normalizing

ron82

New member
does normalizing (so is it called in cubase, raising/lowering audio to a specified amount of db..) lessen the quality of the affected clip? i mean it probably does, more or less.. but when is it critical, or how much of a boost should i avoid?

as for the usage:
after i recorded, i would use it to boost (vocals or whatever should stand out in the mix) it up to 0 db, so i have the highest level possible? also it should be the perfect solution in the mastering stage.. or am i getting this wrong? when do u use this? lol.. :\
 
i see.. but why is that so? if we are all aiming to get our levels as high as possible, to not sound like a whisper compared to commercial cds, then why is that a useless bit of processing? :confused:
 
Well - for one thing - if you're using 24-bit resolution, you don't need to worry about "as high a level as possible".

Second, normalizing doesn't do a thing to help raise your levels. The fundamental issue is that we determine a signal's level based on apparent loudness (which is a function of what we hear, plus the frequency weighting that our ears apply), while normalizing works only on peak levels.

Normalizing is simply a gain-multiplier added to the signal that represents the difference between some maximum (0 dBFS is most common) and the highest peak of the signal. So if the signal's highest peak is -2dBFS, then a +2dB increase will be applied.

However, the fact is that many times, while a song's apparent level is low or moderate, momentary transient peaks will occur at slightly less than 0dBFS - for example, at -0.50 dBFS. So while it is your intention to try and use normalizing to raise the overall track's level, you don't succed, because the gain realized with normalizing in that example would only be a barely audible +0.5dBFS!

The solution to raising level in that example is to limit the transient and then apply a gain, which is not the function of a normaizing process at all, but rather a limiter.

So I say normalizing is a completely wasted process since it does nothing practical to solve the original problem of raising the overall level of a track, while at the same time degrading the digital signal with an additional unnecessary process.
 
I find normalizing very useful. I record tracks from different sources that end up having different volume levels. By normalizing each track, it makes it easier to mix the result on the PC side.

Ed
 
Blue Bear Sound said:
You don't use it... it's a very useless bit of processing.

a bit one-sided. i do a lot of different sources as well (from mini disc field recordings to computer generated sounds) and normalising (or changing the level in general) is a very useful tool.
 
wetteke said:
a bit one-sided. i do a lot of different sources as well (from mini disc field recordings to computer generated sounds) and normalising (or changing the level in general) is a very useful tool.
Then please read my detailed explanation (above) and keeping it mind, try to convince me exactly what benefit you get from using it....
 
Ed Dixon said:
I find normalizing very useful. I record tracks from different sources that end up having different volume levels. By normalizing each track, it makes it easier to mix the result on the PC side.

Ed
Again - if you're mixing anyways (and therefore changing gain using the faders), why are you unnecessarily adding an extra digital generation???

Every single DSP you apply to a signal degrades it slightly due to the round-off error that always results from the mathematics involved, so if you want to maintain as much sonic integrity as possible, you should avoid extraneous processing such as normalizing.
 
i had already read it and there's nothing in it that applies to me. you are right that when you record in a studio you determine your soundlevels at the recording stage, not too quiet not too loud. when you want to alter the dynamics of your recording you use a compressor. normalising would indeed be silly. fine.

but when you don't always have control over the levels of the stuff you're using normalising (or just making a recording a bit louder) can be a very usefull tool.

i didn't tell you wrong, i called your opinion one-sided and i still think it is.


you have a bad temper Mr. bear.
 
Blue Bear Sound said:
Then please read my detailed explanation (above) and keeping it mind, try to convince me exactly what benefit you get from using it....
i thought there was something rather unfriendly about the way you told me i should keep in mind your words and all that. never mind.
we're not all in to the same shit you know, you're obviously a studio guy, my studio is my laptop.

i just LOVE processing!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
 
Thanks for that explanation of normalizing, Bear! I always wondered why SOMETIMES it actually did what I was trying to do, but other times did basically nothing.

Knowing now how it works I would have to say that it is indeed worthless for most tasks...
 
Normalizing is a tool like many others in the recording process. Sometimes you need it and it works well, and other times you do not.

Each has to look at their process and needs and make their own decision as to it's usefulness in their process.

One size fits all generalities are rarely accurate. Saying that the normalize process is "useless processing" represents less that the best advice.

Ed
 
Ed Dixon said:
Saying that the normalize process is "useless processing" represents less that the best advice.
Except that I backed that statement up with solid reasons of precisely why and under what conditions it IS unnecessary.
 
If it represents "useless processing", there would not be any occasions where it was needed. If there are occasions and situations where it is neccessary, then it's not "useless processing".

Saying that normalization is not always necessary and saying that is is "useless processing" are considerably different statements.

Ed
 
Ed, I'll ask you the same question that wetteke didn't answer.

Can you give me a justifiable example of a case where normalizing is useful....?
 
I refer you to posts earlier in this thread.

I think the real issue here is your use of the phrase "useless processing".

The fact is that normalization is a useful process and fits in many situations. Like any audio modification process, there are also cases where it is not needed.

Ed
 
In my reading of "normalizing" it was generally in the context of the mastering process. Trying to get the general flow of songs to be similar in amplitude???

As far as what bear says... I wish I understood a quarter of what all that means. (I will reread that a couple times over.

dana
 
ron82 said:
does normalizing (so is it called in cubase, raising/lowering audio to a specified amount of db..) lessen the quality of the affected clip? i mean it probably does, more or less.. but when is it critical, or how much of a boost should i avoid?

as for the usage:
after i recorded, i would use it to boost (vocals or whatever should stand out in the mix) it up to 0 db, so i have the highest level possible? also it should be the perfect solution in the mastering stage.. or am i getting this wrong? when do u use this? lol.. :\

BB has the idea, i've read many times here to get as close to spec levels to start with and process as little as possible, i notice when i tweak/process tracks, headroom, sparkle and claraty are some times degraded, there is a difference in code between applications i would expect that factors in also, given it's age, build, etc...

it may be what you want though, if it's techno, cyberpunk, etc etc..
it can be an effect within itself then...donno about that though :)

get as close as possible the first rip and use gain, to much processing makes mud pie's...

just a thought, peace...
 
Back
Top