Slitherine

Hi

The MF BG 24 Archer that was disrupted and fighting 3 BG's should have rolled only 1 dice instead of 2.

In the end as hilighted on the console log the
0]BG[7] Helot javelinmen was selected
[2] BG[7] already took an anarchy charge test this turn
[1] BG[7] Helot javelinmen melee attacks BG[24] Median bowmen
[2] BG[7] vs BG[24] combat -> quality re-rolls are enabled
[2] BG[7] melee combat dice no: 2 *1/2 => 1 die
[2] 2(LF vs no LF,frg) (base)
[2] *1/2(frg,SDrd,sdrd) BG[7]
[2] BG[24] melee combat dice no: 4 -2 *2/3 => 2 dice
[2] 4(most BG's are rolling 4 dice in combat) (base)
[2] -2(BG[7] supporting EBGs: BG[1],BG[8])
[2] *2/3(dis,Drd,drd) BG[24]
[2] BG[7]'s melee combat POA: = 0
[2] BG[24]'s melee combat POA: +1 = 1
[2] +1(better armor vs no HW,El,CH,ART,BW) BG[24] vs BG[7]
[2] BG[7] dice roll: [(6)5] (-1 POA [5,6]) => 1 hit..
[2] BG[24] dice roll: 2,1 (+1 POA [4,5,6]) => 0 hits..
[2] BG[7](1 hit) vs BG[24](0 hits) combat => attacker BG[7] wins
[2] BG[24] (defeated) lost 4.07% (41) strength
[2] BG[7] (winner) lost 0.19% (1) strength
[3] BG[24] must pass a melee combat CHT
[3] BG[24] CHT -> quality re-rolls are enabled
[3] BG[24] CHT test score: 2+6 -1 = 7 >= 7 => passed
[3] -1(dis,SDrd,sdrd) BG[24]
[1] BG[7] Helot javelinmen was deselected





Tks!

this is not a bug. the sequence for computing the BG[27]'s dice no is: as you can see the *2/3 modifier for being disrupted appears and it is applied in there.

the result is 4-2 * 2/3 = 2 * 2/3 = 4/3 = 1.33 = 2 (as the result is rounded to an integer using ceiling, same as in old FoG(RB))

Hi Dan,

Tks for the answer.

I have to disagree in here. I tested the same situation on FOG RB and the result is that the BG facing 3 Bg's in melee (two are in support) must roll only 1 dice if disrupted previous the melee combat.

See the screenshot from FOG RB:



As you can see the defender BG rolled only one dice: 3 for been disrupted and minus 2 dice for the two enemy BG's acting as support.

As you say and as the console shows FOG U takes in consideration the disrupted modifier so I think the problem is when the result is rounded:

the result is 4-2 * 2/3 = 2 * 2/3 = 4/3 = 1.33 = 2.
Maybe 1.33 should be rounded to 1?
1.5 or above would be rounded to 2?

Tks!

Tks

There's no such thing as 3 for disrupted. it's only 4 * 2/3 = 2.66 which again should have been rounded or ceiled in order to obtain a 3. And the disrupted modifier is applied after the support one anyway. In FoG(U) combat dice no computations, I'm getting the base dice no, subtract EBG's no of supporting BGs, multiply the cohesion modifier with the strength modifier, and only at the end multiply the (base - sup) integer with it, to keep the integer computations as longer as I can (and thus avoiding rounding errors).

I remember very clearly that I've read somewhere a statement (by Keith most probably) that the dice no computation result was ceieled, and not rounded or floored. I'll have to ask Chris and Robert about this.

I think 2.66 is Ok to be rounded to 3.

Anyway. Probably the best would be to ask the people you said.

But clearly we have a difference in here with FOG RB.

Cheers!

addressed in revision 689

// modified GP.CombatDiceNoCompute() by moving the computation of supporting number of dice modifier after applying the cohesion and strength modifiers to the base dice no