Conflict between VOB and actual DVD viewing

Kevin Mc wrote on 12/3/2010, 4:53 PM
Greetings,

I am using Vegas Pro 9 and DVD Architect 5. In Vegas, I rendered an NTSC 4:3 stream for DVD Architect. I viewed the resulting MPG on my computer screen and it looked quite good. The video is just under 2 hours, so DVD-A had to re-compress it down to 4.5 Mb/s. The resulting VOB file looks good when played straight from my hard drive in DivX. When I burn the project to DVD - it looks terrible. There are jittery motion lines running through every "panning" shot. These shots all look good in the final VOB file.

Any thoughts? I am happy to share more detailed info - let me know what you need.

Thanks,
--Kevin

Comments

musicvid10 wrote on 12/3/2010, 5:23 PM
-- You should render your video only once, in Vegas, using a bitrate that will fit on a DVD. Letting DVDA recompress is a bad idea for many reasons.
-- Use a DVD Architect Video template and a Dolby Digital AC3 Audio template to render two separate files.
-- You are creating an interlaced DVD, so interlaced lines will show up on computer playback in many software players, unless you set the software player to deinterlace. This is normal behavior and there is a wealth of information available by searching these forums and the internet.
TOG62 wrote on 12/4/2010, 12:01 AM
As an alternative to re-rendering if your project is a little over the capacity of a DVD is to let DVDA prepare without rendering and then use DVD Shrink to bring it down to size.
Steve Mann wrote on 12/4/2010, 7:46 AM
"As an alternative to re-rendering if your project is a little over the capacity of a DVD is to let DVDA prepare without rendering and then use DVD Shrink to bring it down to size."

You will still be recompressing an already compressed file, compromising image quality. For best results, compress only once. In Vegas.
TOG62 wrote on 12/4/2010, 7:55 AM
You will still be recompressing an already compressed file, compromising image quality. For best results, compress only once. In Vegas.

True in a sense but the compression used by DVD Shrink works differently from that in DVDA and does preserve quality remarkably well - arguably better than lowering the bitrate.
musicvid10 wrote on 12/4/2010, 8:16 AM
"arguably better than lowering the bitrate. "

Steve is correct.
The only way to reduce file size given the same dimensions and bit depth is to lower the bitrate. It's math, and is what DVDShrink does.

The fact that one implementation may work better than another in some cases doesn't erase this fact; it is still two encodes whether the second one occurs in DVDA or DVDShrink.. A single encode is vastly superior to two, in every case.
TOG62 wrote on 12/4/2010, 2:25 PM
My comment was based on John Cline's contribution in this thread http://www.sonycreativesoftware.com/forums/ShowMessage.asp?ForumID=22&MessageID=629075.

I think it's fair to say that whether or not you get a better looking result using DVD Shrink depends on the source material and also the preference of the viewer.
Steve Mann wrote on 12/4/2010, 3:25 PM
"I think it's fair to say that whether or not you get a better looking result using DVD Shrink depends on the source material and also the preference of the viewer. "

You *cannot* get a better looking result when you re-encode already encoded video - no matter what software you use.

Steve Mann
TOG62 wrote on 12/5/2010, 12:03 AM
You *cannot* get a better looking result when you re-encode already encoded video - no matter what software you use.

Of course that's true. The comparison I was making was between encoding at a lower rate initially or using DVD Shrink after authoring. I have used the latter approach many times and always been satisfied with the result.
Lou van Wijhe wrote on 12/5/2010, 3:39 AM
Kevin,

Couldn't the jittery motion lines in panning shots been caused by field order mismatch? This is something you often see only after viewing the DVD on TV.

Lou
Steve Mann wrote on 12/5/2010, 12:00 PM
The comparison I was making was between encoding at a lower rate initially or using DVD Shrink after authoring. I have used the latter approach many times and always been satisfied with the result.

Fine if you're happy with the results. I just won't deliver a re-encoded product to my clients, which is what DVD Shrink and DVDA would do. It's like passing a xerox of a xerox as the original.

Steve Mann
bStro wrote on 12/6/2010, 10:43 AM
Steve, did you read the thread / post he linked to? What DVD Shrink does is completely different from what DVDA would do.

Rob
Steve Mann wrote on 12/6/2010, 1:17 PM
Requantizing video data is just another, albeit more efficient method of re-encoding the data. It may be more efficient than a brute force re-encode, which is what DVDA does, but it's still a re-encode of already encoded data.

Steve
Former user wrote on 12/8/2010, 11:20 PM
I agree that a single encode is better, but I also have found that I get better quality if I have over an hour and a half video if I use DVD shrink, instead of using a lower bitrate.

You will be hardpressed to see a difference between an original encode and a DVDshrink re-encode.

Dave T2
TOG62 wrote on 12/9/2010, 2:03 AM
My object in mentioning DVD Shrink was to help the OP to get an acceptable result without the potential lengthy process of recoding from scratch. Very large numbers of people use this method satisfactorily.

I believe that when people refer to re-encoding video and its adverse effect on quality they have not truly understood the process. I will not claim to have that understanding either. It does seem clear to me though that that is not what DVD Shrink does. What it does do is to remove some data selectively so that it will have the least visible effect on output quality. As with all types of lossy compression this is a compromise between size and quality.

If you go back to source and encode at a lower bitrate the same compromise arises. Data will also have to discarded or else there would be no size reduction. The real issue, therefore, is which method will yield the best compromise. This is why I said that it depends on the source material and also the preference of the viewer. Some types of material, e.g. with fast motion, may suffer more from reduced bitrate and some users may be more sensitive to the degradation produced.

At the end of the day I can only suggest that sceptics try both methods on the same source material and compare the results. This is a matter of practicality, not an article of faith.