Frogboy Frogboy

AI Test: December 1 – running blog…

AI Test: December 1 – running blog…

Developer Journal on new AI capabilities for Elemental: War of Magic v1.1

World Difficulty: Normal

Map: Tiny

All players (10 opponents) have “normal” intelligence instead of Pariden

Goal: Elimination of all AI players as quickly as possible.

imageI just find that funny. :)

If you guys see anything wrong with the strategy I employ, let me know.

Early Game:

Build town, recruited a couple NPCs, sent sovereign out to gather goodies.

Observation: Need more choices in what to build early on.

First rule in this game: Any sovereign I can kill, I will kill.

Ok, Verga by himself can take out Relias and friends…

Ack, at war with a lot of players now thinking I was going to steamroll.

image I still haven’t figured out why that sort of thing happens.  It shouldn’t.

Building a pioneer. (Winter, 166 AC)

image

image What? No love at first sight?

Built “Toad Ville” near an ancient temple.

Ok, AI sovereign is stuck.   Need to make them intelligently learn to teleport out.  Let’s see how this works..

image

Fixing a deadlock I caused… Grr.

Ok, deadlock fixed.

Grr. It is really really hard to nail down what the battlerank should be.  I thought I had Verga but nope. He killed my whole group.

Going to try doing it via tactical (loading a saved game to try it again to see).

imageShould be easy…right?

 

image

Well crap, Magnar came after me!

image

This isn’t going well.

image

I really hate Verga.

GOT HIM!

I KILLED VERGA!

Magnar’s forces have come for me.

This could go either way.

image

Heh.

imageWith Magnar thrown back, I’m in a much better position.

Magnar is still a threat but my capital (and last city) is no longer in immediate danger.

Building up again.

Need more war music, emailing Mason.

Getting low on candy.

Next…

 

 

I am starting to gain momentum.

image

Pariden shows up.

 

Okay, it’s almost midnight. Time to check in my changes and head home.

..updating…

213,261 views 64 replies
Reply #51 Top

Quoting MartialDoc, reply 34

Quoting TheProgress, reply 20Flaw regarding all 4X games:

The diplomacy is always based on who has the biggest sword (who's got the strongest, largest army). Every AI leader has the same mentality / character: "if my nation can conquer their nation, then it will."

Please set Elemental apart from all other 4X games and do something unique. Give AI leaders some hidden stats such as:


aggressiveness: How eagerly an opponent leader will engage in warfare, quest hunting, expansion, etc.

recklessness: The size and frequency of an opponent leader taking risks such as committing entire armies to one task, or going to war with stronger opponents, etc.

pride: How much the opponent leader is willing to forgive past conflicts with a nation. For example, an extremely proud leader may be stubborn and devote himself to taking revenge on a nation that has wronged him in the past.

empathy: How likely the opponent leader is willing to embrace another peaceful nation and make diplomatic ties REGARDLESS of the strength of their armies. A highly empathetic leader may wish to form alliances with smaller, insignificant nations rather than go to war with them.

 

Really like this idea.  Something like this could make for a much more interesting AI.

 

I agree, I would like to see some individual personalities out of the sovereigns.  Perhaps have 2 or 3 personality types for each Sovereign so don't quite know which one you are going to get.

 

Quoting Napean, reply 36

Oh crap, that's my son!
It seems it's not just the AI that doesn't realize that sometimes

There was also that instance where you intended to move a group of units but only moved one, as you had clicked it to look at stats, then inadverdantly split your force.  Could this be solved by incorporating a "split group" button/function, and groups of units can not be moved independently until they are selected and the "split group" button is pushed?

It's a pleasure to see you at work, Elemental is definately worth the wait (though at present AoW:SM is getting a lot of play while I wait for 1.1 )

 

This might be getting to particular, but the AI should make some errors from time to time.  It does make it fun.  I have always felt the randomness is what makes strategy games fun.  Don't get me wrong, even on "normal" level the AI should make a mistake like that once every 50 turns.

Reply #53 Top

Quoting Frogboy, reply 49
I really REALLY like this idea.
Lets be serious. We have been waiting for such feature for a long while. The same one that is already used in the combat log for the autoresolve. :P Better late than never!!! :O

Reply #54 Top

Ohhh, actually... I really, really like the idea of localized blows. That would give a lot more usefulness to armor parts. And it opens up possibilites: you could make weapons which are better at striking a specific part, have a specific effect happen when too much damage is dealt to a part...

Reply #55 Top

Quoting joasoze, reply 25
Regarding more build options early on and how to stop the easy rushing of AI capitals.

 

Suggested buildings:

Homeguard (not sure about the required material/gold to build - requires two citizens pr city level). This building will take two pesants pr city level and equip them with best available equipment. When city with Homeguard is attacked these will be added to normal units in city. This way one will have some extra defenses in cities without much micromanagement.

 

Snakepit: (Empire)

Early building that will add one snake to city defense pr city level. Snake can not be moved out of city

 

Kennel (Kingdom)

Early building that will add two guard dogs to city defense pr city level. Cannot be moved out of city

 

 

 

All this sounds good, except the best equipment. Maybe, depending on city lvel, the units are different between peasant, then up to like honor guard and stuff.

Also, make walled cities unattackable. Walls could be an improvement from level 2 on, and the first city a nation has automatically has walls, or at least hedge walls.

Walls would be a "unit" by having hp. when that is destroyed, either by seige weapons or a new spell(s) of some kind, they disappear, both tactically and strategically mapwise.

 

In the expansion, maybe have archers behind walls be able to fire over them/stand on them, and units behind walls not able to be attack or to attack unless they go through a gate that can be closed.

 

Also, this (another quote): 

Quoting Werewindlefr, reply 54
Ohhh, actually... I really, really like the idea of localized blows. That would give a lot more usefulness to armor parts. And it opens up possibilites: you could make weapons which are better at striking a specific part, have a specific effect happen when too much damage is dealt to a part...

And armor would be localized... And area striked would depend in how desireable that location is... and how trained the unit is...

This would require a LOT, but if it was done right it would be absolutely fantastic. Maybe an expansion could include more of the RPG Element- allow us to duel another sovereign to the death!! :)

 

And this:

Please set Elemental apart from all other 4X games and do something unique. Give AI leaders some hidden stats such as:

  • aggressiveness: How eagerly an opponent leader will engage in warfare, quest hunting, expansion, etc.
  • recklessness: The size and frequency of an opponent leader taking risks such as committing entire armies to one task, or going to war with stronger opponents, etc.
  • pride: How much the opponent leader is willing to forgive past conflicts with a nation. For example, an extremely proud leader may be stubborn and devote himself to taking revenge on a nation that has wronged him in the past.
  • empathy: How likely the opponent leader is willing to embrace another peaceful nation and make diplomatic ties REGARDLESS of the strength of their armies. A highly empathetic leader may wish to form alliances with smaller, insignificant nations rather than go to war with them.

 

Reply #56 Top

Quoting madtemplar0, reply 55

Quoting joasoze, reply 25Regarding more build options early on and how to stop the easy rushing of AI capitals.


 

Also, this (another quote): 


Quoting Werewindlefr, reply 54Ohhh, actually... I really, really like the idea of localized blows. That would give a lot more usefulness to armor parts. And it opens up possibilites: you could make weapons which are better at striking a specific part, have a specific effect happen when too much damage is dealt to a part...

And armor would be localized... And area striked would depend in how desireable that location is... and how trained the unit is...

This would require a LOT, but if it was done right it would be absolutely fantastic. Maybe an expansion could include more of the RPG Element- allow us to duel another sovereign to the death!!

 

Well if you go down that path you end up with the dwarf fortress damage model which would be a bit much for a strategy game.

Stuff like "The Knight strikes the Thief into the Upper body with his steel Longsword tearing the Skin, tearing the muscle, breaking 2nd false Rib on the right, bruising the sensor nerve, tearing the Liver tearing and tearing the Stomach. A Mayor artery has been hit. Your are spattered with blood. The Steel Longsword is Stuck in the wound. The Thief gives into pain."

 

And No i am not joking - this is a damage-log how it can appear ingame. As long there is no more RPG heavy expansion ofr Elemental i would say that is a bit over the top. 

Anyway what would be nice to literally see is the units Blocking and dodging. I know its some additional work but could make fights a bit more immersive.

Also could it be that the Ai is so short on money and/or resources that it just cant afford groups? I have seen plenty of Ai players that were really low on resources which could have never build a Group of heavy Armed units - especially after i bought them out with some useless "junk" like crystals or horses. I mean it strikes me as idiotic to hand me 1600 Gold just for some crystals which could be used for a decent unit if the money for that wouldnt have been gone now.

 

Reply #57 Top

Quoting sagittary, reply 40
As far as early blitzkrieg, I would say to give cities natural defenders along with associated (new) techs. You could still blitzkrieg (potentially) but you'd be required to heavily forestall any economic development.

 

For instance, for each city level, the city gains 1 defender. This defender is equipped based on a budget (based on city level and buildings) versus resource cost of all available/researched equipment (so that defenders wouldn't just instantly get the best equipment at an undeveloped city). A palace grants +1 defender and +budget.

 

Other buildings would then influence these things and add additional city-only tactical options. These would require an upkeep. For instance, people have mentioned archery towers - perhaps these are build explicitly and act like a spellcaster in battle, allowing you a special spell (hail of arrows) that you can cast (draining resources instead of mana). These would count as mixed damage of physical and magical damage, which would mean the player also needs to balance resistances (or build more in order to zerg).

 

I've also noticed that due to seeding, capitals are more or less the only significant target. There's few other targets of opportunity. In part, this is due to the way the resources are seeded - a lot around the starting location and then falling out there. I would say, just give the palace building itself the various bonuses you want all capitals to have and spread the actual world resources out more so that there's more targets of opportunity. As it is, one goes straight after the capital and in doing so, heavily cripples the opponents economy.

 

I love this idea, so I fleshed it out a little further:

I don't remember the group sizes past squad, so I'm assuming Sethai is correct (squad = 3, party = 9, company = 12).

Defenses added by city level:

lvl 1: 1 foot soldier, 
lvl 2: 1 foot soldier, 1 archer
lvl 3: 1 foot soldier squad, 2 archers
lvl 4: 2 foot soldier squads, 1 archer squad
lvl 5: 1 foot soldier party, 1 archer party

foot soldier: as you'd expect, a defender with armor and weapons consistent with melee combat

archer: ranged weapons specialists

I would love to add a support type at level 5 or with special buildings, but there is no provision for supporting roles in the current tactical combat given available equipment/abilities.

 

graceful degradation of missing technologies:

if squad not researched, replace w/ 2 individuals

if party not researched, replace w/ 2 squads

I'm not addressing company, since 12 is very oddly proportioned after 9

if archery is not researched, replace archers with foot soldiers 1-to-1

equipment: defenders are equipped based on settlement value, such that the cost of equipment = (calculated value of settlement * modifier)

restrictions: defenders cannot have metal equipment if no metal resources are activated in the empire; likewise for items that require crystal

implication: since the value of a settlement grows with settlement size, defenders should be better equiped for bigger and/or better settlements.


Palace:

Adds the Royal Guard to the defense, which is a single additional group of foot soldiers of size Math.min(citysize.defenderSizeRestriction, largest group size researched)

the Royal Guard should have 125% of the typical equipment value

Royal Guard group size restrictions for the Palace:

1,2 : individual (this might be N/A if lvl 3 city is required to build the palace)

3: squad (3)

4: party(9)

5: company(12)

 

Reply #58 Top



edit 2: also, please find some better alternative to teleportation of defeated enemy sovereigns. It just cheapens the impact of battle with them. And when a sovereign is killed, instead of magically removing all of their cities, turn them into independent states or assign one of the sovereign's children as the new leader.


[/quote]

I'm with you there, I'm flabbergasted that the empire explodes, and then you have such an incentive to not kill the leader (take the cities first, save settler resources/improvements)... where in most other conflicts, taking out the leadership is the first priority.  Especially since 4x games normally don't let you kill the leader or have them walking around.  Even the MoM style banishment would be better.

Reply #59 Top

While I was thinking about the combat text popping up (as mentioned earlier).. I was thinking how nice it would be to have a "bone crunching sound" for when a character scores a critical hit. It would add that extra oomph! to successful critical hits. :D

Reply #60 Top

Quoting madtemplar0, reply 55


Please set Elemental apart from all other 4X games and do something unique. Give AI leaders some hidden stats such as:


aggressiveness: How eagerly an opponent leader will engage in warfare, quest hunting, expansion, etc.

recklessness: The size and frequency of an opponent leader taking risks such as committing entire armies to one task, or going to war with stronger opponents, etc.

pride: How much the opponent leader is willing to forgive past conflicts with a nation. For example, an extremely proud leader may be stubborn and devote himself to taking revenge on a nation that has wronged him in the past.

empathy: How likely the opponent leader is willing to embrace another peaceful nation and make diplomatic ties REGARDLESS of the strength of their armies. A highly empathetic leader may wish to form alliances with smaller, insignificant nations rather than go to war with them.

 

Great idea. This would be my preference, as well. It would make me feel like I'm actually playing against a "real" leader instead of a bunch of algorithms.

Reply #62 Top

Quoting Heph_, reply 61
Hmmm where is that promised update? (And "Ai wars 3" for that matter?)

 

I think that this thread technically is AI Wars 3.

Still waiting on that update though. O:)

(Like Brad has nothing better to do.)

Reply #63 Top

Quoting Istari, reply 62

Quoting Heph_, reply 61Hmmm where is that promised update? (And "Ai wars 3" for that matter?)
 

I think that this thread technically is AI Wars 3.

Still waiting on that update though.

(Like Brad has nothing better to do.)
I'm pretty sure Brad's primary focus is AI at the moment, and Kael certainly seems to be doing a good job with what he's doing, so I reckon that in fact Brad doesn't really have too much better to do. Barring higher priorities in his private life, of course.

Reply #64 Top

great videos - wildly entertaining!  :)