|
Post by celebaglar on Jun 9, 2004 23:41:45 GMT
This was something me and Mads noticed when we fought our big showdown in the epic "The Taming of the Fey" game. Nearly 50 full armies were involved, plus a few extras, and we noticed that our mixed armies - the one composed of both riders and warriors - got wiped out practically twice as fast as everyone else. We've since set up a game to test this further, but the battle reports of Farflame, Mad Max and the original geezer from that big battle tell their own stories. (Apologies for the relatively large graphic content) Take note especially of the losses suffered by Trorn, Dreams, Lek and Utarg compared to everyone else. Note that by the time Rorthron fought them, it seems that Lek and Dreams had already lost all their riders. Considering that's only the xenos, when you count the losses inflicted by the other 20+ armies on each side you can see why the mixed armies got wiped out almost immediately. Surely this isn't quite as intended?
|
|
Natmus
Morkin Admin
Fight the power!
Posts: 4,518
|
Post by Natmus on Jun 15, 2004 20:01:20 GMT
Bill and I have had a test game where we tested all possible combinations of mixed armies (1250r/1250w) against either 2500 riders or 2500 warriors, lords and energy equal, attacking or defending.
The clear conclusion is that mixed armies take approximatly double the losses compared to the specialised armies. There is no doubt about it. It would seem that the combat algorithm allow one full attack against each part (rider or warrior) instead of inflicting losses relative to their part of the combined army.
To control these findings we had battles between specialised armies of riders and warriors, which the riders won without trouble, so if riders are better than warriors then a mixed army should be better than a specialised warrior army. Well, they are not!
Finally we had a big battle, with 8 armies on each side. 4 mixed, 2 rider and 2 warrior armies. In this battle it was very clear that the mixed armies took losses approximatly double those of the specialised, so in no time only the 8 specialised armies remained with roughly half their strength.
Clearly, something is very wrong, and it needs to be addressed ASAP. Jean-Yves, Bill and I have documentation for these test battles if you want it.
Ironically, as both Bill and I usually split riders and warriors in separate armies anyway, it was only as we had a large battle where we put everything into it, and some mixed armies eventually had to enter it as well, we noticed something was wrong.
|
|
merlin
Public Area Guest
Posts: 19
|
Post by merlin on Jun 15, 2004 20:14:10 GMT
I would like a recount on all my games then...
|
|
Freiegeister
Morkin Member
'Blasphemy is a victimless crime' - Dawkins
Posts: 1,126
|
Post by Freiegeister on Jun 15, 2004 20:24:15 GMT
I would like a recount on all my games then... Hmm, I thought Blood took Shimeril a little too easy with his small cavalry unit!
|
|
|
Post by celebaglar on Jun 16, 2004 23:34:18 GMT
As far as I can see, there are two issues to be looked at in reference to battle algorithms:
1) Cavalry seems far stronger than infantry. In a straight battle between the two, cavalry will outscore infantry by at least a 3:1 ratio, sometimes more. This gets even more pronounced as the battle goes on and a numerical advantage also appears. IMHO this is too much. Something like a 1.5:1 through to 2:1 would work much better.
2) Mixed armies suffer from "double jeopardy". Thay basically take twice the casualties, once for riders and once for warriors. A mixed army might beat an infantry one with a great deal of luck, though it will usually lose comprehensively. Against cavalry, mixed armies are nothing more than cannon fodder.
|
|
Natmus
Morkin Admin
Fight the power!
Posts: 4,518
|
Post by Natmus on Jun 17, 2004 7:50:36 GMT
A 1) Cavalry seems far stronger than infantry. In a straight battle between the two, cavalry will outscore infantry by at least a 3:1 ratio, sometimes more. This gets even more pronounced as the battle goes on and a numerical advantage also appears. IMHO this is too much. Something like a 1.5:1 through to 2:1 would work much better. To back this up with numbers, in out two test battles the riders won with the following margins: Amelista, who was attacked by 2500 dwarrows, had 1540 riders at the end of the battle, when all the wee ones had perished. Valethor, who attacked an army of 2500 dwarrows, had 1668 riders left at the end of the battle. It seems that the rider armies inflict roughly twice the damage the warrior armies does when they are numerically equal. Perhaps all the algorithm need is a fraction modifier: If an army is half-and-half each segment of that army should only suffer casualties equal to this fragments relative size of the entire army. So the damage is first determined against the whole army, then determined to each segment. Perhaps this priciple could be extended to larger battle. A simple example could be two friendly armies, one of 2000 men and one of 1000, are in battle. One day they suffer casualties of 600 men, and under the current rules they would each suffer 300 of them. And so I think: WOuld it be possible to add a fraction modifier of the total force, so the larger army of 2000 could suffer 400 casualties, while the smaller suffers 200, representative of their relative size?
|
|
|
Post by sparrowhawk on Jun 17, 2004 8:03:02 GMT
Riders do have quite a high hit factor compared to warriors. I could lower it slightly, although I do still want them to be superior fighters so that players have to decide how to structure their armies. I have been considering removing rider bonus in mountains and forests though.
|
|
|
Post by celebaglar on Jun 17, 2004 8:28:42 GMT
Riders do have quite a high hit factor compared to warriors. I could lower it slightly, although I do still want them to be superior fighters so that players have to decide how to structure their armies. I think lowering it slightly would help. At the moment infantry are just cannon fodder unless they're wheeled out in hugely superior numbers, so this accentuates the "pile everyone into one big army" effect. This would be a useful addition, as would giving a racial advantage to dwarves and Fey respectively.
|
|
|
Post by sparrowhawk on Jun 17, 2004 8:33:49 GMT
This would be a useful addition, as would giving a racial advantage to dwarves and Fey respectively. These two races already have movement bonuses in mountains/hills or forests, so I'm not sure whether they should also have battle bonuses. Unless you mean when fighting on those particular terrain types?
|
|
|
Post by sparrowhawk on Jun 17, 2004 8:38:28 GMT
It seems that the rider armies inflict roughly twice the damage the warrior armies does when they are numerically equal. Perhaps all the algorithm need is a fraction modifier: If an army is half-and-half each segment of that army should only suffer casualties equal to this fragments relative size of the entire army. So the damage is first determined against the whole army, then determined to each segment. Perhaps this priciple could be extended to larger battle. A simple example could be two friendly armies, one of 2000 men and one of 1000, are in battle. One day they suffer casualties of 600 men, and under the current rules they would each suffer 300 of them. And so I think: WOuld it be possible to add a fraction modifier of the total force, so the larger army of 2000 could suffer 400 casualties, while the smaller suffers 200, representative of their relative size? It's been a request since Alpha 0.1! I'll look into it again - I've an idea of how to implement it in a manner that should not be too database intensive (the fraction part is easy, the complex part is handling all the record relationships)
|
|
|
Post by celebaglar on Jun 17, 2004 9:28:42 GMT
These two races already have movement bonuses in mountains/hills or forests, so I'm not sure whether they should also have battle bonuses. Unless you mean when fighting on those particular terrain types? I do. Dwarves should be a real problem in the mountains, but at the moment they're just like every other infantry: not very good. Since there are so few of them anyway, it would be good for the game balance to give them a fighting bonus in mountains. Similarly, the forests are the homes of the Fey, so they should be more menacing there than any other race. Being quick is all well and good, but not very useful if you fight like a fairy when you get there.
|
|