mkanar
19th November 2001, 20:19
Experts skip below to my question that starts with *****. The preceeding information is simply a fast run-through of video technology as I understand it.
DV = Digital Video, not Digital Video in general but the codec used in today's Digital8 (D8) and MiniDV camcorders.
Okay, so for a long while (years back) I couldn't understand what all the fuss was about interlaced video....so every odd line gets drawn on the screen and then every even line, which makes up a complete frame; so what's the problem? Ah-ha! I finally realized that the odd and even lines where being captured at different times (in the case of NTSC, I "think" that there is about 0.016683016022368587882791802633247 seconds between each even and odd line capture.
Of course, movies that are *filmed* are captured as complete frames; only 24 of them each second. In the case of NTSC, some of the frames of course have their even or odd lines duplicated and interlaced with the next frame so that it comes out as about 29.9706fps or whatever. However, my point here is that these full (progressive) frames are sent to the TV as interlaced video, but television audiences are generally satisfied with the smoothness of the output even though the source has only 24 full-frames/second rather than almost 60 half-frames per second.
***** My question:
After moving some video from my DV (D8) camcorder to my computer via IEEE 1394 AKA FiREWiRE, I was actually kind of surprised not that the frames were interlaced, but that my digital camcorder continues the legacy method of capturing the even and odd frames at different points in time. To me, it seems as if it would have made more sense to capture both the even and odd lines at the same time. Then, when viewing the video on a computer or non-interlaced display, the video would look good. In the case of outputting this video data to the TV, the even and odd lines would be output to the television as interlaced video, much like a 24fps film, except no need for pulldown since the source would already be 29.9706fps or whatever. It seems to me that compressing progressive frames would be easier than compressing interlaced frames since there are more differences due to the time differential. Furthermore, some of us hope that someday, video will be predominently progressive rather than interlaced. With LCD screens continuously becoming cheaper to produce, and since LCD screens prefer non-interlaced video anyway, I would think that the industry would also be pushing away from interlaced video.
Anyhow, I appreciate corrections/clarifications and/or explainations!
Thanks!
DV = Digital Video, not Digital Video in general but the codec used in today's Digital8 (D8) and MiniDV camcorders.
Okay, so for a long while (years back) I couldn't understand what all the fuss was about interlaced video....so every odd line gets drawn on the screen and then every even line, which makes up a complete frame; so what's the problem? Ah-ha! I finally realized that the odd and even lines where being captured at different times (in the case of NTSC, I "think" that there is about 0.016683016022368587882791802633247 seconds between each even and odd line capture.
Of course, movies that are *filmed* are captured as complete frames; only 24 of them each second. In the case of NTSC, some of the frames of course have their even or odd lines duplicated and interlaced with the next frame so that it comes out as about 29.9706fps or whatever. However, my point here is that these full (progressive) frames are sent to the TV as interlaced video, but television audiences are generally satisfied with the smoothness of the output even though the source has only 24 full-frames/second rather than almost 60 half-frames per second.
***** My question:
After moving some video from my DV (D8) camcorder to my computer via IEEE 1394 AKA FiREWiRE, I was actually kind of surprised not that the frames were interlaced, but that my digital camcorder continues the legacy method of capturing the even and odd frames at different points in time. To me, it seems as if it would have made more sense to capture both the even and odd lines at the same time. Then, when viewing the video on a computer or non-interlaced display, the video would look good. In the case of outputting this video data to the TV, the even and odd lines would be output to the television as interlaced video, much like a 24fps film, except no need for pulldown since the source would already be 29.9706fps or whatever. It seems to me that compressing progressive frames would be easier than compressing interlaced frames since there are more differences due to the time differential. Furthermore, some of us hope that someday, video will be predominently progressive rather than interlaced. With LCD screens continuously becoming cheaper to produce, and since LCD screens prefer non-interlaced video anyway, I would think that the industry would also be pushing away from interlaced video.
Anyhow, I appreciate corrections/clarifications and/or explainations!
Thanks!