Re: Bushido's Evasion to evade something!
cool.
the negative number doesn't need testing. was never meant as an evasion patch, just prevents the script from going to negative numbers, won't affect many people without changing it, but it still shouldn't be sending negative numbers out like that, especially when it's not done calculating the chance. It was tested in the sense it will not crash a server.
The evasion change on page 4 i posted does need testing and i strongly urge someone who can, to test it to make sure it will run, as well as make a patch for it.
here's how the current SVN has it.
bonus = 80
Next line
if bushido is say 120. so add 120 to bonus. 80 + 120 = 200
Next line
Lets say i have no tactics. (add nothing to bonus)
Next line
1.0 + 200 / 500
= 1.4
Which is RIGHT
BUT if i lower the bushido skill it returns a wrong value.
bonus = 80
Next line
if bushido is say 61. so add 61 to bonus. 80 + 61 = 141
Next line
Lets say i have no tactics. (add nothing to bonus)
Next line
1.0 + 141 / 500
return = 1.282 That translates to 28.2% parry chance added to your parry, with only 61 bushido skill, the max at 120 is only 40% it's suppose to be. 60 skill - 120 skill && 16% - 40%
12% higher than it should be.
the change uses different math, and will always return the proper value. should anyway, i went over the numbers a few times.
osd_daedalus;559819 said:I am confused too... I have also reverted that patch for negative values to "ready but needs testing more"
But looks like tam has right. The formula should be correct, that delay coded in that way is ruining all
cool.
the negative number doesn't need testing. was never meant as an evasion patch, just prevents the script from going to negative numbers, won't affect many people without changing it, but it still shouldn't be sending negative numbers out like that, especially when it's not done calculating the chance. It was tested in the sense it will not crash a server.
The evasion change on page 4 i posted does need testing and i strongly urge someone who can, to test it to make sure it will run, as well as make a patch for it.
here's how the current SVN has it.
Code:
if( !Core.ML )
return 1.5;
double bonus = 80;
if( m.Skills.Bushido.Value > 60 )
bonus += m.Skills.Bushido.Value;
if( m.Skills.Anatomy.Value >= 100 && m.Skills.Tactics.Value >= 100 && m.Skills.Bushido.Value > 100 ) //Bushido being HIGHER than 100 for bonus is intended
bonus += 50;
return 1.0 + bonus/500;
bonus = 80
Next line
if bushido is say 120. so add 120 to bonus. 80 + 120 = 200
Next line
Lets say i have no tactics. (add nothing to bonus)
Next line
1.0 + 200 / 500
= 1.4
Which is RIGHT
BUT if i lower the bushido skill it returns a wrong value.
bonus = 80
Next line
if bushido is say 61. so add 61 to bonus. 80 + 61 = 141
Next line
Lets say i have no tactics. (add nothing to bonus)
Next line
1.0 + 141 / 500
return = 1.282 That translates to 28.2% parry chance added to your parry, with only 61 bushido skill, the max at 120 is only 40% it's suppose to be. 60 skill - 120 skill && 16% - 40%
12% higher than it should be.
the change uses different math, and will always return the proper value. should anyway, i went over the numbers a few times.