============================================================...

Prompt

==================================================================== Taric Lv9 E 技能刷新机制仿真与出装路径优化规范 ==================================================================== 1. 基础参数与常数定义 ==================================================================== * 英雄等级 (Level) = 9 * 基础攻速 (BaseAS) = 0.625 * 攻速收益率 (ASRatio) = 0.625 * 前摇比例 (WindupPct) = 0.18 * 前摇修正系数 (WindupMod) = 0.25 * Q 施法时间 (QCastingTime) = 0.25 * Q 基础冷却时间 (Q_CD_raw) = 3 * E 基础冷却时间 (E_CD_raw) = 12 * 9级自然成长攻速成长 (growAS)： growAS_lv18 = 0.34 growAS = growAS_lv18 / 17 * (Level - 1) * (0.7025 + 0.0175 * (Level - 1)) 在 Level = 9 时： growAS = 0.1348 (即 13.48%) 2. 面板属性与普攻耗时计算 ==================================================================== 根据装备购买属性与符文属性级联叠加，计算游戏内实际攻速与普攻耗时： * 实际总攻击速度 (TotalAS, 翻倍被动且最大上限为 3.003)： TotalAS = min((BaseAS + (BonusAS + growAS) * ASRatio) * 2, 3.003) * 单次普攻总耗时 (AttackTime)： AttackTime = 1 / TotalAS * 基础前摇 (BaseWindup)： BaseWindup = WindupPct / BaseAS * 实际当前前摇 (CurrentWindup)： CurrentWindup = BaseWindup + WindupMod * (AttackTime * WindupPct - BaseWindup) * 实际当前后摇 (CurrentBackswing)： CurrentBackswing = AttackTime - CurrentWindup 1. 技能急速与被动减 CD (CDPA) ==================================================================== 根据属性计算技能实际冷却时间，以及被动每次强化攻击的减 CD 量： * 面板实际冷却时间： TotalHaste = BaseHaste + Haste CDR = TotalHaste / (TotalHaste + 100) Q_CD = Q_CD_raw * (1 - CDR) E_CD = E_CD_raw * (1 - CDR) * 被动强化攻击减 CD 量 (CDPA)： CDPA = 1 + Haste / (Haste + 100) 4. 连招耗时与技能冷却仿真模型 ==================================================================== * x∈{X∈Z,x∈{1-10}} * 标准模式下：coeff = 1.00 * 纳沃利模式下：coeff = 0.85 * 连招内攻击耗时: * ComboAttack(0) = 0 * ComboAttack(x) = (x - 1) * AttackTime + CurrentWindup * Last_ComboAttack(x) = ComboAttack(x) - ComboAttack(x-1) * Q技能未准备就绪等待时间： QRemain(0)=Q_CD QRemain_tmp(x)=max(0,(QRemain(x-1)-Last_ComboAttack(x))*coeff-CDPA-CurrentBackswing) QRemain(x)={Q_CD, x % 2=0; QRemain_tmp(x), x % 2=1} Trigger(x)={0, x <= 2; 1 - (x % 2), x >2} QCdGap(x)=QTrigger(x)*QRemain_tmp(x) * Q施法时间 > 攻击后摇硬直等待时间 (nextAWait)： nextAWait(x) = (1 - (x % 2)) * max(QCastingTime - CurrentBackswing, 0) * 连招累计耗时： * comboTime(0) = 0 * comboTime(x) = comboTime(x-1) + Last_ComboAttack(x) + QCdGap(x) + nextAWait(x) * Last_comboTime(x) = comboTime(x) - comboTime(x - 1) * 初始 E 冷却：CD_init(0) = E_CD * E 冷却时间递推： CD_init(x) = (CD_init(x - 1) - Last_comboTime(x)) * coeff - CDPA * 刷新条件与最终耗时计算 * 采用两种标准 (comboTime_final)： * 1：命中即刷新 - 条件：CD_init(x) <= 0 - 结果：comboTime_final = comboTime(x) * 2：后摇内自然刷新 - 条件：CD_init(x) > 0 且 CD_init(x) <= CurrentBackswing - 结果：comboTime_final = comboTime(x) + CD_init(x) 抛开QCdGap，nextAWait，coeff 简化刷新条件: comboTime_final={x*AttackTime,E_CD <= x*(AttackTime+CDPA)} 设定 1%bonusas价值25 1haste价值26.67 数学方法分析 相同经济下bonusas与haste对comboTime_final的影响效率对比