddy on 24bit, and if u want later put a c4 compressor and a l2 limiter. both with minimum settings. nothing more. in case u want to master your things. if u'll be mastered by profesionalist than this is system is even more stable! dont want to say the only one that really works...
and bro. read very carefully what colin says. there is the truth
 Sound:
www.myspace.com/oxidelic
www.myspace.com/setanicmusic
Image:
www.antumbra-studio.com |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Jul 5, 2006 02:32
|
Quote:
|
On 2006-07-04 15:05, Nik wrote:
the point in using the limiter is to contain the peaks....if you put the C1 before the limiter then you would be clipping the C1, which is why i put it after.
the C1 isnt going to lose any 'RMS' if you apply the right settings
|
|
Well the C1 shouldn't clip if you have all your levels set right in the first place. The C1 is non linear so having an L2 in front will mean that the "clipping" caused by the L2 could end up anywhere in your wave form (and not just at the top). It is nearly always best to have the brickwall limiter at the end of the chain.
Also, unless you are using the C1 as a brickwall limiter (which it isn't) you will be loosing level if you put it at the end of your chain.
UnderTow |
|
|
|
Colin OOOD
Moderator
Started Topics :
95
Posts :
5380
Posted : Jul 5, 2006 04:18
|
Quote:
|
On 2006-07-05 02:32, UnderTow wrote:
Quote:
|
On 2006-07-04 15:05, Nik wrote:
the point in using the limiter is to contain the peaks....if you put the C1 before the limiter then you would be clipping the C1, which is why i put it after.
the C1 isnt going to lose any 'RMS' if you apply the right settings
|
|
Well the C1 shouldn't clip if you have all your levels set right in the first place. The C1 is non linear so having an L2 in front will mean that the "clipping" caused by the L2 could end up anywhere in your wave form (and not just at the top). It is nearly always best to have the brickwall limiter at the end of the chain.
Also, unless you are using the C1 as a brickwall limiter (which it isn't) you will be loosing level if you put it at the end of your chain.
UnderTow
|
|
Putting a brick-wall limiter before a compressor can be exactly the right thing to do in some tracking/mixing situations, although when mastering you need a limiter as the very last process to make sure you catch all the overs and get that last .5dB of level.
Mastering - http://mastering.OOOD.net :: www.is.gd/mastering
OOOD 5th album 'You Think You Are' - www.is.gd/tobuyoood :: www.OOOD.net
www.facebook.com/OOOD.music :: www.soundcloud.com/oood
Contact for bookings/mastering - colin@oood.net |
|
|
|
Nik
Error Corrective
Started Topics :
13
Posts :
142
Posted : Jul 5, 2006 13:16
|
Quote:
|
On 2006-07-05 02:32, UnderTow wrote:
Quote:
|
On 2006-07-04 15:05, Nik wrote:
the point in using the limiter is to contain the peaks....if you put the C1 before the limiter then you would be clipping the C1, which is why i put it after.
the C1 isnt going to lose any 'RMS' if you apply the right settings
|
|
Well the C1 shouldn't clip if you have all your levels set right in the first place. |
|
With my method of ‘maximising’ the levels are supposed to be overloading the master output!
METHOD 1. Normally you would keep the levels in check and make sure you get no red light on the master output. You then apply the limiter and pull the gain threshold down until you achieve the require loudness.
METHOD 2. You have the track levels riding higher which peaks the master output. You apply the limiter to block the peaks, but don’t pull down the gain threshold as you have already achieved the required loudness.
The difference is how you achieve the loudness :-
1. Through the gain threshold; or
2. through extra output of combined track levels.
The way I see it is –
Method 1. I have noticed that you can hear the limiter squashing the signal at a lower ‘loudness’.
Method 2. you are making full use of the bits by having extra juice in the track levels.
The way I see it is that the limiter is still ‘limiting’ in either case but with method 2 you are not asking it to make more calculations by pulling the gain aswell....that’s the way is see it but this could be wrong I know  (just trying something different here)
This is why I put the L2 on first as I need to stop the red light before I add any more master plugs.
If I use a C1 could I just add another L2 after if the non-linear thing really makes a difference???
1-0-1-0-1-0-1-0-?-0-1 |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Jul 6, 2006 02:29
|
Ok Nik, I understand your reasoning now but there are a few flaws in your logic:
First up, in theory, having the limiter set at 0 dBFS (in&out) and sending it signals above 0 dBFS _should_ have exactly the same effect as sending lower levels to the limiter and pulling down the threshold (L1/L2/L3) or increasing the input level (Elephant). So your approach doesn't actually give any advantage.
In theory, it shouldn't give any disadvantages either but in practise it can as quite a few plugins (that you might want to use before the L2) will clip on inputs, outputs and accumulators. In other words, there is no practical or even theoretical advantage to your method.
Your idea of working at "The required loudness" and then just catching the peaks with a limiter is actually wrong because as soon as you add the limiter, you are reducing the perceived loudness. (Wether compared to a clean signal or a clipped signal. Clipped signals tend to appear louder than they really are for various reasons). So it isn't a question of dialing in the right loudness for the job and then adding the limiter to grab the superfluous peaks as that step itself actually reduces loudness.
(Important tip to any producer: Loudness is achieved in the mix! A well produced mix can be made even louder in mastering but a bad mix can not ever be made to sound as good and/or as loud as a good mix)
Also, your idea of "making full use of the bits" is wrong. Withing the DAW, the signal is 32 bit floating point (or 64 bit floating point if you use a real DAW , the important thing here is that the exponant part of the floating point number scales the level of the signal while the remaining bits in the mantissa are fully utilised reguardless wether you are at +6 dB FS, 0 dB FS or -120 dB FS. You are not using any more or less significant bits. No resolution is lost. (And anyway, if it was, the end result is CD at 16 bits. Even if you loose 1 bit (6 dB) or even 2 bits (12 dB), it really doesn't matter ater the final conversion to CD format.
Colin makes a valid point about the possible use of a limiter before a compressor but in your scenario, there is no limiter after the compressor so you actually always loose a tiny bit of RMS levels at the end of your chain. Yes you could add another L2 at the end (Anyway, the L2 sucks compared to the Elephant so who cares really? .
Still I think you should have your mix, with no stereo bus processing, peak below 0 dB FS. After that you can completely detroy the sound to achieve "commercial" levels but at least you are starting with a clean mix.
UnderTow
|
|
|
|
Elad
Tsabeat/Sattel Battle
Started Topics :
158
Posts :
5306
Posted : Jul 6, 2006 09:09
|
maximize @ mix (on group for example) is great.
dont do it too harsh (most of the time it shouldnt work realy just on unwanted peaks) .
its like.. the "pro's" say its not so good , but it does sound good.
try take all your rythmic sectin push them into group and maximaize it . get the tresh around original input (lets say its around -10) and then make you own setting of volume (i like perc. to be -12 or 14) , now if u loose some power just take tresh abit more down.
solved all my unwanted peaks/clips
www.sattelbattle.com
http://yoavweinberg.weebly.com/ |
|
|
|
Kitnam
Mantik
Started Topics :
110
Posts :
1151
Posted : Jul 6, 2006 10:16
|
i never maximise while producing on the master-out, because i dont like the sound, maybe i use the wrong plugins, but today i dont have the cash to buy me a sony infiltrator.
what i use and what is very risky too is to produce through a saturator and a bit of usual compresson, if i would ad a maximiser i would destroy the rest of all dynamics. besides of that every mastering engineer explains how horrible it is to get overcompressed trax to master.
colins idea about sending different material to master-studio is very nice too. after years of kid-home-producing i try to be give myself some professionality (as possible). which means that my next master-files-cd sending to the engineer will contain:
- a listening mix (premastered by the artist)
- a production downmix (with saturation, compression)
- a rough splitted downmix (no saturation, no compression, splitted material into kick+bass+drums+percussion and fx+hooks+textures+sounds+synthies)
anyway it wont be possible to send 100% uncompressed signals because i always have compressed/distorted groupchannels like drums and sub. this makes the signal very sensible, i realy better leave maximising to the engineer or the mastering-process. if i make my artist-premaster the maximiser is the last plug in the chain. and only one and only a little bit. in my eyes nothing sounds worth than an overlimited signal. |
|
|
|
PoM
IsraTrance Full Member
Started Topics :
162
Posts :
8087
Posted : Jul 6, 2006 10:57
|
|
a limiter on the master can show you easily what wrong in the mix,is it true or it s just cause you set the volume louder? |
|
|
|
Nik
Error Corrective
Started Topics :
13
Posts :
142
Posted : Jul 6, 2006 13:11
|
Quote:
|
On 2006-07-06 02:29, UnderTow wrote:
First up, in theory, having the limiter set at 0 dBFS (in&out) and sending it signals above 0 dBFS _should_ have exactly the same effect as sending lower levels to the limiter and pulling down the threshold (L1/L2/L3) or increasing the input level (Elephant). So your approach doesn't actually give any advantage.
|
|
i have got better results this way. i havent tried the elephant limiter but the L2 seems to work better like this for me. i'm not sure if it is exactly the same. I'll have to do a proper test on that one some time.
Quote:
|
Your idea of working at "The required loudness" and then just catching the peaks with a limiter is actually wrong because as soon as you add the limiter, you are reducing the perceived loudness. (Wether compared to a clean signal or a clipped signal. Clipped signals tend to appear louder than they really are for various reasons).
|
|
the notion of 'appearing louder' IS 'percieved loudness'.
Quote:
|
(Important tip to any producer: Loudness is achieved in the mix! A well produced mix can be made even louder in mastering but a bad mix can not ever be made to sound as good and/or as loud as a good mix)
|
|
i dont disagree with that at all. but if you are mastering yourself you do have to employ a limiter to get it to the same levels as released dance music.
Quote:
|
Also, your idea of "making full use of the bits" is wrong. Withing the DAW, the signal is 32 bit floating point (or 64 bit floating point if you use a real DAW , the important thing here is that the exponant part of the floating point number scales the level of the signal while the remaining bits in the mantissa are fully utilised reguardless wether you are at +6 dB FS, 0 dB FS or -120 dB FS. You are not using any more or less significant bits. No resolution is lost. (And anyway, if it was, the end result is CD at 16 bits. Even if you loose 1 bit (6 dB) or even 2 bits (12 dB), it really doesn't matter ater the final conversion to CD format.
|
|
ok thanks for clearing that one up.
Quote:
|
Still I think you should have your mix, with no stereo bus processing, peak below 0 dB FS. After that you can completely detroy the sound to achieve "commercial" levels but at least you are starting with a clean mix. |
|
but doing it my way (well, its not really 'my' way cause i know that some production studios employ a mastering processor on the master fx chain in the mix) is that you can tell if you have a bad mix if the limiter starts to distort or squash the sound. you can instantly know whether or not your track will reach those commercial levels or not, and if not, then you can easily adjust the problems there and then in the mix. this is actually more conducive to achieving a good mix........instead of going to and from mastering....to mix.....back to mastering until you get it right.....
you can also easily A B your mix against released material.
in any case, quiet mix or not, you will be applying mastering fx at some stage. the issue is when you apply them and the convienience and flexibility of using those tools.
1-0-1-0-1-0-1-0-?-0-1 |
|
|
|
|
