48 Frames Per Second

 Posted by at 20:28  No Responses »
Dec 162012

The Hobbit: An Unexpected Journey, released this weekend to record-shattering numbers, is outstanding. I saw a midnight show as Thursday gave way to Friday, and I just returned from seeing it for a second time. I look forward to seeing it again with family over the holidays, owning it on Blu-Ray, and devouring all of the behind-the-scenes and extended content as I did with the Lord of the Rings films when they came out.

The show we attended opening night was the standard, 24 frames-per-second, non-3D format. I wanted my first experience with the movie to be about the film itself, the story, its characters, and so forth. I didn’t want to risk distraction by a new film format until I had taken the movie in under comfortable and familiar circumstances first. Today, Cody and I saw it at 48fps in 3D. The new frame rate has met with a great deal of contention and I want to share my thoughts on it while they’re still fresh.

The opening shots in Bag End are weird. While some claim that it “takes time” to adjust to the new frame rate, I don’t think that’s what I was reacting to at all. There were plenty of shots where the higher frame rate was noticeable, but these specific early shots were noticeable and felt off. It’s difficult to precisely articulate, and I would dearly love the opportunity to compare the 24 and 48 fps versions shot-for-shot to see if I can identify the cause, but many of these seemingly simple shots feel sped-up, jerky, or otherwise technically marred. As the film went on, these shots occurred less frequently, but continued to be noticeable, which is why I do not believe it to be a simple matter of acclimatization.

On the flip side, some shots were outright stunning because of the higher framerates. Contrary to a great deal of the criticism I’ve seen, I thought all of the CG shots were greatly enhanced by the higher framerates, gaining astounding clarity and visual character. In these shots, I felt more immersed, not less. Cody concurred with this sentiment, so I know I’m not entirely alone in feeling that way.

On the whole, I’d say the movie had a 10/80/10 split of its length where the framerate made it look amazing, was unnoticeable, or was noticeably “off.”

I have some thoughts as to why, though, that are not as simplistic as “it’s a bad way to make movies/doesn’t look as good/majestic/etc.” or “we’re just not used to it yet!” I think there’s more to it than either of these two camps. What we’re seeing strikes me as a new version of the uncanny valley.

For those unfamiliar with the term, the uncanny valley refers to the sudden drop in humanity a facsimile human likeness has as it approaches reality. Cartoon characters are clearly not really human, but we recognize them as human-like; they do not trigger this reaction. As we get more and more real (e.g. The Polar Express), the characters that are quantifiably “more human” seem more alien. They are human-like, but lack the fine distinctions to mark them as truly human, and thus make us uncomfortable. Lord of the Rings‘ Gollum might be the first CG character to truly breach the valley, and there have been many more since. As one might imagine, CG artists got better at bridging the valley as our understanding of it improved and as technology allowed us to better simulate those gaps.

I think the same thing is now happening with framerate. With 24 fps, we’re subject to images blurring across 1/24th of a second — far more than our eyes actually detect. It’s clearly not “real” to the point where we accept it as a convention of the medium. At 1/48th of a second, things have far less time to blur, but there’s still some motion smoothing going on. Our eyes see at effective framerates of ~60fps or higher, which means this 48fps standard approaches the perception level of our eyes, but doesn’t quite match it. Welcome to the uncanny valley.1

On top of that, though, I think there’s also some aspect of virgin technique at play here. Filmmakers are accustomed to working at 24fps. They know how to light a set, move a camera, set up marks for actor movement, and so on at this frame rate. They know how to make 24fps look good. 48fps is a different tool in the toolbox. It doesn’t work like 24fps and expecting it to do so results in shots that will jar viewers.

A lot of this crystallized during a flashback sequence, which I recalled from the 24fps version to feel somewhat ethereal and lethargic, not quite slow-motion but still sluggish enough to feel like some distant memory. In the 48fps version, I expected this scene might look strange–but it didn’t! It actually looked fantastic! Because they shots and motion had to be slowed down, none of the movements felt jerky or stilted, giving the entire sequence a marvelous quality.

In total, I’m not sold on 48fps yet, but neither would I consider myself a detractor thereof. I think it has great potential for providing films with a greater semblance of life than they have had before, but I think filmmakers need to be careful in its use and adapt their style of shooting (and editing! shots have to be cut more slowly so as to not feel abrupt!) to account for it. In many places throughout THAUJ, they don’t quite manage it, which makes it pull the viewer out of the film. In many other places, they nail it and offer a greater level of immersion than we’ve before had.

I’m curious to see where this leads.

  1. This is a very rough simplification of a much more complex process — both on the film and on the occular side of things. If you’re curious about this, I suggest Googling “frame rate human eye” or reading the wikipedia article on Frame Rate for more information. []
Sep 202010

TIE Fighter case artIf that post title got you excited, I apologize.

For a while now, I’ve toyed with the idea of doing some kind of Star Wars fan film, being both a Star Wars nerd and an amateur filmmaker. One idea that popped into my head recently, while recollecting fond and cherished memories of playing the TIE Fighter computer game, was to adapt the game’s story into a TV (well, web) series. The game was story-driven enough that I think it could work, and had enough characters that it could be interesting. I’m not suggesting I’m going to do this. I barely have the time and energy to do all of the current projects I’ve saddled onto myself, let alone adding something as megalithic as this. But it’s still fun to think about.

According to lore, the TIE Fighter player assumes the mantle of Maarek Steele. Seems like a good choice for the series’ protagonist. As the game progresses, a number of major secondary characters and antagonists are introduced. Among them are then-Vice Admiral Thrawn, the rogue admirals Harkov and Zaarin, and Darth Vader puts in a cameo, too. Including the Imperial officer that briefs Steele before each mission, as well as the member of the Emperor’s Secret Order that provides secondary objectives, might work as well.

In terms of adapting the game, I think I’d first just go through the game mission-by-mission and isolate the major story components from each. These would get woven into the major arc of the series, which itself might even be split into seasons to mirror the distinct campaigns in the game. Once that had been done, the next step would be to pick out key bits of dialog from the game and weave those into the episode script. Nostalgia, man! It wouldn’t have to be line-for-line, but it’d be a fun callback to hit some of the key lines.

I might visit the idea some time in the distance future. TIE Fighter stands as my favorite game of all time (yes, even over WoW), and it nicely dovetails with the desire to do a Star Wars fanfilm. Of course, I’m not sure if I will ever be able to commit the amount of time doing an entire series would require. But hey, it’s fun to dream.