Hello,

When I have a high-res audio recording (say 24/96) that I want to process (EQ, compress, limit, etc) using Soundforge and plugins, what is the best way to do this in a situation that requires a 16 bit, 44.1 KHz stereo result?
I.e.: when and where do I resample?
What is the best tool? SF internal resample option?

I'd guess i'd go: EQ -> Compress -> resample -> Limit
But is this correct?

It's been a while since I used Sound Forge but the "key" point in taking an audio file down from a higher bit rate to lower is dithering.

Dithering is a term which makes the downgrade more tranparent and randomises the errors.

The order of processing makes little difference to the downgrade. Basically you need to process everythying first to a finished level, then dither down.
_________________
Recording Studio Suntans

Hi,

In any situation, the very last things you should do is downsample, and then dither. Once you go down to, say, 16-bit/44.1, you never want to do anything else to your file. Processing a dithered file, or any file in 16-bit will result in sound degredation, especially if you have used dither. If you absolutley must process a file after dithering it once, convert it back to 24 or 32-bit (float), do what you need to do, then go back down to 16-bit without the dither. Also, it's my opinion, but I do not like the sample rate conversion or the dither in Sound Forge. Not only is it not very configurable, but I don't think it sounds very good. I think you would be better off using a plug-in or another software program for this.

As far as processing is concerned, less is more. Listen to the tracks, live with them, if you can afford the time, and decide if and what you need to do to them, if anything.

There really is no set procedure for mastering. We cannot give settings or advice because we cannot hear what the tracks sound like. You, as the engineer must make the calls.

I will say that if you use something like the Wavahammer in Sound Forge, you can get carried away, as compression can be addictive. It's best, if possible, to check your work against the original, unprocessed source to hear just what you are doing.

My advice, since I do own SF? Find and download some other compressor DX plug-ins. If you have a VST Wrapper, you can use VST plugins.

I know commercial CDs are compressed/limited to be LOUD, but try to resist the temptation to do the same. Loud CDs do not make the listener take notice. Rather, the highly compressed sound makes the music dull, robs it of it's dynamics, and turns off the listener after a short time.

Hope this all helps...

Leonard wrote:

Hi, In any situation, the very last things you should do is downsample, and then dither. Once you go down to, say, 16-bit/44.1, you never want to do anything else to your file. Processing a dithered file, or any file in 16-bit will result in sound degredation, especially if you have used dither. If you absolutley must process a file after dithering it once, convert it back to 24 or 32-bit (float), do what you need to do, then go back down to 16-bit without the dither. Also, it's my opinion, but I do not like the sample rate conversion or the dither in Sound Forge. Not only is it not very configurable, but I don't think it sounds very good. I think you would be better off using a plug-in or another software program for this. As far as processing is concerned, less is more. Listen to the tracks, live with them, if you can afford the time, and decide if and what you need to do to them, if anything. There really is no set procedure for mastering. We cannot give settings or advice because we cannot hear what the tracks sound like. You, as the engineer must make the calls. I will say that if you use something like the Wavahammer in Sound Forge, you can get carried away, as compression can be addictive. It's best, if possible, to check your work against the original, unprocessed source to hear just what you are doing. My advice, since I do own SF? Find and download some other compressor DX plug-ins. If you have a VST Wrapper, you can use VST plugins. I know commercial CDs are compressed/limited to be LOUD, but try to resist the temptation to do the same. Loud CDs do not make the listener take notice. Rather, the highly compressed sound makes the music dull, robs it of it's dynamics, and turns off the listener after a short time. Hope this all helps...

Thanks very much for the advice!
What resampler plugin/solution do you suggest?

I do use plugs for slight compression/limiting and EQ.
A plugin that could do 24-bit (or better) resampling would be very nice.

I am also wondering about the headroom I mus keep before resampling. 2-3 dB?
I could fix that by limiting/dithering after resampling.
Any ideas?

Quote:

What resampler plugin/solution do you suggest?

I am a fan of Isotope. Many people also favor Waves.

Quote:

I am also wondering about the headroom I mus keep before resampling. 2-3 dB?

I recommend at least -.3db.

Quote:

I could fix that by limiting/dithering after resampling. Any ideas?

Fix what?[/quote]

ano wrote:

Hello, When I have a high-res audio recording (say 24/96) that I want to process (EQ, compress, limit, etc) using Soundforge and plugins, what is the best way to do this in a situation that requires a 16 bit, 44.1 KHz stereo result? I.e.: when and where do I resample? What is the best tool? SF internal resample option? I'd guess i'd go: EQ -> Compress -> resample -> Limit But is this correct?

Processing is best done before resampling in order to help reduce artifacts caused by the filtering required in order to prevent alias frequencies from occuring. Check out a few articles on Nyquist and anti-alias filters if this doesn't make any sense.

Anyway, the order of processing that I would perform is:

EQ-> Compress->Limit

Then resample and dither from 24 bit to 16 bit.

If using CD architect to create the CD, it may perform the last two steps for you. Haven't used it in a while, so check the manula for this.

Resampling "shouldn't" affect dynamic range. I would however leave the highest value around -.5 dBFS on the final CD in order to prevent overs.

Best,
Tom
_________________
Tom Volpicelli
The Mastering House Inc.
http://www.masteringhouse.com
CD Mastering and Media Production Services

Leonard wrote:

Quote: What resampler plugin/solution do you suggest? I am a fan of Isotope. Many people also favor Waves.

Waves has a resampler plugin?! Would be good news!
What's the name?

Quote:

I am also wondering about the headroom I mus keep before resampling. 2-3 dB?

I recommend at least -.3db.

Quote:

I could fix that by limiting/dithering after resampling. Any ideas?

Fix what?[/quote]
The headroom I leave for resampling. See above.