Open Beta Update 25/7/2023
- Thread starter FangirlCrazily
- Start date
-
We have updated our Community Code of Conduct. Please read through the new rules for the forum that are an integral part of Paradox Interactive’s User Agreement.
Not to a sufficient degree, imo. I still see AIs run around with 3 support units in a single stack or 6+ among three stacks, and it tends to generally have lower-tier units than I do (although not always); but perhaps research bonuses will alleviate that part, at least.
Either way, another area of gain that might not be immediately apparent when it comes to army combat strength is composition. Balancing the number of melee and ranged units (I usually attempt to get a 3/3 balance, with Skirmishers counting as wildcards) and ensuring every stack has precisely 1 support (or 0, if you have no good support units, and certainly if you only have wildspeakers with no T3+ animal units, ffs AI stop stacking five Wildspeakers together only for their summons to get oneshot on approach) does a lot in improving your army's actual combat power, simply because it ensures you have ranged units for damage and melee units in front so your opponent cannot simply run ahead and slaughter the ranged units, while also having a heal.
I know that. What I'm saying is that it's not just manual combat where I perform better than the army strength estimate given, my armies also tend to perform better than that in autocombat. Which means that there are modifiers to the strength of a stack that aren't represented in it's given army strength. In itself, that's not a surprise, as it's impossible to accurately value every single modifier and unit synergy in the army strength, but it does mean that there are ways to increase the strength of an army that you can't see in the army strength. Specifically, if you've got 600 army strength from, say, 6 support units, or if you have 600 army strength from 1 support unit, 2 archer units, 1 polearm unit and 2 shield units, the second is absolutely going to beat the first.
I realize it is difficult, possibly even impossible, to teach the AI to take this more difficult to catch modifier into account, but we can determine the sources of it and teach the AI to attempt to build it's armies in the same way, even if it doesn't see an advantage in army strength as a result.
Capping the AIs army power relative to the strongest Human player just means it is hard to actively lose a game.
If you are playing alone vs AI and you lose an army, they won't take advantage of it to steamroll you afterwards.
Maybe the cap for the AI should only increase and never decrease? So they would only be 30% more powerful than your strongest historical state.
- 4
- 1
Let's be frank here.
Capping the military of AI opponents with regard to what the human player does is child mode.
You need at least one difficulty level where the AI gets free reign. Which is already a stretch, because if the AI has too many units you could just decrease their economic advantages. I have the suspicion that the reason for all this nonsense is that "empire building" is more pronounced now and too many people coming from the empire builders are neglecting military, falsely thinking that early investment in military is wasted, even though you will get a lot of rewards for fights, both in XP and resources.
So even if there is a veneer of empire building, diplomacy, trade, free cities and so on, you STILL need to fight as much as possible and invest into military. If you neglect that aspect you deserve to be hammered. Not pampered.
Capping the military of AI opponents with regard to what the human player does is child mode.
You need at least one difficulty level where the AI gets free reign. Which is already a stretch, because if the AI has too many units you could just decrease their economic advantages. I have the suspicion that the reason for all this nonsense is that "empire building" is more pronounced now and too many people coming from the empire builders are neglecting military, falsely thinking that early investment in military is wasted, even though you will get a lot of rewards for fights, both in XP and resources.
So even if there is a veneer of empire building, diplomacy, trade, free cities and so on, you STILL need to fight as much as possible and invest into military. If you neglect that aspect you deserve to be hammered. Not pampered.
- 1
- 1
- 1
I tend to focus quite heavily on empire building, both because I like it and because I've played a lot of Civ VI, and frankly I still don't think I'm neglecting my military.
That said, I'm eager to see an uncapped AI prove me wrong. Strategy games are the most interesting when your playstyle is challenged and needs improvements, after all.
So far, I cannot remember, in 330 hours of playtime, that I had to make the decision to leave the production queue of a city empty because I needed to build more units. It's always been a case of 'eh, I'll just build the units a bit later, it's not like I'm in a hurry'.
And in more general terms, I agree that the AI, at least on the harder difficulties (and in fact I'd argue, anything from Normal difficulty upwards) should not be limited by (a modifier to) the strength of the player. For the lower difficulties, I can see the advantage because it stops the AI from rolling over a completely clueless player. But on higher difficulties, it's hard enough already to make the AI strong enough to compete with the experienced players, so let's not handicap it further!
That said, I'm eager to see an uncapped AI prove me wrong. Strategy games are the most interesting when your playstyle is challenged and needs improvements, after all.
So far, I cannot remember, in 330 hours of playtime, that I had to make the decision to leave the production queue of a city empty because I needed to build more units. It's always been a case of 'eh, I'll just build the units a bit later, it's not like I'm in a hurry'.
And in more general terms, I agree that the AI, at least on the harder difficulties (and in fact I'd argue, anything from Normal difficulty upwards) should not be limited by (a modifier to) the strength of the player. For the lower difficulties, I can see the advantage because it stops the AI from rolling over a completely clueless player. But on higher difficulties, it's hard enough already to make the AI strong enough to compete with the experienced players, so let's not handicap it further!
- 2
- 1
Please correct the multiplayer error - termination of the time limit for turn when one of the players gave up
The issue I see with capping the AI based on the player's army size is, that it incentivizes the player to not build up backup units, as any unit built and kept in the rear causes the AI to build more units and throw them at the front.
- 2
- 1
Couldn't agree more, specially about the cap never reducing, only ever going up.
I think that approach would make a lot of sense and balance the desire to have a challenge while reducing the chance the AI will tank its economy by mass producing units.
- 2
At least for the highest difficulty: please, remove the proposed unit cap of the AI. We need a difficulty setting that gives a challenge.
And otherwise, make the cap more like +300% and let the cap never decrease.
BTW. I like several changes in the new update ...
And otherwise, make the cap more like +300% and let the cap never decrease.
BTW. I like several changes in the new update ...
Also, keep in mind that the AI players may have vastly different military necessities than the human player. What if the human player starts in a somewhat isolated position underground and doesn't need to invest heavily into military early on? What if they invest into additional heroes to add military capacity? What if an AI has to wage a war against a hostile Free City - or what if two AIs clash? What if AIs have more cities and a larger territory to guard? What if an AI tries to go for some victory that involves stacks out of nowhere attacking - and the human player simply demilitarizes? Release a city or two as a Vassal and simply disbands half their army?
I appreciate the feedback guys but this change has been in since the start of the Open Beta. Giving this feedback now is just too late, we can't make changes to AI Behavior anymore without risking upsetting the entire system and balance.
The question here is, why are you even making back up units?
Keep in mind that the AI will not start disbanding units just to reach the threshold, they stay where they're at until they start losing units. If it's winning combats it will just keep going, even if the player now has less Army Value.
I like having uncommitted reserves in the back, so I can reinforce my frontlines where needed. In addition, they can deal with minor incidents like quest mobs, pillaging marauders or harass enemy troops that snuck by or simply attack from another side without having to recall a stack from the frontline.
Plus, if I'm running a surplus of 1000 gold I feel like I should do something with it.
- 4
@ Triumph-Jordi
Just a question: Don't you think that the AIs need significantly more military might to have a chance at winning against a human opponent, considering battles are limited to 18 vs. 18 units max? At the 18 vs. 18 stage, with roughly equal strength, the human player will win with, well, depending on the game stage, half to two thirds or even three quarters of their units still alive. To win, the AI MUST field the same again.
Now, the problem, here is, to change that, the handicaps would have made sense, since it was supposed to make up for that (the battle handicap). But that has now been shelved.
So we have now this:
Initially people complained about "unit waves" of the AI - too many units to squash, a slog.
You reacted, by limiting the military building capacity and at the same time handicapping the human player in battle.
Then people complained about the handicap being "no fun", seeing the dregs of the AI doing mor damage than the elite of the humans.
But that leaves just the limiting of the military building capacity now, making the AI "too weak" now and "handicapped".
So basically spoken, if you cap the AIs building capacity, you MUST keep the battle handicap in.
Put it in again with the official patch, take the flak that will come and use the time up to the next patch to rethink this.
I'd say, that there is nothing you can do to help the AI being "competitive" that will not gain negative feedback.
In my opinion, there should be difficulty modes to satisfy everyone. No one is forced to play on highest level, just to be able to say, I beat the game on brutal difficulty. So brutal difficulty SHOULD lead to people losing games. I think that the good players, those who have the ability to play on that level, will rather play against overwhelming odds, than being reduced to suddenly dealing no damage in battle anymore. I also think, they would rather lose against Pretender Kings instead of winning, knowing that the Pretender Kings start building up their army only, when the human player is within X% of their capability. You don't need to play Pretender Kings then.
So that will be a tough nut to crack for you.
Just a question: Don't you think that the AIs need significantly more military might to have a chance at winning against a human opponent, considering battles are limited to 18 vs. 18 units max? At the 18 vs. 18 stage, with roughly equal strength, the human player will win with, well, depending on the game stage, half to two thirds or even three quarters of their units still alive. To win, the AI MUST field the same again.
Now, the problem, here is, to change that, the handicaps would have made sense, since it was supposed to make up for that (the battle handicap). But that has now been shelved.
So we have now this:
Initially people complained about "unit waves" of the AI - too many units to squash, a slog.
You reacted, by limiting the military building capacity and at the same time handicapping the human player in battle.
Then people complained about the handicap being "no fun", seeing the dregs of the AI doing mor damage than the elite of the humans.
But that leaves just the limiting of the military building capacity now, making the AI "too weak" now and "handicapped".
So basically spoken, if you cap the AIs building capacity, you MUST keep the battle handicap in.
Put it in again with the official patch, take the flak that will come and use the time up to the next patch to rethink this.
I'd say, that there is nothing you can do to help the AI being "competitive" that will not gain negative feedback.
In my opinion, there should be difficulty modes to satisfy everyone. No one is forced to play on highest level, just to be able to say, I beat the game on brutal difficulty. So brutal difficulty SHOULD lead to people losing games. I think that the good players, those who have the ability to play on that level, will rather play against overwhelming odds, than being reduced to suddenly dealing no damage in battle anymore. I also think, they would rather lose against Pretender Kings instead of winning, knowing that the Pretender Kings start building up their army only, when the human player is within X% of their capability. You don't need to play Pretender Kings then.
So that will be a tough nut to crack for you.
- 1
Honestly, it did strike me as strange that somehow no one decided to talk about it until now...
Still, how possible is my suggestion? If we can't have a split between a Peace unit cap and a War unit cap, then maybe the numbers can be doubled? I feel like half the issue is not that the cap exists, but that the % difference between player and AI is far too small.
I appreciate how hard of a job it is to balance not just game stability but multiple play experiences for different players.
Take your time, we can wait for the next patch for a potential cap removal or, imo the better option, the cap never reducing.
More than anything, I believe the North Star for AI should always be to WIN the match. If it is not realistic trying to do that, than it needs to change. That doesn't mean the AI should be focused in any meta strategy, the same way many players that play to win don't worry about meta at all.
Now, while the AI is trying to win, it should try to roleplay to the best of its ability but never at the detriment of trying to win the match. It is then the job of the game to try to give the player in-game lore -based reasons why the AI will sometimes abandon roleplay to push a victory condition. This is particularly lacking from AOW4 right now and I feel it distinctively when I am in the late game getting DOW from former allies (but never attacked, haven't had a single siege defense in the game yet) and I am left trying to do RP gymnastics to explain the AI behavior so I can suspend disbelief (and fail)
Above it all though, I hope the Triumph folks remember that you all got most of your decisions right and I love AOW4. Cheers.
No, no. The problem isn't actually the cap - there must be a cap, so that the AIs don't get a negative income and retain gold for building and mana for spellcasting -, the problem is the fact that the cap is linked to the military of the human player(s), and not linked with, say, upkeep to income relation of the AI in question, in other words, their economic situation. As a byproduct, a cap based on the human player military makes map features like Pretender Kings useless - why start the map with a warring Triumvirat of stronger Pretender Kings, when they cannot make use fof their starting advantage, but instead have to wait with further arming, until the human player(s) are strong enough to compete?
- 1
I'm not demanding everything is changed now, and I don't think other people are either. If anything, I consider every piece of feedback I give to be a general piece of feedback. If it can't be added to the coming patch (assuming you guys want to implement it at all), that's fine. It can go in the patch after it. Or even the one after that. Or whenever you guys have time to actually add it.
Don't stress too much.
The game is already great, and you guys are already improving it at amazing speed! I can only imagine how enjoyable it will be several years from now.- 2
- 1
I wanted to stress that I agree with Leyrann and mean criticism the same way. I am very happy with the communication between dev team and the fans and I feel heard. Thank you Triumph for your interaction with the fans and all the work you put into the game.I'm not demanding everything is changed now, and I don't think other people are either. If anything, I consider every piece of feedback I give to be a general piece of feedback. If it can't be added to the coming patch (assuming you guys want to implement it at all), that's fine. It can go in the patch after it. Or even the one after that. Or whenever you guys have time to actually add it.
Don't stress too much.
Last edited:
- 2
- 2
Ok, that can happen. Maybe this topic should have been discussed earlier - in hindsight. Let's see how the new patch works out ... many good changes to look forward to. But I am worried about the reduced challenge though.
Yes, that makes no sense. These different starting conditions were one of the reasons I felt I was challenged. Now the challenge became less.