Pennsylvania, USAI appreciate the interest, everyone. Since the community is larger now I am going to release one of my "Timing Files" (every run that I verify gets one) which shows all the counted frames. MyOhMyke, since you just got the Bitch WR I'm going to pick on you here. This video shows exactly the frames that were counted towards your loadless final time.
I make this in a video editor and it allows me to be consistent and precise across all of the communities' runs. If you were making an auto-splitter, the "stop" trigger would be the boss health going to zero, as I use the final keystroke that kills the boss to stop the timer. However, I would wager that it is more difficult to code something that would start and stop precisely on fade in and fade out for the main body of the levels. I looked for another more obvious breakpoint but this is what I settled on after not really being able to find anything more distinct (and at this point I am committed to the method for consistency). The only weird exception to the fading is Coco and Cindy, which for whatever reason does not "fade" in but rather transitions as a blend directly from the loading screen. No other level or boss fight does this.
Anything that involves a human typing is going to really throw off the test more so than the variable we're trying to test for. Still, I decided to frame-perfect synch your two halfs and see if the cutscenes ran identically. They do. (Given the margins for error on the 30 fps video we are working with.)
Please enjoy the relevant part of your test resynchronized. Also I deliberately made no attempt to fix the double sound and now we have double double sound and that's kind of awesome.
I'd have to see the submission to make the call for sure, does it run faster than it's supposed to with post-processing disabled?
There is loading time between the main level and the boss, which is also cut out.
Basically I count the "game" time from start of fade in to end of fade out, and nothing else. That said, I don't expect you to use the leaderboard timing for your splits... I've been retiming everything manually on my own - and though it's a bit of a pain in the ass, I've found that there are significant load time differences between people's runs that make what I have been doing much more fair than a straight-up RTA run.
MD5 hash of file River City Ransom.nes: 29 4e 4f a0 92 db 8e 29 d8 3f 71 e1 37 d1 f9 9f
Produces Generic Dudes on Everdrive.
Holy shit these times are getting low.
That's pretty much SDA timing (first to last frame of player control). I'm new to running the game but for what it's worth I like that ruleset.
I figured we should have a thread for discussion of the first race, and of course I wanted to share my new vid and other media you guys have put out there for it:
Original Live coverage and commentary:
http://www.twitch.tv/infestedriche/c/6222235
Official Race Results:
http://www.speedrunslive.com/races/result/#!/128709
Original Race Posting:
https://docs.google.com/document/d/130dkwwf4kxEyD6LbfWoE5imPZmcDxV4RoD00otpeyU8/edit
Hey guys, just wanted to report that the game / leaderboard that I set up seems to be throwing PHP errors:
The only non-standard thing I did here is that I set the IL table to be the primary display, so it might have something to do with that?
What about just punting to a team of global moderators if X number of days / weeks go by?
Looks like there's already been an admin response here, but I'd like to throw in and say that I would definitely use this feature if you guys were able to work it in.
Was great fun. I'm definitely in next year (if yinz will have me again).
Yo, I submitted a 8:51 and you guys (CartinaCow?) changed it to 8:49. I'm not complaining but I was wondering the proper way to start / stop timer for runs moving forward.
Any chance you can create a 0-10 category? People race this fairly often.
