Morale in Battle for Wesnoth

[turin]

agreed

KISS, not complexity.

[kmj]

The difference between a "fear" unit and adding "morale" is that the latter affects the underlying mechanics of the game everytime anyone plays it, whereas the former only ever comes into effect when scenario designers choose to use that unit.. this is a big deal, I think. I'd definitely prefer the "militia" and "fear" units over changing the underlying gameplay mechanics.

I agree that morale is realistic, but I don't know if any implementation of it in Wesnoth would do justice to Sun Tzu... so, with the KISS model, I say until chess players make their pawns scatter at the sight of a rook, we don't need to implement morale in Wesnoth.

[Christophe33]

In the toned down version of morale, it would affect only lvl0 and 1 units, the other being immune since they are elite/experienced.
A simple version is that when they reach less than 50 % HP and not near a unit with leadership, they won't be able to initiate an attack or moving into an enemy ZoC. Alternate malus would be just a reduction in domage they do (like 50%) or chance to hit the enemy.
In this way it would not be much different than any mechanism you would like to set up for a fear effect except that fear doesn't invove hitting a unit in your version.... Actually their might be some problem with fear...if you try to bring a "fearfull" unit into the ZoC of a "fearsome" unit... would it refuse to do so, or flee immediatly...Or would it be again just a malus in domage/chance to hit?

[Dave]

In the toned down version of morale, it would affect only lvl0 and 1 units, the other being immune since they are elite/experienced.

This would mean that the change would make lower level units less powerful compared to higher-level units. I don't think we need or want a change which has this kind of effect.

David

[Christophe33]

This would mean that the change would make lower level units less powerful compared to higher-level units. I don't think we need or want a change which has this kind of effect.

David

This will mean that lower level units needs the support of leader(ship) to perform well in a tight fight witch sounds reasonnable. Maybe the presence of other friendly units could reduce/delay or completely stop the effect of loss of morale. For example if there are more than 3 lvl of friendly units in any combination in the ZoC off the wounded unit then it doesn't suffer the consequence of being shaken (but it would comes back if it leaves/lost the support of other units).

[questbird]

In the toned down version of morale, it would affect only lvl0 and 1 units, the other being immune since they are elite/experienced.
-Christophe33

Rather than affecting only low-level units I imagine it would be more situational. Let me illustrate with two detailed examples, which I probably should have used in my first post (you might want to refer to it as you read these examples).

The first is a Lawful level 1 unit, say a horseman, attacking an level 1 Chaotic orc in broad daylight. The troops attack each other but neither kills the other. Say the orc and the horseman are both wounded but not critically. At the end of the turn, both units make a morale check since each was attacked and has an enemy in its zone of control. The horseman adds +1 for daylight, +0 for the unit's (level -1), +0 for his wound level, and +0 for enemies above one in his ZOC to a random element (1 to 6) of 4 to give a total of 5. The horseman is unaffected. The orc adds the following modifiers: -1 for daylight, +0 for the unit's level, +0 for the wound level and +0 for enemies above one in its ZOC to a random result of 5, for a total of 4; the orc, too, is unaffected in this case.
Overall for this first example, the horseman had a 1 in six chance of becoming shaken, while the orc had a 3 in 6 chance.

The second example takes place at night. A knight (level 2 horseman), flanked by an elven archer is being attacked by 3 bats (level 1). One of the bats is also flanked by the elven archer but in this instance, for simplicity the bats have not attacked the archer. However the knight has attacked bat2 and critically wounded it.
Here's a brief sketch:

(see attachment below)

This is how it pans out. At the end of the turn the bats and the knight make morale checks. First the knight. He adds: -1 for Night, +1 for 1 ally in ZOC, +1 for level (2-1), +0 for being moderately wounded and -2 for 2 additional enemies in his ZOC for a total of -1 to a random result of 3 giving 2. The knight becomes shaken. The bats must also check morale. Bat1 adds +1 for Night, +1 for 1 ally in ZOC, +0 for level(1-1), +0 for wound level, -1 for 1 additional enemy in ZOC (the archer) for a total of +1 to a random result of 3; Bat1 is fine. Bat2 is in the best position, receiving +1 for Night, +2 for 2 allies in ZOC, +0 for level, -1 for critical wounds for a total of +2 added to a random result of 2 = 4; no problem. Bat3 receives +1 for Night, +1 for 1 ally in ZOC, +0 for level, +0 for wounding for a total of +2 added to random 2 = 4. Again no problem.

So the final result is that the knight is shaken and cannot attack next turn or move into an enemy ZOC (though may still defend himself against attack). Overall for this example, the knight had a 3 in 6 chance of becoming shaken, Bat 1: 1 in 6, Bat 2 and 3: no chance due to their strong position and in spite of the fact that Bat 2 was badly wounded.

So you can see that a knight with backup given the wrong circumstances can have more chance of becoming shaken than a lone horseman (though the horseman placed in the same position as the knight would be even more likely to panic).

Now you may dispute the probabilities, you may dispute the complexity (although those simple calculations would be performed by the computer in a twinkling), however the above examples illustrate and reinforce strategic principles which already underpin Battle for Wesnoth. That is: stick together, don't get isolated and surrounded, watch out at Night (if you're Lawful). Notice that the wounded Bat in the second example was protected from fear by close proximity to two allies. This illustrates another principle: keep your stronger units on the flanks. If the knight had managed to kill the central bat then the morale story might have been very different; each bat would lose the protection of its fellows and lose 1 point from its morale check. Divide and conquer works in Wesnoth too.

This would mean that the change would make lower level units less powerful compared to higher-level units. I don't think we need or want a change which has this kind of effect.

David

Lower level units are already less powerful compared to higher level units but I see what you mean, you don't think that the differences should be emphasised even further. Fair enough point. A morale system would not only affect low level characters however, as I hope my above examples demonstrate.

Quest Bird

[Christophe33]

While I'm in favor of a system to include some effect of morale I think Questbird system involve too much chance/random. In your example 2 the knight and his allies are in a wining position even though it is night, so there should be no chance ofor them to be shaken. I would prefer a deterministic system to check whether a unit is shaken or not with no additionnal random factors.

[Gafgarion]

The prime problem with the lvl 0-1 morale idea, in my opinion, is how do you scare a Walking Corpse? They're mindless zombies and I would think they wouldn't be affected either way.

[Christophe33]

Most undead shouldn't be affected by morale issue . It will add to their mindless undead statue. Actually right now I think that change of AI makes the walking corpse not enough mindless...They tends to flee or avoid attacking instead of coming mindlessly into massive "suicide" attack as they should do.

[questbird]

The prime problem with the lvl 0-1 morale idea, in my opinion, is how do you scare a Walking Corpse? They're mindless zombies and I would think they wouldn't be affected either way.
- Gafgarion

I agree, I don't think Undead would be affected by morale. That would make them quite tough.

While I'm in favor of a system to include some effect of morale I think Questbird system involve too much chance/random. In your example 2 the knight and his allies are in a wining position even though it is night, so there should be no chance ofor them to be shaken. I would prefer a deterministic system to check whether a unit is shaken or not with no additionnal random factors.
- Christophe33

Perhaps a knight vs a bunch of bats was a poor choice , since, yes the bats would probably be beaten (for a start, they are all likely to be critically wounded when they attack the knight, and chances are that the knight would have finished at least one of them off.) But the example demonstrated what might have happened if all three survived and weren't badly damaged.

As for randomness, well that's just the way I was thinking at the time, so that it wouldn't be absolutely certain that morale effects would take place, just likely, just like a tough unit doesn't always kill a weak one if the tough unit is unlucky in its hit rolls. There doesn't have to be a random element though, I suppose, or at least it doesn't need to be as random as I suggested.

Another way of implementing the system (still with a random element) would be to mirror the terrain/defense mechanism where units defend based on their terrain. Well for morale the 'terrain' would be how much fear is around. Say the base 'fear defense' (morale) was 90% for a unit. (it could vary by unit, just as units terrain defenses vary.). Each time they have to make a morale check (basically whenever they have been attacked and are still in contact with an enemy. If they succeed in their morale check, they are fine; if they fail they are shaken.
Apply following modifiers to their 'fear defense':
+10% for good light
+10% for each ally in ZOC
+10% if affected by Leadership
+ ((unit's level -1) x 10)%
+10% if unit is healthy
+10% if unit has Courage/Steadfast trait
-10% for bad light
-10% for each additional enemy in ZOC after the first
-10% if unit is critically wounded

I used the same modifiers and the system is essentially the same (different probabilities), but note the similarity with the defense mechanism and the ability to immediately size up the odds in easily readable percentages of a unit becoming shaken (that is, their morale at any given time). You could even factor in the defensive ability of the unit in the terrain - maybe 10% of their defence. eg, elf in woods adds +7% (10% of 70%) to morale because they feel more secure in their position..

Quest Bird

[quartex]

Too many modifiers. How is the casual player supposed to understand when his unit he affected by fear, and when he isn't when there are so many modifiers? Anyone looking at that system would get really confused.

I love D&D, and that reminds me of a classic D&D system, where you add a series of modifiers together to determine if the unit is afraid or not. But wesnoth is famous for being simple and easy to learn. Consider Dave's solution to a unit's armor class, he doesn't take into account their level, race, or how many units are flanking/surrounding them. All he does is see what terrain they are on. It's a maddeningly simple system. Sometimes I wish it was more complex, when my level 3 champion keeps missing that walking corpse, but because it is simple the system is easy to learn and easy to determine, in an instant, how hard it will be to hit an enemy.

Likewise morale needs to be that simple. For example:

Fear (a level 2 ability): All enemy units of a lower level that are not adjacent to a allied/friendly unit with leadership do 50% less damage when they attack.

That seems simple, easy to counter, not too powerful, and yet effective. The % damage reduction could be changed of course.

[Christophe33]

Too many modifiers. How is the casual player supposed to understand when his unit he affected by fear, and when he isn't when there are so many modifiers? Anyone looking at that system would get really confused.

I love D&D, and that reminds me of a classic D&D system, where you add a series of modifiers together to determine if the unit is afraid or not.

If you think D&D is complex, don't try playing RoleMaster except if you want to improve your mental calculation capabilities. Well, I actually love that game but we ended up developing an exel application to calculate the % of most competances... and we still had tto apply modfication afterward..

But wesnoth is famous for being simple and easy to learn. Consider Dave's solution to a unit's armor class, he doesn't take into account their level, race, or how many units are flanking/surrounding them. All he does is see what terrain they are on. It's a maddeningly simple system. Sometimes I wish it was more complex, when my level 3 champion keeps missing that walking corpse, but because it is simple the system is easy to learn and easy to determine, in an instant, how hard it will be to hit an enemy.

Too simple doesn't mean good. Actually as a player you just click on the enemy unit and look at the display of the % chance for you to hit, the domage, same for the enemy. You don't care whether it comes from a complex calculation involving 1000 parameters or from a simple table as long as the results is clear. I would also expect a higher level unit to have a better chance to hit. Maybe a bonus of 5% /level on the" to hit chance" wold allows to reflect this while remaining simple.
Actually how many of you can figure in advance by mental calculation the amount of domage a unit is going to perform to another one?
I can't even though I know that day/night affect by 25%... and that there is somewhere a table of resistance that can change tthe result a lot.... So I end up trying to place a unit against the enemy and look at the proposed domage...If I don't like it I try another type of unit. After sometime you learn some of the main trick like holy/magic are very efficient against undead while piercing is not...but I still can't tell in advance by how much.
In other word the % of chance to hit is simplistic while the amount of domage done combines parameters that most player can't remember or calculate by themself...why should we calculate in advance...the computer do it for us. Ok, end of this part on hypersimplicity

Likewise morale needs to be that simple. For example:

Fear (a level 2 ability): All enemy units of a lower level that are not adjacent to a allied/friendly unit with leadership do 50% less damage when they attack.

That seems simple, easy to counter, not too powerful, and yet effective. The % damage reduction could be changed of course.

I actually agree on something similar despite my comments on the too simple chance to hit parameters (or lack of it).
The system I outlined earlier for morale is fairly similar and simple:
If a unit lost more than 50% HP (can be another number like 66% i.e when the HP bar turns red) AND not undead (berserk?) the unit will suffer from a shaken morale if there is not a unit with leadership nearby.
An alternative is that "shaken" could be also be countered by the presence of at least 3 levels (or 4 or 5) of friendly units in it's ZoC or if there is more level of friendly units than enemies in it's ZoC. Anyway, healing the unit above 50 % (or 33%) would also emove the shaken statue.
The penalty for being shaken could be a reduced chance to hit, reduced domage or inability to initiate an attack. The last possiblity iss probably the most difficult to implement.

[Dave]

I don't think that an extensive morale system fits well into Wesnoth, and this is why:

Wesnoth is not a 'wargame'. Wargames are those games which try to simulate real battles, and have complex rules. Rules like line of sight, rules about weather, rules about stacking units, rules about surround bonuses and skirmishes, and rules about morale. Wargames usually send all but a small niche of enthusiasts scurrying for the nearest action hack-n-slash game.

Wesnoth could easily be confused for a wargame, because it uses hexes, and few games but wargames use hexes. However, Wesnoth is not a wargame. It's a turn-based strategy game with an RPG feel. It uses simple, appallingly unrealistic rules about units attacking, and knocking hitpoints out of other units. It deals with units on entirely the wrong scale for a battle -- each side in a battle consists of perhaps twenty individuals.

It's a game where lawful units gain an advantage during the day, and chaotic units at night. Where people fight for days without need for rest. Wesnoth thumbs its nose at realism -- it's high fantasy, a chance to escape from reality, not to try to impose endless rules from reality on the game.

Adding a feature to the game is usually done by deciding that certain rules would enchance the game, and then working out an orthogonal way to implement it. This feature is going round the wrong way: deciding we need this feature from reality, and then deciding upon a good way to implement it gameplay-wise.

Sun Tzu is probably rolling over in his grave at how battles are fought in Wesnoth. We don't care, the aim is to have fun, and I think that Wesnoth is a fun game.

As a user on happypenguin has said:

I never liked turn based strategy games, but this in incredible fun. And there's a huge amount of depth in the game, with terrain modifiers, units gaining experience, healing on villages, and more. I never played games like that before, but this kept me up all night.

Let's keep it a game that children and adults can play -- a game that people who normally love action games can drop by and play, without getting confused by why their unit is running away, or why they can't attack because the unit is scared, or why their unit is getting slaughtered when surrounded.

Let's keep Wesnoth simple.

David

[questbird]

Let's keep Wesnoth simple.

David

Well, that's the end of that topic then.

Quest Bird

[Kamahawk]

The KISS of death.l