hycdes / gftool Goto Github PK

Step to Reproduce

1. Open https://hycdes.com/pages/GFT_Echelon.html

2. Add WA2000 to echelon
   

3. Select 承伤测试, set 回避 to 9999, check 敌人能够被杀死, and click 100次平均
   

4. Select 输出测试, set 回避 to 9999, and click 100次平均
   

5. Select 承伤测试 again and click 100次平均

Expected Behavior

The result of step 3 and 5 should be the same. Specfically, WA2000 should take some damage.

Actual Behavior

In step 3, WA2000 took no damage.

Then in step 5, WA2000 took some damage.

如题，建议把少前和方舟内容拆分成不同目录，不要混到一起

芯片的结果可以进行旋转来节省校准，感觉添加一个旋转所有芯片的按钮更人性化？

This has been reported on NGA 3 months ago, but the report seems to be buried in the discussion.

Step to Reproduce

1. Open https://hycdes.com/pages/GFT_Echelon.html
2. Add 维尔德 MkⅡ to echelon
   
3. In 输出测试, set 回避 to 100 and 护甲 to 100.
   
4. Click 计算1次 several times and check if 总输出 is always divisible by 5.

Expected Behavior

The probability of 总输出 being divisible by 5 should be around 20%.

Actual Behavior

总输出 is always divisible by 5.

Game Mechanics

Apparently, this calculator assumes the five bullets in a single shot by a 5-link t-doll either "all hit" or "all miss". The following picture proves the assumption is incorrect:

ZIBI

感谢大佬的工具, 针对打开网址慢的问题, 我做了个Electron打包桌面的封装, 手动更新需要安装git, 剪贴板用快捷键处理(网页内部无法调用)

https://github.com/Kreedzt/GFTool-Electron

可以给大佬的少前工具站架一个镜像站吗，体验上应该比直连GitHub pages好

重装部队练级60→100显示的当前等级和目标等级是40 60

Hi Guys,

Was explaining this for the en gf official discord so I thought I'd dump it here

1. select HOC type
If you tick the check box, M2 combinations will NOT use second group of 5 block chips. However, it'll cause 1 to 3 empty spaces.
2. set output
all combinations: Output all combinations possible from the displayed image.
Prioritised combinations: Output the most optimal combinations for selected stats combinations
3. Stats prioritisation.
Select: mixed total stats, lethality, pierce, accuracy, reload.
4. Filter settings
Tick the corresponding stats for which you want to filter the maximum overflow of stats cap and block.
The button is for sake of convenience and activates/deactivates the value cap (left) and block cap (right)

Example 1
BGM-71 has a stats cap of 190 lethality, 329 pierce cap.
The image sets a stats OVERFLOW cap of 10 lethality and 5 pierce.
Therefore, it will not output any combination with lethality > 200 OR pierce > 334.

Example 2
2B-14 has a reload stats cap of 85, and it correspondingly has a block cap of 6.
2B-14 has a reload block cap of 8. The image ticks a block cap of -2. Therefore, it will not display any combinations with stats block count > 6.

That's a relatively precise (almost sentence to sentence, but worded with grammar as correct as I can manage) translation of:
https://github.com/hycdes/GFTool/blob/master/img/chip/tutorial/cc-2-cn.png

描述：芯片仓库table，最左侧能否新增一列CheckBox，支持多选，这样可以一次性删除多个芯片。
原因：想到这个优化的原因是，我拼完一个重装的最优解后，再给另一个重装拼，为了不重复，打算删掉已经使用的，按序号去找，结果发现芯片仓库一次只能删一个，但是每删完一个，芯片仓库的所有芯片会重新排序，我也得重新找序号，还是有些麻烦，看能否优化一下

3月16日,使用芯片计算器时,出现卡死现象.网页崩溃,浏览器收不到回应.
多次尝试,在给定的芯片数量较少时,不会出现问题.
当芯片数量达到100以上时,最优解计算功能失败.

阵型模拟器梯队预览部分排版貌似乱了，貌似是因为电脑浏览器宽度不够？

截图如下：

Safari 13.1 和 Chrome 81.0.4044.129 均可复现。

deleted

Step to Reproduce

1. Open https://hycdes.com/pages/GFT_Echelon.html

2. Add WA2000 to echelon
   

3. Set 组数 to 99
   

4. Select 承伤测试, Check 敌人能够被杀死, and set 生命 to 1
   

5. Click 100次平均

Expected Behavior

WA2000 takes fatal damage.

Actual Behavior

WA2000 takes no damage at all.

当设置了AK-15时, 无法模拟计算.
提示list_tdoll[i][1].Skill is not iterable, Echelon_sub.js:515
list_tdoll中AK-15的skill为undefined
Windows10 chrome 79.0.3945.130（正式版本） （64 位）