The Dance of the Firefly

The Dance of the Firefly
Directed by Richard Bevis

Let's get this out of the way so I don't have to build up to it; Dance of the Firefly is probably the most beautiful looking film ever created with this program. I mean, have I missed something? Have several releases slipped under my radar in the past year that physically gave viewers orgasms? The response to the sheer visual power of this flick is a little lukewarm. Maybe that's not surprising considering this community's absolutely mind-boggling aversion to phenomenal visuals. There seems to be an instinct to give the cold shoulder to films that exist primarily to explore this aspect of the medium, as if by tossing out the rote recitation of standard 3DMM movie plots in favor of a more visually focused exploration, it cheapens the experience somehow. Maybe I'm jumping the gun. At any rate, this movie was incredible just to watch. I was transfixed, I was absorbed, I was captivated. I forgot I was looking at my own damn models most of the time. If this movie does nothing else, it adapts to the new toolset in a way that harnesses it's potential in a way I wasn't expecting to see for a long time.

This concept has of course been done a hundred times before, so if you're looking for something structurally unique, this is not where I'd point you. Brandon Wiemann's Bruised and Tom Kearns' Color cover this same ground in their respectively sugary and angsty ways. What makes Firefly another flavor of exploring the idea rather than simply a retread is that it knows that the technique doesn't sustain a movie by itself. Anyone can watch the countless visual extravaganzas that have made use of it over the years and try it themselves, but without an actual reason behind it, all they'd be doing is playing parrot. Bevis uses the firefly idea in a way that makes it work, at least on a functional level. Whether it moves you or not is another story entirely, but the sense I got from the movie was one of real exploration, curiosity, and discovery, not simply "hey look everything was greyscale before but now color is coming back". What Bevis is expressing is clear enough if you can disarm your cynicism long enough to notice; the fireflies are not magically reinvigorating a gloomy, colorless world, but illuminating a world that has, over time, become a rote pattern, lost in a peripheral haze.

What makes this effect work is how stunning the greyscale world looks. I've never seen anything that looks nearly this good or indeed this unique in a greyscale techinque. What even closely compares? Dead Heart in a Dead World? Regime of Terror? I bring this up not to unfairly attack what those movies attempted, but because they're the next closest reference point I have to a real, vibrantly imagined greyscale setting in 3DMM history. Without the advantage of the models or v3dmm's fast greyscale techniques, they're not even in the same category. What these techniques allow Bevis to do is to avoid the implied gloom and darkness of earlier greyscale applications in 3DMM and create something that isn't devoid of life at all. Someone mentioned that the greyscale shots look artistically better than the color that eventually flickers into being; I think that's a good point, and something of a dubious one for the effect of the film as a whole, but I'll get into that in a minute. The fundamentals of this concept are unchanged from previous incarnations as I mentioned, but their specific application services the artistic design of the film nicely. The shots of the lanterns being lit, the surface reflections of the firefly light, it's all tied together to be more than just an effect, but actually an experience.

A brief technical note that does bear mentioning: The slowdown in several scenes was extremely heavy. I know this is a slippery slope in 3DMM, because if every movie was designed for the slowest computer in the community, we'd still be using default scenery. But good design in a program that has to render movies on the fly is to optimize as much as possible. With v3dmm's ability to import textures, a massive, lush, animated scene that breaks 95% of the community's computers like that first treeline shot should really be done on a flat tile with multiple textures (avoid one texture with a panning tile if possible, it doesn't look nearly the same). Not every instance of slowdown can be avoided, but think twice about your scene-building methods when things are extremely polygon-intensive.

Those few playback hiccups aside, on that purely visceral level, it's a beautiful and stunning film to just watch and take in, and shot-by-shot it mostly succeeds in building toward something instead of meandering as a footage collection. But something was always nagging me just a little bit about it. It's not a fault of Bevis as a director or Firefly as a concept, at least not to a degree that could be easily fixed and refined. Fundamentally, I think it's just a pang of bad artistic luck, when everything about a concept is done right but you're aware of your distance as you watch it despite the slick polish and obvious care put into the film. Maybe it pushes a little too hard. Maybe it's just a little too serious. Maybe the film itself could have used a little distance, a little LESS passion and absorption; movies this earnest and involved with themselves invariably make people a little uncomfortable. Shit, maybe it's just me.

Bevis must have done something right, though, because usually I can keep this overenthusiastic art critic voice better in check. I'm all wound up and analyzing and poking around because a lot of really simple concepts end up spawning movies that reveal a hell of a lot more possibilities than might have been intended. Sure, it's just a neat little flick about fireflies revealing beautiful landscapes. But man, getting those little suggestions of awe and curiosity from a one-dimensional character following them out the front door and across the hills starts the brain thinking and wondering and before you know it the canvas is wide open for other neat little ideas and wonderings.

Critical Score: 90/100.
Personal Score: 90/100.

[Richard Bevis]

Thanks Aaron - that was a brilliant score!