Missile fire effectiveness question

[Foz]

This is especially true if the intended target is cavalry. Horses are long critters, and their big, vulnerable sides are apparently accounted for in the game mechanics. At least that's been my experience since I accidentally discovers that missile cav running alongside enemy cavalry units dealt a lot more death than units chasing them.

Wanted to clarify this point, and relate it to the problems of the bugged unit animations for 2-handers. Basically, from the behavior of archers shooting at cav (noting that it varies in effect based on the horse's cross-section at any given moment) and the fact that 2-handers with bugged animations are completely ineffective in combat, we can surmise something about the game engine: that it is in fact modeling the entire battle in a physics engine, and determining hits based on battlefield objects intersecting. To clarify, this means the game actively tracks the flights of arrows, the swords of swordsmen, and various other in-game weapons, and determines when to check for hits by figuring out when a damage-dealing object has intersected a man or horse.

Various observations can be made to back up this theory:

1. Missiles can impact men other than they are intended for, even in units that are not being targeted.
2. Archers are more effective shooting into the broad side of a horse.
3. Various bugged units are made worthless by the animation bugs.

#3 immediately tells us that something more complicated than unit-to-unit or even man-to-man calculations must be going on. If either of those were the case, animations could not affect combat at all, and the animation-bugged units would perform perfectly normally while only looking wrong. In short, this point means the animations are actually USED in determining combat results.

#2 Could be explained away by modeling cross-sections from maybe 8 directions, and approximating incoming fire as from one of those 8 directions. However, it's actually just as easy to determine weapons intersecting targets, since in order to depict the battle graphically at all, the game already must be utilizing various elements that represent the arrows and men involved in combat. With that framework already in place (i.e. the battle already exists on a coordinate grid for the purposes of graphics) it's a logical step to use it to determine battle results.

#1 solidifies the point even more. There's no reasonable way to determine who (if anyone) has been hit by a flying arrow except to actively calculate if it intersects anyone. The fact that a stray arrow is capable of killing anyone at all that it contacts indicates that the mechanic is again not stat driven, but modeled as realistically as possible via the actual arrow flight path. Archers even adjust their fire to avoid walls and other blocking terrain elements, further suggesting that the game is treating the arrows as if they are physical projectiles which can be obstructed by anything and everything that might get in their way.

So what does this mean to all of us? Well, primarily, it means that what you see on the battlefield, and what makes sense to you intuitively, is what you get. If you move a unit into your archer's line of fire, the archers adjust the fire over them, or your unit gets shot if the friendly fire is unavoidable. It means that a whole lot of work that we never see has gone into the battle system for this game, and as a result the underlying battle mechanics are a step above the simpler systems usually employed in this genre. In short, it means we have a more natural and realistic field in which to wage medieval wars than games so far have typically been able to deliver

[dopp]

Urgh, that same physics engine in RTW was responsible for my archers shooting their OWN FRONT RANK in the back of the head for a really long time...

[Doug-Thompson]

That exchange was a telling little vinette on the whole forum.

Some of us (the foz 4) are always excited by the potential shown. The rest (dopp) are always disappointed when the latest version of TW doesn't live up to it.

Anyway, I tried my little experiment, and found that putting the unit in big squares or lines made no difference -- because the approching enemy infantry would always veer to the right. That way, their shields covered the the fire from the right and from the front.

This was exactly the effect RTW struggled so much with in phalanxes. It's a pretty logical response.

So, veteran infantry players who have seen this all the time, how do you exploit it?

[dopp]

You might want to be careful about categorizing people so broadly; there is a difference between people on a rant about how the game sux and those who point out where and how the game can be improved.

[Doug-Thompson]

You might want to be careful about categorizing people so broadly; there is a difference between people on a rant about how the game sux and those who point out where and how the game can be improved.

You're right. I remember complaining of being categorized myself a few times. Sorry for the broad brush.

[dopp]

No problem.

I've noticed several complaints about the very high fire arcs of M2TW not quite matching the firing animations, especially for crossbows and such. I think that it's necessary to prevent a repeat of the RTW friendly fire bug.

I thought missile units stopped firing on their own when the enemy routed. I have relatively few friendly fire incidents these days, but then I use gunners mostly.

[Ars Moriendi]

@Katank & Doug-Thompson :
thanks for the tips, guys. I always felt I'm a little "raw" when it comes to using my archers, just letting them fire at whatever's in front of them then pulling them back behind the line and forgetting they exist. Discussions like this make me appreciate the finer points in the M2TW battle engine.

-----

@the_foz_4 :
Your theory is probably true when it comes to missiles, but I doubt, however, that frame-by-frame collision detection computations come into play in hand-to-hand combat. They're necessary in missile fights because you cannot know in advance if the arrow trajectory will connect to its predicted impact point (target moves, obstacle gets in the way, etc.). But in h2h, the attacker and defender are connected already, all you need to do is roll the results based on att/def ratings and display the animations accordingly.
About the bugged units : it is my impression that the animations themselves are okay, it's the AI triggering the attack that is queer. The effectiveness of units is of course lower if the animations take more time (delaying the next roll / combat round).
But I'm straying off-topic here, let's get back to archers :).