Thoughts on 4K for production and distribution.

I have conflicting thoughts about 4K for production and distribution. At one level I’m convinced it’s being pushed on us by equipment manufacturers when there is no real demand: at another I know from experience that there are some non-obvious advantages to 4K. But one thing is clear: the push to 4K is not about a push to improved quality.

As I articulately point out in this interview I did with Rick Young just before the Supermeet at IBC in Amsterdam, if the push was really to have better quality in the production chain that would translate to improved viewing quality in the home, we’d be focusing on greater bit depth per color channel, and full color bandwidth. Instead we have such silliness at 4K with 4:2:0 color at 8 bits!

A true focus on quality would use the extra bits we apparently have (in order to do 4K) not for extra resolution (which there is little to no demand for) but to improve bit depth and color channel detail. Before we create a single 4K camera, we should have 2K/1920 at a minimum of 4:2:2 and 10 bits per channel. If we cared about quality that is.

On the other hand, if you only care about selling new gear, particularly new display panels (because the Japanese display manufacturers are doing poorly financially) then 4K is an easy “sell”, because simple minds equate “bigger” with “better”. We see this already in HD, where 720P has been demonstrated year after year (at IBC) to deliver better images into the home than 1080P of the same material, because of the additional compression required for 1080P. (1080P in a master suite will show the additional quality.) 4K is going to require even more compression, with more blocking and poorer quality color gradients, even with the improved efficiency of H.265.

H.265 is considered to be twice as efficient as H.264, but 4K is 4x the bandwidth, so we’re still stuck with only half the required distribution bandwidth. Even 4K proponent Panasonic had a slide on display at IBC showing their (optimistic) prediction of 4.2% penetration into home for 4K in 2018!

4K for the home is as dead in the water as 3D always was. With 3D it was/is easy to dismiss it. To be honest I’m not a believer in 3D and I think it’s a distraction from the film. Whatever future 3D has in the cinema its future in the home is non-existent.

But 4K for production… Mostly unnecessary and a distraction. Focus on bit depth and color channel for quality through the production channel. Mostly unnecessary does not mean there are no redeeming features.

Yes, there’s the “easy to reframe for HD” argument, which I’ll happily shoot down with the counter argument that the time to get the framing right is in production. That’s the actual job of the Director and DOP. Encouraging sloppiness, and lack of advance thinking, is not a good argument for 4K!

Then there’s the “future proof” argument. For what? Cinema releases have had 2K Digital Intermediates for at least the last five years. (A couple of 4K DI release, but over 95% are 2K.) Has there been an audience push for “more resolution”? Of course not, because people will go see a story shot in DV and blown up to the cinema screen if the story and characters are compelling. (As they have done in the past.) Maybe if you think your project is likely to get an IMAX release there’s an argument for 4K for future proofing, but again, it’s pretty much a bogus argument.

So why am I not adamantly against 4K like I was 3D? Very early in my career I learnt the value of “oversampling at the source”. Even if the destination is a 640 x 360 web video, higher resolution source will create perceptually sharper final image. If that’s what you want. Remember that for most of my career, “video” has always had to have resolution/sharpness reduced to make it more “filmic”.

So, shoot 4K if you must, but don’t fool yourself it’s really about quality. It’s really about gear lust. Don’t be a gear-slut: focus on story, lighting, sound, camera and character. These are the things the audience cares about.

There is another context that “4K” is being used as shorthand for “digital film production” where acquiring at 4K makes sense if there’s any expectation for a long life for the production. This use of the term implies fairly specific workflows at the higher end of digital production for feature films and television series.

All that to explain why Intelligent Assistance is participating in the Larry Jordan’s “PowerUp: 4K & FCPX” seminar on January 14 in Burbank. Discount code “INTELLIGENT” to save 20%! This seminar was originally scheduled for November, but the focus on 4K in the MacPro and FCP X comments, makes moving it back beyond the release of the next version of FCP X prudent.

6 thoughts on “Thoughts on 4K for production and distribution.”

  1. These are some really good points Philip. I recently had an interview with Paul Cameron, a UK broadcasting specialist, who argued that 4K can can be important for movies and the cinema because it gives movie makers and Hollywood an opportunity to be better compared to television once again ( I don’t think it has any chance of becoming the next home video and TV technology, though, and will probably meet the same faith as 3D. Who has space for a 90inch 4K TV anyway?

  2. Thank you for writing this! It’s as if you went into my brain and siphoned this out. It is about gear manufacturers and nothing more.

  3. You state one point of view on home 4K and state it well . . . .

    and yes, it seems too early for the manufacturers to be foisting it on us
    and yes, the ever annoying “resolution for the sake of resolution” drumbeat ignores many other and more important issues re: image quality.

    The other and MAJOR point of view is production . . . .

    Ask ANYone in ANY part of film and TV production and they will tell you 4K is the MINIMUM we should be looking at for A MASTERING format . . . . .

    Granted, and to your major point about home HD and 4K only being 8 bit 4:2:1, mastering is at a higher bit depth and rate and at full 4:4:4 sampling.

    I have seen 4K in the course of my production work and it is VERY impressive.
    It is also the ultimate pig for bandwidth as you rightly point out . . . but it solves more problems than it creates . . . at least now that the processing power needed is becoming more available.

    I have also seem “home” 4K demos and it is also quite good looking.
    It also solves a few problems for “home” monitors which you didn’t discuss . . . .
    it helps some 3D issues [ and 3D is certainly not dead yet ] and it helps
    some aliasing and fringing issues.

    Of course, the elephant in the room is content; the whole discussion is pretty much moot until NATIVE 4K content become readily available . . . . and available at low levels of net compression.

    Feature film post production is rapidly becoming largely 4K centric and that will only increase as the increase in quality is undeniable. . . . . and feature film CAPTURE is moving rapidly to 4K and beyond for the same quality reasons.

    TV post production is already using 4K monitors in control rooms as it can show four HD streams on one display . . . . and 4K camera for sports as digital zooms and slow motion are MUCH cleaner than native HD even with the down conversion.

    So yeah, RIGHT NOW 4K for the home is a moot point and is being prematurely pushed by the display manufacturers . . . . . . but there is pretty much no stopping it.

  4. great post, and good comments. I agree with Greg H that from a production standpoint 4K 4:4:4 10+ bit wil be the next standard that will (hopefully) last a generation or two. I cannot see wanting a home monitor that does not have better color depth or bandwidth than the 1080p monitor I currently enjoy. So, manufactures, if you are listening – keep pushing the envelope with professional and prosumer cameras, displays and components until you get to the sweetspot of 4:4:4 10 bit. Then as that market matures present better options to the consumer market. I don’t need toys, I need tools.

    1. So if that’s the standard why is the only parameter actually being expanded is resolution – the one thing that film never cared about. Remember how we reduced sharpness (resolution) of video to make it more life the pleasing film looK? If 4:4:4 10 bit is to be the standard (the storage folks will love it) then why is there NO move to either in any 4K camera released to date? We all know that 2K (HD) with at least 4:2:2 10 bit would be way better quality and have more flexibility through the production pipeline than the available 4K 4:2:0 8 bit cameras? or 4:2:2 8 bit?

      I do not believe this is a push for quality because the quality is not being provided. Instead we’re being pushed into gear because someone needs to sell panels.

  5. Great, Great, thoughts on pushy manufactures. Yes it reminds me too of the great 3D manufactured push. Build it and they will buy it mentality. Blu-ray standards are a joke at 4.2.0 color compression. Don’t get me started on Blu-rays inadequate menu system in todays high tech world. My Elephant in the room would be; can we please get away from component color subsampling altogether. RGB Raw 10 or 12bit is the native way both the cameras exports it’s signal and display device reproduce the signal in. Buy the way If you think UHHD or 4k is an issue the true Elephant behind the curtain would be NHK’s 8k specs and all those wonderful, how many channels of audio?

Comments are closed.