now that i think about it I think I lost a bet from early 2015 that om couldnt get worse.
in scoreV2, lets say 2 players play this map.I wanted to address you in the Discord regarding this (at least, I think it was you) but had to go off to lunch so I'm gonna do it here I guess:
player 1 misses 1 time on accidentally in the very middle, but thats his only miss.
player 2 isnt as good as player 1, and cant keep up with the 250 bpm at the end, and misses 5-8 times, but he combo'd the rest of the song.
assuming they had similar 200/100/50 counts, player 2s score would be *miles* ahead of player 1, despite playing much worse on the map.
With how the accuracy v2 formula works, 5-8 misses would cause a MASSIVE drop in score, especially if there's not a huge amount of notes. Even with 2000 notes, 5 to 8 misses would cause a reasonable dent on the accuracy side and would likely suffer a reasonable amount against the amount of points that the other player would have lost chopping the song right down the middle. Everyone is making it seem like a combo right in the middle is going to completely invalidate your score and make it so that it's impossible to come back -- the fact of the matter is, a miss right in the middle of the chart is still going to net you half of the combo scoring attribute, assuming that the chart has a constant pace -- for anyone who doesn't realize it, hasn't read yet, or refuses to read due to being completely blinded by a new score system, combo is only going to accommodate for 20% of the score. If a player simply misses and then recovers, their score will suffer, but not as much as you're making it out to be.20% is still massive for something that is as arbitrary as combo scoring. in a system where the difficulty structure of a chart is not taken into account, where you miss should be irrelevant. to put things in perspective, 20% is 200k. the difference between a 600k barely-passed-the-chart play and an 800k "I-can-get-97%-or-higher" play. (of course its near impossible to actually get 0 points from the combo factor but that's besides the point)
The inevitable solution for having a combo-based scoring system is going to require a curve that emphasizes building combo more than it does sustaining it - that is, putting a larger weight of the score for the combo bonus in the first x% of a song's max combo, and then resolving that curve over the remaining portion of the combo.
It's very clear that the weighting is harsh if you find some of the screenshots going around -- it does need to be addressed, as it is certainly not infallible. I don't agree with the modifier bonuses for HD/FL/FI, but I do feel like a slight orientation towards combo allows one of the more key elements of the game to be emphasized, and that's long notes. With the way the timing window has been shifted for V2, the mechanic is a lot more important to be proficient at - and it penalizes the fuck out of you if you can't do it (as it should).
yah misread sorry.
Ankanogradiel wrote:rip best pp System NotLikeThisThis is score system, not pp system