> We have been doing an interactive presentation CD-ROM for a company, and > now they want to move the presenations on the CD to video. The > presentations contain only media that originated digitally. All our > presenations were done in After Effects with 640x480 QT source movies at > 30fps. > > The question I have is what fps do the source movies need to be to > properly render to video? I don't believe that you can just take the > 30fps source and move it to the tape without have fuzziness due to the > interlaced nature of video. I want the clearest video possible, and read > somewhere that I may need 60fps source in order to properly render to > both odd and even lines. > > Does anyone have experience in this area that could lend me a bit of > advice. I'd really appreciate it. Thanks. > > Qapla', > > Jeff If you want to convert digital material to NTSC video at the smoothest possible rate, then yes there should be 60 frames per second (but you would only use half of each frame). However 30 frames per second won't look too shabby unless there is brisk movement on the screen. Your NTSC video won't suffer fuzziness because you only used 30 frames per second, it will suffer smoothness in motion over time (temporal resolution) if there is much movement. NTSC video is 30 frames per second (29.97 actually), but each frame has 2 interleaved fields at 1/60th of a second intervals. In other words, a complete frame consists of two pictures with the first using every other line (odd field 1) & the second filling in the lines in between (even field 2). This tends to give more smoothness without the bandwidth needs of 60 full frames per second. If you transfer 30 frames per second to NTSC, you will end up with a more "Film like" look because there is no motion in any of the individual frames. The new Canon DV camcorder actually has a feature that allows shooting video at 30 full raster pictures per second rather than 60 interleaved half raster pictures per second. Apparently some people actually like this look. BTW, if you use a nonlinear editing system to output your files to video, you will probably have to scale them to the size used in the NLE system to retain dimensionally correct aspect ratios. Most current nonlinear editing systems use rectangle pixels to improve the horizontal resolution. A typical size would be 720X480 or 720X486 for a 4:3 ratio screen. Also make sure that when importing to a nonlinear system, that it doesn't swap the odd & even fields (which would make your pictures look fuzzy). If it has a field order selection, try it both ways to see which one gives you better vertical resolution on a still. Doug Hembruff
D. Plänitz