AI fire at long range [Warning: Spoiler]

[UP844]

AI-controlled infantry does not fire at all at targets at long range: often, this means losing some excellent shots and moreover, makes life easier (too much easier!) for the human player, as he can stay just one hex beyond the normal range and be certain that his troops will never be fired.

I do not expect the AI being devious as one of my ASL pals (who left a building almost unguarded in a RB game... and then blew up both the building and my Assault Engineers inside [:(] using Set DCs) but I'd like to have an opponent as deadly as possible.

In the attached images there are two examples taken from a game of "Buchholz Station". Both U.S. stacks include a leader (C or C*), 3 1st Line squads, 1 MMG and 1 BAZ44.

In case #1, the Germans first moved (green arrows) two squads in a stone building: the U.S. stack fired at them (range = 9 hexes) using the MMG only. The squads did not fire.

Later in the movement segment, half a dozen German squads moved in the lower part of the map (being careful to never move within 6 hexes from the stack). The U.S. squads, again, never fired. Even with 2 or 3 hindrance hexes, movement and terrain modifiers (the infamous FFNAM/FFMO deadly duo [:D]) would have negated or at least greatly reduced their effect (the picture reports the net fire DRM).
[Playing the U.S. side, I would have fired, if only to slow the German advance. Even with a single squad firing each time, the chances of breaking or at least pinning a German squad are reasonable, in my opinion.]

In case #2, the U.S. stack fired at the German units (range 7 hexes) using the MMG only as soon as they Assault Moved in the hex (a stone building). On the subsequent Movement segment. I moved several nearby units (3 squads, 1 MMG and 1 HMG) in the building and in the adjacent woods hex to make a large fire base: nobody fired at them. In the subsequent US Fire, the MMG (only) fired again on the woods hex with no effect. In the German Defensive Fire (and in the subsequent Fire) the U.S. stack was wiped out by German fire.

I would not bother you with other images, but this "normal range" issue also applies to MG: in fact, in "Silesian Interlude" a Russian stack containing an ATR (range=12) and a LMG (range=6) fired the ATR at my infantry (at range between 7 and 12 hexes).

By the way, the issue is much more severe when Germans vs. Russians are involved: the Germans may stand at 5 or 6 hexes and blast away the shorter-ranged Russians without reply.

[UP844]

Excuse me if I am becoming a nuisance, but this bug is taking away much fun from playing TotH [:(].

Today, I played "Silesian Interlude" and by the end of turn 3 there were no Russians on the map. I have been able to wipe out all the Russian infantry by firing at them with German infantry standing just 1 hex over the Russians' normal range. My infantry was in open terrain and the Russians did not fire a single shot at them [:(] [:(].

[Gerry4321]

That is very poor. There is no penalty (other than breaking a MG) for firing so it should fire at long range for sure. In ASL you had to wirry about snipers.

I have the same situation in With a little help from the StuGs. And the Russians have 8 FP with the equivalent of an 8-0 leader. My unit is in a stone building but they need to take that shot.

[Paullus]

Thank you for bringing this AI behavior to our attention. We will off course add this to our list of improvements we want to make to the AI.

[genesismwt]

I think its a slippery slope either way we go. If we know that the AI will most likely shoot at the first target it sees, we will just make the first target appear at long range.

I don't know, I am only guessing, that to have the AI hold its fire at long range was a design decision, as it is probably too easy to bait the AI into long range bad shots.

Maybe, that can be a toggle, like the ending scenario when there are no more opponents. Players have the choice as to whether AI takes long range infantry shots, or not.

[Peter Fisla]

The AI was programmed not to primarily focus on long range fire because I didn't want the AI to waste long range shots when there might be possibility of a human stack being near by.

[UP844]

I understand your point, but the AI never fires at long range regardless of the Fire Phase. Anyway, not every long range shot is necessarily a bad shot. Knowing the AI will not fire will lead the human player to move in the open without fear of being wiped out by a low-factor (and high-negative-TEM [:D]) shot.

A smart player must only stop 1 hex away from the enemy normal range and then may fire forever without worrying about return fire. Russians, which usually have large numbers of shorter-ranged infantry, are especially disadvantaged. On the other hand, they might stand 7 hexes away from Germans and lay down a huge number attacks.

[Gerry4321]

Peter, you can allow them to fire at long range in the Defensive Fire segment (not in the movement segment) - the player cannot game the system then.

[Peter Fisla]

I will see what I can do guys...thanks for your feedback.

[UP844]

Thanks once more, Peter!

By the way, I have a vague notion that checking LOS and deciding if firing or not up to long range (especially during the movement segment) implies additional computations and, consequently, might imply some delay in the larger scenarios. I think, however, that the additional game value (making the AI much more powerful without "cheating") will be well worth it.

Additional rant: I would like to see additional features as anyone else (call me a masochist, but I'd like leading M13/40s of the Ariete Armoured Division in the Western Desert [;)]). As Peter said when the game was released, however, let's focus on getting rid of the bugs affecting the basic mechanics.

[Peter Fisla]

ORIGINAL: UP844

Thanks once more, Peter!

By the way, I have a vague notion that checking LOS and deciding if firing or not up to long range (especially during the movement segment) implies additional computations and, consequently, might imply some delay in the larger scenarios. I think, however, that the additional game value (making the AI much more powerful without "cheating") will be well worth it.

Additional rant: I would like to see additional features as anyone else (call me a masochist, but I'd like leading M13/40s of the Ariete Armoured Division in the Western Desert [;)]). As Peter said when the game was released, however, let's focus on getting rid of the bugs affecting the basic mechanics.

Yes, enhancing the AI adds more processing for the CPU and of course I will be working on optimizing and tweaking the AI. At the same time please be aware that the AI will never be perfect. So when designing scenarios with the AI in mind, play to its strength and avoid its weakness