How to adjust rear channel output during AC3 Encoding

[YZ_-Freak]

I have been struggling trying to figure out how to adjust the output level through the rear left and right channels during AC3 5.1 encoding. I have read the sticky’s and searched, but I do not see anyone else trying to do what I am doing.

Here is what my audio typically consists of:
Audio is from a video camera that has only 2 channels (left and right) of live band performances. When the audio is played through a receiver, there is no output through the center or rear channels. I filter the audio in Cool Edit Pro to remove hiss, EQ, etc. Audio eventually ends up on a DVD

I have tried the Ambisonics method to create 6 mono wavs. I then up the db level on the rear channels by 10 db in Cool Edit, then encode with Soft Encode, but the rear output is so soft you have to put your ear up to the speaker to hear it. I have also tried just dropping the stereo wav file into Soft Encode 3 times, deleting the LFE channel, but I still do not get any noticeable output through the rear channel. Unlocking the audio on the rear channels and adjusting the db settings to a higher level does not make any difference either.

What I am looking to do is duplicate the front L/R channels to the rear, a center channel would be nice. I have tried what to me seems to be about every possible combination of settings in Soft Encode (as well as Digigram, AC3Enc, and BeSweet), but I can not get the results I desire.

These are the basic settings I have been using for Soft Encode :
Audio Service Config:
448 kbps 48kHz (auto checked)
Audio Encoding Mode: 3/2
Bit Stream: Main Audio Service Music and effects as well as Complete main
Dialog normalization: -31 (and –31 to –15)
Audio Bandwidth: 20.30 kHz

Bit Stream:
-3dB center and surround mix
copyright/original bitstream checked
Audio Production info unchecked

Preprocesing
Input filtering: only Digital deemphasis unchecked
Both Surround channel Processing checked

Dynamic Range compression: Music: Standard with no RF overmodulation

The audio does not have to be 5.1, since the audio quality is not that good to begin with. 2/2 will work. Just looking for a way to adjust the volume through the rear channels.
Can someone suggest a different method or point out what I am doing wrong.

[YZ_-Freak]

getting better results with Besweet/Azid
here is how I set up Azid:
http://members.cox.net/usefulidiot/media/azid_settings.gif

I would think adjusting the db settings in Soft Encode for the rear channels would do the same thing, but it does not. Why is this?

The only other "issue" I have is that using the Ambisonics method, the right front channel is louder than the left. This is noticable when listening to the 6ch wav. I have tried both the Pentagon and Surround in the Emigrator setup.
What setting would I need to adjust in the Ambisonics setup to equalize the left/right output?
Or would it be easier to adjust in the Azid settings? (-5 db for ex.)

[McBob]

I had much the same experience with softencode.
Try just using AC3Enc in 2ch interleave. 2ch interleave will just mirror the front two channels to the rear and center.

Surprising nobody suggested it to you earlier.

[YZ_-Freak]

AC3Enc seems to work the best... in 2Ch-Interleaved Input as suggested. However, for me, this is one time the FAQ's and sticky's led me down the wrong path. The wav file I was working on had a avg RMS value of 16. Set the encoder to 16, along with the other settings as set out in the Dolby Digital Encoding thread... and no rear channel output. So I took a 5 minute sample of the video/audio, and encoded the audio 20 different ways until I found a setting that worked. Overall, for my situation (crappy cam audio) these are the settings that I found work the best no matter what the avg RMS value is:
Dialog Normalization: 31 DB graetly increased volume through rear channels. As this was lowered, so was the overall volume. Did not cause audio pumping either.

Bit Stream Mode: ME

Dolby Surround Mode: Dolby Surround encoded even though it was not surround encoded, this setting added more of a surround sound to the audio

Audio-Production Info: unchecked

Preprocessing: Digital De-emphasis only one unchecked

Surround 3dB Attenuation: checked

Dynamic Range Compression: Music-Standard
RF Overmodulation Protection : Checked

Since it looks like I am the only person trying to do this conversion, these settings will probably not be of any use to anyone...

[Sycho]

Question, how did you record the audio, because if it is mostly mono, even using the ambisonics method will not place very much on the surround channels

[YZ_-Freak]

The camera used to record with is unknown to me. They vary by taper. The video/audio is captured through a Canopus ADVC100. Audio always shows up as stereo in Cool Edit. If the video has a good surround stereo audio track, I am using the Ambisonics method to create 6ch wav file, which I split with Besweet, then encode with AC3Enc to 5.1 AC3. But if it is mono/no surround, I skip the Ambisonics method and encodeto 2ch AC3 as described above and get "surround" sound.