Skip to content

HVAA优化及BUG修复 - #155

Open
WayneFerdon wants to merge 694 commits into
dodying:masterfrom
WayneFerdon:master
Open

WayneFerdon wants to merge 694 commits into
dodying:masterfrom
WayneFerdon:master

Conversation

@WayneFerdon

@WayneFerdon WayneFerdon commented Aug 21, 2025

Copy link
Copy Markdown
Contributor
  1. 增加初始闲置竞技场配置非空检测
  2. 修正其他技能的OC判断及选择范围判断,修正条件判断中iscd的值与Readme中示例不相符的问题 (修复技能可释放判断及其他技能技能排序无效问题 #154 未进行冲突处理直接覆盖了,因此按照其中实际有效的内容重新修改了)

@Qiyue0515

Copy link
Copy Markdown

最近转法发现一个问题:
// 先处理特殊的 “先给全体上buff”
let skillPack = ['We', 'Im'];
这里改成先上weaken再上Imperil会比较好,这样血线的压力会小一点

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

最近转法发现一个问题: // 先处理特殊的 “先给全体上buff” let skillPack = ['We', 'Im']; 这里改成先上weaken再上Imperil会比较好,这样血线的压力会小一点

@Qiyue0515 现在这个应该就是先we再im的啊

@DreamSeeker08

Copy link
Copy Markdown

最近转法发现一个问题: // 先处理特殊的 “先给全体上buff” let skillPack = ['We', 'Im']; 这里改成先上weaken再上Imperil会比较好,这样血线的压力会小一点

你好,能分享一个法系的设置给我导入吗?
我不太会弄,更新到最新的v2.90.22.20还是不会释放陷危、虚弱等debuf

@Kotone-Hikaru

Copy link
Copy Markdown

法师好像不能正确的释放im 只会释放weaken 在给每个怪都上完weaken之后就卡在这里了

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

@DreamSeeker08 @Sakuyasaki 之前的debuff逻辑有条件判断时可能会有点问题,分支新上传了一版(2.90.22.22),可以测试看下还有没有问题

@Kotone-Hikaru

Copy link
Copy Markdown

@WayneFerdon 2.90.22.22 imp正确释放了 但是全部上好imp之后不会自动释放攻击,但是鼠标移动到对应的怪上后才会攻击 且有的时候只会用普通攻击 不会释放t123法术

@WayneFerdon

WayneFerdon commented Oct 23, 2025

Copy link
Copy Markdown
Contributor Author

@WayneFerdon 2.90.22.22 imp正确释放了 但是全部上好imp之后不会自动释放攻击,但是鼠标移动到对应的怪上后才会攻击 且有的时候只会用普通攻击 不会释放t123法术

@Sakuyasaki
我这里测试是正常的,普通和高级的都能够释放,不确定是什么原因
不过推测可能是战斗中途启用新脚本,导致ability数据未初始化,战斗中读取技能范围数据出错,这个节点刚好是点击完技能还没选目标(在算范围),如果有mb等脚本的话刚好移上去就选中了?
新的版本(2.90.22.23)增加了非空检测,可以再看下

@Kotone-Hikaru

Copy link
Copy Markdown

@WayneFerdon 确实是因为中途才开启脚本导致的 如果在一开始就开启然后再进入战斗的话就能正常释放t123
不过这个 自动闲置竞技场好像有点问题 还是说逻辑改变了

@Kotone-Hikaru

Copy link
Copy Markdown

AR好像是ok 的但是GF不行 不知道为什么

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

AR好像是ok 的但是GF不行 不知道为什么

@Sakuyasaki 有bug,修复了(2.90.22.24)

@Kotone-Hikaru

Kotone-Hikaru commented Oct 23, 2025

Copy link
Copy Markdown

发现现在的全部上debuff好像逻辑和之前的相比 不一样 之前会优先释放258这3个位子然后再释放没有释放到的
现在是按照优先级来释放的 大部分时候是乱序 会导致有的时候会多1t 理论上来说假如敌人是8个的话 第一次释放 2 5 7 这样应该是最好的 (有机会补到6号位没有上到debuff的情况)其他的按照258释放 一遍然后补漏掉的 比较好

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

@Sakuyasaki
给全体放debuff的两个各加了选项,可以选是按照位置还是权重释放
按位置还是用权重的逻辑,但是统一权重为-1而非按状态计算,保留结合用死亡、已有debuff等的特殊权重进行排除
本来想直接改了,但是We给高权重的先放有时也能降低自己受到的伤害,所以各自处理了(

@Kotone-Hikaru

Copy link
Copy Markdown

@WayneFerdon 感谢 现在自动gf 已经正常工作了 但是debuff顺序释放好像还是没生效 现在还是按照权重来释放的

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

@WayneFerdon 感谢 现在自动gf 已经正常工作了 但是debuff顺序释放好像还是没生效 现在还是按照权重来释放的

@Sakuyasaki 应该修复了(2.90.22.26)
漏了选初始目标也要按顺序了(

@Kotone-Hikaru

Copy link
Copy Markdown

@WayneFerdon 观察了一下 现在的释放 第一次 释放目标1 会给 123上(可能是机制?给第一位上会顺延到后面) 第二次就会释放4 但是这个时候只会给45上debuff 这样3的位置会被重复上debuff 浪费了一个轮次 感觉还是应该改为给2 5 8位置轮流释放

@Kotone-Hikaru

Copy link
Copy Markdown
function getRangeCenterID(target, range = undefined, isWeaponAttack = false, excludeCondition = undefined, forceUseIndex = undefined) {
    // 静态计数器
    if (typeof getRangeCenterID.callCount === 'undefined') getRangeCenterID.callCount = 0;
    getRangeCenterID.callCount++;

    // --- 固定顺序目标 ---
    const fixedIndices = [1, 4, 7]; // 数组索引
    let msTemp = JSON.parse(JSON.stringify(g('battle').monsterStatus));
    msTemp.sort(objArrSort('order'));

    if (getRangeCenterID.callCount <= fixedIndices.length) {
        // 尝试获取固定目标,如果越界或死亡就跳过
        for (let i = getRangeCenterID.callCount - 1; i < fixedIndices.length; i++) {
            const idx = fixedIndices[i];
            if (idx >= 0 && idx < msTemp.length && !msTemp[idx].isDead) {
                return getMonsterID(msTemp[idx].order);
            }
        }
        // 如果固定顺序没有可用目标,回归正常逻辑
    }

    // ---------------- 原本逻辑 ----------------
    if (!range) return getMonsterID(target);

    const centralExtraWeight = -1 * Math.log10(1 + (isWeaponAttack ? ((g('option').centralExtraRatio / 100) ?? 0) : 0));
    let order = target.order;
    let newOrder = order;

    let unreachableWeight = g('option').unreachableWeight;
    let minRank = Infinity;

    for (let i = order - range; i <= order + range; i++) {
        if (i < 0 || i >= msTemp.length || msTemp[i].isDead) continue;

        let rank = 0;
        for (let j = i - range; j <= i + range; j++) {
            let cew = (j === i) ? centralExtraWeight : 0;
            let mon = msTemp[j];

            if (j < 0 || j >= msTemp.length || mon.isDead || (excludeCondition && excludeCondition(mon))) {
                rank += unreachableWeight - cew;
                continue;
            }

            rank += cew;
            rank += forceUseIndex ? -1 : mon.finWeight;
        }

        if (rank < minRank) {
            minRank = rank;
            newOrder = i;
        }
    }

    return getMonsterID(newOrder);
}

让GPT写了一下 我试了一下这样是ok的 可以正确的释放到正确的258的位置()

@WayneFerdon

WayneFerdon commented Oct 25, 2025

Copy link
Copy Markdown
Contributor Author

@WayneFerdon 观察了一下 现在的释放 第一次 释放目标1 会给 123上(可能是机制?给第一位上会顺延到后面) 第二次就会释放4 但是这个时候只会给45上debuff 这样3的位置会被重复上debuff 浪费了一个轮次 感觉还是应该改为给2 5 8位置轮流释放

@Sakuyasaki
gpt这个即使不是全体或按顺序也会强行按顺序,而且目标死亡往后跳一个实际也会将死亡的包含在范围内了,也是不对的

原来的是按照顺序从上往下选需要释放的
但是释放本身可能会失败(被抵抗了)
你的这个可能是放了123,然后选5放456,但是6被抵抗了
例如:

  1. 在2放完123
  2. 在5放456,5失败
  3. 此时因为5需要释放,而包含5的最优的范围为567,就会在6释放567

不过这个也有问题,新的版本已经改为:

  1. 释放全体时会遍历全部需要放的目标
  2. 然后选出包含任意目标中权重最优的(按顺序的话仍是单个权重为-1)
  3. 这样就能优先往后去释放一片目标(无论是按顺序还是实际权重,确保收益最大化)
  4. 然后再进行补漏(失败或者未释放到的)

@Kotone-Hikaru

Copy link
Copy Markdown

新版这个 选择权重好像还是有点问题 在8个怪的时候 有时候会先在2或者6释放,这样123 567都会上上debuff 然后就会空下来4和8 这样会多放一次imp 这里的逻辑感觉如果是要按照释放的debuff到底有没有赋予 然后再对应的改变逻辑好像有点复杂()

@Kotone-Hikaru

Copy link
Copy Markdown

想到一个逻辑大概是就是重下往上检测第一个带debuff的 然后+2就是下一个需要释放的目标
比如给123上好imp了 然后下一个就应该上在5
假如1和3抵抗了 就给2的+2上 那就是345 然后漏掉的之后再补?
不知道能不能实现()

@WayneFerdon

WayneFerdon commented Oct 25, 2025

Copy link
Copy Markdown
Contributor Author

@Sakuyasaki 遍历的时候忘记按照顺序遍历了,导致还是按照权重遍历,后面权重高,先遍历到被记录了,然后遍历到前面的低权重的时候,算出来有效范围都一样,不大于先记录的,所以没覆盖掉
2.90.22.29已改为要按顺序的话就会按顺序遍历,这下应该行了(

想到一个逻辑大概是就是重下往上检测第一个带debuff的 然后+2就是下一个需要释放的目标 比如给123上好imp了 然后下一个就应该上在5 假如1和3抵抗了 就给2的+2上 那就是345 然后漏掉的之后再补? 不知道能不能实现()

从下往上找这个如果之前手动选了最后一个,就会变成直接补漏,所以还是得上面这个修复(

@Kotone-Hikaru

Copy link
Copy Markdown

@WayneFerdon 试了一下好像好像正常工作了 感谢更新

@Kotone-Hikaru

Kotone-Hikaru commented Oct 27, 2025

Copy link
Copy Markdown

发现卷轴使用有点异常 守护卷轴好像 有点问题 同样的条件 一次 gf 其他卷轴都是15左右 守护卷轴用了150
image
且这个好像没有效果 在ar也会不停释放守护卷轴

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

发现卷轴使用有点异常 守护卷轴好像 有点问题 同样的条件 一次 gf 其他卷轴都是15左右 守护卷轴用了150 image 且这个好像没有效果 在ar也会不停释放守护卷轴

@Sakuyasaki 我这里是正常的啊(
没有一直放,而且筛选也是有用的
不过原来的释放逻辑也有点问题,也稍微改了下(2.90.22.30)

@Kotone-Hikaru

Copy link
Copy Markdown

@WayneFerdon 这个有没有什么办法能导出log之类的吗

@Kotone-Hikaru

Kotone-Hikaru commented Oct 27, 2025

Copy link
Copy Markdown
You evade the attack from Macallan. -- Aiga Mizuki crits you for 3810 wind damage. You evade the attack from Google Dancing Israelis. You evade the attack from Hemidactylus. You evade the attack from Storyteller. Youshi153 gains the effect Imperiled. Hemidactylus gains the effect Coalesced Mana. Hemidactylus gains the effect Imperiled. Practiced Ass Spanker gains the effect Imperiled. You cast Imperil.   Cooldown expired for Storms of Njord You are healed for 11583 Health Points. You cast Cure.   You gain the effect Protection. You use Scroll of Protection.   Cooldown expired for Cure You evade the attack from Youshi153. You evade the attack from Practiced Ass Spanker. Hemidactylus gains the effect Coalesced Mana. Hemidactylus gains the effect Imperiled. Practiced Ass Spanker gains the effect Imperiled. Macallan gains the effect Imperiled. You cast Imperil.   You evade the attack from Macallan. Aiga Mizuki crits you for 3487 wind damage. You evade the attack from Storyteller. Hemidactylus crits you for 5077 slashing damage. Your spirit shield absorbs 1553 points of damage from the attack into 24 points of spirit damage. Aiga Mizuki gains the effect Imperiled. Storyteller gains the effect Coalesced Mana. Storyteller gains the effect Imperiled. Google Dancing Israelis gains the effect Imperiled. You cast Imperil.   Google Dancing Israelis hits you for 2676 wind damage. You evade the attack from Youshi153. You are healed for 11583 Health Points. You cast Cure.   Spawned Monster G: MID=282436 (Youshi153) LV=500 HP=174260 Spawned Monster F: MID=110494 (Hemidactylus) LV=500 HP=203444 Spawned Monster E: MID=153284 (Practiced Ass Spanker) LV=500 HP=183060 Spawned Monster D: MID=124000 (Macallan) LV=500 HP=183060 Spawned Monster C: MID=112618 (Aiga Mizuki) LV=500 HP=179060 Spawned Monster B: MID=194240 (Storyteller) LV=500 HP=171860 image 这是一次异常的释放保护卷轴的战斗记录

我试了几次 每次在异常释放卷轴的时候 都会放一个cure

@lukasva77

lukasva77 commented Sep 13, 2026

Copy link
Copy Markdown

@dalai1234 you are using a version over a year old persistent had an update and so many updates happened script is kept up to date here by WayneFerdon

TLDR u are doing it wrong

@lukasva77

lukasva77 commented Sep 13, 2026

Copy link
Copy Markdown

@dalai1234 is it broken on Commit e8d2fb8 which is the latest cause it happened to me more than once that script breaks but it gets fixed quickly if you post some debug logs or some more info

EDIT I meant breaks that way that timer goes bye bye

@lukasva77

Copy link
Copy Markdown

@dalai1234 "This started around the evening of September 11th" you could hunt down the specific commit to help WayneFerdon

@dalai1234

Copy link
Copy Markdown

@dalai1234 you are using a version over a year old persistent had an update and so many updates happened script is kept up to date here by WayneFerdon

TLDR u are doing it wrong

What I mean is that I’ve been using the latest version of WayneFerdon’s script, but recent updates have caused it to stop working, it only runs correctly if I revert to an older version.

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

@dalai1234 更新2.91.231试试

怀疑是遭遇相关数据读取到空值,但是之前没做容错,前几天的某次更新又直接用在不能为空的地方了
清除缓存如果只是篡改猴的脚本数据的话大概是无效的,因为推测报错的地方是在localStorage(而且应该是缺少了数据而不是内容错误)

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

推测是缺少了这个数据,要在devtool中才能看到,新版本已经加了缺少时的兼容

image

@dalai1234

Copy link
Copy Markdown

@dalai1234 更新2.91.231试试

怀疑是遭遇相关数据读取到空值,但是之前没做容错,前几天的某次更新又直接用在不能为空的地方了 清除缓存如果只是篡改猴的脚本数据的话大概是无效的,因为推测报错的地方是在localStorage(而且应该是缺少了数据而不是内容错误)

感谢更新,目前已正常,我之前试过调出控制台清理本地存储和会话存储,但都无效,甚至触发了无限卡语言设置的情况,只有先装v2.90.22.12,再升级2.91.XXX才能正常显示界面。

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

感谢更新,目前已正常,我之前试过调出控制台清理本地存储和会话存储,但都无效,甚至触发了无限卡语言设置的情况,只有先装v2.90.22.12,再升级2.91.XXX才能正常显示界面。

@dalai1234 那应该是了,大概是因为从2.90.xx版本升上来会触发旧版本数据迁移并写入localStorage,进而不会再触发这个问题

@lukasva77

Copy link
Copy Markdown

latest ver.

TypeError: can't access property "tw", arena.data.cached is undefined
cacheIWorTW moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:1106
mainProcess moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:943
HVAA moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:930
dAeHjve_wC8GC85Dr moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:577
dAeHjve_wC8GC85Dr moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:9003
dAeHjve_wC8GC85Dr moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:9004
fn moz-extension://30326764-a8bc-4293-9765-e4fad582e458/sandbox/injected-web.js:1
moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:1
[HV]AutoAttack.user.js:8999:11

@lukasva77

Copy link
Copy Markdown

I'm finished today with HV so don't know if above does anything

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

latest ver.

TypeError: can't access property "tw", arena.data.cached is undefined cacheIWorTW moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:1106 mainProcess moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:943 HVAA moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:930 dAeHjve_wC8GC85Dr moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:577 dAeHjve_wC8GC85Dr moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:9003 dAeHjve_wC8GC85Dr moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:9004 fn moz-extension://30326764-a8bc-4293-9765-e4fad582e458/sandbox/injected-web.js:1 moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:1 [HV]AutoAttack.user.js:8999:11

@lukasva77 Fixed in 2.91.232.

@lukasva77

Copy link
Copy Markdown

You have fixed this I just didn't upgrade from yesterday so I'm stuck on starting tower with nothing happening but is it ok to upgrade script mid battle hopefully gonna give it a try
here's how it looks on firefox

TypeError: can't access property "innerText", gE(...) is null
cacheIWorTW moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:1105
mainProcess moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:943
HVAA moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:930
dTqDGqy9sNbuf$du$/< moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:577
dTqDGqy9sNbuf$du$/< moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:9006
dTqDGqy9sNbuf$du$ moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:9007
fn moz-extension://30326764-a8bc-4293-9765-e4fad582e458/sandbox/injected-web.js:1
moz-extension://30326764-a8bc-4293-9765-e4fad582e458/ [HV]AutoAttack.user.js#23:1
[HV]AutoAttack.user.js:9002:11

@lukasva77

Copy link
Copy Markdown

Seems to be working fine after reload tower came alive

@lukasva77

Copy link
Copy Markdown

Also you have 3 useless entries in commits right now does that happen by accident or is there any purpose

@WayneFerdon

WayneFerdon commented Sep 14, 2026

Copy link
Copy Markdown
Contributor Author

Also you have 3 useless entries in commits right now does that happen by accident or is there any purpose

They're not useless at all but not that usefull so no worries.

@lukasva77

lukasva77 commented Sep 16, 2026

Copy link
Copy Markdown

MOSTLY USELESS POST!!!

#155 (comment)

I have posted before about notifications I would like to add that this is really fast you can double click notifications so It's almost instant better than disabling animations

I think might be worth including in guide or readme somewhere but firefox user base is 3 out of 100 also I only confirmed on windows 10 so 11 might be different so considering chromium and android and linux might be mostly worthless

TLDR fixes native notification speed on windows 10 and almost instant switch to tab

@lukasva77

Copy link
Copy Markdown

I'm getting big lag on HV UI disabling HVUtils fixes it which means HVAA is greatly optimized so disabling HVUtils seems obvious but maybe would it be possible to put the most important enhancements into HVAA I think too much work HVUtils is a large script
What do you think about HVUtils cause I can't evaluate it is it good code bad code or just unoptimized
Also that lag for a few convenient options I don't know if It's worth it

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

I'm getting big lag on HV UI disabling HVUtils fixes it which means HVAA is greatly optimized so disabling HVUtils seems obvious but maybe would it be possible to put the most important enhancements into HVAA I think too much work HVUtils is a large script What do you think about HVUtils cause I can't evaluate it is it good code bad code or just unoptimized Also that lag for a few convenient options I don't know if It's worth it

@lukasva77
The HVUT laggy stuff might be mostly caused by the armory functions (especially the filters)? So perhaps simply disable those might do the help.

@lukasva77

Copy link
Copy Markdown

HMM? I didn't try disabling HVAA and only HVUtils but yeah disabling HVAA and HVUtils is fast almost instant seems two big scripts together is the problem
So HVUtils is also optimized almost as fast as native UI I mean It's almost instant and feels great
So one is best disabled or just manually disable HVAA for a while when selling or buying on the market or upgrading equips

filters set to empty made no difference also I disabled percentage ranges I only have these 3 scripts

@lukasva77

lukasva77 commented Sep 16, 2026

Copy link
Copy Markdown

I gave a look at other users configs posted here how do you do the 2nd line cause everything has by default only number 1 what does this do anyway EDIT oh wait new add I'm dumb
overall most stuff is human readable I could have just imported config that looks the best but it seems just setting up the sleep debuff would be faster than changing everything to my liking

also does stuff like hp,4,10 versus hp <= 10 make any difference some people have some stuff different

also non readable stuff (skillOTOS.151+skillOTOS.152+skillOTOS.153) == 0 what does something like this do

also what does left side in debuffs do the one that says EWF 900 is it for some fine tuning
I will sift this stuff slowly I guess

I think auto config or like optimized defaults could be a thing for persistent only need mage mana optimized and speed optimized and tower also just melee with debuffs or a shield mage with debuffs also should be separated per page you know the 15 left side menus for fine tuning

@lukasva77

Copy link
Copy Markdown

Wait until HV monster db loaded what does this do???

@WayneFerdon

Copy link
Copy Markdown
Contributor Author

@lukasva77

I gave a look at other users configs posted here how do you do the 2nd line cause everything has by default only number 1 what does this do anyway EDIT oh wait new add I'm dumb

Also, try pressing Enter while any formula tab is focused.

image

also does stuff like hp,4,10 versus hp <= 10 make any difference some people have some stuff different

hp,4,10 is a legacy format, and should make no difference in actions

also non readable stuff (skillOTOS.151+skillOTOS.152+skillOTOS.153) == 0 what does something like this do

skillOTOS is the acted times for the given id (151 etc.). Some of these non readable parameters are not as managed as others like targetRank, and are directly get from the raw runtime cache data.

also what does left side in debuffs do the one that says EWF 900 is it for some fine tuning I will sift this stuff slowly I guess

image

It's for weighting while choosing debuff targets.

Most of the non readable parameters can be trace back to specific issues in this repository, as conversations in this pull request or other standalone issues. It's just not that full-filled in listing and translating all of them.
Also, try using the debug console output to help understanding (or test if anything buggy). (@ # are both available for all formula any where)

image

I think auto config or like optimized defaults could be a thing for persistent only need mage mana optimized and speed optimized and tower also just melee with debuffs or a shield mage with debuffs also should be separated per page you know the 15 left side menus for fine tuning

I myself is using 1H/NT rather than mage, so this is still not a planned function.

Wait until HV monster db loaded what does this do???

This is to ensure the formula getting data correctly while using targetDB with aother HV script `HV Monster Database

image image

@lukasva77

Copy link
Copy Markdown

I played around with it on this days tower I think It's better but the only command I understood is monsterAlive > 2 this makes it not waste sleeps when monster number is little other I have _targetBuffTurn_sleep <= 1 _targetBuffTurn_stun <= 1 still not sure what they do but seems to help
also just copying from someone else had given me sleep only on the bottom 3 monsters so I had to test them one by one which is the culprit I wonder if copied settings worked together with some other tab like attack rule

TLDR I think I got some reduction on mana potions and elixirs in tower

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.