On the last bugfix I wrote that the shooting attempt in the battle uses only aiming and leadership. This is wrong – it also uses endurance or hiding, depending on what side you play on. This is a bug that got introduced during reworks on the battlesystem in 1.29 and that is going to be fixed soon. Sorry for the confusion.
Looking through the code I found an ugly bug that went through undetected: The battle map display did mix the defense and attack values for sectors – so in fact, the big tower of the big fort gives a 27 attack bonus, not defense bonus – and vice versa. It wasn’t detected since its introduction in 1.22 (1.23?) because no one actually looked at the original map editor and checked the values accurately – the values were right, it was simply just twisted in the display.
Another bug that I found is within the help: While it was internally documented and implemented correctly, the public documentation does not mention, that shooting and dodging rolls have minimum values: If an attacker shoots, a bonus of 25 is added to his attempt, whereas the defender gets a bonus of 10. This was not properly documented and thus should be fixed soon.
I apologize for these errors that existed for so long undetected.
The bonus gets a bit weakend: First of, the bonus will only affect the sector bonus itself, the tower’s character bonus is not increased. Secondly, the factor is reduced from 40% to 30%. Taking the example of the screenshot here which finally shows the actual numbers:
– Character specs: 31 Points on dodging, 30 on hiding, 100 on aiming and 82 on leading.
– The tower grants +20 on hit chances and 27 on defense and 20 for character’s bonus (this is where I noticed the map bug)
– The attack value is increased per default by 25, the defense value is increased by 10 (this is the game rule that was not present in the help)
– Character premium is off (+30% sector bonus)
Result defense:
31^.4 + 30^.4 + 82^.4 + 20 * 1.3 + 20 + 10 = 3.94 + 3.9 + 5.83 + 26 + 20 + 10 = 69.67
Result attack:
100^.4 + 82^.4 + 27*1.3 + 20 + 25 = 6.31 + 5.83 + 35.1 + 20 + 25 = 92,24
The worker bonus is now really small: in case of a premium player, the defense is raised at a maximum by 12 and the attack is raised at a maximum by 16.2 – so it contributes about up to 10-20% to the bonus. On other sectors it’s respectively smaller. In the given example, the 30% bonus results in a value that is only about 10% higher than the “normal” bonus would be.
The soldier bonus is now also done and it works like this: Soldiers are adding their leading skill points by up to 50% (Premium) on their own leading bonus and their 4 neighbors that are within the same sector. So let’s take the same character and convert him to a soldier and put him on a soldier’s tower of the same level. The bonus calculation is now changed in this way:
Result defense:
31^.4 + 30^.4 + (82 * 1.25)^.4 + 20 + 20 + 10 = 3.94 + 3.9 + 6.37 + 20 + 20 + 10 = 64.21
Result attack:
100^.4 + (82 * 1.25) ^.4 + 27 + 20 + 25 = 6.31 + 6.37 + 27 + 20 + 25 = 84.68
A single soldier is thus a bit weaker on his character tower than a worker. If multiple soldiers of the same values stack, the result would change in this way (the bonus is maxed when 5 soldiers form a cross, the one in the middle receives then 25%*5 of the leader skill):
Result defense:
31^.4 + 30^.4 + (82 * 2.25)^.4 + 20 + 20 + 10 = 3.94 + 3.9 + 8.06 + 20 + 20 + 10 = 65.9
Result attack:
100^.4 + (82 * 2.25) ^.4 + 27 + 20 + 25 = 6.31 + 8.06 + 27 + 20 + 25 = 86.37
Thus, the worker is still in advantage if he’s on his character’s tower. With a premium, the worker’s advantage increases more drastically (as the life points of the soldiers are doing). So the new bonus increases the final bonus just by .5 to 10 points. Given the other values, this is mostly only making sense in case of sectors without a bonus – or when soldiers support players that have nearly no skill points on leading.
Since character classes are now quite important in fortbattles, we decided to display the classes in the overview. We had to sacrifice the small person picture for getting some more details, and the symbols are fairly primitive to simplify identification. We played a bit with coloring the lifebars and other things, but it turned out to be quite chaotic. So this is what we came up with:
