Found while investigating https://github.com/ppy/osu/issues/25719. I wanted to confirm what the reporter was saying about the PP numbers reported being for nomod rather than daycore, and proceeded to find that mania simulate would return entirely different numbers.
Score has not been used for performance calculation since https://github.com/ppy/osu/pull/18749, so the current code's assertion that "only total number of hits is considered currently" is no longer correct.
The "hit statistics from accuracy" approximation thing is very loose and not intended to really be that useful, but it should technically yield the expected accuracy value (there are multiple possible solutions, this just finds one of them). I don't think that function is particularly useful or accurate in any ruleset anyhow.
Found while investigating https://github.com/ppy/osu/issues/25719. I wanted to confirm what the reporter was saying about the PP numbers reported being for nomod rather than daycore, and proceeded to find that mania simulate would return entirely different numbers.
Score has not been used for performance calculation since https://github.com/ppy/osu/pull/18749, so the current code's assertion that "only total number of hits is considered currently" is no longer correct.
The "hit statistics from accuracy" approximation thing is very loose and not intended to really be that useful, but it should technically yield the expected accuracy value (there are multiple possible solutions, this just finds one of them). I don't think that function is particularly useful or accurate in any ruleset anyhow.