Choosing a DVDAuthor front end

[jmemmott]

As I noted in the T2Sami/closed captioning thread, I have added support for DVDStyler to T2Sami but am suffering from some grief getting the captioned subtitles to show up on the final DVD when I use it. My initial assumption was that the two programs, DVDStyler and DVDAuthorGUI, may be applying slightly different default options when they invoke DVDAuthor. If this was true, with better understanding of DVDAuthor, I might be able to add something to the DVDStyler project to get better results. It has also been suggested that differing versions of the DVDAuthor binaries used by the two programs may be the cause and “GUI for DVDAuthor” might be a better choice for a high-end frontend.

I don’t want to support a plethora of frontend programs. DVDAuthorGui makes an ideal low-end choice – very easy to use but it produces a very usable basic DVD with the features I want. In the near future I expect to add additional features to T2Sami that may make it attractive to people beyond the hearing impaired community such as the ability to suppress the infamous VBI noise line in native video without re-encoding. For that reason, I would also like to support a high-end DVDAuthor frontend.

Help me choose. Which one - and how do I get the minimum set of features I need out of that choice.

[dlfl]

My guess is there are a lot more dvdstyler users on this forum, including me, so my vote is for it.

[Dan203]

such as the ability to suppress the infamous VBI noise line in native video without re-encoding.

If you plan to do that by blocking it out via a black bar in the subtitle image don't bother. I already tried this approach and most of the PC based DVD player applications have a margin for the subtitle image and actually don't display the portion which you intended to cover up the VBI trash. If you had another idea for getting rid of it then ignore this message and carry on. :)

Dan

[jmemmott]

If you had another idea for getting rid of it then ignore this message and carry on. :)


A little different and I am not yet certain that it will work either. The gist is that T2Sami uses its own demultiplexing engine to parse the video stream, extract the captioning information and separate the elementary streams. When I was working out the captioning logic, I pushed the implementation down the mpeg hierarchy until I could decode the first slice of each frame so I could look at the "VBI" content. I no longer need to run it at that depth but I did discover that it does not take appreciably longer if I need to do so.

I believe it should be possible to decrease the luminance of the first row of blocks (8 pels) in that first slice by making coefficient modifications as I parse the stream. The random nature of the VBI data in relation to the rest of the data in the frame actually makes this a little easier that it would be if it were all a picture. The visual effect would be "fading" the VBI noise - possibly all the way to black...

Won't know for sure if it will work until I have time to finish the math and try it.

[Das Achteck]

You might want to check the user rating and comments on the three contenders at videohelp.com. You can access these through http://www.videohelp.com/tools?s=2#2.