• 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.
Once they leave land, they turn into ships. Peasant ahips, defender ships(that ram!), archer ships, mage ships...you get the picture? It's a dumb way of expressing naval transport and showing naval engagement. A single "ghost ship" type that carries defenders, warriors, knights, etc is more exciting to use against smaller pirate coves seabosses. and puts real risk into carrying your troops over water. Thats the type of naval tbs I'd like to see.

I try to value the opinions of others, but this opinion of yours is very, very shortsighted when it comes to naval gameplay. I can't agree with you on this, I think triumph can do better. I also don't appreciate the attempt to strawman my observations into a meaningless attempt at assuming what type of game i want to see.

We're not discussing land gameplay yet, this topic is about themes we want focused on in season 3. I would like to see the naval theme explored.

I agree with you but I think they redacted the original naval gameplay because of the AI. Perhaps they are waiting for the AI to improve significantly enough to add it back into the game.
 
  • 1
Reactions:
In general I hope the future AOW4 development is focused more on adding options for synergies, thus new playstyles guarantying even more fun options and replay value.

I don't care about any new skins of races or weirder and weirder mounts. Better AI and for instance improving diplomacy would do more for this game than another form.

I would be okay with one dlc that adds 10 new tomes and plenty of options to hero's customization (new faces, clothes, haircuts, facial hair).

I would love the Dark culture reworked with another (sub)culture with a Dark/Chaos start. If there is another undead theme, like vampires, I'd love to see more uses for the Souls economy other than resurrected Skeletons.

I think Triumph's best dlcs have been so far those that were built around the AoW's lore. I'd rather see more elements framed by the AoW lore's history, adding depth and RPG to the game, than everything goes approach.

Frankly I also see more "water stuff" situational and unnecessary.

Asking for 'better AI' is more or less identical to asking for 'better performance'. It's a reasonable request but necessarily would have to come bundled with content, and it is pretty silly to think that the developers don't already know that players would like either of those things.

There is not going to be the Better AI Expansion, for so many reasons. How do you even sell that? How do you even announce it without the announcement sounding like the the game needs to be fixed rather than expanded on? If I'm doing a dev stream to showcase the Better AI Expansion, how do I actually show that work off?

On an organizational level, Triumph also certainly has different teams working on different aspects of the game. What is the art team doing for the Better AI Expansion? What are the gameplay design teams doing? The map design teams? The narrative teams?

So, 'I don't want content, just better AI' misses the point that if you want better AI, it is going to come with an expansion that includes new content. And ergo it makes much more sense to voice your preference for upcoming content rather than just telling the developers something they already know that is constantly echoed over and over again (and frankly that I think would be said over and over again even if we ever got SkyNet level AI because players are so used to complaining about AI now that I think they'd still do it regardless of how capable their SP opponent was).


And even if all of the stars aligned and you created and successfully marketed the Better AI Expansion... IMHO, a bunch of people would actually hate it. Like, the majority of players would not have fun with it. I think there is a lack of appreciation for just how many people play AoW 4 and think the game is already super hard. That cannot tackle the presently implemented hardest AI.

I've been playtesting a single player tabletop game based on X-Wing, and have been reeaaally shocked at how 90~ percent of people approach play. My sense of what 'normal' difficulty is has been inflated by years of consuming speedrun content and considering myself to be a mediocre strategy game player. But no, I'm in an upper echelon - it's just that the higher you go up the skill curve, the steeper the curve gets. Most people can barely eke out a win against a pretty mid level AI opponent, even one that is only implemented in cardboard.

SkyNet level AI would likely cause most people to stop playing the game. Power users would cheer, but those aren't most players.
 
  • 1
Reactions:
With ai or PvE, there has to be a consensus. You want casual and hard-core gamers to enjoy the game, and thats why we have difficulty levels to give both a fun way to play the same game with different challenges.

AI should be improved in any way that reduces the need to have to cheat against human players. Personalities should dictate their playstyle.

I would pay for AI improvements though, especially if bundled with new forms, units, and tomes.
 
With ai or PvE, there has to be a consensus. You want casual and hard-core gamers to enjoy the game, and thats why we have difficulty levels to give both a fun way to play the same game with different challenges.

AI should be improved in any way that reduces the need to have to cheat against human players. Personalities should dictate their playstyle.

I would pay for AI improvements though, especially if bundled with new forms, units, and tomes.

Having an AI in a commercial game that doesn't need to cheat to present a challenge to a veteran player is an unrealistic expectation. Even something like AI War, where the AI is REALLY good and is helped by having an overwhelming initial advantage against the player, cannot present a meaningful challenge to a veteran player without cheating. SNK bosses only function via cheating and having unfair movesets. Same with Gill.

Classic XCOM only works by giving the aliens cheaty amounts of stats and night vision (which makes thematic sense because they're ALIENS!, but in mechanical terms is still cheating that is really obvious if you look beyond the aesthetics). Darkest Dungeon works by giving the enemies stat cheats and a whole different type of damage that only the player ever needs to worry about that stacks-up over time and is super sticky. Battle Brothers works by forcing the player to manage fatigue (the enemies all fatigue cheat) and limiting their squad size so the dice are always tilted against the player.

And in all of the above instances, the AI STILL cannot challenge a veteran player.

The idea of having a challenging zero cheats AI in a somewhat symmetrical 4X game with all players starting on roughly the same footing, and where the player has access to mountains and mountains of meta data that the AI does not, is absurd. It is an unachievable goal. Personally, I would rather have game development be focused on what is actually possible rather than chasing impractical dreams that might not even improve gameplay.


But honestly I'm tired of this same dead horse being kicked over and over. I'm done with it. people who don't even understand what AI is or how game logic works wanting 'better AI' without knowing what they are even asking for and suggesting since they would pay for it well then obviously it would be a good selling feature for a commercial product (when the reality is that, at best, most consumers don't care).

Personally I hope S3 happens and is full of cool toys. I'd be happy with a slightly smarter opponent but am happy with where the AI is at right now. Certainly we're come a long way from the far-too-passive AI of AoW 4's release.
 
One thing I forgot to say earlier is that I've been loving the crystal dwelling aesthetic, and the stuff that you can get with it. I just wish you could get that kind of aesthetic style for your own faction's buildings. I'd love to be able to get that as a culture type or something.
 
Last edited:
Having an AI in a commercial game that doesn't need to cheat to present a challenge to a veteran player is an unrealistic expectation. Even something like AI War, where the AI is REALLY good and is helped by having an overwhelming initial advantage against the player, cannot present a meaningful challenge to a veteran player without cheating. SNK bosses only function via cheating and having unfair movesets. Same with Gill.

Classic XCOM only works by giving the aliens cheaty amounts of stats and night vision (which makes thematic sense because they're ALIENS!, but in mechanical terms is still cheating that is really obvious if you look beyond the aesthetics). Darkest Dungeon works by giving the enemies stat cheats and a whole different type of damage that only the player ever needs to worry about that stacks-up over time and is super sticky. Battle Brothers works by forcing the player to manage fatigue (the enemies all fatigue cheat) and limiting their squad size so the dice are always tilted against the player.

And in all of the above instances, the AI STILL cannot challenge a veteran player.

The idea of having a challenging zero cheats AI in a somewhat symmetrical 4X game with all players starting on roughly the same footing, and where the player has access to mountains and mountains of meta data that the AI does not, is absurd. It is an unachievable goal. Personally, I would rather have game development be focused on what is actually possible rather than chasing impractical dreams that might not even improve gameplay.


But honestly I'm tired of this same dead horse being kicked over and over. I'm done with it. people who don't even understand what AI is or how game logic works wanting 'better AI' without knowing what they are even asking for and suggesting since they would pay for it well then obviously it would be a good selling feature for a commercial product (when the reality is that, at best, most consumers don't care).

Personally I hope S3 happens and is full of cool toys. I'd be happy with a slightly smarter opponent but am happy with where the AI is at right now. Certainly we're come a long way from the far-too-passive AI of AoW 4's release.
Kev, I'm not a programmer or a software developer. But I am a paying customer that likes to know what is possible and what is not. Maybe you know more about the finer points of AI tuning than I do, so when you say it's impossible to make a competent ai without cheating enabled, then I'll just take your word for it.

It would be nice to have triumph's official opinion on the matter though. Many players have expressed frustration, on this forum and on reddit, playing against an AI that cheats and can pull stacks and fully equipped heroes out of its backside whenever the need arises. I recall seeing a cheat modifer in realm settings which should help players manage these issues for the time being.

Players shouldn't complain about an ai that cheats on the hardest or second hardest difficulties, but the way the ai cheats and how effective the cheats are against the average player might merit some consideration. There doesn't seem to be an agreement on how powerful the program should be.
 
Last edited:
Kev, I'm not a programmer or a software developer. But I am a paying customer that likes to know what is possible and what is not. Maybe you know more about the finer points of AI tuning than I do, so when you say it's impossible to make a competent ai without cheating enabled, then I'll just take your word for it.

It would be nice to have triumph's official opinion on the matter though. Many players have expressed frustration, on this forum and on reddit, playing against an AI that cheats and can pull stacks and fully equipped heroes out of its backside whenever the need arises. I recall seeing a cheat modifer in realm settings which should help players manage these issues for the time being.

Players shouldn't complain about an ai that cheats on the hardest or second hardest difficulties, but the way the ai cheats and how effective the cheats are against the average player might merit some consideration. There doesn't seem to be an agreement on how powerful the program should be.

The AI does not just spawn in units out of nowhere. This has been tested multiple times. They do not pull units out of a hat

Player just get caught out by a mechanic they weren't accounting for or terrain they hadn't scouted and then go complain on reddit about how an army popped out of nowhere even though the AI cannot do that.


As far as what is possible, I will point to chess engines. Hypothetically, having an AI that is as good at AoW 4 as a chess engine is at chess is doable, if a company spent and equivalent amount of money and energy building such a system as Google poured into making an exceptional chess engine.

So now even the most basic chess engines that you can play against for free are many orders of magnitude better at chess than even the very best human players. Humans cannot compete at chess against a chess engine, ever. If you play against one, you will lose.

So... is that fun? The chess community doesn't seem to think so. Nobody will play against chess engines. It's a waste of time; you just lose, and don't really learn anything and can't discuss the game after. They only thing to do with them is pit the chess engines against each other to try and tease out new strategies. Which is kind of interesting, but... is that the experience we're looking for?

And you can always lower the difficulty settings of the AI, but as someone who has done that... it feels wrong. The AI will, at random intervals, just give you material. It will just throw the game, and it will be obvious it is throwing the game. So the win will feel completely undeserved (because it is). Is that the experience we're looking for?


What do you actually want out of the game, and why do you think only a 'better AI' could provide it?
 
  • 1
Reactions:
@Kevin Ross Brown you're making a strawman argument. New DLC's have accompanying patches where the AI issues can be addressed. The point was, besides the potential ideas for new content: I'd rather see improvements to the base game than more forms or mounts.
 
  • 1
Reactions:
The AI does not just spawn in units out of nowhere. This has been tested multiple times. They do not pull units out of a hat

Player just get caught out by a mechanic they weren't accounting for or terrain they hadn't scouted and then go complain on reddit about how an army popped out of nowhere even though the AI cannot do that.


As far as what is possible, I will point to chess engines. Hypothetically, having an AI that is as good at AoW 4 as a chess engine is at chess is doable, if a company spent and equivalent amount of money and energy building such a system as Google poured into making an exceptional chess engine.

So now even the most basic chess engines that you can play against for free are many orders of magnitude better at chess than even the very best human players. Humans cannot compete at chess against a chess engine, ever. If you play against one, you will lose.

So... is that fun? The chess community doesn't seem to think so. Nobody will play against chess engines. It's a waste of time; you just lose, and don't really learn anything and can't discuss the game after. They only thing to do with them is pit the chess engines against each other to try and tease out new strategies. Which is kind of interesting, but... is that the experience we're looking for?

And you can always lower the difficulty settings of the AI, but as someone who has done that... it feels wrong. The AI will, at random intervals, just give you material. It will just throw the game, and it will be obvious it is throwing the game. So the win will feel completely undeserved (because it is). Is that the experience we're looking for?


What do you actually want out of the game, and why do you think only a 'better AI' could provide it?
I want more culture and tome units to implement more historical and fantasy based strategies than what i can achieve now with just single unit monostacks. I also want more hero classes, ways to upgrade hero classes, and expanded skills for them to equal or surpass the old hero system. Lastly, I would like better naval gameplay that makes sea battles interesting, risky, and less convoluted.
 
Just saying "Better AI" isn't really communicating anything, y'know?

You need to be specific. What is the AI doing that it shouldn't be doing, and what isn't the AI doing that it should be doing.
 
  • 1Like
Reactions:
The ability to settle through one or two underground passages, perhaps with forts or 'garrisons'. Or drill holes through narrow bedrock. Some way to allow you to make use of resources one province over, We can flatten mountains, let us drill tunnels.

Also I want elemental driven steam trains. ;)
 
  • 2
Reactions: