FCPX Best Export Format for Cable TV

my 2¢...

ProRes and AIC are editing formats. Delivery formats are mp4, h.264, heic... HEIC is not an available option to you and mp4 is ancient, so use H.264.

If you're going to call and ask the station, find out *which* formats *they* don't have to convert. Broadcast stations are extremely busy and they don't like wasting time transcoding video. Also, they will most likely convert to DVD format when they do which will hammer the quality of any video you send them. Don't think for a minute that they are going to broadcast ProRes 422 or LT which is at best 5 times the size of a corresponding H.264 file. [They don't CARE about quality — they only care that something is ON THE AIR.]

If you want to risk it without asking: my recommendation is h.264/AAC from FCPX as a Master File.

If you are doing your own SD conversions, here's a free Center Cut Safe Zones template you can use:

https://fcpxtemplates.com/center-cut-safe-zones/

(BTW - my local NBC affiliate still broadcasts SD doing this center cut technique and the logos and lower thirds are ALL chopped off at the edges!!!! So if your content is important to you — do your own work! Don't trust the station to treat your content with any reverence.)

Broadcasters usually have specifications for delivery. Ask the services where you are delivering the media to provide theirs.

MtD

If they accept it, LT would be a better choice as it's lossless, I-frame, and less subject to compression errors.

My recommended frame rate in the American market is 30 (HD or higher) or 60 (for 720). In a "PAL" market, 25 or 50. I suppose you can use 24 if it can be accepted (without alterations), but I simply do not buy into the fantasy that 24fps is more "cinematic"... Cinematic effect is the realm of the videographer's understanding of the scenes being shot and not the frame rate of the capture. It has to do with motion blur (shutter speed) and lighting (aperture + shutter speed + speed of film, what I knew as ASA now ISO.) Cinematic is captured, not made. Therefore, any frame rate can be "cinematic". [People are experimenting with frame rates up to 120 these days.]

The first rule is always *ask* the host station all the details of the format they *prefer*. They (stations) are not all equal (more affluent stations will have better equipment). If the quality of your materials matter then by all means - do the work (including the encoding) yourself. All they really want is a file they can plug into their programming schedule.

It's not really an exception... If you *value* the appearance of your media, do everything you can to keep OTHER PEOPLE from re-encoding it because they won't care.

My "rule" is basically a wake up call. STOP ACCEPTING THAT 29.97 (or the others) IS CONTEMPORARY. It's not. It's ARCHAIC. After July 13th of this year, every station in the U.S. MUST broadcast digital or go dark. [However, I think they've figured out a workaround to this mandate: streaming is not broadcast. Broadcast is OTA (over-the-air). So low power stations are making deals with local cable tv providers and continuing to use their old "broadcast" hardware.]

My "rule" is based solely on choice. If you have the choice — do NOT choose fractional. The reasons are many, a few of which are: 23.98, 29.97 and 59.94 are not simply those values — they are infinitely repeating decimals and cannot be accurately represented by digital. It thoroughly messes with timecode (time code is not time!*). The *need* for fractional frame rates DOES NOT EXIST ANYMORE!!! It was an analog hack for an analog problem (fixing interference with waveforms). These *problems* do not need to be addressed in a digital environment.

This guy does a really good job explaining how fractional frame rates came about:

https://www.youtube.com/watch?v=3GJUM6pCpew

I started waking up to fractional frame rates when I wanted to be able to create my own timecode/timing plugins. You can't for fractional rates in Motion, the logic is not there (as I said: timecode is NOT time in fractional frame rate video). Otherwise, for every other frame rate, I can create frame accurate clocks for (I'm confident enough to say it:) any frame rate. Before that, I would use 29.97 because I just thought that's what "everybody still uses" (how "IBM PC" of me). What difference did it make? It's just 30fps slowed down by 1%. Wait... What? How does that work? Is that something the software has to do now — "physically" slow down how fast the video plays? How does a computer handle a number like 0.003003003003003003003003003003003003003003003... Impossible / Not very well. It never really sync's with timecode. That's when I started looking into it.

I invite you to do your own research and formulate your own conclusions. My conclusion is: Fractional Frame Rates need to GO.

If the cable station requires a fractional frame rate. Many still specify 29.97., which a great many devices still shoot, even if they call it 30. It’s not a problem outside North America and Japan. PAL countries use 25 or 50.

I was stunned *this* year to find out that low power stations are still operational using equipment I have not been able to find any information about. I got the info from somebody working at an LPTv station that they *have to* use 29.97 (he may have been lying to make a point — I wouldn't know the difference). Trying to get any information about the equipment they use is harder than finding out anything else on the internet. You can find current equipment available for sale — they all handle everything digitally, but try to find anything about what was used 30 years ago (but still in service)... fuggettaboutit. In other words: I have not been able to find out/verify how or why a 29.97 video source is required for transmission through any NTSC studio broadcast equipment. I shudder to think that old VHS videotape systems might be part of the distribution chain, but that might explain it.

Why do they "have to" use 29.97? Isn't it part of the standard? From the Final Cut help glossary:

NTSC format

The video standard defined by the National Television Standards Committee, the organization that originally defined North American broadcast standards. Analog NTSC video has 525 interlaced lines per frame, a frame rate of 29.97 fps, and a limited color gamut. Digital NTSC video has a frame size of 720 x 486 pixels (720 x 480 for DV and DVD), and a frame rate of 29.97 fps. See also PAL format.

You don't need that much precision:

With 30 fps and a 4-hour program you get 432000 frames.

If you use enough digits to make the frame rate accurate to 1 part in 10 million, then after a 4-hour program you'll be off by 0.0432 frames. So you don't need an infinite number of digits. Also, computers can keep track of divisions. Try this on the macOS calculate: 30 / 1.001 * 1.001. You'll get exactly 30, even though 30 / 1.001 is an infinite repeating decimal.

And you've got the drop frame bit. But yes, if you have to fit your program in a fixed time slot, like 30 minutes, 29.97002997002997... is a pain.

Yes, 30 fps is much easier. No drop frame bit needed.

You asked.

I would have thought most places would like 48K PCM audio.

Just curious. What do you recommend for the frame rate here? Would this be an exception to your integral fps rule?

I couldn't agree more. Unfortunately often the first rule of what the station prefers overrules what the creator prefers.