Choice of Defense

[Celtic_Minstrel]

I think I'd forget about inputting the actual defense weight and just set it to 0, saving the normal defense weights in [variables] for easy restoration later.

[nuorc]

I guess I would need that explained like you would to a windows/clicking DAU...

Would a dau player be confronted with choosing a value for some variable or would there be sth like

Choose behavior on defense:
- aggressive (deal more damage)
- defensive (take less damage)

Is the first part a general setting or prerequisite for the "Change defending weapon" part?
(Maybe "Choose defending weapon" as I might go with the default?)

Would the "Change defending weapon" be a general setting or come up during each attack?

THX!

[Ravana]

I intended that be unit setting, but as it has no in game effect now it is up to you how you make it.

As i wanted that change ai values it is change, not choose there.

[Dwarven_Void]

I think that it is fine as it is.

My logic is that if a mad orc suddenly rushes at you with a big sword, you don't have time to ask your commanding officer what to do - you take an instinctive decision based on what feels best to you.

So going back to the Shaman vs Druid, the Druid can't go "Oh dear, that orcish archer is shooting at me. I'd better stop time, go ask Lord Kalenz what I should do, then come back and use a slow attack" but will instead go "Ahh! I'm being shot at!"

This(rather silly) train of thought to me explains why it's ok for the AI to choose defence attacks.

[JaMiT]

Since I'm interested enough to speculate, but not enough to try things out:

1) What happens if you join a multiplayer game and one of your units advances after being attacked by an AI unit?

2) What happens if you host a multiplayer game and one of your units advances after being attacked by an AI unit?

[nuorc]

This(rather silly) train of thought to me explains why it's ok for the AI to choose defence attacks.

Well, their drill sergeant (me) tells them every sweet day:
priority 1: survive
priority 1b: take as little damage as possible
priority2: on your attack: kill, Kill, KILL!!!!11!1!!!!

So said drill sergeant is not happy when the (sometimes not too clever) ai messes with his orders...

[tuggyne]

I think that it is fine as it is.

My logic is that if a mad orc suddenly rushes at you with a big sword, you don't have time to ask your commanding officer what to do - you take an instinctive decision based on what feels best to you.

So going back to the Shaman vs Druid, the Druid can't go "Oh dear, that orcish archer is shooting at me. I'd better stop time, go ask Lord Kalenz what I should do, then come back and use a slow attack" but will instead go "Ahh! I'm being shot at!"

This(rather silly) train of thought to me explains why it's ok for the AI to choose defence attacks.

This would be plausible enough were it not for the long timeframes involved. Nearly all attacks resolve over a period of an hour, often closer to several hours, as ToD rather clearly shows. If you have anything more than a minute or two to respond it shouldn't be all that difficult to pick a more sensible option. (Also, the aforementioned drill training.)

[Dwarven_Void]

This would be plausible enough were it not for the long timeframes involved. Nearly all attacks resolve over a period of an hour, often closer to several hours, as ToD rather clearly shows. If you have anything more than a minute or two to respond it shouldn't be all that difficult to pick a more sensible option. (Also, the aforementioned drill training.)

Sure but lets face it - Wesnoth seems to run to a timescale of it's own, with the length of hours, day, quarters etc, being bent according to the need of the current scenario. A bit filmsy as a rationalisation I admit, but my point is that time has never yet stood in the way of smooth gameplay (and jerky "can't do anything till they choose an attack" isn't smooth by any measure.)

In regard to the "drill sergeant" point, maybe a better solution would be to allow you to control the general patten of a unit via the right click menu.

Eg/ you right click your druid and tell it to "be defensive" and take minimum damage. This could be done by unit, or by the whole army - basically it boils down to changing AI behaviour in terms of "agression" and "attack weight" values. It could probably be done via lua (although I know nothing about lua).

[MetalKing]

I might be wrong, but I consider 'strategy' to be the general approach on how to achieve victory in the end,
while 'tactics' cover more acute aspects, like attacking this or that unit (first), withdrawing/pushing forward etc.

I don't think you are wrong but the Term Strategy isn't and can't be general mandatory defined (..) but your Understanding seems to be compatible with my. I consider Strategy to be larger in Scope/Range (of: ), Period of Time, Planing and Preparation while Tactic is quite the other Way around, The most surprising Fact is that Tactic can change into Strategy and vice versa if you change the Scope. A FootBall/Soccer-Team can get instructed before a Match with an Strategy to win the Match but while the Match the Players have also to consider tactical Issues as the Player can't rush upfield to take a shot at the Goal like Strategy said while the Ball is rolling downfields. Strategy say play upfield but Tactic may require to run downfield. One is Planing and one is Fact.

So here is another Understanding about Strategy and Tactic but probably there are so many Understandings like People are.



In DefenseLines are sometimes Units not at Edges but even Corners and I don't want that Units as WeakPoint to kill Attackes and thus clean Space for more Attackers. Brief and generally: I want Units which PrimaryObjective is to keep a DefenceLine to survive and it doesn't matter if they can kill an Attacker resp. killing an Attacker can endanger the explained P-Objective.

The Matter Chocie of Defense seems to touch Interests so this Thread's Topic is surely anyway relevant.

[tuggyne]

This would be plausible enough were it not for the long timeframes involved. Nearly all attacks resolve over a period of an hour, often closer to several hours, as ToD rather clearly shows. If you have anything more than a minute or two to respond it shouldn't be all that difficult to pick a more sensible option. (Also, the aforementioned drill training.)

Sure but lets face it - Wesnoth seems to run to a timescale of it's own, with the length of hours, day, quarters etc, being bent according to the need of the current scenario. A bit filmsy as a rationalisation I admit, but my point is that time has never yet stood in the way of smooth gameplay (and jerky "can't do anything till they choose an attack" isn't smooth by any measure.)

In regard to the "drill sergeant" point, maybe a better solution would be to allow you to control the general patten of a unit via the right click menu.

Eg/ you right click your druid and tell it to "be defensive" and take minimum damage. This could be done by unit, or by the whole army - basically it boils down to changing AI behaviour in terms of "agression" and "attack weight" values. It could probably be done via lua (although I know nothing about lua).

I'd definitely agree that it'd be best to pick a preference for high damage dealt, standard, or low damage taken, rather than choosing the defense every time. I was mostly pointing out that it could work the other way too with no real difficulty.

[Celtic_Minstrel]

This would be plausible enough were it not for the long timeframes involved. Nearly all attacks resolve over a period of an hour, often closer to several hours, as ToD rather clearly shows. If you have anything more than a minute or two to respond it shouldn't be all that difficult to pick a more sensible option. (Also, the aforementioned drill training.)

Time of Day can define the length of turns, sure, but there's no fixed value that it assigns to this. If you're using the default schedule, a turn could be considered as about four hours long; if you use the alternate default (24-hour) schedule, it's only one hour. And who knows what it is for the Burning Suns schedule.

[JaMiT]

Well, their drill sergeant (me) tells them every sweet day:

Have you ever noticed how little people listen?

I thought (but cannot find a reference at the moment) that the inability to choose the weapon on defense was intentionally one of the difficulties Wesnoth players have to plan around (with some sort of justification along the lines of what Dwarven_Void wrote). On the other hand, maybe the choice of defensive weapons should be changed so it favors staying alive over dealing damage (for all units -- still not a choice for players, but perhaps a better default). The logic for choosing defensive weapons might be something that was just copied from the choice of offensive weapons. Or maybe not. I haven't been around long enough to know all the design decisions that have been made.

And who knows what it is for the Burning Suns schedule.

UtBS uses 15 turns to cover 2 days (the same 2 days that the standard schedule covers with 12 turns). So UtBS turns are a bit shorter than standard at just over 3 hours each, rather than 4 hours each.

__

My questions above still stand. I have not looked through the entire code base, but I saw a part that suggests the answers to the two questions might be different. If that is the case, the inability to choose advancements on defense could be a bug (not a feature request).

[nuorc]

FootBall/Soccer...

I definitely agree. Often in the beginning, at strategical turn-points (and as arrogant end pose) my units stand on neighboring hexes.
In between, a casual connection around a certain point might be sufficient. Boils down to 'move as a team...' and proper positioning.

Playing sxc, I generally try to harden my units against the average damage received and maximize the damage I deal, while specifically upping both for the next boss to kill (next match). I tend towards over-preparing, so I might kill the end boss with less units than available within one turn, all my units with green health bars, but this being turn 97/100. But if I have a season to win...

In regard to the "drill sergeant" point, maybe a better solution would be to allow you to control the general patten of a unit via the right click menu.

+1

Have you ever noticed how little people listen?

Yes.

one of the difficulties Wesnoth players have to plan around

Trying to master BfW, each peculiarity of the battle system surely is a challenge of it's own.

The logic for choosing defensive weapons might be something that was just copied from the choice of offensive weapons. Or maybe not. I haven't been around long enough to know all the design decisions that have been made.

??? Currently you can choose offensive weapon and cannot choose defensive weapon?

Since I'm interested enough to speculate, but not enough to try things out:

1) What happens if you join a multiplayer game and one of your units advances after being attacked by an AI unit?

2) What happens if you host a multiplayer game and one of your units advances after being attacked by an AI unit?

My questions above still stand. I have not looked through the entire code base, but I saw a part that suggests the answers to the two questions might be different. If that is the case, the inability to choose advancements on defense could be a bug (not a feature request).

I could only add speculation, but I definitely think it should be handled the same way.

Anyway, I don't know why I shouldn't be able to spend the same care & attention to defense as to offense.

Platitude

Offense wins matches, defense wins championships

[taptap]

I thought (but cannot find a reference at the moment) that the inability to choose the weapon on defense was intentionally one of the difficulties Wesnoth players have to plan around (with some sort of justification along the lines of what Dwarven_Void wrote)

I wonder, because it hardly ever matters for the overwhelming majority of units as only a few units have multiple attacks of the same sort. It seems to me to be an unintended consequence (of the imo good choice not to require user input off-turn) more than an intentionally introduced feature. I am comfortable with the current situation, but claiming it is a feature, I don't know. Also, it isn't necessarily the question whether you control it, but it is hard to plan ahead because the AI choice is neither rigid (= often bad but you can calculate with it) nor right (= it often makes good decisions, but you can't rely on it). It might be possible to make AI choice better and more context-sensitive, without giving players control (and thereby complicating MP play endlessly).

The most relevant cases in mainline are elvish shaman level-ups with multiple ranged attacks of which one is slowing, probably in future Arif level-ups in melee, with one slowing attack as well, and the rarer cases were you would prefer not to kill an enemy in retaliation (i.e. choose the lower ctk attack for a Necromancer, Knight, Paladin or so, even though the attack has no special abilities like slow) to avoid freeing the attack spot for another attacker. And of course all those campaign heros who find additional weapons...

[tuggyne]

The most relevant cases in mainline are elvish shaman level-ups with multiple ranged attacks of which one is slowing, probably in future Arif level-ups in melee, with one slowing attack as well, and the rarer cases were you would prefer not to kill an enemy in retaliation (i.e. choose the lower ctk attack for a Necromancer, Knight, Paladin or so, even though the attack has no special abilities like slow) to avoid freeing the attack spot for another attacker. And of course all those campaign heros who find additional weapons...

That's a pretty good summary. So a design that allowed picking between several priorities for defense (without actually pre-choosing defending weapon, which might not really be desirable) should probably have these settings:

• Maximum CTK, maximum damage dealt: For times you just want to blast enemies away
• Minimum CTD, maximum CTK, maximum damage dealt: Approximately the current approach; blow things away, but without dying
• Minimum damage taken, maximum damage dealt: Avoid serious injury if possible, but hurt the enemy bad
• Minimum damage taken, minimum CTK: For cases when heavily outclassed, where killing an enemy just opens up two more to come in