NTSC - 29.97 versus 30 fps?

[SCDVD]

Is there any reason why NTSC is 29.97 fps instead of a nice round number like 30?

[bendixG15]

from wikipedia .....
Color information was added to the black and white image by adding a color subcarrier of 39375000/11 Hz (approximately 3.58 MHz) to the video signal. For technical reasons the addition of the color subcarrier also required a slight reduction of the frame rate from 30 frames per second to 30/1.001 (very close to 29.97) frames per second, or 227.5 subcarrier cycles per scanline.

[edDV]

It was all about making a color system that worked within the existing 525/60 monochrome system without causing video or audio interference for existing monochrome TV sets. Turns out the subcarrier frequency for 30fps caused severe beating with audio and monochrome picture distortion.

This was important for composite video and RF NTSC transmission. S-Video (Y/C) and component analog YPbPr carry luminance, chominance and audio on separate wires. In those cases 30fps can work fine in theory. However, 29.97fps is always used to allow easy conversion to and from analog NTSC.

to be precise

"More experimentation revealed that if the frame rate was dropped just 0.10001 % down to 29.97 frames/second then the resulting horizontal scanning frequency would be 15.734264 kHz (29.97 x 525), and the chrominance subcarrier would be 3.579545 MHz (455/2 * 15734264). Thus the audio/color subcarrier difference would be 920.455 kHz (4.5mHz - 3.579545 MHz). Turns out this is very close to being an odd multiple of one-half the scan rate. (15734.264 / 2 = 7867.132 * 117 = 920.4544 (almost an exact odd multiple - the 117th multiple in fact !) So by dropping the scan rate from 15,750 to 15,734.26573 and selecting the 455th/2 harmonic for the chroma subcarrier, effectively interleaved the chroma with the luminance information, which greatly reduced the visible interference."

quote from http://www.videointerchange.com/pal_secam_conversions.htm

Or pictorially, chroma fits in the spaces between luminance harmonics.


http://members.aol.com/ajaynejr/vidres2.htm

Separation of luminance from chrominance is best done with a comb filter.
http://members.aol.com/ajaynejr/vidcomb.htm

[SCDVD]

edTV, Thanks for such a complete and clear answer. I knew there had to be a good reason for 29.97.