There's a movement to eradicate fractional frame rates such as 23.98fps. But will it ever be possible with the large broadcast infrastructures and legacy equipment that is in place?
Counting all the way to thirty in whole numbers is something that’s often confused for TV engineers. The story of why 29.97 (and change) happened is well enough known – when colour TV was introduced, the audio channel generated beat patterns with the colour information in the same way that two parallel fences produce interference patterns when you drive past. Tweaking the frame rate changed the relationship between the two signals – changed the spacing on one of the sets of fence bars, in our metaphor – so problem went away.
Unfortunately, that did mean that every hour of video was short 108 frames. What can it possibly matter? We’re hardly recording these signals, after all. It’s 1952. We’re just going to broadcast it so people can watch, then go to the milk bar for a frosted malt with a girl in saddle shoes and a poodle skirt while listening to The Chordettes on a transistor radio, daddy-o. Then some hip cat razzed everyone’s berries with the novel concept of video editing, turning NTSC timing into a debacle that broadcast engineers have formed religious sects just to lament thou shalt not count to thirty in multiples of one divided by one thousand and one.
No more 23.98
The self-flagellation appears to have reached fever pitch in the last few months with the advent of nomore2398.com, an initiative to Math.ceil those fractions. While the people involved are careful to create distance between their employers and the initiative itself, which is not officially sponsored by a large distributor, one key voice appears to be JayDee Vandenberg, someone who’s very senior in post production at Disney, giving the idea some clout.
The thing is, the desire to move away from fractional frame rates is not one that lacks clout in the first place. Essentially the entire world of film and TV engineering agrees that the situation is absurd. It was discussed extensively when the world was moving from standard-definition to high-definition. Nothing about HD has any connection to the issues which made the whole thing necessary in the first place (actually, most of standard definition didn’t either). Hushed gatherings of dedicated heretics sometimes gathered at places like NAB and various SMPTE conferences to discuss the fact that it might be nice if the movies learned to count in whole numbers.
The reason it didn’t happen then is the same reason it isn’t likely to happen now: because of downconversion and backward compatibility. The problem isn’t that modern hardware won’t do it; even not-so-modern TVs now routinely support a wide variety of frame rates, and while computers, tablets and phones often do a hack job of converting the rate at the display, most people won’t notice or care. The problem is that many HD broadcasts will also be downconverted and simultaneously broadcast in standard def. Rescaling the frame is really easy. Retiming the whole thing by one frame in thirty-three and a third seconds, without introducing hideous artefacts, is not easy.
As we recently discussed, no NTSC-M transmitters are now running in the USA. It’s nice to think that this raises the possibility of all the distribution channels which are using fractional rates moving to the nearest whole number. Probably, most of their gear is capable of doing that, and reasonably easy to reconfigure. The issue is the definition of “probably” and “reasonably” and the amount of work it would take to iron out all the inevitable glitches in something as large and complex as an entire broadcast television network that is currently working perfectly well. Then there’s the issue of worldwide distribution, which is a very big deal for some productions. It’s likely to be a while before we can confidently say that nobody needs 29.97.
Yes, this is obviously a good idea. Nobody disagrees, but we should not overlook how hard it will be to make it happen. It’s a shame we didn’t hold our noses and dive in back in the late 90s, but it’s easier to fix now than ever, given that the tech to bridge the inevitable technological gaps has never been more available. It’s difficult not to to conclude, either way, that this is a nettle worth grasping.