Seihon's normal run, and framerate discrepencies
8 years ago
United States

So I checked out his run to see if there were any strategies I might have missed...and was quite surprised to see Rumia taken down over 6 seconds faster! So I looked a little further, first to see if my game was lagging, but then I noticed that his game was running at 62.50 FPS instead of the usual 60.

Running some numbers, his game is running a little over 4% faster than mine. 2:01 Rumia vs. 2:07 is 121 seconds vs. 127, but 121¤62.5 (7562 frames) vs. 127¤60 (7620 frames). This is without taking into account decimals, variances in split timing, or any potential improvements, but it still shows that for every minute of gameplay, he gains 2.5 seconds. That's...kind of an unacceptable disparity. There's three options here:

  1. Force 60 FPS. This is the easiest and best option. Alternatively, if he can't get the game to run at 60 FPS, a framerate adjustment could be done (20:31 = 1,231 seconds, 1,231 seconds ¤ 62.5 frames per second = 76,937 frames, and 76,937 frames at 60 per second = ~1,282.3 seconds, which is 51 additional seconds. 20:31 + 0:51 = 21:22).
  2. Allow any framerate. This is a bad call since the vpatch can set the framerate to literally anything: I would grind out a 120 FPS clear, and that would be unassailable without also clearing at 120 FPS, or close to it.
  3. Leave the boards as is. This is also a fairly bad call since so much of the game autoscrolls for free timesave just by running the game incorrectly.

These sorts of framerate decisions would likely end up applying to every game, so choose wisely (hint: option 1 with faster runs disallowed is the easiest for moderation, option 1 with the adjustment is the easiest for the players, options 2 and 3 are both very flawed)

What I usually do is scale it to 60 fps. Unfortunately I overlooked it this time by accident, sorry. I adjusted Seihon's time to what it would be at 60 fps. Playing at a higher framerate provides no advantages to my knowledge, so it seems ok to adjust for that case.

Playing at a lower framerate would make the game easier, however, so I propose limiting slowdown % (for games that provide it)+framerate difference to 3% (58.2 fps), as long as the user is making every reasonable effort to run the game at 60 fps (still scaling to 60 fps unless it passes the 3% point).

I plan on adding this to the rules for all the games if people find these guidelines to be reasonable. I also added timing rules to all of the non-fighting games.

Game stats
Latest threads