I find it tempting to write that Leon Arnott’s Animated Screenshots is the If We Don’t, Remember Me of video games, but I’m not sure if that’s exactly true.
Somehow Gus Mantel’s IWDRM, through its slight and carefully controlled animation of film stills, creates long, silent, haunting moments that feel like an extension of the movies they’re from, without being direct excerpts. Arnott’s work, as far as I can tell, comprises literal moments from the games they quote, and as a result feel less like subtle new interpretations of an existing work and more like — well, animated screenshots, really.
Time in a videogame moves naturally in loops. Sit your character still, and the world does in fact stop moving, the clouds drifting past while the candles flicker their four-frame animations in their sconces — forever, or at least for as long as you care to wait. Play a boss fight passively, and watch as the screen-filling terror reveals itself as a predictable, on-rails process, ultimately powerless.
The two sites do share similarity in their surprising use of the animated GIF as a vehicle for quietly contemplating, and even discovering, works in other media. (Does the animation above make you as curious to play The Extinct Bird as it does me?) Definitely worth a browse, in both cases.
I was not familiar with IWDRM until just now. I was, however, aware of three frames, which is basically its equal opposite. And a few of my posts (such as today's) are in some respect influenced by Shot Context. But when I started it, my only real intentional influence was Anna Anthropy's time tunnel.
I don't have much of an artistic ethos to my tumblr, or really think of it as a "work" in the same way that, say, Dead End Thrills is. All of my GIFs are created from screen-recorded emulator footage. My strongest guiding principle is in being close to 100% faithful to the actual game footage as possible. This reflects not just a subversive commitment to milking a full 6Ofps out of a medium not known, designed or intended for that framerate, but also the fact that I don't actually have as much capacity to post-process the footage as I like. There are currently just two cases where I have modified/filtered the footage, and those were inspired primarily by filesize constraints, and only secondarily by aesthetics.
My other principle is that of having the footage loop as cleanly as possible. This prevents me from using, among other things, more Knytt Stories footage (due to its heavy use of non-deterministic particle effects). My VVVVVV post is about as far as I will go with revealing seams in my footage. (Of course, in rare cases I discard this entirely in the interest of capturing single, highly extravagant cutscenes.)
As for my choice of scenes... essentially I just pick whatever feels particularly illustrative or indicative of the game in some respect, be it a standalone instance or a series of patchwork scenes, and which actually contains some animation. My first Mega Man 2 post shows how some of its stages provide interesting/scenic looping animations more easily than others.