The Jurassic Park Lesson on Why 3D is Unpopular

Back when I started this blog, I was well aware that I was writing to a niche market of people who, like me, see 3D as a potential improvement to the film industry. I still stand by that. I want to see 3D flourish into the great cinematic experience it should be. That said, if 3D was the lump sum of its awesome potential, I wouldn't need this blog to defend it, and my recent voyage to the cinema to see Jurassic Park 3D led me to understand why 3D isn't appreciated.

Barring a few conversion errors (transparent objects and between trees being the largest offenders), the 3D conversion in this film was a fantastic technical achievement. When those conversion errors weren't present, I would not have been able to easily identify this as a non-native 3D movie, thanks to the care they obviously took to accurately map people and objects. Given the number of details they got right, I am permissive of the errors that persisted. Regarding the cinematic benefit, 3D didn't always help (some scenes were framed so that the 3D version drew your eye away from the action of the scene due to the convergence/focus technique I described in a previous post), but when it did help it felt like the movie's natural state was in 3D (in particular, I think of that vegetarian hacker kid dangling from the air vents above a hungry velociraptor). They made the best possible 3D conversion for this film, and the studio should be proud of what they created. It was extremely good, but they were contending against external elements outside studio control that fought my attempts to enjoy it.

The first was the critics. Now, admittedly, most critics gave Jurassic Park 3D favorable reviews, even for the 3D. But, as I've seen with critics before, nobody has ever talked qualitatively about why the 3D was good. And unfortunately, the first critic I found to actually give some sort of qualitative assessment claimed that, "like most 3D conversions, it looked like a cardboard cutout." (Forgive the lack of a reference; I don't want them to receive traffic, and therefore ad money for such a ludicrous statement.) So, we have an outright lie declaring the film poorly produced technically, or no information to go on at all about attending this movie. I imagine a number of people were filtered out of wanting to see the film at this stage.

The second problem was the theater itself. What hurt me especially was that I wanted to attend a premium showing of the film. Consequently, I purchased tickets for a showing labeled under Cinemark's XD branding (which is supposed to compete with IMAX as a premium screen size and projector quality). The theater I attended was instead using the XD branding for a film that was clearly being shown in a regular movie theater rather than an XD one. On top of that, I never recieved a pair of 3D glasses (fortunately I had snuck in my own pair of higher quality glasses). I wasn't charged the XD price for the showing, but I am still annoyed that I was unable to attend what I intended to when I arrived, and the staff provided me otherwise poor service.

Speaking of which, the third problem was also the theater's fault. Jurassic Park 3D makes heavy use of floating windows to keep the 2D film decisions of the movie from hurting you eyes in the 3D version. But there's a problem common enough that Disney created a tutorial video for theater employees specifically telling them not to do what they did in the theater I attended. Basically, in 2D theaters, employees will sometimes cut off the edges of the movie because it's difficult to set up the movie precisely and there's rarely critical information there in 2D movies. But floating windows exist entirely in the edges. Consequently, I had the worst of both worlds: frame violations and black bars on the edge of the screen.

Now, this isn't meant to be a post about how bad that theater is. Typically when I've attended this theater, my experience has been great (I would still recommend it to any of my friends as my preferred theater). But it highlights 3D film's biggest weakness: its complexity. There are so many layers to a successful 3D exhibition that if any portion of it is inadequate at all, 3D becomes a painful experience. Everything has to be just right, or audiences will find it unacceptable (especially since it's billed as a premium service), so there's no room for laziness or incompetence in the 3D business.

I think that may also help to explain why 3D TV is failing to gain traction as well. While it fixes the problems I had in the cinema by being completely in control of exhibition, there are few channels available to those interested in 3D (Comcast, for example, has a 3D demo channel and ESPN 3D). Those that do air commercials from whoever's brave enough to advertise on such a slim market. The people who advertise resort to the cheapest possible conversion for their commercials (I saw one where the actors in the ad looked like anorexic zombies because their necks had no depth whatsoever).

That said, this doesn't mean 3D is bad, or that it will ultimately fail. Jurassic Park 3D is doing very well at the box office, and 3D TV purchases are way up this year. This market will survive, but only if it insists on some quality control.

Could a simple protective film be the future of glasses-free 3D viewing?

I didn't come up with that title. I stole it from one of those ad-in-disguise news articles that's been going around Malaysian news sources. There's an example here, but I'll give you the gist of what they're saying, and then answer the question from the headline.

Developed in Singapore by Temasek Polytechnic (TP) and A*STAR's Institute of Materials Research and Engineering over the course of two years, the film, called EyeFly 3D and set to go on sale globally in May, contains 500,000 microscopic lenses that can render static and moving images in stereoscopic 3D, yet it is only 0.1mm thick, making it the same size and thickness as traditional screen protectors.
It won't take a user's existing 2D images and films and convert them magically in front of their eyes, but it can take any content shot in 3D and display it as such, without the need for 3D glasses.

They use a lot of numbers and buzzwords to make what EyeFly is doing sound impressive and science-y. And while what they're doing is cool, they are over-hyping their product just a tad by pretending it's a revolution. This is not "the future of glasses free 3D viewing," but a re-application of old technology we've seen in a number of other products (The Nintendo 3DS is the first to come to mind, but less famous examples include the HTC EVO 3D, the LG Optimus 3D, The Fujifilm W3 Finepix 3D Camera, or the sticker covers on the sleeve of 3D Blu-rays).

This technology is what's known as a parallax barrier, in which precisely cut slits in a screen (or ridges in the case of those 3D stickers that 3D Blu-rays use) are used to direct images individually to your eyes. In the case of EyeFly, they've created a film that you can put over your existing smartphone to allow this kind of image separation. They've also created software for the iPhone that will allow you to transform normal stereoscopic video into an interleaved video format that will allow video to play on your phone after you put this film on it (although they're planning an Android version for the Galaxy S2 as well).

It's a technology that works, and a good way to cheaply upgrade from a 2D to a 3D smartphone. It will present the same problems I've previously highlighted with autostereoscopic displays, including the narrow viewing angle, and halving your phone's resolution. Halving your phone's resolution could also affect the text displayed on it, by the way. I'd have to actually see the product in action to know by how much, but difficulty reading is something to look for in reviews and to take into consideration before buying.

But I digress. The conclusion here is that this isn't revolutionary technology. It's economy class technology that will get the job done if you're looking for a cheap 3D smartphone. It is something I'd look into myself (assuming their software ever supports Nexus line Android smartphones and tablets, or some 3rd party makes universal android software for them), but not something I'd recommend without first seeing it in action. EyeFly looks like an interesting product, but that company is really misinforming the public about why their product is so interesting. That makes me suspicious about their ethics as a company, so my advice is to wait for reviews to know if that product is worth buying. Hopefully they just hired a marketing department that doesn't know what they're marketing.

Why The Focus Convergence Problem Is Actually a Good Thing

I'm referencing an article written by Roger Ebert that heavily references Walter Murch. In that article, the two argue that 3D cannot be a viable format due to the way it changes how your eyes work. It's not something that should be impossible for your eyes to do (it requires using your exterior and interior eye-muscles separately), but it's understandable how that can be stressful to your eye muscles. It's similar work to trying to lift your ring finger when the rest of your fingers are clenched.

Modern stereographers seem to try and solve this dilemma by spending the majority of a movie converging on the object at which you're meant to be looking. Inf films that do this, you're mostly converging and focusing on the same object, so your eye muscles are once again in unison. Frequent 3D cinema goers will therefore naturally look at the focal object, since it's the one that provides the least resistance for their eyes. Once you understand this, and once audiences have been trained to accept this, this makes it one of the most powerful tools that 3D cinema gives a filmmaker.

I noticed this recently when I was watching Hotel Transylvania in 3D. I didn't expect much from the 3D because the director of the film has gone on record saying he only made it 3D because the producers of it forced him to make it 3D. But there was one scene that (probably unintentionally) really stood out.

In the film, Dracula is attempting to plan his daughter's 1,018th birthday, but is being shown up by a human hiker that randomly wandered in and quickly became the life of the party. The scene shows all of the hotel's monster guests crowded around the human hiker, while Dracula is screaming in the background and vying for their attention. When this shot was composed in 3D, the crowd and the human were converged, standing right where the screen was, and Dracula was placed firmly into the positive space.

What interested me about this scene was that Dracula was moving, making noise, and by all means should have been drawing attention to himself. In fact, I tried to pay attention to him, but I found myself experiencing resistance. Because he wasn't converged and the other characters were, I felt like I actually couldn't look at him.

At that point it dawned on me. I can use 3D to hide things in plain sight. While art has typically involved the use of line and color to draw a viewer's eye to a desirable object, 3D has the ability to draw a viewer's eye away from an unintended object. This, of course, works the opposite way for objects coming into the negative space (which may explain why people are under the impression that negative space is the only part that actually makes something 3D), with negative space objects drawing attention away from where the viewer's eye naturally wants to rest.

Admittedly, this isn't a brand new discovery. Alfred Hitchcock already knew about and applied this concept to 3D film in his movie, Dial 'M' for Murder, and I know of at least one other person I've personally met who has noticed this before. Still, it's an interesting concept that is not commonly understood by those inside or outside of the film industry. By creating areas that are actually uncomfortable for your eye, the filmmaker is given complete control to hide or emphasize subjects of a scene. And this can be used to effect some powerful emotions.

For example, let's say you're hypothetically creating a movie about a ninja, and he needs to sneak into his target's abode unseen. You could converge on objects other than him with him in positive space, and he'll feel like a superb expert of his art, since the audience feels compelled not to see him. Conversely, if you want to make him appear vulnerable, you can converge on him with guards in the positive space of the shot, or even shoot with him protruding into the negative space. Suddenly, he's calling attention to himself and not seeing his enemies coming. You worry for him, because you feel him slipping in his stealth.

Or assume you've been tasked with remaking Cipher in the Snow, or some other movie with a protagonist that doesn't draw attention to himself. Telling his story with him in the positive space of the picture will immediately communicate how withdrawn and lonely he is, before a line is even uttered or a gesture made to make him into an emotionally withdrawn character. He's withdrawn emotionally because he appears withdrawn physically.

All this would not be possible if we managed to somehow solve the focus and convergence issue, because objects would no longer cause discomfort in these kinds of shots. You'd be able to stare at Dracula too easily, making him lose his desperation. 2D film is therefore too distracting for certain messages, while 3D makes distraction a non-issue. And to me, that's an amazing quality, not a problem challenging its existence.

So with all due respect to Mr. Ebert and Murch, I disagree with your assessment.

3D Realism

There's an idea that I've been meaning to test (but have never got around to). It has come from a few thing. First, nearly everyone I've interviewed seems to believe that 3D makes an image appear more realistic. Technically speaking, that's not true (there are so many ways that 3D film differs from the way we see 3D objects in real life that I intend to exploit in anything I make), but the fact that people have this sensation is why 3D is generally perceived to only be good for action and horror films (which is ludicrous when you consider the success of films like Hugo and Life of Pi, but I digress).


Second, I realized at one point in my life that my favorite film, Inception, would only work in 2D. This is in no small part because of the Penrose Steps sequences in that film. The Penrose Steps are an impossible staircase that can be drawn in a 2D space to look like they're infinitely descending, because they use inconsistent depth cues to create the illusion of descending. It's impossible to recreate in a 3D space without actually altering the depth. That's the reason it wouldn't work in Inception. Inception was based around existing in a dream world where the logical rules of existence are suspended and replaced with dream logic. The Penrose Steps can exist in this kind of warped logic world because they are impossible but appear possible, which gives the feel that even the impossible is possible that the Inception dream world needs. If converted to 3D, the warping depth needed to create them would both stand out like a sore thumb and call attention to the impossible dream logic, so that it is no longer immersing you in the dream world. 2D doesn't have these limitations, since the Penrose steps can safely exist in a 2D world.

Third, I noticed simply from writing this blog how difficult it is to think about objects in 3D. I don't entirely understand it. I know I watched Finding Nemo in 3D. But every time I think about it, images from a 2D version of that film pop into my mind. It makes thinking about, dissecting, and analyzing stereography require accessing a memory about how I felt about what I was seeing rather than actually recalling the scene composition itself (which is really hard to do, by the way). Eventually I came to the conclusion that this means that while I see things in 3D, I think about things in 2D. I also regularly browse YouTube videos for criticism of 3D to debate, and one of the most mystifying claims people make is that, when they think about it, they don't really remember being wowed by the 3D after the initial wow factor wore off. I take them at their word, but I place more emphasis on the remember factor. I do know from watching 3D that the 3D images never disappear, so I'm led to conclude that people lose the concept of how wonderful a 3D image is because it's only possible to experience 3D, not recall it from memory.

The broader conclusion from all 3 observations is that people react to 3D films in a different way than they do with 2D films. Perhaps this is what people refer to as immersion, but I would describe it differently. In my mind, 3D films make you feel like you're watching something that actually happened, while 2D films make you feel like you're recalling something somebody told you. Put another way, 3D films make you feel like you're actually there for something, while 2D films feel like a memory. This means that 2D films are wholly capable of immersing you in their world, story, etc. But they embed themselves in you like an out of body experience. 3D films don't do this. Rather than feeling displaced in time, 3D makes everything feel current, real, and impactful in a more direct sense than the general human approach that 2D film provides.

To test this, I want to make a 3D film that incorporates both 2D and 3D footage. I wouldn't be the first to do this, but I've never seen anybody else do it this way. I once saw a low budget 3D film from the CMF3D film festival that combined 2D and 3D footage, but I felt like they did it backwards. In their film, called "Collapse," a man takes a drug that causes him to start hallucinating, with the hallucinated world presented as 3D footage (with no other differences). It felt wrong, because the 3D footage felt more real, and his insistence that he was damaged for seeing things this way seemed counter-intuitive and made the less realistic feeling 2D images stick out more than the hallucinations (which were supposed to be the weird part). Instead, I want to make a film that uses 2D to directly convey the idea that some scenes are thoughts, or visions and imaginings. If I'm right, the 2D images will not stick out the way they did in Collapse, but instead flow and create the sensation that you're thinking about things when the 2D hits, and experiencing things when the 3D hits.

Right now, though, this is only a theory. I need people willing to work with me on this experiment, but so many people have been turned off of 3D lately that nobody else seems interested in making this thing. However, if I can find a group to work with me on this, and it turns out I'm right, it could be a very helpful step toward creating a definitive 3D film grammar. Imagine all the questions this could answer about why 3D works or doesn't work in a given title! Imagine how much more useful this would make 3D as a tool for a filmmaker!

If you've ever asked me, I've always said that 3D movies should be shown exclusively in 3D, and 2D movies exclusively in 2D. The reason why is that I have thoughts like this one about 3D and realism vs 2D and imagination that make it seem that certain techniques just don't work in both mediums, and that both 2D and 3D film are both being held back from their full potential by being forcefully shown in both formats. But that's a topic for another day. For now, I'm just thinking about techniques, and wondering if this one I've never seen before would be valuable.

The Value of Negative Space

It's been a while since my last update. I have been working on this blog; I've just scrapped everything I've posted because none of it has been good enough yet. Hopefully this won't be another post like that, because whether it is or not I'm posting it anyway.

First, a little context for what I'm talking about. In 3D film, you have the ability to render objects in front or behind the screen. In order to be able to describe the things in front and behind the screen, stereographers needed a way to address 3D space. They decided to name the space behind the screen, "positive space," because it increases the amount of space between the audience and the object. Conversely, everything that gets closer to the audience, or "pops out" of the screen, is said to be in the "negative space." I bring this up because there's been this trend lately to deem any use of negative space as "gimmicky." But such a blanket statement is too closed-minded. There are so many reasons that negative space can and should be utilized in 3D movies!

Now, I'm not talking about the lowbrow reasons, meant just for the fun of seeing things protrude from the screen that you can then touch. I should clarify. When I say lowbrow, I mean appeal entirely for the novelty. Things like the scene early in Journey to the Center of the Earth where the lead character spits into a sink, just so the water can splatter well in front of the screen. With time, the novelty of this effect fades. The thing is that these shots aren't always inherently meant for the purpose of novelty, and should not be resisted just because they exist.

Not All Popout Was Meant to Be Gimmicky

The first reason I say this is because of the 3D version of Finding Nemo (which not enough people have seen). If you've seen either version, you'll recall that early in the film, there's a scene where Marlin is exhibiting some comical cowardice by over-cautiously exiting and entering the sea anemone where he lives. The shot is framed so that, in the 3D version, Marlin and Nemo swim from out of the anemone and into the negative space. Conventional wisdom of 3D films would suggest that this scene must have then been shot to exploit the novelty of the 3D effect. While it does exploit it (very amusingly, I might add), the idea that it was framed specifically to exploit it is absurd. Finding Nemo was originally created as a 2D film, and then was later re-rendered in 3D as an afterthought to capitalize on the new 3D boom. They never originally intended that scene to be experienced in 3D. Hence, it couldn't have been composed specifically for 3D. It's just the most logical way to frame that scene, whether in 2D or 3D.

3D Framing Effects an Emotion

But more than just negative space occasionally making the most sense, negative space can be used to positive effect (if you'll pardon the pun). First, because negative space literally brings an object closer to the viewer. This enables filmmakers to have a tool in their tool belt that typically has been only available to theatrical productions; position away from or close to the audience. That may seem unimportant at first glance, but how close a person or thing is to you affects your emotions toward them. If they are further away than it seems they should be, they feel emotionally unattached, withdrawn from your world (to use a word that applies both literally and emotionally, they appear distant).  But if they're too close to you, they feel invasive, and you feel uncomfortable.

Let's take, for example, Hugo (a rightly celebrated champion of 3D cinema). There is a scene in which the titular character, Hugo, is confronting a grumpy toy maker to retrieve a journal that the toy maker had taken away. He instead finds that the toy maker has burned his journal. To compose the emotion of this scene in 3D, Martin Scorsese decided it would be useful to show a scene where Hugo is holding his ashen journal out into the screen, the camera facing upward, and the ashes floating into the negative space. The ashes come extremely close to the viewer (uncomfortably so). This sense of discomfort is then subconsciously transferred back onto Hugo, with whom we empathize at the loss of his journal. It's a beautifully composed scene, which doesn't work as well in 2D, and most certainly wouldn't be possible if 3D were forbidden to utilize negative space, as so many critics are calling for in their haste to declare 3D a useless gimmick.

Involving the Audience

Another valuable use of negative space is the mere fact that negative space is the area that the audience inhabits. Hence, unless a floating frame is negating that effect, using the negative space is a way to invite the audience to explore their own role in the events unfolding on screen. I discovered this when playing the video game, Batman: Arkham City. There's a scene in which Batman confronts Mr. Freeze about creating an antidote to a poison that the Joker used to poison Batman. Mr. Freeze decided to use antidote as leverage to force Batman into serving his own goals. Mr. Freeze crescendos his power over Batman by dramatically holding the antidote in front of him while he crushes it. The 3D version frames this so that Mr. Feeze's hand and the antidote reach in front of the screen. The first time I played, I subconsciously decided to try and reach for the antidote, as taking the role of Batman in this scene made me feel desperate. But I wasn't quick enough. Mr. Freeze crushed the antidote between my fingers. I played again at a later date and was a little quicker that time. Even then, however, my fingers passed right through the vial. Either way the message was clear. Despite my best efforts, that vial was unobtainable. I was weaker than Freeze, and that makes me desperate.

Not every situation is identical to the one in Arkham City, but there are plenty of situations in which a filmmaker may want to include the audience in their film world by utilizing negative space. Perhaps they want to create a similar helplessness by showing failure to grab an outstretched hand in negative space. Perhaps they want to invite you into the magic of a birthday cake's candles by holding it close enough for you to blow them out yourself. It all depends on the script. But the point is filmmakers and critics do film a disservice by immediately declaring all use of objects protruding into negative space to be gimmicky. Negative space is potent when applied judiciously. And, to be honest, I'd like to see more of it.

Glasses No Longer Needed

This is just a quick update. You remember how in a previous post I said glasses free 3D TVs hadn't caught on because you could only view them from one angle? Well, I neglected to mention Lenticular displays. See, there's this type of display that can do glasses free 3D from any angle, but it requires a lot more information that two video images, and is not compatible with current dual video films. However, it is now more accurate to say that it was incompatible.

You see, at CES 2013, Dolby demoed a brand new 4k resolution (higher resolution than most theaters currently show movies) lenticular 3D TV that uses internal software to convert stereoscopic inputs into lenticular outputs. In case you haven't been following along with my geek speak, that means there now exists a theater quality 3D TV that you don't need glasses to watch.

It'll probably still feel weird to people with binocular vision problems, but this is really exciting. Dolby have just solved two of the three problems I hear most often when people criticize 3D: The need to wear glasses, and the dimming of the image. Headaches for people with binocular vision deficiencies could still be a problem, but all-in-all, this may be the exact thing 3D needs to make it mainstream (assuming it's reasonably priced).

Go Read This Other Post

I just did a guest post on another blog. This is it. Go read that, please. It might as well be posted here.