Normalize or Amplify?

Hey guys--
I have a waveform that is not very hot but it sounds clear.
Should I use normalize to boost or amplify or a combination.
I will try all these things when I get to the studio, but wanted
to get the expert input and maybe a tip or 2

Thanx a bunch

Hi Pauxier,

Well, normalising means this: CEP is going to scan the file for the loudest part, after that it's going to amplify the whole file to the point where the loudest part is just below the point of clipping (assuming you use 100% normalising).

So there you have it. Normalising is the same as amplifying - only it prevents the file form clipping.

-- Syntax Error

Thanx for that,
it almost sounds like a compressor limiter.

Indeed. I'm affraid a lot of people think just that. But normalising doesn't have anything to do with the dynamics of your file.

Glad I could be of help.

-- Syntax

it almost sounds like a compressor limiter.

Syntax is right - it's just amplification that finds the loudest part of your file, and moves the whole file, using that loudest point as a reference, to the new level that you have set for a maximum. So everything gets louder or quieter by the same amount.

So if you simply Amplify, clipping could occur?

I understand

Thanx again

So if you simply Amplify, clipping could occur?

Technically, you can normalize to any value, including over full scale, which means that you can clip regardless of which you use.  
If you notice in "amplify", there's a "normalization" box.  If you insert a peak value into that box, like -1 dB (for example), then hit "calculate now", you'll see the amplification value rise or lower to whatever amount will leave your file at -1 dBFS.  Simply, if you know what peak level or percentage you want your file to be at, it's much easier to find using normalization than calculating (or guessing) the exact amount of cut or boost needed.  However, if you instead know how many decibels or percent you want to cut or boost from the file, amplification is the best way to get you there.
Both are very useful, but in the end will give you the same results.

Awesom info
I'll try these things as soon as I can

So if you simply Amplify, clipping could occur?

I understand

Thanx again

It does what post78 says - well it does in 16-bit mode. In 32-bit mode, you can normalise to any value you like, even way above 0dB, because it's working in 'floating-point' mode. Okay, it will sound distorted, but only because the level's overloaded your soundcard. We are talking about a potential 1500dB dynamic range here...

The really neat bit is that if you then re-normalise your file back to 0dB, the clipping will go away as if by magic!

At some point, we will be able to bring you more information about how this works, but for the moment, just take it that it does!

FWIW, I "normalize" all tracks to 96% (which is my own preference) when I'm tracking.  Volume is cumulative and I find that by doing this when building sessions, that I don't have to do any major antenuating with the track and master volumes.  Surely... your mileage WILL vary!  

SteveG -
quick question about the 32-bit no-clipping thing.  Though I use 32-bit mode and am able to bring off the chart spikes down into range, I cannot see them if i zoom out vertically before I normalize them.  Everything is just cut off at 0 db.  Should I be able to do this? Thanks

KO

SteveG -
quick question about the 32-bit no-clipping thing.  Though I use 32-bit mode and am able to bring off the chart spikes down into range, I cannot see them if i zoom out vertically before I normalize them.  Everything is just cut off at 0 db.  Should I be able to do this? Thanks

KO

I put all my tracks through the dynamic processor on de-esser medium, then equalize through the geq, and then normalize to 80%...that usually leaves it sounding very crisp and clear....just my 2 cents

SteveG -
quick question about the 32-bit no-clipping thing.  Though I use 32-bit mode and am able to bring off the chart spikes down into range, I cannot see them if i zoom out vertically before I normalize them.  Everything is just cut off at 0 db.  Should I be able to do this? Thanks

KO

Not quite sure what  Scottshu975's previous post relates to - but what you are seeing is correct - the visual display is limited at this point. It has nothing to do with the audio, though, which definitely isn't limited!

You can prove this for yourself by deliberately overloading a 32-bit track, and then, whilst it's shrunk and still looking like a block of green, renormalising it to 0dB and seeing what happens. As if by magic you get your correct-looking track back!

Thanks Steve

I guess I just didn't understand the point of be able to zoom out to see positive dB values if the audio information can't be seen there anyway. Thanks again.

KO