I think that’s a decent compromise.
Very useless buff though. The donjon’s problem is not ‘utility’. That, it provides plenty already: can fire arrows, create units. The problem is that Sicilians are actually very vulnerable to tower rushes, as donjons are hard to wall in, and at 200 stone a piece, you won’t be able to build them at the same rate as towers – meaning that you can build 2 towers to range each other for 250 stone, for donjons this is 400 stone. 400 stone is a significant investment in feudal age.
1 Like
Not true. Unless your talking about Castle age sergeants, no they don’t. OP is talking Feudal regardless. They have even stats in feudal age with men at arms. The only different is maa have 1 higher attack then sergeants, and sergeants have 1 higher armor than maa, but equal when fighting. Run a scenario, whoever gets first hit wins.
They are better in castle age, by 1 armor over the long swordsman, but i would still not trade them against long swordsmen as it’s not cost effective. If you take even resources, the long swordsmen win (they benefit from supplies where serjeants don’t).
So to the OP point, they are in the feudal and castle, just really expensive infantry compared to barracks units, and it’s not until elite that they have somewhat of an edge over champions, but the cost difference still doesn’t add up and champions are way more cost effective.
It appears even, just open up editor and make them fight to see who wins. Serjeant everytime in feudal
I did, 7 times. I even did 2 serjeants vs 3 maa which you claim will win. It’s not even close. maa win every time
Are you sure you have both players in feudal age, and set the PC personality to “none” (otherwise they act dumb and might not attack sometimes in tests).
Ill redo my tests and get back with you in a few hours
video of my test. For real took your advice and still see similar results. What does none personality do?
1 Like
None personality does what is says. The opponent AI stops behaving. So only unit AI works.
The unit AI is simple. If you see an enemy in sight, chase it on aggresive stance and attack it.
With the normal AI, the AI thinks of exploring the map with a military unit just as it would do in standard games. This happens even if it is getting attacked.
What @SauravT said for the “none” personality. You have to put in on for testing units against each other otherwise tests are really inaccurate. You have a few areas in the test where two units didn’t even attack each other but stood there. Also, in at least one 3v3, the maa came out ahead. Your results of the serjeants coming out ahead on the larger groups could be simply because the maa are pathing differently and trying to explore. Do a test on none for personality, and put a 9 serjeants vs 10 maa. The maa will come out ahead. If you flip it (10 serjeants, 9 maa), the serjeants will come out ahead. Even tests are hard to decide on which unit is better simply because it can be the flip of a coin who comes out ahead., and sometimes with staggering numbers (like 5 or so leftover for one side).
Thanks for taking the time to do the video! 
OP asked for small buff. So I suggested one. If he asked big buff I could suggest 5 more.
no it doesn’t.
1 serjeant has 45 health, 5 attack, and 1melee armor.
1 man at arms has 45 health, 6 attack, and 0 melee armor.
that means in each attack a serjeant is going to do 5 damage to a man at arms and take 45/5 9 hits to kill that man at arms
and the man at arms will with each attack do 6-1 armor for 5 damage and take 45/5 9 hits to kill that serjeant.
they both attack at the same rate. the only way the serjeant is almost beating 2 man at arms is if the man at arms are spending more time walking around then attacking.
so you’re doing something wrong.
1 Like
I feel like this idea is wrong because first, it would make beating donjons just obnoxious, and two, if you buff them in feudal like that it makes it so that you can’t buff nor donjons nor serjeants elsewhere.
After putting ai to none maa when out numbering serjeant defiantly win. I didn’t know that trick. Even numbers serjeant win. Uneven numbers maa win
you have to run multiple tests to account for different variables like unit pathing, who gets the first hit, etc. simple numbers shows that man at arms vs serjeant is going to come down to who gets the first attack in.