I have to acknowledge that the question is difficult. If we wanted to define Vault in such a manner that runs accepted to it retain a sensible amount of gameplay, how would we do that without the rules governing this becoming overly arbitrary, complicated and ridden with complexities and endless lists of exceptions? It's not an easy question.
There is certainly value in glitches that utterly break the game and perhaps even skip the entirety of it, but it may be problematic if that's the only published version of the run.
I think this is something that should be decided pretty soon, because it's very probable that such runs will start being submitted at some point.
So what happens if a run is submitted that just skips the entire game? I think that technically speaking it fulfills the requirements of vaultable any%, and thus it will replace any possibly existing run of that game. Which means that there would essentially be no speedrun of that game, for all intents and purposes. Is that really what we want?
For me, a mistake-ridden playthrough that uses entertaining tricks and glitches and tries to be efficient and innovative even when it fails to be the quickest is still more of a speedrun than something that skips the entire game.
FWIW, regardless of how we call the actual input/replay/whatever files, we can still refer to our publications as movies (which is arguably a more suitable context for the word, considering how the publications are presented).
@Spikestuff: an encode on YouTube also contains the notes, game information, author, rerecord count, and subtitles, in addition to allowing players to watch the movie. (So it doesn't contain all the things you mentioned, but it does contain most.)
GBA/DS
Feos, be my guest. I don't like the way it looks, and it's a speedhack, but that part of the emulator core is not likely ever to change again, so its a fine time to speedhack it. Just don't do it for the mappers first, that's too big a job.
This will make it so the space before the test has background color, but the space after it doesn't. Is this something I should put on your bug tracker too or intentional?
It would be trivial to overclock the whole thing but that would correspondingly speed up sound. I'm not sure I can seperate out the sound without creating spaghetti, so yeah fceux 2.2.3 is probably your best bet for NES.
Hey uh. This sounds odd, but did something broke with RAM Search in the interim builds for GBA? Basically, when I search for certain values, it displays a result with TAStudio closed, but for some reason with it opened and playing a movie, it shows 0 results despite the RAM Watch indicating otherwise. Example: 1. Open the game 2711 - Drake & Josh (U)(Rising Sun).gba 2. Open TAStudios and then open https://drive.google.com/open?id=0B-2O13fpsnI4Zy16NmpYU01wNzA The movie should start up at frame 19,000-ish 3. Add this address in EWRAM to RAM Watch: 03069E. It is an unsigned 2 byte address. It should display 1023 at frame 19,000+ or so 4. With TAStudios still open and the movie still playing, open RAM Search and change the settings to find a 2 byte unsigned address in EWRAM 5. Change it to search "Equal to" and the value to "1023". Click Search For some reason, despite the fact the address above is in RAM Watch and displays 1023, RAM Search claims there's no results. Oddly, if I played back a bk2 file instead, it seems to work fine. Is this happening to anyone else?
Bummer.
Speaking personally, I would probably push for double-obsoletion right now while still allowing an optimized 2p any% run to have a shot at being re-published if the audience finds it entertaining.
I think you've about hit the nail on the head, Samsara. Let this one obsolete both any% runs, with the caveat that "2p any%" is still a valid category by our metrics.