Author
|
Audio normalization
|
dharmatronic
IsraTrance Junior Member
Started Topics :
21
Posts :
130
Posted : Aug 26, 2009 17:34:30
|
Hi,
A friend of mine was trying to explain me why we must normalize every piece/sample of sound we get to -3 or -2 db for each track. Considering that every sample has different peaks how can we suppose that normalizing all samples to i.e. -2db will bring same volume for all of them??
Quite confusing…
Be good!
|
|
|
sideFXed
IsraTrance Junior Member
Started Topics :
22
Posts :
430
Posted : Aug 26, 2009 18:54
|
I don't think you really have to do this. you got per channel a volume knob, that's the same as normalisation. normalisation checks the wave and turns the volume up/down to meet the requested volume. It doesn't change the dynamics, so it won't hurt BUT you will have to redo the gain structure of your track.
not worth the trouble. it certainely isn't a must.
  soundcloud.com/epsylohm |
|
|
Euphorica
IsraTrance Junior Member
Started Topics :
23
Posts :
97
Posted : Aug 26, 2009 19:21
|
I think it's bullshit, just use your ears!
If everything has the same volume then there are no dynamics in your music and it can not sound as a good ballanced mix.
  http://soundcloud.com/euphoricapsy |
|
|
Fuzulu
Fuzulu
Started Topics :
58
Posts :
1096
Posted : Aug 26, 2009 21:42
|
it just gets ur sound upto that -2db or -3 level u set there ... does not matter how loud the sound was before ... if u set it -2db it will increase ur sound enough to touch -2db |
|
|
Elad
Tsabeat/Sattel Battle
Started Topics :
158
Posts :
5306
Posted : Aug 26, 2009 22:33
|
if the point is to get all the samples as loud as you can then mix with the channel faders.. there is a point to that.. somewhere..
but imo it will just probably will make you CLIP your sound in the fx chain.
if you talk about bounce channels closest to 0 then i agree , its getting the best out of your sound - theoretical part is that you lose -10db in the edges of the the sound (reverb tail etc) but if anyone realy hear 10 dbs in the bottom of the -144db mix then hurray for him , i really cant hear it.. somewhere around -70db its allready 99% gone... so...
  www.sattelbattle.com
http://yoavweinberg.weebly.com/ |
|
|
dharmatronic
IsraTrance Junior Member
Started Topics :
21
Posts :
130
Posted : Aug 27, 2009 10:40
|
Tks. Next time I will consider all commented here.
@Elad, I think I didn’t get the theoretical part of your post. If you could explain it better or just leave a link so I may check it later.
=)
|
|
|
Elad
Tsabeat/Sattel Battle
Started Topics :
158
Posts :
5306
Posted : Aug 27, 2009 12:00
|
hit snare on 0 db and have 10 seconds reverb on it
reverb over time gets weeker , lets say in 5 seconds its on -40db and in 8 seconds its -90db. at 10 seconds it reach the bottom of your -144db (24bit)
hit the snare on -60db with same reverb. after 5 seconds you get to -1000db and on 9 seconds its all gone...
not real numbers , but thats the general idea.. same for delays and release of synth..
in the end it does matter when all the sounds are totaly open
alot of the times there is something very small that ruin your mix but when over doing it.. it gets not so small.. like mixer noise on -90db is ok if its on your lead synth.. but record all your sounds with that noise and it will be very loud in your mix...  www.sattelbattle.com
http://yoavweinberg.weebly.com/ |
|
|
dharmatronic
IsraTrance Junior Member
Started Topics :
21
Posts :
130
Posted : Aug 27, 2009 12:40
|
cool... much clear now!
tks |
|
|
Fragletrollet
Fragletrollet
Started Topics :
111
Posts :
1748
Posted : Aug 27, 2009 21:37
|
I thought this gain-stagin was way more important in the analog realm, where you want a high signal to noise ratio... in digital enviroments, the "amplifier" doesnt impart any color on the sound, so it doesnt really matter (as long as all sounds are under 0dbfs in a 24 bit enviroment...?)
We have to differentiate between soft-synths and recording synths in the analog realm, before the conversion to digital. Am I right?
Undertow, where are you
Id rather not normalize, dont see the point. Just make a balanced mix, then "master" it with a compressor and a brickwall limiter...
  http://www.myspace.com/fragletrollet
http://www.myspace.com/unknowncausesound
http://www.fragletrollet.com/ |
|
|
Dancing_Zombies
Started Topics :
2
Posts :
35
Posted : Aug 28, 2009 18:33
|
|
Elad
Tsabeat/Sattel Battle
Started Topics :
158
Posts :
5306
Posted : Aug 28, 2009 21:55
|
Quote:
|
On 2009-08-27 21:37, Fragletrollet wrote:
I thought this gain-stagin was way more important in the analog realm, where you want a high signal to noise ratio... in digital enviroments, the "amplifier" doesnt impart any color on the sound, so it doesnt really matter (as long as all sounds are under 0dbfs in a 24 bit enviroment...?)
We have to differentiate between soft-synths and recording synths in the analog realm, before the conversion to digital. Am I right?
Undertow, where are you
Id rather not normalize, dont see the point. Just make a balanced mix, then "master" it with a compressor and a brickwall limiter...
|
|
digital age includ nord and virus and korg's and guitars with amp emulators. there is cables with all that , there is noise too. like i said one lead with -90db noise is allright i guess but try it with 5 channels recording the same noise.. bothering.
vst only doesnt matter really... well the tails still disapear sometime earlier but 24bit is allready enough depth to eliminate those little things from mind..
  www.sattelbattle.com
http://yoavweinberg.weebly.com/ |
|
|
Fuzulu
Fuzulu
Started Topics :
58
Posts :
1096
Posted : Aug 29, 2009 00:09
|
good info in here |
|
|