Quality of Video Burned to DVD

vegasnewbie wrote on 1/15/2004, 3:18 PM
Hello, I am a Vegas Newbie and I am working on the "try out" version of DVD Architect. Because of this, the program naturally does not allow me to see the final output of the test projects that I have created. I have noticed that the preview quality of the video is a little "jerky" compared with the original tape. I think I have chosen the highest video quality preview available, and I have also selected the appropriate PAL settings to match the video that I have captured in the .avi file.

Before I buy the programs, I wondered if someone could confirm that the video quality in the final version of the project, as burned to DVD, would be better than the preview quality that you see when running Vegas 4 and DVD Architect. If not, would there be any reason why the preview video is a little jerky compared with the video on the original DV tape?

Thanks very much for your help.

Fred

Comments

farss wrote on 1/15/2004, 3:37 PM
THe final quality will depend upon many factors, this is the same for any authoring process.
The reason the preview is jerky is because it is decoding video from one file and audio from another if you've already encoded. Reducing the size / resolution of the preview window or using a faster machine will reduce the jerkiness to some extent.

Either way though you'll not see any of it in the final result. The oreview is mostly so you can check things like menu operation and chapter points.

If you really want to see how a DVD is going to look and sound the only way is to make one and play it back on a set top player. Even playing in a PC based DVD player is very different to what you'll see on a TV. DVD +/-RWs are a good way to do this.
wobblyboy wrote on 1/15/2004, 6:49 PM
I have been pleased with the quality of DVDs made in DVDA.
Jay Gladwell wrote on 1/15/2004, 7:22 PM
Both Farss and WobblyBoy make valid points. When done right (and that begins with shooting quality video to start with), the results can be excellent. I am extremely happy with the quality that Vegas + DVD delivers. It comes highly recommended in my book!

J--
vegasnewbie wrote on 1/15/2004, 8:05 PM
Thanks very much for these replies. I guess that to get the best video quality on a DVD, your project needs to be no longer than 60 minutes? In other words, with a DVD TV recorder, you can only get 60 minutes' worth of the highest grade of video. Does this also apply to DVD Architect?

Regards, Fred
Fleshpainter wrote on 1/15/2004, 9:11 PM
You can do a quick test of output quality by looping 10'ish seconds of your timeline and rendering it to file several times using different settings each time (call the rendered files "quick test A, B, C, etc" or something and write down what settings you used for each.) Then play them with various players. This allows you to see the difference side by side of the all the optional formats and settings.
The "Draft (Auto)" will give you the smoothest movement and the "Best (Full)" is most sutable for when you are paused and want to take a look at details up close such as when chroma keying.
kentwolf wrote on 1/16/2004, 12:16 AM
Personally, I've done 120 minute analog video to DVD and it looks just as good as the original.

With MiniDV, I've decided to cut my max down to 90 minutes.

Looks terrific...yes...with DVD-A.

The Main Concept MPG2 encoder does a very good job.
farss wrote on 1/16/2004, 12:24 AM
With DVDA you can use VBR mpeg2 and ac3 audio. Firstly this cuts down the amount of space used by the audio to leave more for the video (read you can encode at a higher bitrate). Also VBR gives better use of the available bandwidth.
My main objection to STB DVD recorders is trying to add chapter points can be a real apin and no real choice of menus etc. Mind you I've only used the Philips unit, maybe the other brands are a bit better.
If all your looking to do though is make DVDs from off air material sure Vegas and DVDA would eat that up but seems a bit of overkill to me. If you're looking to do something a bit more creative then you're sure in the right place.
jopereira wrote on 1/16/2004, 2:04 AM
Right now I'm trying Procoder 1.5 and it seems VERY good using 5000 bps bitrate (2-pass, Mastering quality).
It took 5 hours to encode a 55 minute clip...

In Speed mode the quality is about the same as MC encoder and my 2.4G PIV does it in real time (55 m clip = 55 m encoding).

Jay Gladwell wrote on 1/16/2004, 4:42 AM
"I guess that to get the best video quality on a DVD, your project needs to be no longer than 60 minutes?"

No, that's not the case here.

J--
planders wrote on 1/16/2004, 7:12 AM
I've recently put together a DVD containing a 2-hour live performance and a large gallery of behind-the-scenes photos for a local theatre group. The quality is frankly way better than I had expected.

I suppose I shouldn't be surprised, though; the DVD spec originally called for a 130-minute running time (i.e. a typical 2-hour movie with closing credits added) on a single-layer disc, and many of the early commercial releases did exactly that. Very important in the "dark ages" of 1997, before the details of dual layer disc manufacturing had been resolved.

As the format progressed and people began acquiring larger and better TVs (not to mention PC players running at upwards of 1600x1200 resolution), demand for higher bitrates led to the result that these days pretty much every movie (regardless of length and special features) ships on a dual layer disc.

Even the 60-minutes indicated for maximum quality video is only an estimate; depending on the content and your encoder settings, it's entirely possible to fill up a DVD with a 30-minute video.
riredale wrote on 1/16/2004, 11:43 AM
Planders: Good points, but a small correction on the last one. The maximum bitrate for the DVD-video format is around 10Mb/sec, including audio. Doing the math results in 60 minutes of video. If all one had was 30 minutes, there is no way (that I can see, anyway) to increase the bitrate further to take advantage of the disk space. Here's the quick rule: Final combined bitrate = 600 / number of minutes. In other words, if you are putting 90 minutes on the DVD, your target bitrate is 600/90=6.7Mb/sec. AC-3 audio takes .2 of that, leaving 6.5Mb/sec as your average video bitrate.

By the way, the DVDs I've made look pretty clean, but it will be very difficult to get as good as the Hollywood boys do. For one thing, they only have to deal with 24 frames per second, not the 30 over here in NTSC-land. Secondly, their source is progressive, not interlaced, which adds additional complications. Finally, the killer: their source material (35mm film) is very clean, and almost completely noise-free, while our stuff has lots of video noise. Since MPEG2 relies heavily on frame-to-frame redundancy, video noise really kills quality.
farss wrote on 1/16/2004, 12:28 PM
The other thing that makes the Hollywood stuff so much better is that it hasn't been through DV 25 compression. Top end mpeg encoders take their input through SDI.
planders wrote on 1/16/2004, 1:26 PM
Thanks for the clarified math. I forgot about the bitrate "ceiling". Still, it's always important to remember that running time for a DVD is much more dynamic than on analog media. It's nice to have the ability to control the tradeoffs made by the encoder in order to optimize the disc to the project.

Interesting note: by your handy formula, I see that my 2 hour play has a bitrate of about 5 Mbps. I've seen posts in the past by people who have been very satisfied with the compression quality as low as 3 Mbps. Obviously, you'd want to run a test or two before producing that 3 hour indie epic...