Proposed new timing changes for BTD 1 and BTD 2
3 years ago
Sweden

Hello, I'm here to propose some new timing rules for when the timing for runs should end.

Why?

Recently there was some discussion about the top 2 runs in BTD1 glitch%, with two runs that were of a very similar time. For the discussion see: https://www.speedrun.com/bloons_tower_defense/thread/aft7e I misstimed my run by timing a local file while youtube seems to introduce some lag, but for fairness sake timing should ofcourse be done by the youtube videos. The problem with ending the timer at "congratulations" is that it fades in, so if the video is blurry, what resolution is used, and how readable the text should be is all pretty much arbitrary choices as anyone who has timed bloons TD runs would know.

The solution?

End the timer on the first frame on which no projectile particles are visible on screen after the last bloon is popped. That's darts, tacks, bomb explosions etc. https://imgur.com/a/y6U2NpM for an example. The downside of this is that sometimes darts will fly for many frames before leaving the screen, but atleast it's objective when time should end.

Norway

The way I timed things was by using spikes. When they disappeared, that seemed to be on the first frame that the "congratulations" screen started to fade in. This is a really consistent way to time it, but it relies on having road spikes present, which is not utilized in all categories

Your proposed solution seems fine to me. As far as I know, these projectiles disappear before road spikes, meaning if you re-time old runs, all of them should get a slight improvement, if anything. So even if this timing method introduces a minute level of RNG, its consistency more than makes up for it imo.

Also is there a discord for this game? The invite in the forums doesn't work

Edited by the author 3 years ago
Gerbungis and KILLKID like this
Sweden

@Desch I don't think we have a discord atm, atleast not one i'm in. We switched to these rules in BTD1 now and I'm retiming some runs, and I noticed timing my records locally as opposed to from the youtube upload took off like 0.5 seconds. So my record should be like 0.5 seconds slower I think, if you get something close we can retime the runs using https://slashinfty.github.io/yt-frame-timer/ .

Gerbungis likes this