While the AI in 1.3 has seen improvement, I believe the following helpful advice gleaned from various threads over the last weeks might help to greatly improve its performance. In the hopes that His Right Honourable Msgr. Bradley Wardell might grace this meagre thread with a perfunctory perusal, I herewith scribe:
1. AI Unitbuilding and Unit Design:
- Suboptimality: AI designs units with traits it cannot use, for example Wraiths (Resoln) with Plate Mail Proficiency. Less severe: early-game defenders with Plate Mail Proficiency.
- Suggestion 1: AI should design separate unit Templates for each faction, noting some of the most common strengths and weaknesses within each faction.
- Suggestion 2: The AI should rely MUCH less on "upgrading" existing units, and simply upon designing new ones as technology becomes available. Increase the number of templates and individual designs to reflect various stages in technology, particularly the early game, in which armor is not available
- Suggestion 3: AI unit designs should become available to the AI to build from after certain triggers have been activated, most notably, the discovery of a prerequisite technology.
2. AI Citybuilding:
- Suboptimality: The AI does not plan or use its cities effectively. Example of usage: Conclaves producing units while Fortresses produce research; available enchantment slots in Conclaves include Aura of Might when producing Wealth. Examples of planning: Tiles without essence are chosen for Conclaves.
- Suggestion 1: The AI should favor settling tiles with Essence more than those without. The AI should plan to make Conclaves out of villages with at least 2 essence, otherwise it should make towns or forts.
- Suggestion 2: AI should use fortresses primarily for generating units, not producing research; AI should make fewer conclaves, build more research- or mana-generating structures in conclaves, use conclaves primarily for generating research or mana, train fewer units in conclaves; AI should build more food- or wealth-generating structures in towns, use towns primarily for generating food or wealth, train fewer units in towns
- Suggestion 3: The AI should not settle tiles with 6 or fewer combined resources (unless there are 3 or more harvestable resources within range and no other tiles are appropriate). AI should greatly favor settling on tiles with 9 combined resources, even if there are no harvestable tiles nearby; AI should favor settling on tiles with 8 combined resources, even if there are no harvestable tiles nearby; AI should favor settling on tiles with 7 combined resources if they contain 1 essence, even if there are no harvestable resources nearby.
3. AI Championbuilding:
- Suboptimality: The AI will frequently equip a "natural" mage, such as Procipine, as an axe-wielding warrior, resulting in greatly ineffective champion.
- Suggestion 1: AI Sovereigns should have more detailed templates according to which the AI chooses equipment, traits, and behavior. Some sovereigns, such as Procipine, should focus much more on using magic spells.
. AI Tactical Behavior:
- Suboptimality: The AI will often choose suboptimal behavior for its champions in tactical combat. The AI *will* spend mana on strategic unit blesses, it appears, but seems much less likely to spend mana tactically. The AI will choose (also for human player champions with a "beserk"-causing spell or item) to keep a melee unit back with archers instead of forward.
- Suggestion 1: The AI should be much more willing to spend mana to cast buffs and debuffs. The AI should be more willing to cast spells.
- Suggestion 2: The AI should be more willing to place its melee champions in a position to attack in future rounds, also helping to "defend" its back ranks.
5. AI Strategic Behavior:
- Suboptimality: The AI will often not move its units (at all), but keep single units just outside of cities, or near enemies / ruffians, getting them picked off easily. The AI will much too frequently undervalue peace. The AI will leave unattended lairs, treasure chests, or enemy outposts uncaptured. The AI will much too frequently rely on mediocre combinations of chaff units produced in highly unproductive towns or conclaves or villages to defend its vital interests. The AI relies on units to attack more than it does Champions backed up by units. The AI will build "stacks" of scouts to "support" its champions, instead of ranged units (to support melee champions) or melee units (to support ranged champions). The AI does not compliment its champions (1 melee, 1 support) or its troops well (support champion needs fitting troops, and needs to cast spells to support them).
6. AI Pathfinding:
- Suboptimality: The AI is greatly hindered by the current rule mechanic disallowing most non-hostile entities to occupy the same tile simultaneously, causing them to go for dozens of turns out of their way in the faulty hope of thusly being able to bypass a bottleneck. (This also hinders pathfinding of human units.)
- Suggestion 1: change ruleset to allow non-hostile entities to occupy the same tile simultaneously; non-hostile units can occupy same tile as other non-hostile units and structures.
- Suggestion 2: Should a destination be "blocked" by (hostile) units, the unit pathfinder should not find a route-around, but a different desirable destination.
Link not displaying in above post for some reason
No idea why the images still aren't working, they work when I go directly to the links.
N/A
The 3rd one, the custom faction AI had a power graph that looks like Ythril on that mana graph. It was at like 330 when the rest of us were at 150 or less because it was conquering Resoln and Ythril cities left and right at the beginning of the game.
It was basically the definition of runaway AI. It dominated the game from turn 1 till turn 200+ with at least double the score of the next highest the whole time. It was positioned such that it basically had half the map for free while the rest of us fight for space amongst ourselves on the other half. Out of all the factions, I would think that would have the highest mana graph.
Do warrior AIs always start with some ridiculous pool of mana like that?
Do you think the AI is just playing the custom faction wrong? Is it assuming X sort of AI when it should be assuming Y sort of AI?
My guess is that it was just using lots of unit enchantments and never got anywhere with per turn generation, as is usually the case.
When you play higher difficulty than challenging the AI get extra bonuses, giving them all the same starting mana would be such a bonus, as well as more gold. So, i wouldn't consider bonuses to the AI a bug when playing higher than challenging since they have said the AI do not get these until they play higher than challenging. One of the reasons I like this AI, because I can play a game without the AI cheating its brains out to play a game.
All my suggestions are in my mod. Please try it and ask me what you need to make AI better. https://forums.elementalgame.com/450250/page/1/
I still think the OP applies, and most of the suggestions there should be fairly straightforward to implement and would greatly enhance the AI without further cheating.
In my mod AI only units are very cheap and easy to build, they also reduce unrest by -20. These are AI only units. Human player should know other methods.
Another idea, is to make AI only city militia instead of common city militia which automatically appears in an AI city and reduces unrest.
Another idea works only for Expert level difficulty and higher. All AIs receive a bonus mega-noble trait (let's call it Happy nation) and live with lower unrest.
I would agree that the AI does generally try to go really wide and the game does greatly punish that.
Anywhere they can find a settleable space they generally settle in it, even if it is a horrible space and no player would ever put a city there.
I can understand the idea behind that, because given enough time cities usually pay off and because AIs can't really be taught complex strategy like the kind you would need if you wanted to teach an AI where not to build cities.
It doesn't really seem fair to make the AI get -20 unrest per defending unit, though. The unrest penalty is there to limit the viability of going really wide and it seems like it is doing its job like it is supposed to.
Just letting the AI expand infinitely with no drawbacks is kinda iffy to me.
Is there a way to add a calming trait? Something that would do -3 unrest and some way to make the AI use it in place of some other trait that it is using now?
That seems like a pretty fair way to go about fixing their unrest problems to me.
So you want to use this trait too? AI will never be as good as player. You have to decide what you want Challenging AI but with chets, or equal rights to human player and AI, no cheats... We already have this second variant.
All proposals to "change AI" I think are very expensive to devs.
I don't think I would use the calming trait very often because I am not an ICS kind of player, but all the AIs are ICS AIs so it should help them a lot, especially since they produce many more units than I do.
It just seems like Cleric would be better as a unit trait than as a building.
If I am playing with henchmen, I will often make lots of commanders and put them in cities to keep my unrest down.
I usually play humans and I do often end up going that route, but it's kinda unfair that only humans can do it. It seems like a strategy that should be open to everybody if they are willing to devote resources to it.
The best help for AI is the help which doesn't need any actions from AI. Who knows, will it act so or not?
Here's something dumb the AI does. As Magnar they love spamming slaves and slave archers in particular. [e digicons]:'([/e]
Slaves are for backing up existing elite units and wearing the enemy down before hand. You should never build just slaves. I can kill an entire stack of slave archers and take no damage as all those units have like 10 initiative and no health. Slaves are cannon fodder not damage dealers so why build slave archers. Seriously the AI needs to build some elite Quendar units, the game rewards elite units not masses of weak ones. A good strategy would be some elite Quendar archers or macemen backed up by slave defenders to tank.
Secondly, my current pet peeve is how the AII never builds any production, troop training, or unrest buildings. This allows me to easily out produce them midgame. Seriously I always build workshops and bell towers first in brand new cities as it allows them to develop quickly and catch up. In fact these are some of the few buildings I rush. The AI seemed to like building troops from it's low level border cities, even the newly settled ones. I don't ever build units from a newly settled city. It takes forever and stunts the cities development.
This is because AI has low priority for building type "Workshop". This priority is fixed in my Patchwork mod https://forums.elementalgame.com/450250/page/1/ Now AI has better production race.
I am actually interested in the economics of workshops for the AI.
Is it really that beneficial for them to even make workshops?
A faction has to obviously spend time to build the things and it adds to a base of production and pays itself back over time.
Does it really impact the AI? Are we sure that the AI is producing things according to the same rules the players are?
It sure seems to me like the AI produces a whole lot more than I do at the baseline.
If it is producing like I am, then it is getting more done by not building workshops.
If it isn't producing like I am, then it sure seems to get enough done by just doing what it is doing.
Maybe the AI isn't limited by the production stat or by the unrest stat like I am?
Maybe the production and unrest stats are not really what controls the AI's production and maybe the reason players do better than AIs is not really related to those things?
AIs use productivity and unrest like players. Unless you are playing higher then Challenging and then they get a boost but they still follow the rules.
When I play the AI cities are always far behind mine. I usually have much higher level improvements and although the AI has a lot of units my upgraded fortresses can pump out a unit with chainmail in 3 turns midgame.
The thing that throws me off is that the AIs seem to be completing things at a reasonable speed when I look at their cities.
If they were grinding their economy to a complete halt by not building production buildings and not building unrest buildings then it would seem like they would get to the point where they couldn't complete anything in any reasonable amount of time.
To support the viewpoint that AIs play by the same rules, I can say that the few times the AIs do really really well in the graphs they are often playing factions I created. The one time in particular that I remember the AI just going nuts was when they were using one of my factions in particular where I designed a whole lot of henchmen type units for it. Because of that, they made dozens of henchmen in the game. If 20% of those became commanders with unrest reduction abilities and they put them in every city as defenders (entirely possible) then it would have been hard for them to have anything other than 0 unrest anywhere.
It probably does help that those units couldn't die (only rack up handicaps) and that that AI more or less had half the map to expand into uncontested. For that matter the act of them dying would just power up their cities as the henchmen got immobilized at home and the cities were forced to 0 unrest immediately.
We obviously play on different maps.
I never see stuff taking 50 turns for the AI to make, but it's hard for the AI to get more than 4 or 5 cities on the tiny maps that I play on.
With me they keep reasonable production speeds and still never go anywhere in terms of power.
If their production varies inversely according to the map size then the best way to help the AI is to play the smallest possible maps.
Indeed, the smallest possible maps give the player the greatest likelihood to get a bad start. That by itself gives the AI plenty of win %.
There are many great features available to you once you register, including:
Sign in or Create Account
