From David:

While I was at Vancouver, I didn’t see Screen International for Sept 15, but I’ve caught up since I got back. SI reports extensively on Bong Joon-ho’s The Host. Budgeted at $11 million, it has become the top-grossing Korean film of all time (in unadjusted dollars). It was released in July, and 38 days later it had 12.37 million admissions in its domestic market, beating The King and the Clown. The subsequent Asian rollout has yielded a total gross of $77.8 million as of 15 Sept. The Host will open in Europe in November, and in the US, distributed by Magnolia, in late January 2007.

Mr Bong was kind enough to give me a CD of the soundtrack, and it’s a lively and varied score. I recommend it. When we first met in 1995, he told me that Film Art was widely pirated in Korea, so at Vancouver I gave him a copy of The Way Hollywood Tells It. This is the source of his comments in the inscription: “To David Bordwell! Thanks for your amazing book! (My legal version copy!) Good luck!” Maybe some of his luck with The Host will rub off on me….

Kristin here–

I have finally caught up with Cars, the new Pixar animated film. Not brand new, exactly, but it’s still playing second run on the big screen here in Madison. Good thing, too, because it would be difficult to appreciate its technical virtuosity on DVD. (It’s due out on DVD on November 7.)

For me, part of the fun of watching a Pixar’s film is to try and figure out what technical challenge the filmmakers have set themselves this time. Every film pushes the limits of computer animation in one major area, so that the studio has been perpetually on the cutting edge. In Cars, that area is light and reflections. The comic scene of Tow Mater running around backwards has a breathtakingly flashy effect, literally, when he runs into a forest and can be tracked only by the rapid bursts of light that come through the trees.

The reflections are dazzling at times. By choosing highly polished cars and trucks as characters, the filmmakers forced themselves to devise ways of showing light realistically bouncing off their painted surfaces. This happens in virtually every scene, but the moment when the refurbished town of Radiator Springs turns on its array of neon lights in the evening is a real tour de force. The vehicles parade up and down the main street, and the reflections run over their surfaces from every side. (This segment and the design of the town’s drive-in restaurant irresistibly recall the appealing look of American Graffiti.)

Cars builds on the methodical technical progress Pixar has made over the past decade.
Perhaps the greatest technical challenge in this kind of animation comes in “rendering,” or adding surface texture and color to images. In the early 1990s, Pixar invented RenderMan, a program that made a huge leap forward in the sophistication of this process. It was used for the dinosaurs in Jurassic Park (1993), which included relatively few shots because rendering was so time-consuming and complicated. RenderMan has since become one of the most basic tools for creating CGI (computer-generated imagery), and it can be seen among the technical credits of almost any big effects-heavy film, including The Lord of the Rings (2001-2003) and Pirates of the Caribbean: Dead Man’s Chest (2006).

Many versions of RenderMan have come out since its invention, and the studio’s animated features have been the driving force behind its progress–though its short films also provide early testing grounds for new developments. In 1995, the studio released the first feature-length movie made entirely with CGI (computer-generated imagery), Toy Story. At that point, rendering anything beyond colored, smooth surfaces was impossible. Toy Story revolved around toys precisely because they could look reasonably realistic despite such limitations. The challenge then was simply to make a full-length film with CGI and to make it an absorbing, amusing story.

Objects with more complicated surfaces, especially composed of many tiny objects moving independently but alongside each other, required technical innovations. In a bug’s life (1998), it was realistic grass. Monsters, Inc. (2001) went a step further and created believable fur. (Final Fantasy: The Spirits Within came out that same year, touting its creation of realistically moving human hair, but it lagged behind the sophistication of Pixar’s big fuzzy blue monster.)

Substances that move in complicated and random ways have always been tough to animate—especially water and fire. Disney’s 1940s features, which were of course drawn animation, were great partly because the studio had the resources to conjure up realistic water (for the sea scenes in Pinocchio, 1940) and fire (in Bambi, 1942). Pixar pushed RenderMan to create extraordinary water effects in Finding Nemo (2003).

The Incredibles (2004) didn’t focus on one single challenge in the way that most Pixar features do, but its main accomplishment was to create a strong 3D look to the sets and characters while finding stylized designs for the first human cast to populate one of the studio’s features.

By the way, the surface that had remained the most difficult to simulate realistically using CGI—human skin—was finally achieved by two other companies. One was ILM (George Lucas’ special-effects company Industrial Light & Magic) when it created the infamous Jar Jar Binks in Star Wars Episode I: The Phantom Menace (1999). The other was Weta Digital, which animated Gollum in Rings.

In Cars, the reflections in the distinctive surfaces of painted cars was enabled by another upgrading of the RenderMan system, adding a ray-tracing capability—a capability that also assisted in creating realistic shadows and other tricks of lighting. (For a discussion of many aspects of the making of Cars, check here.)

Overall, of course, Pixar’s uninterrupted streak of hit features stems from the fact that all this technology is put in the service of smart, funny, well-constructed stories. I’ve seen some reviews suggesting that Cars isn’t quite as amusing or engrossing as many previous Pixar films, but I don’t think it suffers at all in comparison. There are so many puns, both verbal and visual, that one has to be very alert to notice them. There’s a running gag about a naive car being overjoyed when the famous racecars keep calling him by name—not remembering that he’s sporting a personalized license plate reading “Fred.” Every time that happens, it gets whisked over so quickly and in such action-packed compositions that it would be easy to miss.

Another thing that struck me about Cars, and this has nothing to do with the technology used, is the extraordinary stylistic differences between the two main environments in which Lightning McQueen, our hero, finds himself. The racing-world scenes are, predictably, fast and lively: very quick editing, the hero’s visions of the crowd, superimpositions, camera movements, and loud, loud sound create a hectic pace. The sweeping desert landscapes and sleepy little town, on the other hand, have gentle music and sound effects, a much slower cutting pace, and a general leisureliness. Yet the result did not bore me in the least, for the design of the surroundings and the group of eccentric vehicles that “people” Radiator Springs provide a different sort of enjoyment—and indeed a bit of relaxation after the visual and sonic bombardment of the opening. Other films have contrasted different setting by using stylistic techniques, but I can’t think of one where the gap between them is so broad.

I note that the supplements listed for the DVD being released in a month don’t include any real making-of documentaries. Presumably a special edition will come out later that will feature some, and maybe then we’ll get to witness some of the technique behind all those lights and reflections. In the meantime, if you haven’t seen Cars and live someplace where it’s still in a theater, give it a try.

More from Vancouver from DB:

Bong Joon-ho, one of the most talented Korean directors working now, has in a few years proven himself adept in many genres. His first feature was the charming comedy Barking Dogs Never Bite (2000), and he followed that with one of the best recent cop movies I know, Memories of Murder (2003). Now The Host has broken Korean box-office records and won tremendous praise at Cannes last spring.

Naturally, there’s been a buildup of interest for the three screenings of The Host scheduled during the festival. The opening one is sold out, and the others are nearly full too. This morning there was a Forum discussion with Bong, moderated by Tony Rayns, and it proved to be a very interesting conversation.

Bong had wanted to make a monster movie ever since his childhood, when he looked out his window at the Han river and imagined a creature like the Loch Ness monster rising out of it. Eventually he was able to summon up the money to do so. The budget for The Host ran to about $11 million US, nearly half of which was used on special effects. (Bong points out that the ordinary Korean film is budgeted at about what he spent on fx.) He contracted the CGI out to several firms, including Peter Jackson’s Weta Digital and The Orphanage, a San Francisco company.

To save money Bong cut several monster shots, instead simply suggesting the creature’s presence through other means. He was inspired by Spielberg’s handling of Bruce the shark in Jaws: faced with Bruce’s constant mechanical failures, Spielberg used point-of-view shots and Williams’ score to signal when the shark was nearby.

Will there be a Host 2? Bong says that if there is, he wouldn’t be directing it. He envisions the possibility of something like the Alien series, where different directors turn each installment in different directions.

Tony Rayns set up the context for the discussion with his usual aplomb, and the audience coaxed Bong into wider comments. One listener asked what makes Korean audiences so eager to support their local movies. Tony pointed out that Korea has the most cosmopolitan and film-loving population in Asia, and Bong talked of the expansion of the market, with The Host going out on more than 600 screens. Also, Tony added, Korean movies tend to be very good.

I can’t refrain from a personal note. Bong greeted me warmly, and he reminded me that we met in Hong Kong in 1995, when his breakout short, Incoherence, screened at that festival. At that time he told me that he had read the (pirate) Korean translation of Film Art: An Introduction. I was happy that our book might have contributed a little toward his film career, and he cheerfully autographed my Vancouver catalogue with a little tribute to the textbook. Sometimes I forget that film researchers can affect filmmakers.

I was pleasantly reminded again at tonight’s reception. There I met Reg Harkema, an Ontario director, who became obsessed with Film Art‘s discussion of La Chinoise and nondiegetic inserts….so much so that Monkey Warfare, his film in this festival, is full of them! (Have to catch that.) And Ho Yuhang, director of the Malaysian movie Rain Dogs, knew Film Art but was more interested in my Ozu book. His autograph in my catalogue reads, “My friend, that f*cker, bought the only Ozu book left in the store. Damn!” Yuhang is also a big fan of film noir and he’s now scripting a crime movie.

Apart from hobnobbing with directors, I saw the disturbing docu Rampage, about black family life in one of Miami’s most poverty-plagued neighborhoods. All the young men, including one serving in Iraq, want to be rappers, and the most talented of all is only 14. But can he and his brothers survive gang warfare? I thought that the editing and sound were a little too aggressive, even somewhat sensationalistic, but as the film goes along it raises very tough issues concerning filmmakers’ ethical responsibilities. The question of whether the presence of a film crew changes the situation it’s filming is brought to the surface with really unsettling results.

I ended the day with pure fun. Tokyo Loop is a string of animated shorts, in varying styles, all aiming to comment on life in Japan’s metropolis. I’ve long thought that animated filmmakers don’t get enough credit, because we forget that they have to acquire an enormous understanding of how creatures and objects move. I was reminded of this again in seeing Tokyo Strut, a record of human and animal movement conveyed solely by dots of light, and the very funny Dog & Bone. Other filmmakers record movement, but animators have to know how to create it.

Finally, a greeting to Marlene Yuen and Ted Tozer, who despite my best efforts to hide my operating tactics, spotted me counting shots in screenings at Vancouver last year. Ted, if this interests you, check the CineMetrics website on the first page of this site.

In Film Art: An Introduction (Chapters 1-3) we argue that a movie’s form engages the viewer actively. As a result, we try to show how formal choices can shape the viewer’s response. For instance, a filmmaker who wants to tell a story tries to arouse curiosity, suspense, and surprise (along with other emotions, of course), and narration–the flow of story information–helps him or her do this (Chapter 3). A nice confirmation of this point is offered in Christine Vachon’s new book, A Killer Life: How an Independent Film Producer Survives Deals and Disasters in Hollywood and Beyond (Simon and Schuster).

In the script of One Hour Photo, director Mark Romanek carefully set up an opening stretch that slowly built suspense about Sy, the photo clerk who will take an unhealthy interest in his customers. But then came the studio marketers, who hinted at the film’s premise in advertising. As a result, preview audiences knew a little more than they were supposed to about where the story was going. In test screenings, the first act (the first 25-30 minutes) seemed to drag.

After consulting with Francis Ford Coppola, Romanek decided to begin the film late in the story, with Sy being arrested as a criminal. But the audience isn’t told exactly what he’s done. Then the plot flashes back to the original opening material. “This one change,” says Romanek, “rendered the first act more compelling. The first act played out almost exactly as before, but now the audience is paying closer attention. They’re now put in the position of trying to discover clues as to what Sy might’ve done. They’ve gone from passive viewers to detectives of a sort. And the first act came alive again” (p. 232).

Vachon’s book offers other intriguing examples of how filmmakers try to shape viewers’ responses through choices about form and style. It’s also a fascinating survey of the daily life of an independent producer. Vachon produced Safe, Boys Don’t Cry, Far from Heaven, and other widely admired films. And she never lets the reader forget that financing and ticket sales drive even the low-budget sector.