I created titles and credits for a video on DVD as 720x576 PNG images, which I read into Cinelerra and dragged onto the timeline. The MPEG2 file was rendered using a yuv4mpeg stream piped to mpeg2enc. At some point I noticed that the image quality of the still images in the MPEG2 video seemed to be better when the images were fed to mpeg2enc using png2yuv. I wondered if this was because of the various mpeg2enc options I had set with the video material (not stills) in mind, so I tested using "mpeg2enc -f 8 -o %" and nothing else. With input from png2yuv, the video looks pretty much exactly like the original still images on the computer screen, but when the input comes from Cinelerra, the contrast of the image seems to be quite a bit higher, with a some of dark tones reduced to black.
Is there a logical explanation for this? Is the yuv stream from Cinelerra not supposed to be the same as that from png2yuv, if the original still image is the same? The settings were 720x576 at 25 fps, RGBA 8 bits, bottom field first, for Cinelerra and the same for png2yuv, except progressive instead of interlaced. I'm using mjpegtools 1.9.0 on Pentium 4. Mikko _______________________________________________ Cinelerra mailing list [email protected] https://init.linpro.no/mailman/skolelinux.no/listinfo/cinelerra
