Why 3D doesn't work and never will (2011)
rogerebert.com
rogerebert.com
I am sad to Ebert go, his reviews defined the industry, but towards the end of his career he became increasingly hostile toward anything that wasn't "classic" cinema as defined by films which he reviewed during the peak of his career.
By his definition, films were art. New narrative vehicles such as 3D, VR, and even video games were not films and therefor not art.
In this guest post by Walter Murch, he complains that 3D films force the brain to converge focus at objects at unnatural depth of fields - this inability to switch depth of field quickly prevents 3D films from working. It's an uncrossable technological chasm.
I would counter that this argument is flawed on two critical elements.
First, the first film to make use of shifting depth of field was famously Citizen Kane in 1941 - before this, the ability to rack focus just wasn't a thing. Yet films 'worked' before Citizen Kane. To posit that films need this ability to 'work' is to discredit the first 40 odd years of film.
The second is the argument rests on our inability to solve this technological problem - a dangerous and often repeated statement in the history of film. Just as other old men were critical of the viability of sound or color given the enormous challenges the first versions of these technologies had, Ebert/Murch here were critical of 3D at the time when companies like Magic Leap and Lytro were just showing the first signs of promise in crossing this chasm.
"3D doesn't work and never will" is an interesting predication along the lines of "locomotives traveling twice as fast as stagecoaches are palpably absurd".
I also became a lot more touched by 2D, either games, or painting. Something is impossible to reenact better through a more realistic tech.
Nowadays I also realize that movies are enhanced theatrics. It's very much about planting seeds in the audience, the mean is not the primary factor.
Or look at YouTube, the quality of videos is stunning nowadays. The biggest problem with 3D video is that currently almost nobody has the means of consuming, let alone creating 3D video. The second problem is distribution/payment. It's still shockingly difficult to discover great indie stuff on Steam, YouTube, Spotify, etc. To a degree, that's because most people only watch content from huge media corporations who of course want you to keep watching their stuff - but it can't be the whole story, especially with e.g. Steam.
I have my own age, but I don't give a damn for HD+ videos. It was cool to see demo videos of coral and landscape in stores but ... even the best tv show or movies won't bring me much joy in 4K. That's optimizing the wrong parameters IMO.
3D could be of help, but so far 1) people didn't find how 2) people barely tried. I remember being disappointed by Avatar because the story was so obvious and rehashed... I went to see a few 3D movies (beside thor I think none surprised my senses). And last year Avatar 1 was on TV. I was shocked, the framing was so amazing. I thought, 1) Cameron knows how to film action 2) the live virtual set he had for Avatar made his 3D 'animation' much more tactile/tangible. No other movie had this, they were average 2D shots with more tech. Reminded me how Scott's Alien 1 has more depth than most 3D movies in the first 10 seconds. Something how they approached movie making, real sets (something star wars went back to, ironically after being one of the major step into virtual sets).
For video games, I can't deny that the 3rd dimension brings something, but for aesthetics or story telling it's moot. The beauty is elsewhere. And some games are doing just that, they will toy with representation and mix hires 3D environment with pixelated characters. Its oddity makes the games interesting and mysterious.
I really do believe that our brain likes 2D representations, and attach emotions and meaning to them; paintings and old games were tickling this sense. The hyper realism (which I ran for too) is tickling another sense, and not the right one.
ps: I just found this on twitter, https://thesixfifty.com/game-changer-lasting-legacy-of-the-a..., it actually touches some aspects of what I'm feeling
I'm not a fan of feeling sick at the theater.
The Nintendo 3DS has a 3D slider, that lets you adjust the intensity of the 3D or just go 2D. I can handle it for extended periods on the low setting, but have stopped using it entirely after one session where, upon putting down the game, I couldn't handle reality. It was like my brain had adjusted to the incomplete 3D effect described by this post, and the real world suddenly physically hurt to look at. It only lasted about 10-20 seconds, but it was more than enough for me to not want to use it again.
I think you're conflating the focus of the movie camera on the original scene with the focus of the cinema viewer's eyes on the projection screen.
What Murch is referring to is the focus "setting" of the human viewer being set to the distance to the projection screen. The focus of the movie camera and depth of field at that end has little to do with it.
His point is that the [2D] cinema viewer can sense the flat focus setting of the fixed distance to the projection screen, but at least other depth cues (such as parallax) are also fixed and therefore consistent. 3D viewing involves dynamic parallax but the focus distance to the projection screen remains fixed, and it is the difference that causes problems for the viewer.
The bokeh of the movie camera has nothing to do with this aspect.
I notice in Ebert's reviews an appreciation of films as mere entertainment in cases where is what they intend to be, and an appreciation of films as art if that is what they aspire to. In fact, one of my annoyances at his (often excellent) reviews is that he seemed to apply consciously lowered standards when he thought that a film wasn't aiming too high, which led to him recommending films that were close to mediocre, because they hit the mark they were aiming for. There is a certain integrity in this, of course.
Ebert's complaints about 3D movies were not the grumblings of an old man but largely justified: they are dimmer and usually worse than the 2D version. I've seen stunning exceptions, though: the best is an IMAX documentary called Hubble 3D, which is the best use of 3D film I've seen and superb all around.
I think he did a great job voicing his own standard, which I think makes a lot of sense, in his review of Shaolin Soccer (https://www.rogerebert.com/reviews/shaolin-soccer-2004):
> "Shaolin Soccer" is like a poster boy for my theory of the star rating system. Every month or so, I get an anguished letter from a reader wanting to know how I could possibly have been so ignorant as to award three stars to, say, "HIDALGO" while dismissing, say, "Dogville" with two stars. This disparity between my approval of kitsch and my rejection of angst reveals me, of course, as a superficial moron who will do anything to suck up to my readers.
> What these correspondents do not grasp is that to suck up to my demanding readers, I would do better to praise "Dogville." It takes more nerve to praise pop entertainment; it's easy and safe to deliver pious praise of turgid deep thinking. It's true, I loved "Anaconda" and did not think "The United States of Leland" worked, but does that mean I drool at the keyboard and prefer man-eating snakes to suburban despair?
> Not at all. What it means is that the star rating system is relative, not absolute. When you ask a friend if "Hellboy" is any good, you're not asking if it's any good compared to "Mystic River," you're asking if it's any good compared to "The Punisher." And my answer would be, on a scale of one to four, if "Superman" (1978) is four, then "Hellboy" is three and "The Punisher" is two. In the same way, if "American Beauty" gets four stars, then "Leland" clocks in at about two.
> And that is why "Shaolin Soccer," a goofy Hong Kong action comedy, gets three stars. It is piffle, yes, but superior piffle. If you are even considering going to see a movie where the players zoom 50 feet into the air and rotate freely in violation of everything Newton held sacred, then you do not want to know if I thought it was as good as "Lost in Translation."
The critic that seems to be most compatible to my tastes, and relatively close to Ebert, is James Berardinelli from http://www.reelviews.net. I almost never feel different about films than he does, so it's a great way to filter the deluge of films.
Similarly, Alan Sepinwall, who currently writes for Hitfix, seems to be be very close to my tastes when it comes to TV Shows. Sadly, Hitfix is one of the worst sites I've ever visited, but I bother going there for his take on new and old shows.
(His re-reviews of classics like The Wire are really nice, by the way. Lotta stuff I missed that he points out and makes me appreciate the show even more).
Up front (and especially a front corner) the bottom-top and center-edge of the screen is different enough in depth that if I'm trying to take in the whole canvas my eyes are working on adjusting focus basically all the time.
Lytro and Magic Leap have developed technologies that record and display the entire depth of field, respectively, rather than a particular point of focus. The technology simulates how light enters the eye at different angles, so that as you change focus, the light field is there for you to focus on, no matter the depth. It's like a volumetric recording of sorts.
If this technology were to make it to a movie theater, it may or may not require a 2d screen, but the point is that it overcomes the issue you talk about.
Not to mention, I don't think Magic Leap has released any products.
3D still exists, some are required to shoot in it or have it converted in post, but it seems more a strategy to capture more profit from non-American markets than to actually punch forward any sort of cinematic innovation. If 70% of your blockbuster's money is going to come from overseas, might as well mine it for all it's worth.
I helped convert a ton of movies to stereo from 2011-2013 in the height of all this. It was a nice pay day but the companies involved had to keep moving to different tax-efficient locales and using cheaper labor to actually make money. Unless you were able to get a director shoot in 3D, a much more complex task, then you get stuck with the labor-intensive conversions to add your $5-10 to a ticket price.
But hey, maybe it will work out. I'll be happy not to pay for it.
That was just a quick google search and I’m not familiar with that work specifically, but I am aware of other VR teams at the big tech giants and at universities that are actively making inroads.
You’re absolutely right: this is an example of Clarke’s first law.
Here's the direct link for people that prefer it:
https://arstechnica.com/gaming/2015/08/uncross-those-eyes-re...
Oh, sorry, I didn’t even notice the link changed on me. That’s what I get for pasting links from a mobile device.
Would there be a way to make them without so much emphasis on age? Not that I disagree that people do stereotypically get a bit more resistant to new stuff as they get older ... including myself ... but at least for me I find that it comes from a more nuanced place than what is implied with your repeated use of age as a synonym for the doubters, the lovers of the classic, the skeptics of the new ...
Anyway, really, I don’t think your comment comes off the way you meant it. It feels a tiny bit aggressive/dismissive to me.
Real quick, this Citizen Kane bit is misleading. Welles's use of rack focus was noteworthy, but it was his use of deep focus that was the true wonder. It is much more noteworthy and famous, and possibly what you are meaning to refer to. It was a feat of engineering and lighting at the time, and not at all a paradigm shift as much as notch on the linear progression of film engineering.
Murch really did write the book on film editing. In The Blink Of An Eye is a short and to-the-point text book on film editing, an offers a simple theory as to why it works, which can be figured by the title. The theory seemed too simple to me the way he explained it, but I've come to understand he essentially means that film cuts are not jolting because we are used to blinking. Whether he is correct or not, the nuance of this understanding would basically argue there is a threshold for which the jolt could be overridden by an evolved necessity for being able to blink (and possibly other similar less-than-linear occurances in optical living).
Murch goes on, in the book, to apply this theory and he argues that elements such as narrative, beauty, concept and other cognitive intrigue can widen the threshold. All in all, he makes the what is probably the best argument so far for how and why the engineering of filmmaking magic works.
His book is the seminal textbook on film editing and worshiped by filmmakers generation after generation for it's timeless truths, because these truths offer a simple way for filmmakers to make decisions in their craft. They don't encroach on any style or technical achievement. Despite the practical delivery, they are essentially theoretical psychology.
He also pioneered digital editing.
I don't believe Murch is immune to Clarke's first law, but he is no Ebert. Murch is no critic. He is an engineer, and while this post doesn't get into the specifics, his claims here are supported in his battle-tested texts.
If that's the case, we could be losing something and gaining other things. That is not unheard of in the history of art and technology, and something we should probably all accept.
As for the "inability to solve this technological problem", that could be true. I don't think we can solve all technical problems. The opposite is saying we can solve all technical problems.
The best quote for me in this article comes at the end: "[...] if the film story has really gripped an audience they are "in" the picture in a kind of dreamlike 'spaceless' space. So a good story will give you more dimensionality than you can ever cope with."
This seems exactly right to me. A captivating film already sucks you in. 3D just doesn't improve upon that for me.
This means that you cannot look around things, or change occlusion by moving your head around. As others have mentioned, you also cannot change your focus. Until 3D becomes actual 3D, it will always ring hollow.
It's possible that Oculus may be the platform that achieves true 3D, but I suspect until we arrive at a place where the "holodeck" is a fully functional universe, we won't get there.
Media is evolving (look at games, TV, youtube, and instagram). 3D will find it's place within the evolution of media.
One area where the long form exceeds the 90 min format is the depth of characterization it allows, and accompanying narrative attachment.
Soap operas have relied on this effect for years, allowing them to succeed despite low production values.
Coupled with the superior craftsmanship of modern episodic TV, is a pretty "gripping" experience.
The extended narrative structure also allows more justification for the characters actions to emerge naturally. In a shorter film, a lot of the characters' dialog has to serve as exposition, resulting in dialogue that sounds contrived.
I've also noticed shorter movies reliance on tropes to condense the timeframe of the narrative action so extremely. Things like voice-overs, montages, heavy make up to show aging, conventional narrative structures, the aforementioned expositionary dialog, (sub) titles.
The fact a narrative that spans years can be communicated effectively, in a 90 min film, is powerful evidence that the art of filmmaking does not use a visual medium to simply create an experience that relies on the verisimilitude to lived reality.
This highlights what I see as a...quirk of human inclination to try and recreate, or rather present as alternative perspectives, imagined realities.
Personally, probably related to my middle age, the prospect of technology to create virtual reality has minimal appeal.
In the other hand, technology that allows enhanced perceptions into actual reality is highly enticing, and seems to be an underserved area of human experience. I'm thinking of things, for example, like cognitive enhancing drugs. Or tools that can help control out of balance behavior.
The 3D video captures qualities of the cave paintings you'd never realize in 2D.
And virtually no person alive is going to get to see these paintings in person so it's the closest you'll ever get.
In the director's commentary – which you should really watch, if only to understand what's going on – Wim Wenders tells a bit about 3D cameras, their challenges and opportunities and how much they improved so that he felt comfortable using them "for real" for the first time.
Not that I'm a fan of 3D films anyway, but that's an artistic rather than technical preference.
I’m not saying VR will have no role, or that it’s impossible to make a great 3D movie, I just think they’re niche technologies and will have niche roles to play for a long time to come. Never say never, but anyone who thinks either is on the cusp of massive mainstream adoption or changing the ways we make and consume media is fooling themselves.
I've seen others experience that visceral reaction as well, and generally they tend to pivot and become a bit of a believer in the tech. But we'd all be crazy to think it's temporally just around the corner. The tech hurdles are significant. But definitely not insurmountable IMO.
Saying "good 3d is hard" is very different from saying "it doesn't work and never will". I'm sad to see good 3D on the decline. When done well it's rather spectacular.
Can't comment on 3d on a TV -- never done that. Avatar is the only 3D movie I can remember seeing in a commercial theater. There may have been others, but they weren't worth remembering.
It could make my eyes cross and I'd still pay that premium to keep people quiet more often than not.
In real life you can normally look round a scene, focusing on objects of different distances as desired. But with a 3D film you can’t. You can’t look at an object which seems to be nearby and have it come into focus, as the camera may be focused elsewhere. And by trying to focus on something which isn’t there, everything else goes out of focus as well, confusing the brain and causing the migraines and nausea some people have.
Also the 3D glasses don’t fit over my normal glasses, leading to a really awkward experience.
I don’t think 3D films will be truly successful until we have light field or holographic displays, where you can focus on anything at will.
It will also mean developing a new visual language for this kind of film, where the viewer can decide where to look, rather than be guided by the filmmakers.
It’ll be exciting, and these kind of films will hopefully look quite different to what we see now.
While I agree with this, I've to say I generally prefer 3D versions of the movies for that extra oomph. There is no particular reason left to visit theaters otherwise. Especially it's just super cool on IMAX (I mean "real" IMAX). Movies like Space Station 3D would miss out on too much fun without being 3D.
Maybe the 2D shifting depth of field is easier because you're more used to it.
It may be that as you look from the foreground trooper to the out of focus background troopers your brain attempts to refocus but fails, which feels funny. But if the camera operator and editor correctly anticipate your attention and focus on the right thing, you don't need to refocus or re-verge.
As for 3d, i think its comfort is for the next generation of movie goers to decide on.
Ultimately your brain adapts to what you expose to it the most.
To name drop another movie, I'd highly recommend Dredd to anybody that has a 3D TV available.
At the very very least, it has never distracted me from the the viewing experience. So I don't understand the widespread revulsion for it.
The article is absolutely right, they got incredibly greedy with what little good content they had.
The strange thing is that it seems that different people can have different opinions about how immersive the experience is; even when trying the same simulation with the same VR gear.
So the issue Murdoch is discussing would be relative to the viewer. This is obviously a problem in a mass medium, where making accessible films, and films accessible, requires mind boggling effort.
Someone would need to come up with a true light field display. This would make steroscopy work for the whole area in front of the display. Paul Debevecs team made a daring proposal a couple of years ago to build a light field display from an array of Pico projectors, each one acting as a single pixel of the display. The viewer would need to look directly into the projectors. It seems that this project never got off the ground, though.
3D effect was so small, seemed like to size of a foot ball in front of my face, with a 2D backdrop on the main screen.
Seemed like they had filmed with the cameras too far apart.
Very disapointing.
Between the absolutely paper-thin plot, and the 3D itself, I lasted about 15 minutes, well within my theatre's 30-minute money back limit.