réactions à la news 20 minutes de mastering offertes par MaximalSound

Sur tous les morceaux, le niveau de sortie, en mode numérique, est bien trop élevé (-10 LUFS) par rapport aux recommandations des plateformes type soundcloud. Et surtout j'avais des peak jusqu'à -0.6 dB.

Citation de laurend :

Pas de plug-in ici, mais un soft maison. La description générale est disponible ici.

Ok intéressant donc si j'ai bien compris on a un enhanceur suivi de 32 compresseurs de bas niveau (comme le waves maxxvolume) qui agissent chacun sur une bande, et enfin un limiteur.
Cela m'a l'air un peu agressif comme traitement, pourquoi séparer le signal en autant de plages de fréquences ?
Tu es sûr que cette séparation ne dégrade pas le signal ?
Si on compare le signal divisé au signal original (via un null test) obtient-on un silence parfait ?

// (C) 2006, Michael Gruhn.

// NO WARRANTY IS GRANTED. THIS PLUG-IN IS PROVIDED ON AN "AS IS" BASIS, WITHOUT
// WARRANTY OF ANY KIND. NO LIABILITY IS GRANTED, INCLUDING, BUT NOT LIMITED TO,
// ANY DIRECT OR INDIRECT, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGE ARISING
// OUT OF THE USE OR INABILITY TO USE THIS PLUG-IN, COMPUTER FAILTURE OF
// MALFUNCTION INCLUDED. THE USE OF THE SOURCE CODE, EITHER PARTIALLY OR IN
// TOTAL, IS ONLY GRANTED, IF USED IN THE SENSE OF THE AUTHOR'S INTENTION, AND
// USED WITH ACKNOWLEDGEMENT OF THE AUTHOR. FURTHERMORE IS THIS PLUG-IN A THIRD
// PARTY CONTRIBUTION, EVEN IF INCLUDED IN REAPER(TM), COCKOS INCORPORATED OR
// ITS AFFILIATES HAVE NOTHING TO DO WITH IT. LAST BUT NOT LEAST, BY USING THIS
// PLUG-IN YOU RELINQUISH YOUR CLAIM TO SUE IT'S AUTHOR, AS WELL AS THE CLAIM TO
// ENTRUST SOMEBODY ELSE WITH DOING SO.

desc:5-Band Splitter
desc:5-Band Splitter (Splits In Low:1+2,Mid:3+4,High:5+6,UberHigh:7+8,SomeMore:9+10) [loser]
//tags: processing routing
//author: LOSER

slider1:200<0,22000,1>Crossover 1 (Hz)
slider2:2000<0,22000,1>Crossover 2 (Hz)
slider3:5000<0,22000,1>Crossover 3 (Hz)
slider4:8000<0,22000,1>Crossover 4 (Hz)

in_pin:left input
in_pin:right input
out_pin:low output left
out_pin:low output right
out_pin:mid output left
out_pin:mid output right
out_pin:high output left
out_pin:high output right
out_pin:uberhigh output left
out_pin:uberhigh output right
out_pin:somemore output left
out_pin:somemore output right

@init
cDenorm=10^-30;

@slider

freqHII = max(min(slider4,srate),slider3);
xHII = exp(-2.0*$pi*freqHII/srate);
a0HII = 1.0-xHII;
b1HII = -xHII;

freqHI = max(min(min(slider3,srate),slider4),slider2);
xHI = exp(-2.0*$pi*freqHI/srate);
a0HI = 1.0-xHI;
b1HI = -xHI;

freqMID = max(min(min(slider2,srate),slider3),slider1);
xMID = exp(-2.0*$pi*freqMID/srate);
a0MID = 1.0-xMID;
b1MID = -xMID;

freqLOW = min(min(slider1,srate),slider3);
xLOW = exp(-2.0*$pi*freqLOW/srate);
a0LOW = 1.0-xLOW;
b1LOW = -xLOW;

@sample

s0 = spl0;
s1 = spl1;
s0 = (tmplHII = a0HII*s0 - b1HII*tmplHII + cDenorm);
s1 = (tmprHII = a0HII*s1 - b1HII*tmprHII + cDenorm);
spl8 = spl0 - s0;
spl9 = spl1 - s1;

spl0 = s0;
spl1 = s1;
s0 = (tmplHI = a0HI*s0 - b1HI*tmplHI + cDenorm);
s1 = (tmprHI = a0HI*s1 - b1HI*tmprHI + cDenorm);
spl6 = spl0 - s0;
spl7 = spl1 - s1;

spl0 = s0;
spl1 = s1;
s0 = (tmplMID = a0MID*s0 - b1MID*tmplMID + cDenorm);
s1 = (tmprMID = a0MID*s1 - b1MID*tmprMID + cDenorm);
spl4 = spl0 - s0;
spl5 = spl1 - s1;

spl0 = s0;
spl1 = s1;
s0 = (tmplLOW = a0LOW*s0 - b1LOW*tmplLOW + cDenorm);
s1 = (tmprLOW = a0LOW*s1 - b1LOW*tmprLOW + cDenorm);
spl2 = spl0 - s0;
spl3 = spl1 - s1;

spl0 = s0;
spl1 = s1;

Les erreurs évidentes de mix sont-elles réellement corrigibles au mastering ?