TMPGEnc DVD Author 3 and GOP Re-encoding

[Incast]

I've recently moved from TMPGEnc DVD Author 1.6 to 3.

For me this move is essential as the new version gives my MPEG2 material the correct ITU-601 colorimetry flag without re-encoding, which no other software does to my knowledge.

However, the 'smart transcoding' feature is a real nuisance. With the correct settings I've managed to prevent it re-encoding up until the final few GOPs of each track. But once it gets here it seems to always insist upon re-encoding. Quite how much it does tends to be quite sporadic, ranging from one to three gops.

This is hardly earthshattering, but I don't particularly want to re-encode any material unnecessarily.

I've done a lot of hunting on the web and found this in a post from TMPGEnc at videohelp. The user is discussing version 2, but that's pretty much identical to version 3 when it comes to the trans/re-encoding capabilities.

If the disparity in the timecode is significant enough, then it may be required that some of the GOPs in the video are re-encoded, usually the ones near the end. This is to again ensure that compliance is met and that playback is possible on as many DVD players as possible.

I'm not convinced by this, I've encoded an enormous amount of test footage tonight from a wide array of sources and it consistently feels the need to re-encode the final few gops. I've yet to find material that it leaves completely untouched.

The material tested has come from commercial DVDs, DVD-Rs recorded from standalone players, DVB and MPEG2 files encoded in Procoder 2. At least one of those should be completely standards compliant.

Is there any logic behind why it is doing this? I'm happy for it to re-encode, if there's a reason for it. I've authored many discs on DVD Author 1.6 and I can't say I've ever noted a problem at the end of the disc.

[urvieh]

I came across this a while ago. There are two solutions: One is to enable XDVD support. This means it will not re-encode at all. The second is to take the elementary streams into Cuttermaran, turn on the encoding feature with QuEnc, HCEncoder or the old TMPGEnc 2.5 and tick "enable dvd-compliance" or what the English equivalent will be.
If the stream is compliant, it will just copy the files to the place you've specified before. Otherwize it'll re-encode the frames that need it and not more.
HTH.

[Incast]

Thanks for your post. Regrettably neither worked for me, I just tried them.

XDVD support appeared to reduce what was re-encoded, but it still re-encoded at least one GOP at the end.

I then used Cutermaran and selected the DVD compliant stream option. It outputted an identical result, and regrettably the re-encoding remained.

I've even tried demuxing material that it's already authored and making it author again. In that instance DVD Author 3 blindly re-encodes a second time. I'm starting to fear the software is simply flawed in this regard.

[urvieh]

I did a test tonight and have found out that it's doing the same thing here. I wouldn't regard it a major hazzle. Anyway, could it be it is a sort of finishing each track for dvd authoring purposes, a kind of indexing, re-enumerating of frames or something like this? Then I wouldn't be re-encoding but restructuring ...

[Incast]

Whatever it's doing TMPGEnc value it, because version 2 has exactly the same issue.

The results of this do seem to allow more fluid motion between tracks than before. But this might be my perception.

[urvieh]

I found out that I wasn't able to play dvd's via drag'n'drop in vlc if they had been done with Pinnacle Studio 10.7. They worked fine if I had had DVDA 2 or 3 do their job.