Netflix Originals: Production and Post-Production Requirements v2.1
backlothelp.netflix.com
backlothelp.netflix.com
Cameron likes long pans over beautifully detailed backgrounds. Those will produce annoying strobing at 24FPS if the pan rate is faster than about 7 seconds for a frame width. Staying down to that rate makes a scene drag.
Now, Cameron wants to go to 4K resolution and 120FPS.[1] Cameron can probably handle that well; he's produced most of the 3D films that don't suck. He's going to give us a really nice visual tour of the Avatar world. For other films, that may not help. "Billy Lynn's Long Halftime Walk" was recorded in 3D, 4K resolution and 120FPS. Reviews were terrible, because it's 1) far too much resolution for close-ups, and 2) too much realism for war scenes. Close-ups are a problem - do you really want to see people's faces at a level of detail useful only to a dermatologist? It also means prop and costume quality has to improve.
The other issue with all this resolution is that it's incompatible with the trend towards shorter shot lengths. There are action films with an average shot length below 1 second. For music videos, that's considered slow; many of those are around 600ms per shot.[2] They're just trying to leave an impression, not show details.
[1] https://www.polygon.com/2016/10/31/13479322/james-cameron-av... [2] http://www.cinemetrics.lv/database.php?sort=asl
Why do home videos have that "home video" look? The biggest reason is the 60 fps frame rate - it just doesn't feel cinematic. Even the 48 fps of the Hobbit films felt too "lifelike" and not cinematic enough.
A lot of prominent directors, as you've mentioned, say they'd like to move towards a world with higher frame rates. But that'll be a bitter pill to swallow for a viewing public that unconsciously believes "cinema" means 24 fps.
3D in particular is very difficult to watch at frame rates as low as 24 fps - a big reason it makes so many people nauseous, and a big reason so many directors are saying we need higher frame rates.
High resolution may not be a huge positive but it is definitely not a negative. There's nothing inherently cinematic or better about low resolution like there is about 24 fps, and if excessive sharpness feels jarring in a scene, the cinematographer can elect to use a lens with a softer focus.
And the strobing effect you mention - unless we're talking 3D (where motion blur feels wrong), a low shutter rate and consequent good amount of motion blur easily avoid strobing.
(Film is really dead. There are some directors still making noise about shooting on film, but it's digitized for editing.)
It just doesn't look film-y. I suspect it won't be until VR that we'll see proper high framerate.
Its just such an oddity that I don't think people will take the risk, given the expense in retrofitting cinemas. (plus virtually no TV is actually capable of properly doing 60fps)
just one point:
>Adding grain in post-processing is on the way out.
thats with us to stay. most film grain from 2008 onwards is digital (yes even on film films) because most will have gone through a VFX pipeline with DI. Grain is stripped out and put back in after.
grain is a good visual cue for a number of things, just like desaturation of colour. its a tool thats not going to go away
I too find 3D gimmicky often, but probably because the technology varies grandly between production, cameras and theater displays. On the other hand, there are 3D movies every day in every big cinemas and 3D TVs as well. So I'm not sure we can say that it wasn't a great success.
> High frame rates haven't been successful for several years
The number of cinemas that can display 48fps is not great, the number of cinemas that can display 60fps is zero? So I don't know how you can say that "High frame rates haven't been successful for several years".
Actually if you look on Youtube, high FPS videos are successful.
> At which point it will look like a youtube video to many, not like a movie in the cinema
There are some great youtube videos out there, don't know why you're saying this. Cinema is what you're defining as 24fps, sure because you're used to it. If tomorrow we start watching a lot of 60fps movies then you will define it as Cinema. Objectively 60fps is better for action movies anyway, the rest will follow.
any cinema that can do 3D can do 48FPS, at least.
RealD uses a single projector, with an electrically controlled circular polarising filter on the front.
This is why 3D in cinema for any kind of action is terrible, because you get juddering nastyness.
To get round this, some places project at 96 FPS(well I thought it was 144, but that might be the limit of the projector where I worke)
You'll have to explain to me why they weren't showing The Hobbit in 48fps then. You sometimes had to go to a different country to see it in 48fps.
3D has two major problems. First is that the technology sucks. The glasses are heavy, bulky, and don't do a particularly good job of filtering out the opposite eye's channel. Second is that filmmakers don't understand how to do 3D at all. Every 3D film I've seen loves to add parallax where there should not be parallax. They don't understand that binocular depth perception only works out to a few dozen feet, which causes anything with observable parallax to be perceived to be nearby, and that in turn causes large objects to look tiny. Seeing a spaceship or airplane or mountain that looks like a toy because the filmmakers decided to "pop" it out of the screen is the exact opposite of a cinematic experience.
High frame rate doesn't have this problem. The technology is good, and using it properly in films doesn't appear to be a failing.
Fake parallax that just turn epic scenes and scenery into tabletop models.
It's obvious, but why do they ruin their efforts like that. Don't they watch their own movies after post 3d editing?
I think even Avatar made it too far. I think I've whatched some animated films that didn't blow totally, but almost every other film that I have seen in 3d was a disappointment.
Either they are huge, to get two cameras side by side, or they have a half-silvered mirror arrangement (with colour disparity)
Add to that the rigs wobble (vomit inducing) and the distance between the cameras is far to wide, it all looks a bit poop, or requires a huge amount of post work to make fly.
So the normal thing to do is manually cut out each object (rotoscope) and adjust the divergence to place it in 3d.
every object, every frame.
it mostly looks a bit poop.
Not to mention is normally done quickly, like clash of the titans was converted in ~1 month.
for documentaries and sports, yeah, its brilliant.
But for "film"? it sucks.
why?
Because it looks like a play, but with bad acting. Everything that we have learnt, subconsciously, about film, is based on 24 FPS. Any action of any narrative substance in a modern film is linked to a slowmo. This relies on 24FPS. Because things are smooth, we register it as different.
Now, I suspect where High framerate will be a thing is in VR. But thats a new medium, in the same way the talkies were.
ultraHD is 8k @ 60fps.
its great for sports and nature documentaries. films look like plays. actors look stilted and wooden.
There is also not nearly as much tweening as you might expect. Sometimes animating on ones just cannot accurately give you the same effect as letting a persons brain fill in the missing info. Which is why watching The Hobbit in 48fps pulled me out of the movie at some points; I appreciated the extra clarity, but there were details that would have otherwise just been a blur that became distracting.
That's a good explanation which fits with my experience. I wonder, could that aspect of 48fps be mitigated by bumping up motion blur (i.e. lowering shutter speed) during such moments.
I assume people who think like that would be very marginal minority of movie experts. I don't believe average viewer would even think of such argument.
Personally I do think people will get over it eventually, especially for less "cinematic" stuff. Or possibly variable framerate will become a thing, and directors will choose different framerates for different effects.
Though, I heard some people say that scenes in old LOTR movies looked more realistic. This was mostly because Hobbit used more CGI though, for example orcs in Hobbit movies were all CGI but before in LOTR they had real actors to play orcs.
I personally didn't get that feeling either. Just got a feeling of "it looks better". Especially that dragon scene.
> variable framerate will become a thing
It already is a thing with slow motions, but it is not what you're thinking of.
Have you ever tried to do slow motion with a 30fps video? It doesn't look good. So your idea will probably look bad, inserting 24fps sequences in a 60fps movie will just look laggy.
I liked it so much that I even went out of my way to buy a TV that had a built-in frame interpolation that's been said to be better than most other high-end TVs.
So I think you're quite wrong about 4k being "too much resolution for closeups" or being "incompatible with... shorter shot lengths".
Therefore, the best way to interpret "Bitrate of at least 240 Mbps (at 23.98/24 fps) recording" is probably that if you shoot at 24 frames per second, then you must have a bitrate of 240 Mbps. Not that 24 frames/second is the only framerate allowed.
I'm surprised they don't ask for VFX sources to be archived though. ST:TNG and Babylon 5 both suffered badly from loss of the original VFX.
What do you mean by this? Is it common practice to "remix" a show when it's rebroadcast? I just assumed they more or less press play on a .mp4.
To make a high-definition format (e.g., blu-ray or HD streaming) they'd go back to the original 35mm masters and re-convert them. But then all of the visual effects had to be re-added on top.
There are various articles that mention quality issues with the remastering:
http://trekcore.com/blog/2015/08/netflix-brings-vfx-fixes-to...
http://trekcore.com/blog/2012/11/review-star-trek-the-next-g...
For DS9/VOY and for Babylon 5.
So collecting VFX shots in a pre-rendered state is notoriously difficult. Doesn't mean it's not worth trying. But you'll probably end up with various decomposed elements (models, rigs) and not something you can easily and quickly re-output in 4k, HDR, etc.
its just cheaper and simpler to re-do it from the clean footage, or perhaps get some of the original models.
Yeah, Star Trek (VOY/DS9) suffers really on Netflix - I wonder what they chose as source - straight DVD rips? The intro of VOY, for example, has weird color artefacts when the deflector dish of Voyager flies by the camera.
One thing intrigues me though - albeit likely a function of my lack of knowledge on the matter - do these requirements implicitly rule out shooting on film for Netflix?
(I mean, I'm sure that ${hollywood-bigshot} could negotiate, but for Joe Public..?)
http://variety.com/2017/film/news/bong-joon-ho-working-with-...
At first, Darius Khondji, my cinematographer, and I wanted to shoot ‘Okja’ on 35mm, but Netflix insisted that all Netflix originals be shot and archived in 4K,” Bong said in an interview with Variety
That's just BS on Netflix's side though. They still feature all kind of non-Netflix movies that are shot on film.
In terms of resolution, no. Standard 35mm can produce a good quality 4K transfer (depending on the speed of the film you're using), and if you shoot on VistaVision (which is where you use 35mm film but rotate it 90 degrees) can go even higher.
Of course, the cost to do so is massive, and it's getting larger ever year. The only TV show I know this year that was shot on 35mm was Westworld.
Joe Public can't afford to shoot on 35mm for his/her Netflix original series even if they wanted to. Netflix would almost certainly only let an extremely experienced A-list director shoot on film, both because they'd be able to negotiate it but also because on film you can only usually afford a 3x overage (that is, you shoot three times what you end up using), whereas with digital that can go as high as 10x. It takes a really disciplined director and DP to manage that.
Even a 10:1 shooting ratio was pretty low for indie features in the 35mm days. Nowadays it's not uncommon to see shooting ratios upwards of 30:1 on digitally acquired productions. (Although that number varies a lot with the director.)
Having seen 1080p stretch and play nicely on a 30 feet cinema screen, and not being much worse looking from regular Hollywood titles even for front seat viewing, I don't see the allure of 8K even for "future-proofing".
Sure, monitors and tvs might improve their resolution in the future. But I don't se human eyes improving much (regarding angular resolution vs distance) or houses getting any bigger to fit a 30ft tv.
4K is good for reframing (cropping) and higher detail, but after some point enough is enough.
Sure the increase in resolution is "imperceptible" in terms of individual elements building up a particular frame, but to assume a visual cortex can't discern increased detail is just arrogant and borderline laughable
But reality is that we're seeing higher color depth ("HDR"), higher framerates, higher quality sound streams (e.g. Dolby Atmos and beyond), and even technologies we aren't envisioning today because the hardware isn't "there" yet.
If 8K is only a resolution bump, then boring, but it may be much more than that. 4K is.
Shoting on lower resolution gives you better color depth/accuracy AND higher framerates to chose from.
It gives better color because (all other things being the same) signal and color accuracy degrade with higher resolution -- less photons "hit" each diode on the sensor.
And there's literally no digital cinema or amateur camera that doesn't allow for much higher framerates in lower resolutions than it does on higher resolutions. That's because the higher the resolution the more taxing it is for the sensor and processor in the camera to keep up with high frame rates. 4K x N fps means 4 times more volume than 1080p x N fps.
Lastly, sound is totally orthogonal, because that's recorded off camera with special recorders.
So we might be getting better color depth and sound quality lately, but that's not because we moved to higher resolution.
Seriously, sure, 4K is enough for output but who says it's enough for input? As long as sensors keep getting better, the industry will keep finding ways to take advantage of it until it's essentially required.
Imagine a future where you can zoom in on any detail as well as you could with a high-res sensor at capture time?
It's not necessary for today's viewing experiences, but we know little enough about what is going to become popular that I wouldn't put ANY bets on "4K" being enough forever.
Not really. For one, 4K already makes life more difficult, adding huge trnascoded files, slow renders, proxy workflows, etc. That's on a pure technical level.
On an aesthetic level, it's a bad habbit too and cropping in post is a lazy copout or an emergency kludge. Setting your frame is more than just "getting what you want in the final output", it affects parallax, compression, etc. Croping from 2x the frame is not the same optically. And deciding in post means a less well thought out frame in the first place. So, it's nice as a emergency kludge to save ones ass, for documentary, ad work or news, but not so good for movies.
Perhaps action movies won't be much improved with better resolution, but I'll want screencasts, browser windows, nature documentaries or really any detailed image to be shown in 4K at least.
Actually I'd want screen to stop being "TVs" or "computer monitors", it should be just a high def big rectangle than can be used to display anything. For that we'd need to come at least to retina level of text rendering from any comfortable distance. That should mean 8k as standard def in my book.
I would love to see this as well.
My ideal "television" has at least 8 HDMI inputs plus a couple of composite or component video inputs. There would be no cable/antenna/tuner input. There would be buttons on the remote for power, volume, input, plus menu and enter to access the settings. There would be a handful of audio outputs, and a switch to disable the internal speakers, which would allow it to feed into an audio system for those that like to have that separate. It would come in a variety of sizes, but only in 4K or 8K resolution.
This doesn't have a tuner or any TV-specific features, nor does it have any of the Smart TV stuff. People who need those are not the target market for this TV.
I wish this was a thing that existed.
I bought my 4K TV at $800 - at RRP of $1800 I would have been really angry but I can buy a 4K HDMI switch and get over it.
The problem with this is that things you want in TVs are different than things you want in monitors. Its very rare that you need a 120+hz refresh rate, 1ms response time television, but you do often want that in a computer monitor, and response time especially is very, very expensive to reduce (modern monitors often have <5ms response time, televisions will often have 10x that).
That's because monitors are mostly about static images, and are viewed in a much smaller distance than TVs.
8K presumably provides the same benefits but to a greater scale; you can scale it down to 4K, you can 'zoom in' on parts of the shot (such as a subtle detail, a pickpocket stealing a wallet for example) without having to use a second camera or record the shot twice, and so on.
When you film in 1080p it's actually a bit more zoomed in than you are expecting because it's really using the whole 4k+ sensor and applying some fancy coding to that to pan and zoom around in that space to provide some stability.
It works incredibly well!
4k and 8k means a completely different use model, like a giant wall-sized screen in the same living room. It's the difference between a movie theatre and an IMAX movie theatre.
When that happens, cinematography can expect to be different, sorta like IMAX movies have different cinematography. Sometimes they even composite multiple HD films together on screen without losing resolution. This would actually be extremely useful for sports, where the background shows the entire field at 8k, with windows showing stats or replays, etc..
my bet is that they are receiving so much submissions, and actually buying all of them, that I guess the company, or a few clever individuals, saw the "consulting" or equipment renting/financing potential and will try to monetize on selling cookie cutter packages.
Cameras like the Sony A7s or the Panasonic GH5 are great for the low budget film, but if Netflix green lights your project, you can afford much better cameras. Unless the project calls for really small and unobtrusive cameras, in which case Netflix would most likely approve the use of whatever camera best fits the project.
If you are not filming in a place where equipment rental is free like the US, GH5's become very inviting as your second (and third) camera.
Not really, at that level everyone in Hollywood is renting their cameras anyway. Indeed, one of them cameras on that list (the Panavision DXL) is only available to rent, you can't buy it.
It is reasonable to say you want a painting that is 4 x 8 and fits on your wall. But dictating that all paintings must be painted with Kolinsky Sable brushes is unreasonable.
4k 10-bit log EXR ACES deliverables, sure thing.
But please don't tell Roger Deakins he can't use that camera[1].
http://nofilmschool.com/2012/09/roger-deakins-talking-about-...
Here's a sample "A Day in Kyoto" shot at 4K 120fps raw:
Crazy expensive but obviously given the value of those reruns the cost made sense.
It's just a matter of which shows are worth the cost and time to do it.
I can't find it now, but I believe they even redid some of the early CG effects by just rerendering the old models with modern software. It's worth preserving that kind of stuff for future generations. One of my favorite video games was rereleased over a decade later with updated graphics. Sadly they had to recreate everything from scratch because the original files were lost. And the original models were very high detail. They just converted to 2d and down sampled a ton because of the limitations of PCs at the time.
References: http://ecee.colorado.edu/~ecen4242/marko/TV_History/related%... http://electronics.howstuffworks.com/first-hdtv.htm
http://dpp-assets.s3.amazonaws.com/wp-content/uploads/specs/...
https://drive.google.com/file/d/0B9DJydDVOVKKLVdCdlF2cFVDVEE...
...again, a bit less detailed.
* c300 mk ii 6998
* c500 18995
* c700 28000
* Varicam 35 12949
* caricam lt 16500
* red dragon 32520
* red weapon 49955
* red helium 49500
* panavision dxl cant find
* sony f55 28990
* sony f65 cant find
* sony fs7 12949
* arri alexa 65 cant find
* ursa mini 4.6k 5995
* ursa 4.6k 4995
Sony don't manufacture the F65 anymore, so you won't find a current list price. It originally retailed for $65k for the camera body, and $20k for the SR-R4 recorder, so basically $85k. [1]
[1] http://nofilmschool.com/2011/09/sony-officially-prices-f65-t...
I was putting some quick foundational numbers in there... I would expect all of those prices to be 5X or more than what you see after all the extra stuff that one would need to have a complete setup might be...
let alone time, training, education, talent, etc etc etc...
Well, its died down now, but it used to be a proper cult. old $work used to make software that was a key step in transcoding from their shitty jpeg2000 "redfile" to something more usable.
yes they had high resolution, but they had rolling shutter and were still quite expensive. Not to mention poor colour reproduction.
It's also interesting how in charge of this trend Netflix is; they're right out front (Amazon is doing interesting things too, with their pilot voting feature). They're providing the formula for how to make shows for Netflix (which would probably be acceptable across the board at Amazon and whoever else is doing original programming at this level and on this scale).
I really like this kind of openness about technical requirements, because it feels like an open invitation to make things. This, more than any hand-wavey "There's more opportunity for small productions than ever before!" kind of statements feels like an actionable piece of knowledge and a part of any business plan that involves making content for this new market. Democratizing that knowledge by posting it on the web is super cool. It seems small, I guess, but it's a cool indicator of where the market is going.
If I were fresh out of film school, or currently working at a production company, I'd be looking at that and thinking, "I bet I could raise the money I need to put together a pilot to these specifications." You'd still have to put all the pieces together to make a good show/film like you always have, but the technical side of it looks really achievable today in a way that it never has.
Sure, if you wanted to shoot beautiful cinematography, you'd want more lens options. But, if you're running on a budget, you could shoot a whole film with that one lens, and it'd still look better than most television shot in the 90s-00s.
I think some of this will change with time as people get more used to the look and the production gets better (effects, costumes, etc.). Music is also an artificial construct from the get-go, even live, but most narrative film is meant to reflect real life in some way, shape, or form. That means that anything out of place is caught immediately by our expert eyes.
All that said, when I was in the industry, one thing I noticed was how easily you could gloss over minor visual details like set dressings or small props (think food on plates), but even the slightest error in sound production was enough to destroy the effect.
It's funny that 2D stereo still works so well. Perhaps that will also change again, with binaural recordings and object-based audio, but that's something I don't see the music industry using anytime soon.
There are some good DVD-A or SACD's with 5.1 audio, but they are niche and expensive.
Well, some people do. But unlike 24 fps vs 60 fps movies, mp3/cd has huge usability improvements against vinyl (not to mention it's free if you pirate or stream), so there's no real competition.
But take musicians for example: even most young electronic musicians, born entirely in the digital era, prefer analog synths for their warmth and bite, or try to emulate it on the PC with external processing or purposeful degradation of the signal.
Or take guitars. Who likes a clean, undistorted signal? Even for jazz, people still prefer a coloured, saturated tone.
Long version if you're interested in why people some people may think vinyl sounds better than CD: http://wiki.hydrogenaud.io/index.php?title=Myths_(Vinyl)
It's hard to compare these two arguments since "CD sound" is the standard these days, whereas the "technically inferior" 24fps is still the cinematic standard. So, while people may label 60fps video as "unnatural", nobody would label a CD recording as such. Also, there are people who've never listened to vinyl to this day, whereas almost anybody will be used to the "film" look of 24fps, which has been the de-facto standard for decades.
The warm sound of vinyl on the other hand is just poor sound quality masquerading as something desirable.
I'd argue it is color grading and wide aspect ratio more than anything that makes movies feel like movies, but 24 FPS and it's less lifelike movement also does make them feel distinct.
On the other hand, no one argues that sports look a lot better in 60 FPS. For home movies I take, I also want at least 60 FPS, because it preserves reality better.
But movies aren't about preserving reality.
We saw the second one in 24fps 2d (we mostly avoid 3D because we rarely care enough about it to pay extra, but it was the only way to see the first one in 48fps) because we both hated the first one's presentation so much, so I don't know whether we'd have experienced it in that one.
I've never had the same impression from TV or home movies which is 50/60 fps.
I for one, don't like that kind of high frame rate motion -- it looks as fake as bad CGI graphics.
I saw all the Hobbit movies in 24fps, and—especially (at least, this is where it first really stuck firmly I in my mind as wrong) in the underground scenes in the first, but also in parts of the other two—had a similar impression.
I think it had to do with some other element of cinematography particular to that film series, though it's quite plausible that 48fps exacerbated it.
Maybe it would be possible to shoot at 60fps or even higher and then dynamically adjust the frame rate to keep the 24fps but smooth some scenes as required?
Film or video shot at high framerates will necessarily have to be shot at higher shutter speeds than films shot at a lower fps. This has an obvious visual impact on the image.
As you need to add VFX, basically all content is shot with as fast a shutter as possible, and you add the motion blur back in post. And, as you rightly mentioned, when you do that you can choose any shutter speed in post.
Dealing with motion blur when rotoscoping is a bitch, btw.
Making it look good is akin to polishing a turd.
Films at 24fps are precisely the same thing. Their lack of detail and realism is precisely what has made the medium so successful over the last 100 years.
I love watching 48/50/60 FPS for psychedelic things or music visualization (and games of course ! like most of gamers) but it should be used carefully and intentionnally by the content director / producer. Some scene/content would require high motion blur and some others scene/content high framerate...
It's a great example of a happy accident, or serendipity.
Ang Lee shot Billy Lynn at 120fps and Sony Pictures had every intention to distribute it in 120fps. In fact they created a version at 24fps for legacy sites, 60fps as well as a 120fps version. At the last minute the studio pulled the 2 HFR versions and released only the 24fps version (note: not pulled for technical reasons) I've seen 20 minutes of the 120fps version and personally did not like the look.
It remains to be seen if James Cameron can utilize HFR in a manner that the general public finds appealing. He has stated that the Avatar features will utilize high framerates in some way.
I can believe that some scenes might benefit from the visual effect of a lower framerate, but I can also imagine other scenes, such as complex action and long horizontal pans might benefit from higher framerates. Also I think the "too real" feeling might work quite well to convey grittier settings.
I do believe someone will crack this nut (as Cameron did with 3D). But it's going to take the right project and very creative filmmaking techniques. Personally, I think the first one that works will be a sci-fi in a sterile setting, so the HFR will work with the narrative, not be a distraction.
True, Post Production is not the largest piece of the pie, but it still adds up to millions of dollars.
An average person can't even tell the difference between 720p and 1080p.
Camera Requirements
4K Resolution:
• Camera must have a true 4K sensor
(equal to or greater than 4096 photosites wide).
The fact that they allow UHD in the IMF master makes it even less reasonable, because downscaled 4K-> UHD footage will have lower resolution than the native UHD footage.
On the other hand I definitely applaud their focus on having good backups and preserving the original source material.
cinema 4k is 4096 wide by 2160 https://en.wikipedia.org/wiki/4K_resolution#Digital_cinema and predate quadHD by quite a way.
firstly the better cameras are proper 4k, QHD cameras tend to be a bit rubbish. Secondly there are many more tools for dealing with cinema 4k, than QHD
Some movies have it worse than others, and maybe it's my own fault for training myself to look for it with interactive games, but there are definitely times where it stands out as an outdated technical limitation.
Conversely, high frame rate stuff "feels" so much more real. It's hard to explain, but if you look at side by side comparisons, the higher frame rate is definitely preferable. The guy who did the tech demo of full motion video on a 1981 PC had some interesting points about this. He did experiments to find the optimal trade off between frame rate and resolution. He found higher framerate with lower resolution was better than the reverse, and had some nice examples of it. PC gamers have known this forever and are obsessive about high frame rates.
Most movies are still presented at 24fps because higher frame rates apparently look different. The Hobbit was released in 60 fps and the reception by critics was mixed. Personally I don't see it but I'm no expert.
48fps, not 60fps.
Sure. Why not specify a progressive 60fps though. Every second fkin phone can do 60fps these days...spec'ing future netflix catalogue at 24fps seems very odd.
When you start increasing that in YOUR film, you lose a sense of familiarity in your audience and their (subconcious) willingness for suspension of disbelief probably goes away with it.
Really? I don't see any tangible reason why a low framerate would be superior to a higher one. Even on a very basic level high framerate is superior to low framerate and 24 is well...the absolute minimum.
Bandwidth and data transfer, for example - also many devices eat through battery like nothing else when you feed them with high-framerate stuff, and old-ish TVs may not support high-framerate material properly.
The bitrate is specified in the specs so that's not it.
>old-ish TVs may not support high-framerate material properly
These are the specs for how it's shot not how it's broadcast.
The Arri Alexa is eliminated by these specs, for crying out loud. The single most popular camera amongst high end feature cinematographers.
This is driven by some misguided belief that input resolution == output resolution AND that resolution is the measure of quality.
I really hope they get their head out of their asses on this at some point.
It's good to have quality standards, and thank god they aren't Turner Classic Movies (the fuck was that all about??). But these specs are as arbitrary as saying all of your food must be cooked in copper cookware.
We tell stories, not pixels.
It isn't a matter of which camera is better, but that the camera is a part of the creative process of filmmaking and is not strictly tied to the quality of deliverables. It is a brush with which an artist chooses to paint, for creative and practical reasons.
For a company like Netflix to control quality the goal must be to capture the creative intent with the highest fidelity[0]. Keeping in mind that the pixels serve the content, the content does not serve the pixels, and pixels are just one of very many things that serve the content. This can and should be done by specifying the quality of the result, the deliverables, not the tools used in the creation process.
[0] Fidelity as apposed to quality is an important distinction (that seems to be lost on Netflix). "Pi", "Saving Private Ryan", "No Country for Old Men", and "Deadpool" have different objective measurements of image quality, but they can all be presented to an audience at high fidelity.
It sounds like Netflix could use someone like you.
They are the McDonalds of movies, and a McDonalds is all about standardization and efficiency.
If only allowing for cameras that deliver a narrow and predictable band of technologies let's them work more efficiently, and if they are able to make and save money that way, it makes perfect sense for someone who works at scale to do exactly what they are doing. They are a market marker now, and thus can afford to play around with previously established norms.
It's true good image quality does help, but my point is they already have both simply by specifying the delivery requirement.
What they are doing is more like a company that says they won't buy software that's written on anything other than emacs, or Notepad.
>Unified Spec: With version 2.1, we made the decision to unify the two previously separate Features/Series and Docs/Coms documents into one overall set of requirements.
I'm in post on a project shot mainly on the Canon C100 (not on the approved list). We will be talking to Netflix at some point about picking up the project and I have no worries about in not being in 4k or shot by an approved camera. If they like the project and want it, the camera format won't matter.
And, these are not onerous requirements. They ensure good quality but don't impose significant costs. There are several cameras here for under $10k, which was unheard of even a decade or two ago; when I worked in television in college (20 years ago), the cameras in the studio cost ~$80,000 each, and the decks we edited on cost ~$25,000 each.
If you're making a show specifically for Netflix, you follow these guidelines; if not, you can make it however you want, but these aren't terrible or adding significantly to the cost of a production. I mean, there are several line items on most productions that'll be more than the camera. And, in post-production, resolution and file size barely matters today; a computer can chew through 4k almost as easily as 1080p, and multi-terabyte hard disks are cheap-as-free.