-
Rendering for Best DVD Quality
I need some help rendering an MPEG-2 file from Vegas suitable for writing to a standard 4.5 DVD with DVD Architect. So far, I have not been happy with the quality of the results of my attempts, and would like to cut to the chase with some expert advice.
I captured a good-quality commercial VHS tape using Sony Capture in Vegas 6. The video is DV format, 29.97 interlaced. Quality looks fine on the PC monitor, with minor color noise, but when rendered to DVD and viewed on a TV the dark scenes are very “boxy” (lots of rectangular blocks all over the place).
I am rendering MPEG-2 files using the MainConcept MPEG codec, variable bit rate, 2-pass settings. Bit rates are max 7,550,000; average 5,750,000; min 2,300,000; DC Coefficient = 10 – per a bit rate calculator on-line at DVD-HQ.
I used Mike Crash’s Dynamic Noise Reduction plug-in to reduce the noise, which works ok. I had it set at 10. (I also tried Neat Video Demo, which seems to work better but adds its watermark on the lower right of the screen. I have used it in the past by reducing the video image in size so that the watermark is outside the image boundary, rendering to another DV file, then cropping the watermark out on the final render to MPEG-2. But this may be hurting image resolution, since I am resizing the image twice in the process.)
I have found through experimentation that reducing brightness and increasing contrast tends to reduce the “boxy” effect. (Is that called “banding?”) I have done this through the Color Correction, Levels, Broadcast Colors, Brightness and Contrast, or other plug-ins. However, this tends to give a murky final image that I’m not too fond of.
I have been looking at the Video Scopes histogram and RGB parade in Vegas, but I do not have the understanding of what I’m seeing that gives me confidence in the adjustments I have made.
Would deinterlacing help? It tends to be a lengthy process. I have tried deinterlacing the video using Crash’s Smart Deinterlace while maintaining the frame rate (29.97). Converting to 24p with 3-2 pulldown seems to add motion blur.
That seems like enough questions for one post. I look forward to excellent advice, rules-of-thumb, best practices, cookbooks, or most anything else you throw my way. Just knowing there’s a place to ask questions helps. Thanks in advance!