DVD Overburning

Originally Posted by Tsunade

like leaving over 500+ MB free...

Sometimes doing this is the best way to get a DVD that stays fully readable even after a long period of time. DVD burning quality usually gets quite worse at the end of the disc compared to the quality in the middle area (inner area is another case, because of the initial power calibration). So if you leave some megabytes at the end you're more likely to get a overall better error rate on the disc.

Jam it back in, in the dark.

So what about placing Vorbis inside an AVI (which is also highly non-standard) - Vorbis is implicit VBR, but quality is better than MP3 (even in VBR mode). So you've a reason to include VBR audio into your ripped movies, it simply helps audio quality.

Concerning the AVI problem: AVI and even the new OpenDML AVI format don't have support for (any?) VBR material and the only way to integrate the material into the AVI is using a very dirty hack (which can cause trouble and VDub tells you so). AVI is outdated, most of the recent features aren't supported natively and some video codecs like h264 can only be integrated in a VERY limited form.
The mp4 container support is quite large on standalones, you should try remuxing the movie into a mp4 - also resulting in fewer overhead, ergo smaller filesize of the movies. If you want even fewer overhead, then try matroska.

cya
liquid

There's nowhere I can't reach.

Originally Posted by T1249NTSCJ

I just find it kind of a waste seeing as how movie studios apply a audio mix for DVD release or what have you and then select to use VBR when the original mix is CBR. But that's just my opinion, I don't bother altering the audio when I encode DVDs to h.264. For the majority of movies out there, the DD5.1 track is always 448 all the way through.

Lets say your source material has two audio tracks (both ac3 5.1 format, lets say two languages) and you want to keep both. Now you can either recompress the tracks (and apply downmix) or keep them like they are.
But if you're space-limited then the best idea is to apply a 2-channel downmix and recompress them with a up-to-date audio compressor. Now lets say you want to use mp3 as compression, because it has a wide support. Lets recall that we're space limited, so we want to keep the datarate below 448kbit/s for both audio-tracks. So we aim at a average datarate of 192kbit/s for the main audio track (you choose one) and 128kbit/s for the second audio track (maybe because the alternative language isn't such important, but you don't want to discard it fully).

Now think about CBR/VBR again. If we compress both tracks using a constant bitrate the overall quality is much worse then if we use VBR, because the bit allocation is a lot more efficient this way (in VBR mode). The codec assigns more bits to the parts of the audio that needs them, and less bits to the quite parts or parts with low complexity.

Also remember that if mp3 and ac3 were technically equivalent you would need 448kbit/s of datarate when encoding to mp3, if you don't want to loose audio quality. mp3 doesn't do this, because peak rate is 320kbit/s.
Now consider CBR - you never get any near 320kbit/s because the datarate is fixed. But with VBR the codec at least has the chance to push the rate up for a while (when coding in complex audio areas).

So if you're reencoding audio data VBR is always better because bit allocation is kind of smart in this mode. Thats one reason vorbis only has VBR modes and does not allow a fixed rate.

cya
liquid

This thing is sticky, and I don't like it. I don't appreciate it.