welcome to the machine
I would be a lot less opposed to the publication of this run if the visually obvious mistake didn't exist. Regardless of how we time it, if TheVoid improved it by another quarter-second or whatever then I think that would go a long way towards pacifying fears of people submitting unoptimized runs of short games. If we can remove that fear then we only have to worry about crappy game choice... but frankly, given SDA's current game selection, that's already a lost battle. I still think this is the dumbest game so far for running but w/e.
In short, yes I think that we need to use smaller time increments for this game and other short games. We already implement a sort of leniency based on run length with longer single-segments: they're allowed more mistakes. That leniency gets smaller and smaller with run length until suddenly it balloons again when we start pushing fractions of seconds, with the justification of "well it wouldn't make any difference in the final time anyways". Screw that. It doesn't make any sense. I thought we were speedrunners, not stat whores. Either use smaller time increments or make it eligible for rejects in verification or something, but since it's come up I think it's time to rewrite it.
Posting this here instead of in Peaches' thread because my request directly affects yhtbtr.
In short, yes I think that we need to use smaller time increments for this game and other short games. We already implement a sort of leniency based on run length with longer single-segments: they're allowed more mistakes. That leniency gets smaller and smaller with run length until suddenly it balloons again when we start pushing fractions of seconds, with the justification of "well it wouldn't make any difference in the final time anyways". Screw that. It doesn't make any sense. I thought we were speedrunners, not stat whores. Either use smaller time increments or make it eligible for rejects in verification or something, but since it's come up I think it's time to rewrite it.
Posting this here instead of in Peaches' thread because my request directly affects yhtbtr.