[0.80] Late game is really very boring

I think it would help if the cooldown of monster abilities are reduced from 5 turns to 3 turns and boss monsters should have many abilities.

i agree that the late game has a tendency of getting boring. i don't really know a good solution, and i don't remember a 4x game that didn't suffer the "boring late game" syndrome. once the early intense part of the game is over and you established a strong base, the snowballing starts and exponential growth ruins late game balance. AI either can't keep up or (on higher difficulty settings) plays by a different rule set (handicaps/advantages) that of course doesn't really make it smarter - just more annoying. once you start running away with the game, you might as well quit and start a new game. at least, FE/LE have the master quest, which is pretty hard and a somewhat decent ending, but other than that, it's pretty similar to other 4X games - not really much to do with that nice little empire you worked so hard for. and of course FE also has the mastery spell - that's not really epic, but at least you can finish/win the game properly without quitting. 

Well, the master quest is sort of a joke to me honestly because of the weakness of big single monsters. SURE, it could be a really hard fight, certain combos, like a high lvl mage with KILL, just make it a joke of a fight.

And yeah, a lot of 4X games do have a similar problem, but i don't think I've ever seen it quite this bad before. Part of that is simple that the game offers so much at the end, really cool spells, awesome plate armor, etc. that I've honestly never really used, cus... whats the point? The game lets me basically unlock the ability the reshape the world, and my response becomes, meh, why bother.

The other big problem is that with unit desinge and optimized cities giving units big buffs, the AI has a whole new way to suck. In a game like Civ for instance sure I could become the main super power late game, but i could still have challenging fights, since my end game units were still on par with the enemies end game units. In FE:LH at end game (or really mid-game) my units are optimized both in how and where they built to a degree which makes fighting pointless. Even in a game like Gal Civ, I could still have epic and fun end game wars, I might not bother conquering  every planet/city of the enemies, but I would fight the war, in LH, I don't even fight the war.

The one of the biggest things for me I think is that end game tactical combat becomes really boring, I normally can win a fight without the enemy even really going. Tactical combat is at the heart of LH, its what sets it apart form other 4x games, if at the late game the battles are still fun, even if the overall situation is pointless, I will keep playing, but once the battles stop being fun, I stop playing. 

Boring late game is one of the most serious FE/LH issues. It comes out mainly of insufficient faction diversity, early break even point making the player invicible and lack of any power-tiders that can AI use to switch power balance back (remember that Iconian late game technology run?).

I have been curious since FE beta how Stardock deals with this and although there were gradual improvements, still more need to be done...

yeah i agree about the tactical combat. the early fights are rather intense, but once you can field some custom units, it quickly gets very uninteresting - charge - cleave - victory. some of the traits are probably too powerful, unbalancing the whole tactical combat system and turning it into a one turn "click to win" game. also, weapon damage vs. HP ratio gets worse over the course of the game. HP are roughly the same, but damage multiplies with higher tiers of weapons. armor helps somewhat, but not all AI units have armor equal to your weapon tech. in the beginning, it takes a few turns of careful maneuvering to kill them with spears and clubs, but once you get maces and battleaxes, most fights end instantly. 

I agree, the basic problem is stack quality. In my games the AI has far bigger armies then me but numbers mean nothing in this game. They can't even use those numbers for a war of attrition because the player hardly looses any unit against those numerous but powerless AI stacks. In the three games I played so far I never made it to high level techs because the game is basically over after the first war in the midgame. After that you I was the strongest faction and ended the game with the spell of the making.

Another problem is the slowness of units on the strategic map. I just watch where the AI is heading and with my road system I am usually quicker then them and can reinforce my cities before the AI arrives. Spells that obcure the field of view in a large area, quicken spells for whole armies or spells that make armies temporaily invisible could help if these were made faction specific and therefore only partly available. 

Overall the AI needs either better algorithms to create more powerfull stacks or game mechanics that superiority in numbers meaningfull.

Well I think solution might lay in racial-unique late game technologies...as in Galciv II. Bringing some really uber-powerful stuff to late game. Kind of enhanced Jugs or (more) enhanced Golems Or armor, spells, weapons, summoned creatures... that are able to really turn the tide, once the AI gets to them (and player still not).

I agree 100%.

The only entertaining aspect of the late game is to conquer a wild-lands. But even then, it doesn't matter if you conquer them or not. 

I think with the game as it is, one solution to this problem would be to either add continents so that you can explore and discover new AI Factions that have not been beat up on by other AI Factions, or perhaps add a dimensional portal that could transport you to a whole new world.

I think what contributes to this problem more than anything is the behavior of the AI factions towards each other. As long as this game remains a "every man for himself" attitude, nothing will change. If Factions were able to forge alliances early in the game then the mid-to-late might be more interesting.

Perhaps both of my ideas combined would allow for a great mid-to-late game. 

In any event FE/LH needs some serious attention to improve this problem.

I'm actually kind of trying this out in a game I just started. I've put the AI factions into to teams, so hopefully they will spend a little less time fighting each other to death over nothing and maybe build up a bit more.

I think one thing that would help a ton is if the AI was also a better at both making and using their sov/champs. Currently these are never really a challenge to me, they are poorly designed (picking seemingly random traits from multiple class at will with no thought to synergy and using horrible equipment, even when they are carrying better stuff) and in tac combat they never really do much. At lvl 12ish my caster sov is wiping multiple enemy units off the map a turn with fireball, or debuffing whole armies, or my assassin is almost impossible to hit and can get like 150 damage from crits. Meanwhile, even though enemy champs are normally higher lvl than mine, I don't I've ever seen one use a fireball, or anything crazy.   

Excellent post,

I agree entirely. I'm an empire builder at heart (sounds like OP is the same) but there is little point building an empire and strong heroes etc because there is nothing to do with it once you reach past mid game.

I think fixing this will require lots of changes and you've mentioned some of them. Just improving the AI to the point where a war between a strong AI and a weak AI will pretty quickly result in the strong AI conquering the weak AI would make a big difference. Imagine if late game when you've taken out one enemy neighbour you next need to fight Kraxis (or whoever) who has bulldozed through 3 other factions and taken all their stuff?

Although even then as various people have said it won't mean much unless Kraxis (or whoever) could use their resources/heroes well and use multiple strong stacks sufficiently well to cause a real threat to a mid-late game player super stack.  Achieving this will require a lot more work I think.

This is a good thread. I will just add that real meaningful alliances between AIs would work towards the same results as having AIs be better at fighting each other. They don't really form lasting relationships. Part of the problem is that the tech for Alliances is so late. I think in general AIs need to go for sweet and short wars, and really avoid prolonged ones.

Speaking to that, does anyone know if the AI is scripted to act differently when its at war or at peace? I ask, because I wonder if those times when the AI is at war with me even though I have no idea where the AI even is and we never see each other units (which is a whole other problem), is the AI wasteing all its production during these periods building units that it can never use against me and thus crippling itself?

To my mind there are two issues at play that could,  without changing the underlying game, make late game more interesting. 

First and most straightforward the AI just isn't consistently good at the game. It doesn't take goody huts. It doesn't farm spawn points. It doesn't field armies worth a damn. It doesn't pick attributes with any purpose. And it doesn't use its most powerful spells

.There'ss a rat's nest of interconnected issues and I don't envy Brad the task he's set for himself. 

But if I could only do one thing to improve the AI it would be to teach it to put all its best units into the most powerful army possible. The stack of doom is crucial. Of course not being able to properly formulate tactical strategy means that the AI will only win when it has an overwhelming power advantage. 

The second and perhaps easier issue is the problem that monsters become far too easy once you've picked up enough levels. As the OP said, solitary units are very vulnerable even if they have 1000 hit points.

The solution I would suggest is to implement more dynamic skills. Skills that scatter enemies. Skills that summon allies. And passive effects that deter some if the spammier techniques (slow or shrink). There's nothing new in this idea, and the bosses will always be vulnerable to strategies, but the battles should at least aspire to be interesting. 

Since TL;DR, I'm going to put an idea up front:

Here's an idea - once that tipping point is reached and the game can tell its toast.  

• A mega hero arises and unites the opposing nations - both empire and kingdom
• For the next X turns he and his band are traveling to super spiffy mountain to do something heroic
• Upon success the hero and all surviving (some could heroically die) sidekicks get more powerful
• If not stopped, for the next Y turns he brings the thing he got by being heroic back to a random city.
• If not stopped, for the next Z turns he consolidates power, the hero and all the surviving sidekicks get more powerful
• If not stopped, the hero and surviving band chase the Sov down from the random city to wherever the Sov is (somehow cloudwalk becomes mysteriously disabled)
• A final fight ensues - with the winner claiming the world.
• Ta Da - holy crap end game

Note: Probably have X, Y, Z and "gets more powerful" and/or # of sidekicks determined by difficulty level.  

One other factor that is critical and mentioned above - I have a 4 essence Fortress that I've put +production, +initiative, +attack, +hit points.  Its my only fortress, why would I want to build units anywhere else unless in drastic need.

But ANY unit built in that fortress is going to have:

+3 move and +3 attack (level 2 upgrade) - which I personally think is better than "go first"

+25 damage to damaged units (if I ever get to level 3 I think)

+4 initiative

+4 attack

+4 hit points  (which is rare since I usually play empire)

So even if I build a base spear dude late game, he's going from 6 to 10 attack, 6 to 10 hit points, and 20 to 24 initiative...

Needless to say I'm going to build my units for initiative and attack and only put on enough defense that doesn't impact my initiative.

I'm going to put them on a warg for an extra +2 initiative

I'm probably going to spend the crystal to give them +1 initiative

I'm going to spend a trait to give them +2 initiative

Another trait to do +15 weight and +1 attack to free up some encumberence (not that you can see it anymore)

I'm going to use my 3rd trait to give them +3 attack for anything less than my probably 29 initiative

Then first turn I have high initiative, extra attack and movement - and will probably go twice compared to the AI.

How is the AI supposed to compete with that? 

I've also focused my production and I may be able to spit out a high defense attack machine in a turn or two once the war machine gets rolling.    And one of these units can usually wipe the floor with multiple NPC units - if not a whole stack if up to light plate/chainmail.

First turn is KEY, especially with multi hit units (cleave, impale).

Initiative is dangerous.  I don't think it needs to be capped, but I think every unit should have an action every turn. 

Charge is dangerous.   I don't think it needs to be nerfed, but we need anti-firstmove traits. A la, on first turn prevent enemy from moving more than 1 tile to come into contact (hard to explain, but if 8 away, can only move 7 on first turn.. can't charge right in).  Call it spearwall or something.  Maybe require spear.

Swarm is dangerous.  Not sure it needs to be nerfed, but should have anti-swarm mechanics.

I also agree late game could use some work.

PERSISTENT METRICS - to compare game to game

Better boss fights a la OP

Some better way to finalize/kill a civ - needing to clear each individual city is just too much drudgery.

    Was it galciv that had the mechanic that if an enemy felt they were overmatched they would surrender to someone and it might not be the person beating on them?    i.e. They lose a few major battles/cities and realize they can't win.. they throw their lot in with someone else.

    Might have actually been EWOM, can't remember but vaguely remember it might have had to do with dynasties.

Here's an idea - once that tipping point is reached and the game can tell its toast.  

• A mega hero arises and unites the opposing nations - both empire and kingdom
• For the next X turns he and his band are traveling to super spiffy mountain to do something heroic
• Upon success the hero and all surviving (some could heroically die) sidekicks get more powerful
• If not stopped, for the next Y turns he brings the thing he got by being heroic back to a random city.
• If not stopped, for the next Z turns he consolidates power, the hero and all the surviving sidekicks get more powerful
• If not stopped, the hero and surviving band chase the Sov down from the random city to wherever the Sov is (somehow cloudwalk becomes mysteriously disabled)
• A final fight ensues - with the winner claiming the world.
• Ta Da - holy crap end game

Note: Probably have X, Y, Z and "gets more powerful" and/or # of sidekicks determined by difficulty level.   

Well you can force the AI to surrender to you. On the normal setting they will pretty much do this when they have one city left (I don't know if they have to have most of their units killed also or what). You get control of their Sov. and all their cities, units, champs, etc. are destroyed. It basically helps speed up end game a little bit, but doesn't make it any more fun really. Also the AI doesn't seem to ever surrender to other AI players, not that it would help them much, since the cities and units are gone. 

Alliance tech being too late in the tech tree and AI being not able to use it is one of the most serious late game flaws. It was discussed even in the FE beta phase, as far as I remember. At least since the last AI update, we have some AIs that are not constantly in war with each other all the time and are able to trade etc.

One example of spiced, memorable late game from my favorite Galciv II. In one of my best games, two late power blocks emerged: me (in the lower middle of the galaxy), Altarians in the uper right corner and distant Thalans vs. mighty, but feebling Torians occupying huge left section of the map, Drath, smallish empire just bellow Torian space and Iconians, in the right-low corner. My empire directly bordered all of them (though Torians were a bit distant), so should the cold war proceed into hot, I would be pretty much first to hit.

Anyway, I felt strong enough to wage two front war and I basically wanted to maul Drath, since they backstabbed me before, but not occupy them, and to push Iconians out and také their border worlds. But would I be able to sustain all-out war? It took me a lot of provocations and influence stations to make Drath declare war on me. Being impatient, I declared war on Iconians immidiately. After few turns, pretty much everybody entered the fray. And while I was able to maul Drath military, deter Torian onslaught and také some Iconian colonies, I was simply not able to undertake the decisive blow to push particularly Iconians (with their late game tech run) out of the main powers. My allies, however, ended much worse, with Thalans being reduced to few planets (later surrendering) and passive Altarians even loosing few colonies to - by that time - weaker Iconians.

And what was the conclussion? While I was rearming for the upcoming global war nr. II, Iconians managed to produce incredibly strong battleship (uber-late tech) and in almost completed secrecy, prepared for attack. While whole empire was holding a fest and most officers were on holiday, Iconians stroke back and in few turns not only overran my frontier positions, but also slaughtered most of my fleet and occupied many of my core worlds. It was too late for my allies to be of any help. With Drath vultures entering the war to get their part, surrender was the only option...

Now this story reminds me of two things. First that I have to install Galciv once again and play it a little bit

And second that I would really love to have such memorable late games in LH!

I remember that in Master of Orion 2 if you refused to be named by the galactic council then all AIs united in a single player and went to war with you. Perhaps something like that could be implemented. Like when it's clear that the human player is far ahead all the AIs band together.

I think this is crucial and almost single-handedly might prevent the "Oh, I guess I'll take this stack of doom and wipe out this random AI now cause I feel like it" attitude. Knowing that your actions have diplomatic consequences, and knowing that these consequences will actually affect you, is so important it's not funny.

The issue I have with this idea, and props for having the idea!, is that its basically the same as the Master Quest victory, and doesn't really fix the issue. What this does is just make end game faster, not make it any more fun. I would still have the big issue of, so I built all this stuff and this civ... what now? Couple that with the fact that late game battles aren't really that fun... I just don't see this fixing core issue.

I think the best way to fix the core issue here is to make improvements to how the AI handles things.

I think they could add a civ's vs. the world mode pretty easily that might make things more interesting, even for good players. Basically, throw the kitchen sink at the players, AI as well, and have multiple different master quests that are the only way to win (so each game is not the same master quest). Monsters can stream out of the wildlands, events can pop up and terrorize civilization; basically just eliminate the checks that stop monsters from being a real PITA (such as no city-razing).

Quests should be easy to make. If they don't have a quest maker that can churn these out as fast as you can write them, then they need to invest in that tool. The art assets are a factor, but really reuse here is completely understandable. This way you can get several (or more) master quests which make this player vs world games more unique and fun. 

The civ's AI is not really going overcome the crippling hurdles (to the victory hump) anytime soon, so this mode would be an easy to implement way to appease players who are good and player's who are not very competitive(since they will essentially be cooperative).