Can you compare 2 channels? -Sound Forge - PC - MAGIX Community Forums Archive

Subject:Can you compare 2 channels? Posted by: blacksmith Date:1/23/2005 5:09:49 PM SoundForge 6.0e(Build 237) I have a Stereo file recorded from a Mono LP (1964). I want to compare the Left/Right channels and get a result that is ONLY what is common to both. Is there a simple way to do this? Since it is a Mono source, anything not common to both channels must be noise, and that's what I want to eliminate. Can yo help?
Subject:RE: Can you compare 2 channels? Reply by: Chienworks Date:1/23/2005 7:05:15 PM No. Sorry. You can easily subtract one channel from the other and the result would theoretically be just the noise. However, it would be the positive noise from one channel and the negative noise from the other mixed together. There wouldn't be any way to know which channel any particular bit of noise came from. So if you then subtracted this noise result from the original channel you'd be removing that channel's own noise and adding in the noise from the other channel. If you were to compare the amplitudes of the samples, sample by sample, between the two channels you can easily identify where they are different. However, there is no way to know what the original amplitude without the noise was to begin with. For example, if one channel's sample is, say, 18532 and the other is 18603, you don't know if the original signal was 18532 or 18603 or some other value. The added noise has distorted both channels by a different amount, but you don't know those amounts. The only way to do this is to compare the recording with the original noiseless signal. But then, if you had a copy of the original noiseless signal, this entire excercise would be unnecessary.
Subject:RE: Can you compare 2 channels? Reply by: Sonic Date:1/24/2005 9:00:31 AM Use the Channel Converter to remove the center channel, invert the result, and mix back with the original. J.

No. Sorry.

You can easily subtract one channel from the other and the result would theoretically be just the noise. However, it would be the positive noise from one channel and the negative noise from the other mixed together. There wouldn't be any way to know which channel any particular bit of noise came from. So if you then subtracted this noise result from the original channel you'd be removing that channel's own noise and adding in the noise from the other channel.

If you were to compare the amplitudes of the samples, sample by sample, between the two channels you can easily identify where they are different. However, there is no way to know what the original amplitude without the noise was to begin with. For example, if one channel's sample is, say, 18532 and the other is 18603, you don't know if the original signal was 18532 or 18603 or some other value. The added noise has distorted both channels by a different amount, but you don't know those amounts.

The only way to do this is to compare the recording with the original noiseless signal. But then, if you had a copy of the original noiseless signal, this entire excercise would be unnecessary.