This isn't a super common mod compared to every other one on the list, it's probably not worth the storage (and memory in case of stable) implications. We can look at revisiting this once we have actual spinner difficulty considerations
* Test theory crafting
* Place in more appropriate place
* fix a bit better
* Move things around
* Reduce diff
---------
Co-authored-by: StanR <hi@stanr.info>
* Rebalance aim and speed
* Rebalance star rating
* Attempt further speed balancing
* More balancing
* More balancing
* Buff aim a bit
* More speed balancing
* Global rebalance
* Speed balancing
* Global rebalancing
* More speed balancing
* Buff aim
* MORE BALANCING
* Revert "Rebalance star rating"
This reverts commit f48c7445e1.
* Decouple velocity change bonus from wide angle bonus
* Replace sin with smoothstep
* Set multiplier back to 0.75
---------
Co-authored-by: James Wilson <tsunyoku@gmail.com>
* initial commit
* changed HD curve
* removed AR variable
* update for new rework
* nerf HD acc bonus for AR>10
* add another HD nerf for AR>10
* Update OsuDifficultyCalculator.cs
* fix speed part being missing
* Update OsuDifficultyCalculator.cs
* rework to difficulty-based high AR nerf
* move TC back to perfcalc
* fix nvicka
* fix comment
* use utils function instead of manual one
* Clean up
* Use "visibility" term instead
* Store `mechanicalDifficultyRating` field
* Rename `isFullyHidden` to `isAlwaysPartiallyVisible` and clarify intent
* Remove redundant comment
* Add `calculateDifficultyRating` method
---------
Co-authored-by: James Wilson <tsunyoku@gmail.com>
* Buff precision difficulty rating in osu!
* Fix position repetition calculation
* Fix aim evaluator crashing, move small circle bonus calculation, adjust the curve slightly
* Refactor
* Fix code quality
* Semicolon
* Apply small circle bonus to speed too
* Fix formatting
---------
Co-authored-by: James Wilson <tsunyoku@gmail.com>
* implement stuff
* fix basic issues
* rework calculations
* sanity check
* don't use score based misscount if no scorev1 present
* Update OsuPerformanceCalculator.cs
* update misscount diff attribute names
* add raw score misscount attribute
* introduce more reasonable high bound for misscount
* code quality changes
* Fix osu!catch SR buzz slider detection (#32412)
* Use `normalized_hitobject_radius` during osu!catch buzz slider detection
Currently the algorithm considers some buzz sliders as standstills when
in reality they require movement. This happens because `HalfCatcherWidth`
isn't normalized while `exactDistanceMoved` is, leading to an inaccurate
comparison.
`normalized_hitobject_radius` is the normalized value of `HalfCatcherWidth`
and replacing one with the other fixes the problem.
* Rename `normalized_hitobject_radius` to `normalized_half_catcher_width`
The current name is confusing because hit objects have no radius in the
context of osu!catch difficulty calculation. The new name conveys the
actual purpose of the value.
* Only set `normalized_half_catcher_width` in `CatchDifficultyHitObject`
Prevents potential bugs if the value were to be changed in one of the
classes but not in both.
* Use `CatchDifficultyHitObject.NORMALIZED_HALF_CATCHER_WIDTH` directly
Requested during code review.
---------
Co-authored-by: James Wilson <tsunyoku@gmail.com>
* Move osu!catch movement diffcalc to an evaluator (#32655)
* Move osu!catch movement state into `CatchDifficultyHitObject`
In order to port `Movement` to an evaluator, the state has to be either
moved elsewhere or calculated inside the evaluator. The latter requires
backtracking for every hit object, which in the worst case is continued
until the beginning of the map is reached. Limiting backtracking can
lead to difficulty value changes.
Thus, the first option was chosen for its simplicity.
* Move osu!catch movement difficulty calculation to an evaluator
Makes the code more in line with the other game modes.
* Add documentation for `CatchDifficultyHitObject` fields
---------
Co-authored-by: James Wilson <tsunyoku@gmail.com>
* Move all score-independent bonuses into star rating (#31351)
* basis refactor to allow for more complex SR calculations
* move all possible bonuses into star rating
* decrease star rating scaling to account for overall gains
* add extra FL guard for safety
* move star rating multiplier into a constant
* Reorganise some things
* Add HD and SO to difficulty adjustment mods
* Move non-legacy mod multipliers back to PP
* Some merge fixes
* Fix application of flashlight rating multiplier
* Fix Hidden bonuses being applied when Blinds mod is in use
* Move part of speed OD scaling into difficulty
* Move length bonus back to PP
* Remove blinds special case
* Revert star rating multiplier decrease
* More balancing
---------
Co-authored-by: StanR <hi@stanr.info>
* Add diffcalc considerations for Magnetised mod (#33004)
* Add diffcalc considerations for Magnetised mod
* Make speed reduction scale with power too
* cleaning up
* Update OsuPerformanceCalculator.cs
* Update OsuPerformanceCalculator.cs
* add new check to avoid overestimation
* fix code style
* fix nvicka
* add database attributes
* Refactor
* Rename `Working` to `WorkingBeatmap`
* Remove redundant condition
* Remove useless variable
* Remove `get` wording
* Rename `calculateScoreAtCombo`
* Remove redundant operator
* Add comments to explain how score-based miss count derivations work
* Remove redundant `decimal` calculations
* use static method to improve performance
* move stuff around for readability
* move logic into helper class
* fix the bug
* Delete OsuLegacyScoreProcessor.cs
* Delete ILegacyScoreProcessor.cs
* revert static method for multiplier
* use only basic combo score attribute
* Clean-up
* Remove unused param
* Update osu.Game.Rulesets.Osu/Difficulty/OsuPerformanceCalculator.cs
Co-authored-by: StanR <castl@inbox.ru>
* rename variables
* Add `LegacyScoreUtils`
* Add fail safe
* Move `countMiss`
* Better explain `CalculateRelevantScoreComboPerObject`
* Add `OsuLegacyScoreMissCalculator`
* Move `CalculateScoreAtCombo` and `CalculateRelevantScoreComboPerObject`
* Remove unused variables
* Move `GetLegacyScoreMultiplier`
* Add `estimated` wording
---------
Co-authored-by: wulpine <wulpine@proton.me>
Co-authored-by: James Wilson <tsunyoku@gmail.com>
Co-authored-by: StanR <hi@stanr.info>
Co-authored-by: StanR <castl@inbox.ru>
* scale misscount by proportion of difficult sliders
* cap sliderbreak count at count100 + count50
* use countMiss instead of effectiveMissCount as the base for sliderbreaks
* make code inspector happy + cleanup
* refactor to remove unnecesary calculation and need for new tuple
* scale sliderbreaks with combo
* use aimNoSliders for sliderbreak factor
* code cleanup
* make inspect code happy
* use diffcalcutils
* fix errors (oops)
* scaling changes
* fix div by zeros
* Fix compilation error
* Add online attributes for new difficulty attributes
* Formatting
* Rebase fixes
* Make `CountTopWeightedSliders` to remove weird protected `SliderStrains` list
* Prevent top weighted slider factor from being Infinity
---------
Co-authored-by: tsunyoku <tsunyoku@gmail.com>
* basis refactor to allow for more complex SR calculations
* move all possible bonuses into star rating
* decrease star rating scaling to account for overall gains
* add extra FL guard for safety
* move star rating multiplier into a constant
* Reorganise some things
* Add HD and SO to difficulty adjustment mods
* Move non-legacy mod multipliers back to PP
* Some merge fixes
* Fix application of flashlight rating multiplier
* Fix Hidden bonuses being applied when Blinds mod is in use
* Move part of speed OD scaling into difficulty
* Move length bonus back to PP
* Remove blinds special case
* Revert star rating multiplier decrease
* More balancing
---------
Co-authored-by: StanR <hi@stanr.info>
* Move difficulty calculation fields from `Slider` to `OsuDifficultyHitObject`
* Remove redundant check
* Use `LastObject` where possible
* Update tests
* Make `LazyTravelDistance` `double`
---------
Co-authored-by: James Wilson <tsunyoku@gmail.com>
See
https://discord.com/channels/188630481301012481/1097318920991559880/1334716356582572074.
On `master` this is actually worse and shows thousands of pp points, so
I guess `pp-dev` is a comparable improvement, but still flagrantly
wrong. The reason why `pp-dev` is better is the `speedDeviation == null`
guard at the start of `computeSpeedValue()` which turns off the rest of
the calculation, therefore not exposing the bug where
`relevantTotalDiff` can go negative. I still guarded it in this commit
just for safety's sake given it is clear it can do very wrong stuff.
* Calculate hit windows in performance calculator instead of databased difficulty attributes
* Apply mods to beatmap difficulty in osu! performance calculator
* Remove `GreatHitWindow` difficulty attribute for osu!mania
* Remove use of approach rate and overall difficulty attributes for osu!
* Remove use of hit window difficulty attributes in osu!taiko
* Remove use of approach rate attribute in osu!catch
* Remove unused attribute IDs
* Code quality
* Fix `computeDeviationUpperBound` being called before `greatHitWindow` is set
* Make aim accuracy scaling harsher
* Use deviation-based scaling
* Bring the balancing multiplier down
* Adjust multipliers, fix incorrect deviation when using slider accuracy
* Adjust multipliers
* Update osu.Game.Rulesets.Osu/Difficulty/OsuPerformanceAttributes.cs
Co-authored-by: James Wilson <tsunyoku@gmail.com>
* Change high speed deviation threshold to 22-27 instead of 20-24
* Update tests
---------
Co-authored-by: James Wilson <tsunyoku@gmail.com>