Slashman wrote:
I did see the Drakken with several armies of Driders in my game and one or two Justicieres.



While I like that the AI is using minor factions more, I think that it should limit itself to not more than a couple support units per stack.

I noticed that they use large number of minor faction troops, as well. I wouldn't mind it if they are using strong minor faction troops in place of inferior faction troops -- but yes, it doesn't always seem to be the case.

Antistone wrote:
Remember that you only need 9 techs to unlock each era, so there will probably be some early-game techs that you never get, even in the endgame.



Canal system is one that I often skip, personally; +1 industry on terrain with industry is pretty good if your city is huge (AND surrounded by certain terrain types), but in the early game when you've just got 1-2 districts that's at most 7-10 industry, and probably more like 5. It costs 250 industry to build, so it doesn't even pay for itself until ~50 turns later! (Longer with winter, since it only works in summer.) And that's before considering the dust you're losing in upkeep every turn (even if you consider industry, say, twice as valuable as dust, that adds another ~12 turns before it pays for itself), or the other things you could have researched or built instead. And then in the late game, when your cities are all larger, Canal System is better, but it also has a lot more competition...



So I wouldn't necessarily fault the AI for never building a Canal System. (Note: I don't have the expansions. Don't know if that affects the balance of this particular structure.)



(If you mean they haven't build ANY buildings at all from era 2 or later, then that's more concerning.)



Similarly, building faction units requires research, and so does assimilating a second minor faction. Again, you don't research every tech; you pick and choose. If the game is well-balanced, then those techs should sometimes be worth getting, but also sometimes not. If you can get away with using your starting faction unit plus one minor faction unit, that's not necessarily a bad plan.



(In fact the game is not perfectly balanced, and there are some faction units I'd be much more likely to unlock than others, but other players could give a better analysis of those details than I could...)

Yeah, I wish if there are more reasons to use main faction troops. Roleplay-wise, I would love to have one shaman in each army as WW -- but power-gaming wise, doing so is penalizing. I wonder if faction troops could be unlocked automatically by tech milestones (say, first additional unit unlocks at having five technologies, and second at era2). Or, tie them with technologies that are independently useful -- increased stack size or bonus to all main faction troops, for example.

Pondera wrote:
ever since I heard that the AI had been improved the last expac, I was thrilled, but work still needs to be done on it. In my last game, I waged a crusade against the Wild Walkers, and noticed that they had bought every mercenary in the store and were making very thorough use of the Ended minor faction unit. However, they did not use a single Agache Shaman or Tenei Walker. Cut over to the Vaulters, next on my list for genocide, and they were using nothing but marines and driders of all things, with a vast majority of their unit composition depending on said Driders. Looking over at my Argent Mage ally, I saw a similar effect with those Lizard Horsemen units and their warlocks. Seems the AI only uses one minor faction unit and their starter unit for everything.



Also, I noticed that they still aren't developing their cities as much as they should be. At Turn 240, they hadn't even built a canal system in their capital city yet.

A save would be appreciated, yes - not much we can really say otherwise.



In terms of unit technologies: remember that every technology you take means a different technology that you're not taking. Previously the AI would essentially unlock all military technologies first, but as a result it didn't have the economy required to really back its posturing up with anything. Units were unlocked early, but they were unlocked at the expense of other technologies. I'm confident that the way technology, including unit unlocks, is chosen now is superior to how it was chosen previously but, again, I'd be more than happy to look into your save game!



edit:

Slashman wrote:
Difficulty levels do NOT change AI behaviour. They only add heuristic based bonuses.

Well actually the harder difficulties levels are now less cooperative in diplomacy and more likely to declare war - it's something we added in the update.

You should peddle the link to the Twitch stream in which you go over these changes, wilbefast, it'd be great!



http://www.twitch.tv/amplitudestudios/v/26718516

wilbefast wrote:




Well actually the harder difficulties levels are now less cooperative in diplomacy and more likely to declare war - it's something we added in the update.

OK I did not know this, but this isn't really more varied play in terms of what the original question was aimed at. I think he was asking if the AI employs more/different strategies at higher levels.

Pondera wrote:
So you haven't seen this kind of odd unit composition? Maybe the problem lies in that I was using the even more epic mod. I've reset my game to standard and am about to give it another try.

Well if you have downloaded the last update, I have made no AI changes in my mod and I dont think the problem lies in that !

ever since I heard that the AI had been improved the last expac, I was thrilled, but work still needs to be done on it. In my last game, I waged a crusade against the Wild Walkers, and noticed that they had bought every mercenary in the store and were making very thorough use of the Ended minor faction unit. However, they did not use a single Agache Shaman or Tenei Walker. Cut over to the Vaulters, next on my list for genocide, and they were using nothing but marines and driders of all things, with a vast majority of their unit composition depending on said Driders. Looking over at my Argent Mage ally, I saw a similar effect with those Lizard Horsemen units and their warlocks. Seems the AI only uses one minor faction unit and their starter unit for everything.



Also, I noticed that they still aren't developing their cities as much as they should be. At Turn 240, they hadn't even built a canal system in their capital city yet.

Unfortunately I don't have a recent save that demonstrates this, but I've found that the AI behaves strangely in Island based maps.



I was playing on Endless difficulty.



Early in the game (Turn 20ish, pre Era-II), the Drakken declared war on me only knowing where my capital was through their racial. It turned out their island was right next to mine, but in the entirety of the game, they had never sent a single unit to my island.



Around turn 50, the Necrophage settled on my island. At this point my military was non-existent. I only had an army containing 4 units wearing Tier 1 iron armor. I walked up to their newly settled city, declared war and attacked as it only had militia inside. I captured the city. The following turn, the Necrophage were scared and immediately asked for a truce despite their army being >10x the size of mine in terms of military score. They wanted truce so desperately that I was able to take some tech and resources as well. Orange was the top score in the game at this point as well.



I managed to get to around turn 200 without a single unit ever nearing my coast short of the Necrophage settler that settled my land all the while having only the 4 unit tier 1 iron armor the entire time. I do recall this same behavior before this AI patch as well so it's not a step back in any way, but I would have hoped that they would be able to figure out how to take advantage of vulnerable races in the case of an island map.



I'll aim to send along a save game when I restart the game again on the same settings.

I have no input for AI empire/army management as I do not play any significant SP anymore.



I do think the AI is significantly worse then before when you auto against someone who manuals. Two big examples.



- Healers healing themselves when they are full health instead of another unit that actually has taken significant damage (drakken generals are CONSTANTLY doing this). "Buffing" also follows this pattern...Units should not buff if the net gain is less than what an attack would be.



- Target priority - seems like the AI has no ability to rank a units threat level, rather instead, relies on it's weapon RPS to choose targets. This leads to a lot of horrific unit targeting. Even the most beginner player would know to target the most dangerous units first. If the threat level is close...fine...use your best RPS to pick a target. If you have 3 militia on the field hitting for 30 and 3 ranged units on the field hitting for 100 there should never be a question of which units should be targeted. This is so basic and it's so disappointing it's not happening in game. It really feels like a big step back....it's absolutely game breaking when this happens against another human being.

I've made note to look into the battle targeting - what I really need are specific cases where there is an issue. I haven't, for example, been able to reproduce a healer healing itself. I've no doubt the issue exists though because you're not the only person who has brought it up. If you can send me save I'd really appreciate it.

re healing itself: I cannot say if I´ve seen it happen after the patch, although I know it happened; but I´d like to point out the case Bigballs describe (manual v auto) is known to have some GUI issues showing the auto player exactly what´s happening.

I think that's always the case, but I feel that the AI update was very good.

Nasarog wrote:
I think that's always the case, but I feel that the AI update was very good.

I second this. Strategically speaking, the AI was able to keep pace with me, and in some cases surpass me, as BL through a good portion of the game. Of course BL's Dust nature makes that less easy in the latter stages because I can buyout everything including population (maybe BL are a bit overpowered).



One thing I am a bit concerned with is weakness of AI armies. It seems like the unit design aspect of the AI is still a bit slightly lacking. This is troubling when you take into account that the AI sometimes has more than enough resources to outfit armies with superior equipment.

I've made note to look into the battle targeting - what I really need are specific cases where there is an issue. I haven't, for example, been able to reproduce a healer healing itself. I've no doubt the issue exists though because you're not the only person who has brought it up. If you can send me save I'd really appreciate it.

This is really almost impossible from our end since we cant save/pause mid battle. Most of the issues that i have affect 1-2 units and are not repeatable when i reload the autosave. Would non-repeatable saves be of any help?

Slashman wrote:


One thing I am a bit concerned with is weakness of AI armies. It seems like the unit design aspect of the AI is still a bit slightly lacking. This is troubling when you take into account that the AI sometimes has more than enough resources to outfit armies with superior equipment.

This. I have a supremacy only game that is around turn 200. The Endless AI has all technology unlocked, tons of strategics, and the the economic advantage. I finally unlocked Serum of Iteru, so we should be on equal footing... but im cutting up the AI's units and empire easily.



Example Save: AI attacks me twice, auto-battle gives me easy wins, manual battling gives me easy wins. Why is it so easy? Then the AI attacks again, with reinforcements that were left out of an early battle...
FIGHT REDS.zip

wilbefast wrote:
Hmm... yes I'm afraid there are changes to the AI file structure that mods will need to take into account.

What do you mean by that ?

I think the A.I needs to have a similar mechanic that CivBE had in its last patch.The game basically lets you play out a science,wonder or spaceship win without doing anything to stop the player or other A.I's.It is pretty boring just waiting 40 turns to win.My neighbor the roving clans had a military 2 times by size.

Ashbery76 wrote:
I think the A.I needs to have a similar mechanic that CivBE had in its last patch.The game basically lets to play out a science,wonder or spaceship win without doing anything to stop the player or other A.I's.It is pretty boring just waiting 40 turns to win.My neighbor the roving clans had a military 2 times by size.

This would be perfect. Is there any way to at least partially mod it in? Thanks in advance.

Hi,



I noticed a strange behavior regarding AI and peace proposal.

When an AI breaks a peace, you can usually, in the same turn, propose peace again (and have it accepted) without giving any resource nor dust. It feels a bit strange : if it had a good reason to break peace why would it accept it again so easily?



I don't know if it is related to the recent changes because I used not to bother about researching diplomatic manse before.

I'm using peace treaties on higher difficulty levels when I play cultists as a way to keep converted villages and focus on attacking only one AI at a time.

However, it almost feels as an exploit due to the way the AI behaves.

thurse wrote:
Hi,



I noticed a strange behavior regarding AI and peace proposal.

When an AI breaks a peace, you can usually, in the same turn, propose peace again (and have it accepted) without giving any resource nor dust. It feels a bit strange : if it had a good reason to break peace why would it accept it again so easily?



I don't know if it is related to the recent changes because I used not to bother about researching diplomatic manse before.

I'm using peace treaties on higher difficulty levels when I play cultists as a way to keep converted villages and focus on attacking only one AI at a time.

However, it almost feels as an exploit due to the way the AI behaves.

It is best to provide saves for stuff like this so the devs can take a look at actual examples.

Slashman wrote:
It is best to provide saves for stuff like this so the devs can take a look at actual examples.

I do have an example of this issue actually, it's one of the things we're working on at the moment.



It's an issue we had in the old system and it apparently hasn't been entirely fixed by the new one, because we still evaluate how much we want to propose a term differently from how much we value that term if it is offered to us. This is so that, for example, Roving Clans can propose to remove a market ban at a price.



I think the best solution might be to very brutally force a negative evaluation of any term that contradicts something we want to propose. This wouldn't break the RC's behaviour (or other similar behaviours) and should fix issues where opening borders is seen positively just after borders have been closed, for example.