An easy way to customise Besweet command-line in DVD2SVCD

[Math]

Hi all,

Almost all SVCD I made so far with DVD2SVCD (many thanks for this great piece of software!) have some "problems" with the audio. Indeed the dialogue are far less audible than the ambiant movie music, and so often I have to play with the remote to adapt the volume (all my SVCD have audio converted from (usually) 6ch AC3 to stereo) . This was not the case for the original DVD;

After reading many posts on this subject and after some personnal experiments, I discovered that this issue could be easily solved with BeSweet:

- The dialogue are usually in the CENTER AC3 channel. A solution is then simply to set downmix Center Level to +3db or +6db. (the default behaviour is to use BSI value, which is usually -3db, ie the new setting gives a 6db-9db gain in the dialogue).

- Another solution is to use the BOOST feature of BeSweet, which applies a dynamic range compression algorithm

Personally I apply both solution with +3db DCL and Dg boost gain 2. The first option is available in DVD2SVCD but sadly the second one is not. In order to circumvent this , the usual solution is to run DVD2SVCD, wait for the conversion to finish, then re-run BeSweet with the wanted options and restart the DVD2SVCD conversion at Muxing&Cutting.


However there is a far more simple solution, and above all, automatic: Just create a batch file that will call BeSweet with the proper settings.

- To do this, simply create in the BeSweet folder a file called BeSweet.bat that contains the following line:

besweet %* -boost( /b2=2 )
(For DOS agnostic, the use of %* allows for transfering all the original command-line parameters to the executable)

- Also, in DVD2SVCD GUI, update the 'audio' tab to point to this file BeSweet.bat instead of BeSweet.exe.

This way, the boost option will be added to the original set of option specified in DVD2SVCD. Obviously the same trick can be used to include more options, to make some pre-processing before BeSweet execution...

Hope it helps.

[markrb]

Just curious. Are you aware of the many options you can use if you click on the down arrow next to the BeSweet Command line on the audio tab?

You mention a -3 setting. There is one on the options that you can change.

You also mention using a boost function on the compression algorythm. Well you can alter that as well. It is set at normal unless you change it. It can be set at Normal, Light, Heavy or Inverse.

I don't know enough about the inner workings of BeSweet to know if this is the same as what you are talking about, but I just want to make sure it isn't.

Mark

[dvd2svcd]

@Math: You're right, boost isn't supported. But I gotta say, what an elegant solution you got there :) I never knew that there was anything like %* (%1 %2 and so on I know). Thanks :D (Maybe +3db DLC is a better default setting)

[Math]

Indeed the option tab was well hidden (!) but I already spotted it a couple of days ago when I decided to give a serious look to that sound problem. There are lot of options but not the boost.

At first, I tried all the three relevant settings (ie. Normal, Light and Heavy) and none of them gave me satisfaction. In fact, I could barely hear the difference between them. May be I was using it incorrectly? That's why I tried the "boost" feature that brought at once good results, even for low gain. I selected the Dg boost variant because I had the feeling that it distorded lesser the signal than other algo which quickly saturate (I've to admit however that I didn't make much tests on them).

FYI boost is indeed a different option than the one above. Boost is applied after signal resampling and before MP2 encoding (controlled by the -Boost ( ... ) option in BeSweet) whereas the option above is a setting in the AC3 decoder azid (option -azid( -c normal/light/heavy/inverse ) ). Both are DCR algo AFAIK, but as I said, boost works better for me.