Sound Forge Compression

[timejunkie]

Well I purchased Har-Bal and I love it.

Just a newbie question...after I use IntuitQ I then import the EQ'd file into Soundforge to add compression. The overall volume increases depending on the compression setting I use.

Is there some way to keep the overalll volume the same until I take the file back to Har-Bal for loudness matching.

Sorry if this is a silly questions, it just seemed like there was no need to match loudness in Har-Bal after adding compression in SF.

Am I missing something here...

Thanks!
Chip

[HarBal]

As you've found out, if you use compression then the overall level can change quite substantially depending on the unity gain setting of the compressor (i.e the input level for which the compressor gain is 0dB). The unity gain level needs to be set to the correct value for there to be no loss in level however the correct value is not trivially determined. YOu'll have to use your ears or take it into Har-Bal again as you suggested.

When I add a compressor stage to Har-Bal I plan to have the unity gain level set to maintain the overall loudness automatically. Can't say when It'll be introduced though. There is quite a bit on the boil at the moment.

Regards,


Paavo.

[timejunkie]

Thanks Paavo,

I just want to understand correctly.

What I am doing is importing the Har-bal eq'd wav file into Sound Forge Wave Hammer. I usually use a pre-set of Medium or Smooth compression.

The threshold is -20db and the ratio is 2:1, output gain is 0db. Scan mode is set for RMS.
Is the threshold what I need to adjust to still apply compression without increasing volume?

My point is that I want to use Harb-Bal "Match Loudness" process after adding compression however the file is already louder after adding compression in SF.

I apologize but I'm a little confused on this point.

[HarBal]

No, the threshold isn't what I was referring to. It was the unity gain level. I'm guessing most compressors don't make that adjustable but rather have it fixed and compensate with a gain control instead. It is best explained graphically but I'll try using a table.

Let us assume you are using a 2:1 compression ratio and the unity gain level is set to -10dB. Then for the following RMS inputs you should have the corresponding RMS outputs.

Input Output
0dB -5dB
-10dB -10dB
-20dB -15dB
-30dB -20dB
.
.
.

For an input level equal to the unity gain level of the compressor the output level equals the input level. For levels above the unity gain level the output is less than the input and for levels below the output is greater than the input. For expansion it is the other way around.

From the above it should then be obvious that if your track average level is above the compressor unity gain level then the apparent loudness will come down because of an overall attenuation. For average levels below its the other way around. If you want to compress the dynamic range but keep the overall level approximately fixed you need the unity gain level to be around the track average level. That is what I was getting at and that is why you've found out that you can't easily maintain matched loudness levels after you apply compression.

Regards,


Paavo.