Cohesion drop anomaly

[Cunningcairn]

The unit circled is not in a block that is classified as an obstacle despite it saying it is in one and has attacked a unit that is not in a block that is classified as an obstacle. Yet when attacked in the flank by the bowman to its left it does not drop a cohesion level. Why is that so?

No cohesion drop.jpg (792.08 KiB) Viewed 1294 times

[Paul59]

The Assyrian Guard are protected by the ditch, so do not drop cohesion when attacked by the archers from across the ditch.

Ditches are tile side objects, so they protect units on both sides of the tile edge, as long as the attack crosses the ditch.

[rbodleyscott]

As Paul says, all “obstacles” affect tile sides and not whole tiles. Fortifications only work one way, but ditches work either way, depending on who is attacking [across] the obstacle.

Any unit attacking with 45 degrees of across the obstacle tile edge count as attacking across the obstacle.

The tile tooltip shows “obstacle” if any of the tile’s 4 sides are affected by an obstacle.

[Cunningcairn]

Thanks for the explanation. So if a unit behind a ditch attacks across it then its flanks will still be protected from other units attacking it across the ditch from another square? I remember reading that if you attacked from inside an obstacle you were no longer protected by that obstacle. Is that not the case?

[rbodleyscott]

Thanks for the explanation. So if a unit behind a ditch attacks across it then are its flanks will still be protected from other units attacking it across the ditch from another square? I remember reading that if you attacked from inside an obstacle you were no longer protected by that obstacle. Is that not the case?

If you attack across an obstacle you lose all POA benefits that that obstacle provides.

If an enemy then attacks your flank across another obstacle, that obstacle will still prevent the attack from counting as a flank attack (if your unit is foot), but your unit won't count any POAs for the obstacle (because it is flagged as attacking).

That is the way it is coded. It could have been coded the other way, but if you want some rationalisation for the way it is, the logic would be that even if the unit is not actively defending the obstacle, the obstacle will slow down the enemy flank charge enough to permit a response by the nearest ranks.

If you are unsure what will happen in a particular case, check the detailed charge tooltip.