Dolby Pro Logic question

[MyC sPiTTa]

Hi,

I've read through as many posts as I can find but I'm still not 100% sure about some things:

I have my soundcard (stereo-out as far as I know) connected to a home stereo with a Dolby Pro Logic decoder (not Pro Logic II – all the posts I have read are only about PL2, which is why I’m posting). As I understand the Pro Logic decoder makes 4 channels from a 2-channel signal.
I'm using Gordian Knot to make DivXs with VBR mp3s so I have a couple of questions:

1) If I use the 2-channel AC3 track (pre-downmixed) from the DVD and make a stereo VBR mp3, will this have Pro Logic audio?
2) If I use the 6 channel AC3 track and choose Dolby Surround Downmix (when making a project file in DVD2AVI as part of Gordian Knot), will the mp3 give me Pro Logic audio?

Basically I'm asking whether the surround sound signal will be carried over into the mp3, or do I need to encode the mp3 in a special way?

If I'm wrong above, then how would I actually do it?
If I'm right, then do I need to play my DivX through any particular player or will the mp3 stereo signal be decoded to Pro Logic by my stereo anyway?

I'm using Gordian Knot but I don't yet know much about making audio formats other than what the default output is from following the Doom9 Gordian Knot guide.

Thanks in advance.

[DJ Bobo]

To 1: yes
To 2: yes

You don't need a particular player.

To be 100% sure that the surround informations are still in the MP3, use stereo instead of joint stereo. That requires of course a bitrate of 160kbps minimum.
Otherwise, they will be there, but not garanteed (the joint stereo mode in Lame is very advanced, so it preserves surround informations very well)

[MyC sPiTTa]

Thanks, that's helpful.

I don't use joint stereo but I didn't realise you needed a minimum bitrate for stereo. I've usually used 128.

I just use Gordian Knot for the audio at the moment so that's BeSweet - I don't know if that uses Lame or not. And I just use the parameters from the dropdown box in GKnot like it says in the guide (-L -3db -c normal). Not sure what these do yet.
I assume you can load in an mp3 you've created with any program outside of GKnot so I'll look around at how to do that.

Thanks.

[DJ Bobo]

You can of course use Stereo with 128k, but the quality will be quite bad, that's why I said 160k minimum.
Generally, all bitrates under 160k should be encoded in joint stereo, for better audio quality. All bitrates starting from 160k can be encoded in both modes, there won't be a big difference in the quality, but using stereo you have a better spatial separation and you keep the surround garanteed.

The (-L -3db -c normal) are AZID settings (AZID is the program that decodes the AC3 to WAV), and have nothing to do with the MP3s.

You can of course let GKnot do the video encoding only. So you encode the audio separately and multiplex it with the video later in NanDub (that's what I do the whole time, though I don't use GKnot, I make it all manual :D)

[user]

Hi,

unfortunately what DJBobo said, is a little bit unclear.

So here are short answers:

mp2: for keeping surround: use stereo, not joint stereo.

mp3: surround is kept with joint stereo !
Use --alt-presets described here: http://www.hydrogenaudio.org/forums/showthread.php?s=&threadid=260

Just select a bitrate or a bitrate range for VBR.



The surround information is primarily in the stereo-wave.
So if you have a 2.0 ac3, it can contain surround, or not, it depends on the producer.
Every stereo sound could contain surround.



DPL vs. DS2:

DS2 is fully compatible to DPL.
Both are fully compatible to stereo-only.

So if you produce today a stereo downmix from a 5-channel source, then select DS2.

If you play it through stereo-hifi, it is well.
If you play it through Dolby surround-hifi, you will have normal surround effects, the same as if you would have encoded in old DPL.
If you play it through Logic7/DPL2/Circle surround decoder, then you will enjoy the whole DS2-surround information....