In reference to my exploit report here: https://www.games2gether.com/endless-legend/forum/22-gameplay-and-ai-issues/thread/6035-critical-endless-siege-exploit



Due to this there needs to be some sort of written rule that the players can inforce where the game does not.



I'm posting here to get some feedback on my original construction which looks something like this:



"In a siege battle, if the attacker uses retreat, he has to immediately cancel the siege on this city with all armies, and may not siege it again until one round end has passed."



---- There is one small loophole in this rule, which I want to draw your attention to. Players could wait until a few seconds are left on the round timer to start a siege. This should be considered expoitive behaviour and dealt with in the same way as not following the rule. What this means is that either the city defender has no time to attack the sieger, or even if the battle does happen, the round timer will probably run out. ----



So if you attack in turn 2, battle is resolved during turn2, you must cancel all sieges and may siege again in turn3.

If you attack in turn 3, but the end of turn timer runs out before the battle ends, you must cancel in turn 4 and may siege again in turn 5. (( This means you can still exploit sieges every 2 rounds, so this is really not the wanted outcome ))





Any feedback is most welcome.

Sir-Rogers wrote:
"In a siege, if the attacker uses retreat

I'm not sure what this means. Do you mean the attacking player cancels the siege, the besieging army is attacked, then retreats or the attacking player loses a siege? Only the defender in a battle can retreat.



I think you mean "If the besieging army is attacked". In this case the most elegant solution would be to say "a besieging army cannot retreat from an attack". I dislike the Retreat mechanic anyway and this tactic demonstrates why.

Propbuddha wrote:
I'm not sure what this means. Do you mean the attacking player cancels the siege, the besieging army is attacked, then retreats or the attacking player loses a siege? Only the defender in a battle can retreat.



I think you mean "If the besieging army is attacked". In this case the most elegant solution would be to say "a besieging army cannot retreat from an attack". I dislike the Retreat mechanic anyway and this tactic demonstrates why.

Fixed my post, the word battle was mising, which caused the confusion. The retreat mechanic is very valid.



Assume you would be sieging a city where you have 6 units sieging and 3 units are defending. Now the defender buys out 6 units and attacks you with a total of 9 units against your 6.



You would lose, if the rule were as you mentionned above the attacker would lose his whole army. That is not fair towards the attacker, it is totally unfair actually.



My version is better as it doesn't discriminate against either party, it is just a fix to the siege exploit.

OK, makes more sense.



Probably requiring a siege to cost the action point, and retreating taking it away (which I think already happens) would be easy to implement...

Hopefully it's not hardcoded, because retreat-siege needs to be adressed.

If you have multiple units in your city can't you make a one-unit army, attack the sieger (with backup from the garrison), garrison your one-unit army, then repeat? If you can't it's kind of a silly strategy (mostly for broken lords), but it's pretty easy to fix; if a sieging army retreats/is defeated then the siege is broken for that turn, which makes a fair bit of sense on both a mechanical and flavor standpoint.

Mawootad wrote:
If you have multiple units in your city can't you make a one-unit army, attack the sieger (with backup from the garrison), garrison your one-unit army, then repeat? If you can't it's kind of a silly strategy (mostly for broken lords), but it's pretty easy to fix; if a sieging army retreats/is defeated then the siege is broken for that turn, which makes a fair bit of sense on both a mechanical and flavor standpoint.

Reinforcements still use their attack if they participate in a battle.



Problem is the retreat-heal trick. You siege, force the sieged city to attack, call retreat on your attacked unit (battle cancels, retreating units loses 50% max hp), then dust Heal the attacked unit to full and the sieged army lost their attack.



Several ideas to address this:



1-. A city can only be sieged once per turn (a city that starts their turn sieged counts). Retreating from any siege breaks the siege..

Optional: A city can be sieged once per turn... per FACTION.



This would address the problem, but still has some issues against lords.

The optional rule makes alliances still able to perma-siege, but I feel this is somewhat OK. It should be much more dificult to break sieges from several factions.



2-. A defending army that breaks a siege gets a Morale boost and recover their action point.

This will grant the defending army a counterattack. The defending army could still retreat twice... Getting killed (retreating deals 50% max health damage). In Broken lords case, don't know if BL armies can Dust Heal when it's not their turn (stil, it will cost them double the amount of dust).



3-. A city breaking a siege will be granted "siege immunity" for several turns.

How many turns might depend on several factors, some ideas could be:



- Based on remanining fortification points (ex: 1 turn per XXX points, minimum 1 turn). Nice since this makes you decide if you want to wait a little more for extra units to ensure victory, or to try a faster push for extra immunity turns, but several factions might have high advantages here.

- Based on the defensive improvements built on the city (not all buildings add to the immunity, and advanced ones may add more than one turn). I actually like this idea since you may be more interested on getting defensive improvements that are otherwise a little lackluster, and older buildings with low fortification points may still be useful.

Propbuddha wrote:
Probably requiring a siege to cost the action point, and retreating taking it away (which I think already happens) would be easy to implement...

I don't think this is a good idea. Rather than using the action point here, it should just be a different flag called "retreated". if you have retreated in a round, your unit is flagged as having retreated. and as such cannot siege until next round