Saturday, June 21, 2014

How's the 3D in Edge Of Tomorrow?


Background:

Following hot on the heels of Top Gun's 3D re-release, Edge Of Tomorrow gives audiences another change to see Tom Cruise's million-watt smile in 3D. The film is based on a Japanese novel called All You Need Is Kill, though it deviates somewhat from the story's plot. The tale itself is a hodgepodge between Groundhog Day, Looper and director Doug Liman's own The Bourne Identity, while also calling to mind the rewards-based repetition that modern video games offer their players. More importantly than all that though, how is the 3D in Edge Of Tomorrow? Read on in our 3D-focused review to find out all you need to know!

Post-Converted 3D:

Edge Of Tomorrow was post-converted from 2D to 3D by Prime Focus World. Their approach to this film seems to be unique in 2014's crowded slate of post-converted films. Rather than hide the technique with a 'native lite' approach, they appear to have been directed to dial up the stereo at pivotal moments and dial it down in others. The beginning sequence is seemingly intentionally dull, before the visuals are dramatically opened up in a chaotic battle scene. We've seen visual 'kicks' before (think of the difference in The Wizard Of Oz between Kansas' monochromatic colours and the Technicolor world of Oz) but this is the first time we can recall one being initiated by a post-conversion team so strikingly.

The beginning sequence of Edge Of Tomorrow has 'flat' feeling 3D

The key to their success appears to be, perversely, how much of the opening sequence is kept close to its native 2D source. This lulls the audience's eyes into a relaxed state, before the post-conversion team shockingly uses scaled negative parallax to literally yo-yo soldiers in and out of the frame. It's a refreshingly brazen technique, and serves the narrative well; our anti-hero is being thrown from a comfortable world into a stark and gruesome reality. The shock is appropriate, and not as gaudy as it could have been in other contexts. The post-conversion team served the narrative well here, and it's likely the shock they were able to convey could not have been achieved if the film were shot in native 3D.


How's the depth of the 3D?

The depth varies wildly throughout the film. The battle scenes opt for a 'deep focus' that allows you to see for miles into the distance. Or, at least, you could if the camera kept still for long enough. The battle scenes are the most interesting visually, in that they provide an arresting focal point for each of Tom Cruise's character's (Cage) loops. One character dies as an object drops from the sky on him, and then in Cage's next iteration through the loop Cage saves the man before the object drops, and then the next loop Cage doesn't bother. Knowing the exact geography of the scene is important for Cage, and it's incredibly important the audience knows the impact of a character standing even a metre from their 'usual' position.


The film-makers attempt to show Cage's growing confidence grow with each new iteration. The camerawork is shaky and chaotic in his first few attempts at battle, and then slowly grows more steady and assured as he begins to learn the ropes. Unfortunately, this also means that the audiences eyes are expected to context-shift with each new iteration of Cage's battle. Your perception shifts from loop to loop; progressing from only seeing a few feet of depth behind Cage in his first loop, through to being eventually able to see for miles behind him as his skills develop. After enough rapid-fire edits and changes to the way the action is filmed, your eyes might wary and tire from all this visual change.


Does the 3D 'pop'?

As mentioned earlier, the opening battle scene is the primary example of negative parallax being used. Soldiers are thrown from a moving plane into a battle field below them, and some are accidentally suspended to the plane by rope. As the plane descends, they're flung at the audience in a yo-yo like technique that we wished had been included in The Amazing Spider-Man 2. The remainder of the film uses this type of technique relatively rarely, with the only 'popping' elements usually being restricted to debris and out of focus guns aimed at the camera. Again, this kind of inconsistency can contribute to some amount of eye strain, as audiences eyes attempt to process the inconsistently applied visuals.

Dull night-time visuals, with a lack of background contrast

Did it make sense to add 3D to Edge Of Tomorrow?

The film-makers made a lot of good choices when considering the idea of Edge Of Tomorrow as a 3D film. The action is largely set in daylight, and the characters contrast well against their backgrounds. However, they made a few missteps once the second act is over. The climactic set piece is largely set in the evening, and it's very hard to differentiate the dark-suited soldiers from their background. Their enemy is cloaked in shadow and similarly hard to pick out visually. Without the kind of bright neon offered in films like Prometheus or Tron: Legacy, these night-time scenes suffer visually, particularly if your local multiplex hasn't adjusted the brightness of their projector to accommodate 3D projection's light loss. Just in case you hadn't noticed a recurring theme in our review; this narrative and visual choice can also lead to eye strain in audiences.


The film itself

Edge Of Tomorrow is one of the most arresting and relentlessly entertaining blockbusters to have been released in recent years. Simply put, it's a gem of a sci-fi/action film. If you're a fan of shoot 'em up video games, this is the film for you. If you're a fan of the enigmatic and interesting stars, then this is further proof that they're A-list talent. If you've ever bemoaned the influx of comic book movies and sequels in the American Summer movie-going season, then you'll find Edge Of Tomorrow's originality refreshing. If you ever watched an action film and wished the female lead were tougher and more heroic than the male, then Emily Blunt's steely performance is going to make you feel better about Hollywood. It's a joy of a movie, let down only by a somewhat anti-climactic final minute of footage. If there were Oscars for crowd-pleasing films, this would be nominated in several categories.

If we had to archive one version, should we save the 2D or the 3D?

If you have the choice, see Edge Of Tomorrow in 2D. Buy the 2D Blu-Ray. Avoid the 3D version unless you're really curious. It's disappointing for us to say, but the 3D version is a noble failure. There's a bunch of great examples of stereo on display, but the eyestrain we suffered while watching can be traced back to several poor visual choices that other film-makers would do well to learn from. Just because you can post-convert from 2D to 3D, doesn't mean you should.

Sunday, June 15, 2014

How's the 3D in X-Men: Days of Future Past?

 


Background:

It could be argued that the first X-Men (2000) film began the modern-day phenomena of blockbuster superhero franchises. 4 films and 2 spin-offs later into this series, X-Men: Days of Future Past arrives in 3D, directed again by the series' original helmer, Bryan Singer. This is not the first time we've seen Wolverine and his pals in 3D, so we've a lot of precedent to judge this effort against. How's the 3D in X-Men: Days Of Future Past? Read on to find out!

Native and Post-converted 3D

X-Men: Days Of Future Past is a hybrid stereo film. Interior and dialogue scenes were largely shot in native 3D. Several key sequences were post-converted sequences from a mono source. Other shots are largely CGI in nature, and the responsible effects houses were able to natively render their footage to stereo. It's therefore quite a confusing film to critique. To do the job properly, one would need a shot list or production schedule. We'll do our best though.


As a rule of thumb, most "first unit" sequences (character beats that would typically be directed by the director himself, and featuring the A-list stars) set indoors are clearly shot in native 3D. An observant audience member will be able to tell the difference. The trick is to look for subtle details; strands of hair, the length of an actor's nose, or even just compositional details like the layers of props on a table. In these scenes, it's interesting how at ease your eyes become. The camera is often locked down, actors are interacting in multiple panes of depth, and the stereo feels completely natural. You're there, say in Professor Xavier's mansion, with the other X-Men. These sequences were shot on ALEXA Ms.

Credit: http://www.arri.com/news/alexa-xt-on-x-men/
"Second unit" sequences (eg: the type of moment that doesn't usually require an expensive star, such as a shot where a hand picks up an object) are often show in 2D, and later post-converted to 3D. The same is true of large sequences that are shot on older film stock, using cameras from the 1970s, where the footage was later converted to a subtle form of stereo too. The conversion is handled well, and there are no glaring errors in the added depth. Aside from having marginally more eye strain, most audience members won't be able to tell the difference. What's interesting though, is the 'feeling' one gets from this rapid-fire transition. According to editor John Ottman, he edited the film in 2D, while the mono footage was being converted to stereo, because the cuts between the two formats were so regular and jarring that he couldn't process the visuals consistently. While audiences are thankfully spared from experiencing this, we at 3Defence definitely came away with a feeling that the film lacked a stereoscopic 'continuity'. Post-conversions are great, until they're compared to native 3D in the same sequence. These sequences were shot on ALEXA XT's.

Of course, as usual with most Hollywood blockbusters, there are also shots in the film that are 'rendered 3D'. In shots where the majority of the action is a composite of shots with digital characters (for example, a few shots of Sentinels in the beginning sequence), it's more than likely that the effects studios in charge of the shots provided the stereo rendered 'natively'. So, X-Men: Days Of Future Past is a hybrid of techniques.

How's the depth?

As mentioned above, the visual continuity is inconsistent throughout X-Men: Days Of Future Past. That is due to a number of reasons, in addition to the differing methods of filming. For one thing, large sections of the film's schedule were aggressively organised around the ensemble cast's availability. For another, the film went through several rewrites (even during production), major edits, and even rearranged which actors were part of key scenes.


So, ignoring the inconsistency of depth used in the film, we can look at a few shining moments where depth is used really creatively. The stand-out sequence, both from an entertainment perspective and a technical perspective, is a prolonged slow-motion one where Quicksilver single-handedly disarms and defeats several foes. The character's power is to be able to "move and think at superhuman speeds." Consequently, like The Amazing Spider-Man 2 before it, we're offered another version of a bullet-time like sequence, where a character is able to interact with a chaotic environment where objects are suspended in mid-air. Aside from being hilarious, the sequence allows the character dozens of objects to interact with and run around, on various panes of depth. It's an engaging sequence visually, and single-handedly worth the price of admission.

 
Another fantastic sequence is provided in the way Blink's powers are used in the film's first big action set-piece. She can project an entry and exit portal, that allows characters to jump through a kind of wormhole. In 3D, we get to see characters leap from one side of the room to the other, to startling effect. Just as your eye learns where a character is in the scene, the character is transported to a different spacial plane, forcing your eye to try and catch up. It's a power visual, that is used well in the beginning and end sequences of the film.


Does the 3D 'pop'?

There are a few instances where debris, water, dust and other particle-based elements come very close to the edge of the screen. Generally the film treats the screen's boundary as a safe barrier between the audience and the action. If you're after yo-yo styled pop-out effects, X-Men: Days Of Future Past is not the 3D film for you.

Did it make sense to add 3D to X-Men: Days Of Future Past?

In theory, it makes sense to add 3D to any X-Men movie. Their characters are colourful, with a variety of interesting powers, and in Days Of Future Past they interact across time and space. In practice though, this film spends a lot of time in a murky and dark post-apocalyptic future, and in 3D these scenes are distractingly gloomy. 3D projection's light-loss was evident in many sequences throughout Days Of Future Past, and we didn't feel the film-makers did enough to counteract this. Films like Prometheus and Tron: Legacy have worked around this issue with bright neon-flavoured yellows and blues. Days Of Future Past's 'fire-light' orange didn't have the same effect, and made the image softer than we'd like.

The film itself

X-Men: Days Of Future Past is a great film... provided you have watched at least four key films from the series beforehand. It's assumed at this point you know the key characters' powers, without any real introduction or back-story. Most characters get an arc of some description, though the impact of that arc is more meaningful if you've seen the other films the arc relates to. In that sense, this is a perfect comic book movie. We wouldn't advise casual viewers check this out, unless they're prepared to do their homework ahead of time! If you have done your homework, you'll be rewarded for as much as effort as you were prepared to put in. The complexity of the lore on offer here is getting seriously dense, and it's a treat to see your favourite characters navigate unfamiliar terrain in the way we see here. At least as of our first viewing, this film vies with X-Men 2 for the title of 'best in the series'.

Should we archive the 2D or the 3D version?

Archive the 3D version, if only for the beauty and hilarity of the Quicksilver heist sequence. Seeing that in 3D is worth it, if you were umming and ahhing about paying the extra dollar for premium 3D ticket prices. There are parts of X-Men: Days of Future Past we weren't fans of, from a stereo perspective, but that one sequence is going to go down as one of the 'all time great' 3D moments.

Sunday, June 8, 2014

How's The 3D in Godzilla?

Background:

Godzilla has been around for 50 years. He's been in almost as many movies. 2014's Godzilla is Hollywood's second attempt at bringing the monster to life. Thankfully, this version was guided to the big screen by Gareth Edwards, whose 2010 film Monsters went a long way with a very low budget. His skills behind the camera have been recognised already, with Disney shoulder-tapping him to helm the next stand-alone Star Wars film. So, how did his transition from micro-budget to massive-budget film-making go? More importantly, how is the 3D in Godzilla?


 Post-Converted 3D:


Godzilla was shot in 2D on Arri Alexa Studio 4:3 cameras, using Panavision C Series anamorphic lenses. Effects studios like Weta Digital, Moving Picture Co., and Double Negative provided the CGI, and The Avengers' cinematographer Seamus McGarvey lensed the action. The film was later post-converted to 3D by two studios, Vancouver-based Gener8 and Stereo D. Their work here is subtle, but effective. Where Godzilla's post-conversion really shines is in the interior dialogue scenes. Rather than flatten out the stereo in such scenes, the stereo compositors add several layers of depth. For example, a doorway might frame a shot, with a concerned child layered in front of it, while a distraught parent paces several feet away in an adjoining room. In fact, doorways are used repeatedly to block characters from one another, and the motif afforded the post-conversion team a great way to separate actors from one another in three dimensions. A lot of the film has a naturalistic 'hand-held' approach, which can't have made the post-conversion easy in shots like this (the doorways would be constantly moving, which requires a lot of rotoscoping). The teams did a great job.

Bryan Cranston's character, behind a door

How's the depth?

Godzilla is the second big-budget 3D monster movie we've seen from Warner Brothers & Legendary Pictures. The difference in approach between Guillermo Del Toro's Pacific Rim and Gareth Edwards' is interesting. It's hard to know if Edwards was explicitly guided by the studio to avoid a similar aesthetic to Guillermo Del Toro's picture, or if he just followed his nose towards a different approach. Perhaps answering that question, Legendary employed their own 'stereoscopic consultant' (though that might have been a supervisory role between the two conversion studios). Del Toro was at pains to show the scale of his monsters, from the monsters' perspective; rearing above skyscrapers, shot from phenomenal heights.


Edwards instead shoots his monsters from the perspective of the panicked humans below them. He communicates this scale in several interesting ways. He places humans on top of tall objects like skyscrapers or bridges. He throws humans out of planes, wearing parachutes. He has awestruck humans at ground-level, craning their heads to see the monsters towering above them. He has humans descend into deep caves, and lean over the edge of craters. The effect achieved is that we feel engaged in each shot. The depth of field is appropriately massive, but with a human in the near foreground and a disaster in the distant background we are invited to feel 'like we were there', and wonder 'what would I do in this situation?'


Does the 3D 'pop'?

There are several inventive usages of negative parallax in Godzilla. The title sequence is a series of military records being redacted, over the top of World War 2-era footage that has been aged and scuffed. The redaction process reveals the names of actors and crew members, and occurs in the deep foreground, in front of the action below. Eventually, the title card of 'GODZILLA' disperses into ashes, and the debris floats out towards the audience. It's a nice way of breaking the 'wall' of the cinema screen, without also breaking the audience's attention from the narrative of the film.


There are other memorable moments where the 3D 'pops'. In the crowded theatre 3Defence saw Godzilla, several audience members ducked and flinched when objects hurled towards the screen. As buildings are toppled, a cloud of dust and debris inevitably follows, roaring from several hundred feet of the way to eventually engulf the audience and their movie theatre. This effect is used to particularly strong effect in a nuclear power plant's meltdown, where a group of scientists attempt to outrun a cloud of radioactive material.


Did it make sense to add 3D to Godzilla?

A post-converted movie, filmed hand-held, set largely in evening scenes, where the monsters are often off-camera? No, it didn't make much sense to convert this film into 3D. This film challenges a lot of our traditional definitions of what is 'appropriate' for 3D distribution. We're glad that the studio took the risk though, and that Gener8/Stereo D pulled off such interesting interior scenes.

The film itself

Your reaction to Godzilla will depend on the history you bring to the theatre with you. If you're a longtime fan, you'll spend the first half of the movie frustrated you're not given much monster action, but then love the last 30 minutes. If you're a casual fan, you'll chuckle at a few references and generally have a good time. If you're a novice to the world of Godzilla, the performances of the humans might grate, and the episodic nature of the film might wear at your attention span. Hopefully you're somewhere in the middle of these perspectives, but it's not likely that any one person will be entirely satisfied by their experience. We hope there's a sequel, and that it addresses a few of these concerns.


Should we archive the 2D or the 3D version?

This is a tricky question. 3Defence is going to side with the 3D version in this case, but after much consternation. The literal framing devices used - where doorways or humans are in the close foreground, separating us from the action far away - demand to be watched in stereo. In 2D, they look like the action is being obstructed. In 3D, they look purposeful, giving us a sense of scale. The actors have been positioned, just like they were in the original Godzilla films of olde, as proxies for ourselves. In 3D, this effect works. In 2D, it's annoying. See the 3D version if you can, on the biggest screen you can find.

Saturday, June 7, 2014

How's the 3D in 'The Amazing Spider-Man 2'?

Background

The fifth Spider-Man film in 15 years, The Amazing Spider-Man 2 is also Spidey’s second 3D film. We were reasonably underwhelmed by the 'original' The Amazing Spider-Man, despite it being shot in native 3D. How does the most recent sequel fare in comparison? Read on for all the details on the film’s post-converted 3D, its depth, effects and whether or not we should save the 2D or 3D versions for archival purposes.

Post-Converted 3D

Unlike its immediate predecessor, The Amazing Spider-Man 2 was largely shot on film, and post-converted to 3D. This task fell to Legend 3D, and Stereo Supervisor Ed W. Marsh. The company has a mixed record with its stereo conversions. Some of their earlier work left much to be desired (eg: Alice In Wonderland and Green Hornet). Like much of the conversion industry, they've upped their game in recent years (eg: Top Gun, Man Of Steel). The Amazing Spider-Man 2 is another  massive step forward for them; average viewers won't realise this film was post-converted 3D.

A film, made on film!
That's a misleading sentence though, because much of the film is not 'post-converted' in the truest sense. Due to the web-slinging nature of the film, large sequences of the film feature hundreds of composite effects shots, most of which would have been digitally rendered directly to stereo by the effects studio in charge. Consequently, most of the action scenes you see in The Amazing Spider-Man 2 are 'rendered 3D', and are therefore as 'native' as the computer generated character they feature.


The movie is a warm looking one, perhaps afforded a softer image by filming on film. Of course, even in 2D you can tell the difference between a digitally shot image and one filmed on celluloid. We mention the film's warmth here primarily because it makes for a nice change, amongst many of its contemporaries. The scenes with Peter & Gwen are radiant, in a way that is as much about the leads' performances as the technical goings-on behind the camera.


 How's the depth of the 3D?


The Amazing Spider-Man from a few years ago had a depth problem. The action scenes were vibrant, but the dialogue scenes were not. Its sequel has learned a few things from this. Half of the dialogue scenes are just as boring as they were last time around, but the other half are staged in locations that emphasize depth. For example, Peter Parker catches up with a - rich - old school friend, and they spend a few minutes talking in a room, separated by an oppressively large stairwell. They then go to a river's side to continue their discussion, which is staged fairly traditionally. But the boys then begin a stone-throwing competition. Each stone thrown goes further and further, and eventually Peter's powers let rip and his stone is seen skipping across a hundred foot of water. These are slights of the hand, allowing for an engaging image while also getting through truckloads of exposition nad 'character development'.


Then, of course, there's the action scenes. When 'Spidey Sense' is triggered, the viewer is allowed to see the world as Spider-Man sees it; the camera can roam in and out of space and time to focus on what's important. As time slows down, and the camera speeds up, the action takes on a balletic quality. This is the most nimble and agile Spider-Man we've seen on screen yet, and his contortionist nature is all the more miraculous in 'bullet-time'. These scenes frequently use focus, color and camera movements to convey geography and choreography. For example (this requires bullet-points to communicate the progression, sorry):
  • Spider-Man might begin a shot in mid-air in the foreground,
  • the camera then whips over to a falling pedestrian, 
  • the camera zooms in on something that endangers that pedestrian (eg: an electrified handrail),
  • the focus then is pulled again and we see Spider-Man, now in the background, engineer his web slinging perfectly,
  • his web then reaches out and prevents a death in the foreground.
This type of sequence repeats itself in key moments throughout the film. Each iteration shows more and more variety of depth, building towards a pivotal showdown between the hero and his adversaries. The showdown, high above the ground, uses stereo-emphasised depth to communicate the peril our hero faces. When a web misses its target, we feel as pained as the hero does. The Amazing Spider-Man 2 uses the bullet-time technique to make for a more engaging 3D image, while enhancing the narrative too. The slow motion allows our eyes to take a rest from camera blur and post-converted characters, and when 'real time' comes back the shock of reality is sometimes made intentionally jarring. Its an overtly manipulative technique, and it works well in 3D.

Does the 3D 'pop'?

The 3D in The Amazing Spider-Man 2 is largely in keeping with the modern stereo aesthetic; aggressive negative parallax is used sparingly. This seems like a missed opportunity. Electro's sparks could certainly have flown into the audience, and Spider-Man could have swung over our heads a few times. That gimmick was essentially what made people pay to see Spider-Man: Turn Off The Dark and it would work well in this series too. It's not fashionable to have this opinion, but in 3Defence's view, this series has a 'dork' of a main character, and it'd be in keeping with his nature to make a visual joke or two using negative parallax.

Did it make sense to add 3D to The Amazing Spider-Man 2?

It absolutely made sense to distribute The Amazing Spider-Man 2 in 3D. The primary villain is bright blue neon coloured, and he fights with startling blue blasts of electricity that illuminate any night time scene. The primary hero is primary coloured himself, and - when he's not battling Electro - is often shot in daylight, in broad and open exteriors. When he's web slinging, there's a deliberately vertigo inducing quality, as we focus on our hero in the foreground while a busy background of skyscrapers rapidly pass behind him. This is the most '3D appropriate' entry in the series so far.

The film itself

The Amazing Spider-Man 2's tagline was "his greatest battle begins". The issue 3Defence has with this film is that "his greatest battle" should have been split into more movies. There's so much going on in The Amazing Spider-Man 2 - much of it interesting and of worth - that the film feels rushed and chaotic. Perhaps if an extra half hour were added to its running time, it would have allowed things to settle. Perhaps rogue story elements like Peter Parker's parents' research could have been eliminated to free up more character development. In any case, the film is highly engaging, it just feels like a second draft; bloated with great ideas and poorly cut down for coherency. The "greatest" thing we could say for the film is that it is the closest we've come since Sam Raimi's own Spider-Man 2 to perfect web-slinging action. If you've ever read a Spidey comic, you owe it to yourself to check the film out, just for that alone.

If we had to archive one version, should we save the 2D or the 3D?

The 3D version of The Amazing Spider-Man 2 is the definitive version to archive. Unlike its predecessor, the film-makers made great decisions here, and designed an engaging 3D experience that you'll remember long after you leave the theatre. The depth of field afforded, and emphasised, by stereo gives the audience a reason to watch the film with glasses on. Legend3D should be proud of their work.