ekgame / storasbot

A Discord bot for osu! leaderboad tracking and notifications.
6 stars 2 forks source link

DTHR is screwed up yo #5

Open NameAlreadyTaken6 opened 7 years ago

NameAlreadyTaken6 commented 7 years ago

DTHR pp is calculated incorrectly by the beatmap examiner.

Currently, it appears that OD is capped at OD10 for DTHR calculations. (My best guess: the code checks to make sure OD<=10 when HR is applied, but it does so at the wrong time.)

Cases I looked at below.

(OD<6 works correctly, since 5.9+DTHR < 10)

OD5: https://osu.ppy.sh/b/1182877 https://osu.ppy.sh/b/1148060 OD6: https://osu.ppy.sh/b/291562 https://osu.ppy.sh/b/129170

(OD6-8.4 underestimates the acc pp, but still have a steeper curve than DT-only)

OD7: https://osu.ppy.sh/b/323875 https://osu.ppy.sh/b/132490 OD7.5: https://osu.ppy.sh/b/1089041 https://osu.ppy.sh/b/999692 OD8: https://osu.ppy.sh/b/153771 https://osu.ppy.sh/b/172070

(OD>8.5 calcalutes a shallower curve for DTHR than for DT, since 8.5+DT > 10)

OD9: https://osu.ppy.sh/b/1119328 https://osu.ppy.sh/b/1053378

Also, in general, DTHR gives a too-low value for strain pp. (By about 15%, usually.) May or may not be related, but noticed it by accident while putting together this post.