Design: Heroes

Kael, Chalid,

I have tried a part of the night to do something about skinning Govannon. It's ugly.:cry: He looks like a drunken down and out tramp who had vomit on his clothes.:sad: I'm light years away from beeing just a poor "skinner".:(
Even if i would try again, you should ask somebody else. The result would be better and Govannon would be ready for a next release more quickly.
Sorry.

About the Hippus Hero, why don't you give him the "Command" Promo ?

The Frog.
 
Maian said:
I've been thinking about heroes for a while. I've been reading how players typically use them and noticed how I've used them myself.

Namely, how players are much more reluctant to send their heroes into combat than typical units. How combat-oriented heroes often have less use in combat than spell-caster heroes, just because they are thrown in only when the odds are overwhelmingly in their favor.

This is wrong. By the very nature of its scarcity, heroes end up being unheroic. One has to wonder how they got to be heroes in the first place when they fight less than their fellow soldiers.

By giving a unit the status of "hero", a player will naturally start becoming a bit attached to it. This happens all the time in RPGs so it's not surprising it happens here. So it also makes me frustrated when the hero does get killed even with overwhelming odds. Not fun at all.

On the other hand, the danger of the hero being killed does add a tactical consideration. Protecting your hero should be important, and killing enemy heroes should be major victories.

This is the heart of the problem. The goal here is to make heroes important and non-expendable, yet also encourage them to play a heroic role in campaigns, instead of standing back and let the pawns do all the work.

So how do other strategy games that integrate heroes do this? They do so in a couple ways. One way is to make the hero more of a general/spellcaster that is only killed when the army is killed, ala Heroes of Might and Magic. Lots of games have the concept of hitpoints plus tactical retreats, allowing players to recall heroes when they are losing in the middle of a battle. Some games allow resurrection of heroes at varying costs. RTS games give units 100% chances of winning over weaker units (for example, pit a hero against a lowly foot soldier and the hero will always win).

I'm not sure how this can be done in Civ4. Chance needs to play far less of a role for heroes. In particular, simulating tactical retreats in battles would be ideal. Concerning resurrection, I'm not in favor of that since that diminishes the importance of killing enemy heroes.

I cant speak for the others but the fact that I dont risk my heroes in battles they are liekly to lose doesnt make their action unheroic. They turnt he tides of wars, Saverous alone saved my empire from collapse in my last game. But he did it without ever fighting a battle he had less than a 95% to win (except for one 60/40 fight against Rosier that he won!).

But its not about the risk in the battle you choose to join. Its about the risk in exposing the hero. Do you move them forward to kill that archer, know that it will be deep in enemy lines and the ai could gang up on it afterwards.
 
Deathling said:
How about:
10~12 Strength, 4~5 Movement, starts with Air III and Summoning (or Divine if you feel like creating a few new spells, like change wind current for +1 movement for all units in tile and +15% offense), can summon a recruiter with Dimensional I after pillaging hamlet-town/razing a city...

I know it's kind of absurd, a mercenary with magic, but I really felt like the Air side of Hippus wasn't shown enough. Even with the Hippus summoner, he only has 1 Air spell and no connection to the mounted mercenaries.

Edit: Just wanted to add that instead of Dimensional I you could use Rebase from vanilla.

I had another idea for Magnadine. Magnadine could become a melee unit in a range of 1 tile away from an enemy city. This could be a problem for coding, but atleast he would be able to defend himself after attacking (unlike all the other raiders)... He could only have the magic attacks as a melee unit, this would make him much more powerful while raiding, which would make sense, as he's a mercenary.

Mounted Magnadine: 3 Strength (yes, that low), 5~6 movement, starts with Hero and Inspire Followers (allows recruiting, I guess).
Inspire Followers could make mounted units (according to era) appear out of the blue. As they appeared they would begin following Magnadine around. These units wouldn't belong to any civilization and would play after Hippus' turn, after winning a battle they could convert to Hippus or something.

Magnadine: 13 Strength, 1 movement, Starts with Hero, Summoning/Divine (see above) and Wind III (or whatever).
 
Kael said:
I cant speak for the others but the fact that I dont risk my heroes in battles they are liekly to lose doesnt make their action unheroic. They turnt he tides of wars, Saverous alone saved my empire from collapse in my last game. But he did it without ever fighting a battle he had less than a 95% to win (except for one 60/40 fight against Rosier that he won!).

But its not about the risk in the battle you choose to join. Its about the risk in exposing the hero. Do you move them forward to kill that archer, know that it will be deep in enemy lines and the ai could gang up on it afterwards.

There are actually two problems here, both caused by the unexpendable nature of heroes. First, heroes aren't used that much. I'll expand on this more later. The second is that it's extremely frustrating when you lose your hero to a battle that's overwhelmingly in your favor. It's frustrating whenever this happens, but moreso for units you cannot rebuild.

There should be a way to make sure your hero doesn't die in a battle that has very good odds. Where defeat does not necessarily mean your hero is killed. Ever heard the saying "losing makes you stronger"? That can never happen for heroes right now. The whole concept of tactical retreat within battles doesn't exist.

Personally, I kinda want all units to have a good chance at withdrawing, but I recognize that Civ4 is an empire building game first, and a tactical turn-based wargame second. Prolonged fights between units may detract from the rest of the game, and the massive number of units already simulate withdrawing (killing a unit can be seen as just destroying part of an army, leaving the rest to withdraw the next turn). However, heroes are an exception here. They can't be replaced. They should be important enough to warrant the additional attention and complexity. In fact, RPG elements would be nice for heroes.

As for the first point - that heroes don't fight as much as others - I still think this doesn't feel right. Perhaps they can change the tide of war, but so can any other powerful unit. There should be something distinguishing a hero from simply being a powerful unit. Heroes just don't feel heroic to me. Sometimes I get more attached to a unit that somehow constantly defies odds and wins over and over again - something I would never risk sending the actual "hero" to do. Maybe that is the missing element. Heroes shouldn't become heroes until they prove themselves to be heroes.

Alright, I'll lay out out my suggestions. Not fully thought out, but good enough for now.

1) Some units should morph into heroes once they reach a certain level. For example, let's say an elven archer reaches level 4. Then it becomes Gilden Silveric. Once that happens no other archer can become Gilden Silveric (for obvious reasons) even if the hero dies.

2) It should still be possible to directly train heroes under certain conditions. Heroes like Typhoid Mary, which aren't based off any unit, can always be trained. Heroes like Gilden Silveric can only be trained after a certain amount of archers are produced (or some other constraint, maybe tech?). This ensures that aggressive civs don't have too large an advantage over passive civs, since both types will still have ready access to heroes.

3) It should be possible to upgrade heroes. Going on with the Gilden Silveric example, once you have access to elven longbowmen, you can upgrade Gilden Silveric to a longbowman. This is the RPG-equivalent of changing equipment.

4) When attacking or defending against a stack of units without an "anti-hero unit", the hero should have a 100% chance of withdrawal. "Anti-hero" could be some sort of promotion granted to most heroes and assassins and their ilk. Each level in the anti-hero promotion in the enemy stack would reduce the withdrawal chance by 20%. Or if that's too complicated, just let the anti-hero promotion reduce withdrawal by 50% regardless of level or the number of anti-hero units in the stack. Note: I have no clue if this can be done in Civ4.

5) When you have an anti-hero unit attack a stack of units that also has an anti-hero unit, your anti-hero unit will attack the strongest non-hero anti-hero unit, instead of the strongest unit in the stack. If the only anti-hero unit in the stack is a hero itself, then your anti-hero will attack that hero. So the order of preference goes: non-hero anti-hero unit > anti-hero hero > any other unit. This sounds complex, but it does add an additional layer of tactics. Anti-hero units serve both as hero killers and defenders this way. Note: Again, no clue if this is possible in Civ4.

6) After each battle a hero participates in in the same turn, the hero's chance of withdrawal is reduced. This ensures heroes can't simply survive repeated attacks in the same turn.

7) Heroes with flanking promotions would get > 100% chance of withdrawal so it would take more anti-hero units or more battles in the same turn to corner them.

8) An equipment system and random items spawning around the map would add an RPG feel. Basically something similar to Heroes of Might and Magic.

The goal of the above suggestions are to make heroes a more integral part of the game. They make the player more attached to them, make them a centerpiece in military campaigns, and something the player spends plenty of attention on. At the same time, I'm keeping in mind that the additional complexity needs to be worth the buck, so I wonder if there are simpler ways to acheive the same effect.
 
Great Ideas!

1. Your creation of hero ideas are quite good.
2. I think that the equipment system will stay out until shadow. (a later release)
3. Yes, basically any attacker/defender modification is possible with python, so the anti-hero stuff could work.
 
Hi,

I agree. Great ideas.

Creating a hero after one of your unit won enough battles, either defensive or offensive, is a good idea. I just think that level 4 is too low. Probably 6 or 7 would be better.

I also think that because Hero are unique unit they add to be protected against "common" units. On the other hand why not a anti-hero promo to hunt this guys ?

To finish, improving a Hero as explain with the elf one (don't remember his name) is a very good idea.

The Frog.
 
I always thought units should upgrade to heroes at a certain point, I just thought that they shouldn't become existing heroes. Level 4 is way too low, maybe more like 10~17 (by the hero and unit class, offensive units gain xp faster).
 
Nice to see people like the ideas :)

That level 4 thing is arbitrary. I chose it for the sake of example. In fact, the "hero level" can be one of the perks of a type of civilization. For example, it may be lvl 6 for aggressive civs and lvl 8 for others.

Also, with all these other bonuses for heroes, there should be no exp bonuses for heroes. Perhaps trained heroes would retain the 1 exp per turn bonus until it reaches the "hero level", at which point the bonus goes away, assuming trained heroes start at level 1.
 
On the note of upgrading hero's, Basium is pathetic. I went through the list of heroes in the Civilopedia, and he only can really theoretically beat Loki if he can pin him down or Corindale, assuming he doesn't get wasted by Corindales spells first. Not to mention plotwise he's supposed to be a fallen solar. At least give the guy immune disease and an upgrade so you cant screw his whole civ over with a marksman. And yes Corindale is useful offensively, the only difference between him an archmage is that he comes earlier has hero and is arcane, and has that peacemaker move. Sheesh, why the heck can a pacifist learn Meteor Shower, Tsunami or Defile.
 
Thunderwing said:
On the note of upgrading hero's, Basium is pathetic. I went through the list of heroes in the Civilopedia, and he only can really theoretically beat Loki if he can pin him down or Corindale, assuming he doesn't get wasted by Corindales spells first. Not to mention plotwise he's supposed to be a fallen solar. At least give the guy immune disease and an upgrade so you cant screw his whole civ over with a marksman. And yes Corindale is useful offensively, the only difference between him an archmage is that he comes earlier has hero and is arcane, and has that peacemaker move. Sheesh, why the heck can a pacifist learn Meteor Shower, Tsunami or Defile.

Basium is buildable from the start, he has no required techs or resources. He may not be able to defeat Eurabatres in combat, but I would wager that having a iCombat 5 unit with the hero promotion at the start of the game is a bigger advantage than getting Eurabatres in the late game. He is certainly isn't pathetic, if anything he is overpowered.

As for Corlindale, he starts out with spells in the non-letal spheres, but if the player decides to turn him into a killing machine that is their option (the AI will generally upgrade him in the spheres he already has). We typically don't enforce roleplaying considerations.
 
I think Loki is really usefull as he is. I cheated him for me in the last thirty games i played... ok i used him to verify if the thing i introduced worked but neverthelesss... quite usefull with his spiing abilities :D
 
More of my musings...

Just realized that a defending unit can't withdraw, so I need to think up something different for hero defense (if needed). While it could be possible to code in defensive withdrawals, I'd prefer to avoid that since it's a large change - need to figure out where unit withdraws to, how it effects gameplay, and whether offensive and defensive withdrawals should require different promotions.

There needs to be someway to ensure that a hero in the middle of a stack is "safe". Suppose we a hero in a stack, and an enemy attacks the stack, selecting the hero as the unit with the best odds against it. This gives the enemy a chance to kill the hero, even if the hero is supposedly well-defended. I'm not saying that the hero shouldn't have been selected - if the hero is the best unit, it should indeed be the one defending - but it should not be so easy to kill this way.

Maybe defensive withdrawals really are a good way to handle this situation. It could be implemented as following:
1) Suppose an enemy attacks a stack of units and selects a unit that can withdraw. The unit loses but wins the withdrawal role. It survives and doesn't move anywhere.
2) Suppose an enemy attacks a lone unit that can withdraw. The unit loses but wins the withdrawal role. It survives and retreats to the safest plot. After digging a bit into the SDK, I found CvPlayerAI::AI_getPlotDanger, which seems to be a good way to find this safest plot.

Should these rules apply to all non-hero units that have withdrawal bonuses as well? I'm inclined to say no, since it provides a huge advantage and could be unbalancing.

Last thought: the "anti-hero" promotion definitely needs a better name but I can't think of one. "Anti-hero hero" is a bit of a mouthful :) Can't think of anything that encompasses both assassins and heroes, so they may just need different promotions that share the same anti-hero effect. Also not sure if it makes sense for assassins and the like to defend heroes, but it does simplify things. The alternative would be having two separate promotions for hero defense and hero killing.

EDIT: Does Civ4 already have retreating/defensive withdrawals? If so, I haven't seen one in the longest while. I've been playing various TBS games lately and at least one of them has it...just not sure which one :p
 
Maian said:
More of my musings...

Just realized that a defending unit can't withdraw, so I need to think up something different for hero defense (if needed). While it could be possible to code in defensive withdrawals, I'd prefer to avoid that since it's a large change - need to figure out where unit withdraws to, how it effects gameplay, and whether offensive and defensive withdrawals should require different promotions.

Maybe somekind of pushback addition to the hero promotion (but only while in a city, outside it could be a withdrawl)...

There needs to be someway to ensure that a hero in the middle of a stack is "safe". Suppose we a hero in a stack, and an enemy attacks the stack, selecting the hero as the unit with the best odds against it. This gives the enemy a chance to kill the hero, even if the hero is supposedly well-defended. I'm not saying that the hero shouldn't have been selected - if the hero is the best unit, it should indeed be the one defending - but it should not be so easy to kill this way.

Maybe lowering the defense of the hero outside of the homeland, while inside the hero not only recovers the lost defense, but gains like 10%... I know this isn't the best solution, but if you have the hero in a stack it won't really matter, well, unless a stack is attacking your units (meaning a large stack).

Maybe defensive withdrawals really are a good way to handle this situation. It could be implemented as following:
1) Suppose an enemy attacks a stack of units and selects a unit that can withdraw. The unit loses but wins the withdrawal role. It survives and doesn't move anywhere.
2) Suppose an enemy attacks a lone unit that can withdraw. The unit loses but wins the withdrawal role. It survives and retreats to the safest plot. After digging a bit into the SDK, I found CvPlayerAI::AI_getPlotDanger, which seems to be a good way to find this safest plot.

Should these rules apply to all non-hero units that have withdrawal bonuses as well? I'm inclined to say no, since it provides a huge advantage and could be unbalancing.

I agree, but, not only would it too hard to find the safest plot, it would also not make sense. If someone attacked you and you decided to run away, you'd be running in the opposite direction.

Last thought: the "anti-hero" promotion definitely needs a better name but I can't think of one. "Anti-hero hero" is a bit of a mouthful :) Can't think of anything that encompasses both assassins and heroes, so they may just need different promotions that share the same anti-hero effect. Also not sure if it makes sense for assassins and the like to defend heroes, but it does simplify things. The alternative would be having two separate promotions for hero defense and hero killing.

How about "Bounty Hunter"? It makes sense. I mean who else specializes in killing specific people? Since there's an Assassin unit, Merceneries are going to be in the game sometime in the near (hopefully) future and Hitman just doesn't fit FfH, I ruled them out.

EDIT: Does Civ4 already have retreating/defensive withdrawals? If so, I haven't seen one in the longest while. I've been playing various TBS games lately and at least one of them has it...just not sure which one :p

Haven't seen it either.
 
@Maian:

We added a function that makes sure Heroes (and magic users as well) are selected later when defending. So when they defend it is because they have a considerably higher chanche of winning. Basically it works like:

Chance of the best normal 60% to win
Chance of the Hero 65%

-> The nromal unit defends

Chance of the best normal 50% to win
Chance of the Hero 80%

-> The Hero defends

At the moment its implemented on a very simple basic but we might further refine the system. (eg when a hero has only 20% and the next unit has 5% the hero defends at the moment. One would wish that in that cas the normal unit defends to save the hero one more turn).
(Oh and this should be true for non heroes with really big levels, too)

And Vanilla has no defensive withdrawals.
 
Deathling said:
Maybe somekind of pushback addition to the hero promotion (but only while in a city, outside it could be a withdrawl)...

? Not sure what you mean by pushback.

I agree, but, not only would it too hard to find the safest plot, it would also not make sense. If someone attacked you and you decided to run away, you'd be running in the opposite direction.

What if the opposite side also has an enemy unit? Then should the retreating unit go right or left? In any case, this is already implemented in FfH for Loki in the FFHFindClearPlot function. It apparently scores the plots in the area around the retreating unit, depending on whether a plot is empty, has friendlies, or is a city, and adds a little random factor.


@Chalid:

Is the code for that in the SDK? I couldn't find it anywhere in the Python code.
 
BTW, when Loki is attacked, the attacking unit doesn't move into plot where Loki previously was. That doesn't make sense and appears to be a bug. From what I can tell, the code for moving Loki is in the cannotAttack function, but the cannotAttack function is apparently called at the start of combat, which is too late to get Loki out of there (if the function returns true, Loki doesn't get a chance to run away and gets clobbered).
 
Back
Top Bottom