h1or4dme8/osu-lazer
1
0
Fork 0
mirror of https://github.com/ppy/osu.git synced 2024-12-05 10:23:20 +08:00
Code Issues Releases Wiki Activity

fixed NaN issue in deviation calc

Browse Source
This commit is contained in:
Givikap120 2024-06-27 16:44:09 +03:00
parent 198f35dc63
commit 0788f18406
1 changed files with 26 additions and 25 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

51
osu.Game.Rulesets.Osu/Difficulty/OsuPerformanceCalculator.cs
View File

@ -322,6 +322,9 @@ namespace osu.Game.Rulesets.Osu.Difficulty
int okCountCircles = Math.Min(countOk, accuracyObjectCount - missCountCircles - mehCountCircles);
int greatCountCircles = Math.Max(0, accuracyObjectCount - missCountCircles - mehCountCircles - okCountCircles);
if (greatCountCircles + okCountCircles + mehCountCircles <= 0)
return double.PositiveInfinity;
// Assume 100s, 50s, and misses happen on circles. If there are less non-300s on circles than 300s,
// compute the deviation on circles.
if (usingSliderAccuracy || greatCountCircles > 0)
@ -421,41 +424,39 @@ namespace osu.Game.Rulesets.Osu.Difficulty
relevantCountMiss *= mistapsMultiplier;
}
if (relevantCountMiss < speedNoteCount)
{
double n = Math.Max(1, speedNoteCount - relevantCountMiss - relevantCountMeh);
const double z = 2.32634787404; // 99% critical value for the normal distribution (one-tailed).
if (relevantCountGreat + relevantCountOk + relevantCountMeh <= 0)
return double.PositiveInfinity;
// Proportion of greats hit on circles, ignoring misses and 50s.
double p = relevantCountGreat / n;
double n = Math.Max(1, speedNoteCount - relevantCountMiss - relevantCountMeh);
const double z = 2.32634787404; // 99% critical value for the normal distribution (one-tailed).
// We can be 99% confident that p is at least this value.
double pLowerBound = (n * p + z * z / 2) / (n + z * z) - z / (n + z * z) * Math.Sqrt(n * p * (1 - p) + z * z / 4);
// Proportion of greats hit on circles, ignoring misses and 50s.
double p = relevantCountGreat / n;
// Compute the deviation assuming 300s and 100s are normally distributed, and 50s are uniformly distributed.
// Begin with 300s and 100s first. Ignoring 50s, we can be 99% confident that the deviation is not higher than:
double estimatedDeviation = hitWindow300 / (Math.Sqrt(2) * SpecialFunctions.ErfInv(pLowerBound));
// We can be 99% confident that p is at least this value.
double pLowerBound = (n * p + z * z / 2) / (n + z * z) - z / (n + z * z) * Math.Sqrt(n * p * (1 - p) + z * z / 4);
double randomValue = Math.Sqrt(2 / Math.PI) * hitWindow100 * Math.Exp(-0.5 * Math.Pow(hitWindow100 / estimatedDeviation, 2))
/ (estimatedDeviation * SpecialFunctions.Erf(hitWindow100 / (Math.Sqrt(2) * estimatedDeviation)));
// Compute the deviation assuming 300s and 100s are normally distributed, and 50s are uniformly distributed.
// Begin with 300s and 100s first. Ignoring 50s, we can be 99% confident that the deviation is not higher than:
double estimatedDeviation = hitWindow300 / (Math.Sqrt(2) * SpecialFunctions.ErfInv(pLowerBound));
estimatedDeviation *= Math.Sqrt(1 - randomValue);
double randomValue = Math.Sqrt(2 / Math.PI) * hitWindow100 * Math.Exp(-0.5 * Math.Pow(hitWindow100 / estimatedDeviation, 2))
/ (estimatedDeviation * SpecialFunctions.Erf(hitWindow100 / (Math.Sqrt(2) * estimatedDeviation)));
// If precision is not enough - use limit value
if (pLowerBound == 0 || randomValue >= 1)
estimatedDeviation = hitWindow100 / Math.Sqrt(3);
estimatedDeviation *= Math.Sqrt(1 - randomValue);
// Then compute the variance for 50s.
double mehVariance = (hitWindow50 * hitWindow50 + hitWindow100 * hitWindow50 + hitWindow100 * hitWindow100) / 3;
// If precision is not enough - use limit value
if (pLowerBound == 0 || randomValue >= 1)
estimatedDeviation = hitWindow100 / Math.Sqrt(3);
// Find the total deviation.
estimatedDeviation = Math.Sqrt(((relevantCountGreat + relevantCountOk) * Math.Pow(estimatedDeviation, 2) + relevantCountMeh * mehVariance) / (relevantCountGreat + relevantCountOk + relevantCountMeh));
// Then compute the variance for 50s.
double mehVariance = (hitWindow50 * hitWindow50 + hitWindow100 * hitWindow50 + hitWindow100 * hitWindow100) / 3;
// Adjust by 0.9 to account for the fact that it's higher bound UR value
return estimatedDeviation * 0.9;
}
// Find the total deviation.
estimatedDeviation = Math.Sqrt(((relevantCountGreat + relevantCountOk) * Math.Pow(estimatedDeviation, 2) + relevantCountMeh * mehVariance) / (relevantCountGreat + relevantCountOk + relevantCountMeh));
return double.PositiveInfinity;
// Adjust by 0.9 to account for the fact that it's higher bound UR value
return estimatedDeviation * 0.9;
}
// https://www.desmos.com/calculator/no1pv5gv5g