Cutoff Rule Change

[McLeopold]

Because a bot with a large hive count cannot be taken if they do not move off of the hill, we will stall this cutoff if there is any dead ant on top of a hill not owned by the dominant bot. This gives the dominant bot time to drain the hive count to 0 and score for razing the hill.

This has been updated in the specs.

Here is an example game of a bot that got robbed of a win.

[Zaph]

[Parasprites]

Just to clarify, this only applies to the Ants Not Razing Hills ending, right?

Also, I think it should apply if there's a dead any within 1 square of the hill. Because it is possible for the defender to move off the hill, but since moves are simultaneous, the attacker can't know that. As an attacker, you have to wait until the hill stops spawning, even if they move off the hill, because it is risky to gamble that they will move off the hill on a particular turn.

[McLeopold]

Yes, other cutoffs could be invoked, like the turn limit. This makes gathering food equal to 1 per hill a viable defense.

If the cutoff timer is invoked, then you have 85% or more of the ants and can afford to suicide attack the hill. If he does move, the game would end right there. If not, it was a 1 for 1 trade and you still probably gained more in gathered food.

If the scenario is more complicated where the defender has free ants away from the hill, then that bot deserves to waste your time by making you guess. It would probably be best to just suicide every enemy ant and hill at that point to move into the end game and close it out.

[fourmidable]

I have a related question. When evaluating if the dominating players as 85% of the ants, are the ants in the hive considered, or only on the map?

It seems to me that some games stop very quickly even when forces are balanced on the map, but there is a large unbalance in unhatched ants. On some maps, it is possible to collect food much faster than spawning ants, leading to huge hive population. This is a problem as it is not possible to know the number of unhatched ants of the opponents, and thus when the "not razing hills" might trigger. Even knowing time is short, it might not be possible to mount a successful attack with the ants on hand. I believe the intent of this cutoff would be best served by considering only the hatched ants against the 85% threshold.

[McLeopold]

hive counts are included