• Civilization 7 has been announced. For more info please check the forum here .

Why does civ 6 feel so much easier than the older civ games?

Civ Chemist

Chieftain
Joined
Jun 9, 2019
Messages
98
I started playing with civ 5 and have been playing 6 since the R&F expansion.

This summer I played civ 3 for the first time. I struggled a lot at the beginning and had trouble even winning on the lower difficulties. Since then, I slowly climbed the difficulty levels and was eventually able to win my first couple of games on deity (even managed to get a lucky Sid victory on a very easy archipelago map). But deity still feels very difficult, and I only really have a chance of winning with a very good starting position and/or powerful civ.

I also picked up civ 4 a couple of weeks ago and just barely got my first immortal victory the other day.

After playing these older games, it seems to me they are a lot more challenging than civ 6. Maybe it’s just my inexperience with these games, and civ 3/4 veterans would disagree. If I had to compare the difficulties (with my current knowledge), I’d say civ 6 deity feels as difficult as civ 3/4 emperor (if not easier).

So, my question, why are the older civ games so much more challenging? Do you think this was a deliberate decision from FXS? Making the game easier would give more players the satisfaction of beating it at the highest difficulty? Or is it simply a consequence of a more complex game (i.e. harder to create an AI capable to deal with the complexity)? Or is it the change of some of the game mechanics (I’m thinking of the change to 1UPT or the removal of tech trading?

What to you all think, or am I the only one feeling this way?
 
It’s not just you. The AI in civ 5 and civ 6 are not that good. The probable reason is the 1 unit per tile rule (1UPT) that civ 5 and civ 6 adopted. Before Civ 5, the AI could build stacks of doom, and was programmed to do this well. It’s much easier to execute : put 20 units in a stack and head for an enemy city. Civ 5 and 6 require players to build melee units, ranged units, support units, place them properly in relation to one another, and maneuver them through mountains, hills, etc. The AI isn’t good at this.

In addition, the latest patch of Civ 6 overvalues science, and the AI cannot support a big army because it builds too many research buildings.

Those are the usual answers. I also find that the Civ 6 AI is just passive after about turn 50. I’m never really worried about hostile action if I get 4 or 5 cities out there.
 
Part of the shift between games is the stack vs. 1UPT. I always felt that the AI could do better pathfinding in 3 and 4. Units are free to move on top of, or through, other units. Coupled with the AI's bonuses for building units, the AI can more easily field a large army and deliver it to your doorstep. The rules for combat in 3 and 4 don't have the same focus on ranged/bombard units that 5 and 6 do; nearly any land unit can conquer a city.

On the empire building side, both 3 and 4 have sliders that let the human player change the focus of their economy with just a few mouse clicks. Cultural expansion -- the size of the city's footprint -- happens organically/automatically, allowing the player and the AI to focus on other choices that affect productivity. Nearly every decision is reversible: Civ 4 civics may be changed more easily than 5's social policies; Civ 3 governments may be changed; buildings may be rushed or sold.
 
The AI can’t really do 1UPT well

The AI can’t handle walls well

The AI cannot deal with Ages, Loyalty, or any of the other add on systems from R&F and GS
 
Agree re 1UPT but also the game is much more complex now with District planning, adjacencies etc. Also, let's be clear, Civ doesn't have anything approaching AI it has computer opponents whose actions are governed by scripting / XML values. (That's not a dig at the developers by the way just pointing out a misnomer, a game with AI would be out of the price range for most of us).
 
Vox Populi's AI is competent with 1upt. Too bad we won't be seeing such for Civ 6!
 
I truly believe this game has so much mechanics, so much stuff to manage that it's just too hard for the common player, let alone the AI to run an empire perfectly and efficiently on every front. Common mortals will lack the attention span necessary to commit efficiently to all the mechanics or just be bored by them (I never efficiently used a military engineer for once). For the AI to be carnage incarnate we would have to go back to the original Civ: build units and attack, rinse and repeat. I prefer the more open way of winning and playing of this iteration of CIV. I think the one true advantage the player has over the AI is the ability to focus on what we reason will work towards a victory type and stick by it. I, however, seldom play like this...I go with the flow, aim for luxury resources, trade the extra and figure out what kind of victory shall I aim for or just try to keep up with the AI for winning a DV, SV or CV, unless a really want an RV victory, that's easy to focus on, though not always easy to win, at least for me.
 
I think it really depends on your skill level. All civ games share the fact that once you get really skilled, the only question remaining is "how fast you can win". It becomes a "number of turns required" game rather than a question of winning.

To me, although Civ4 still being the crown jewel of the civ series in my heart, the beauty of games like (SMAC and) civ 6 is that the sheer number of levers you can pull makes turns more interesting. Even if it feels easier because of the AI.

I guess the trade off is that you can't really set up a civ6 map to be as hard as a Civ4 map, but it makes it up by having more variety. Horse rushes by whipping your cities to the bone gets old at one point. Not quickly, but it does.
 
There are two main aspects.

First, the game is extremely mechanical. There are optimum ways to do most things. It's not like FPS games where many ways of doing things are different but produce different results and changing things up is essential to avoid being countered. Do you go sniper or shotgun? While it depends on your map and skills, it also depends on what your opponent is doing, how the game is progressing, etc. Your choices are meaningful, and the same choices won't produce the same results. While there are bad strategies and tactics, there are no optimum or perfect ones. It's more like an art, trying to find what works best in the specific situation.

Civ 6 is different - there is an optimum decision tree. There is one main strategy - spamming cities - and all you're doing is optimising the build order. Once you've found something close to the optimum strategy, it's easy. You're just tweaking to shave off the number of turns before you're victory comes. It's more like a science - there is a correct way, and it's just a matter of finding it and then repeating it. There is an ideal decision tree, and finding an approximation of it is enough to make the game easy.

Second is the AI. It tends to be good for the first little while, but thanks to the AI science focus bug, it just dies after a while and poses no threat. So long as you're still standing at Turn 100 (and probably earlier, but definitely at Turn 100), you're fine. The AI literally cripples itself. The AI has other issues too, but that's the big one.

Point 2 is fixable. It's likely to be at least partially fixed and if you're on the computer, you can do it[the partial fix] yourself. It's not likely to be completely fixed at this point, but it's doable. Point 1 is inherent to the board game nature of the game. You have fixed ways in which things interact, so there is an optimum decision tree, once you find it, you're golden. There is no way to get around that and still have a game recognisable as Civ 6.
 
First, the game is extremely mechanical. There are optimum ways to do most things.

Every game is extremely mechanical if you are an experienced player. Do something for long enough, and even seemingly random human behaviour becomes predictable enough. This has already been proven by Google's Starcraft 2 AI. By using optimal early build strategies the AI would crush any human player that didn't retort to unexpected mid-game decisions.

Back in the day when playing Americas' Army (early 2000s), I could lob a grenade at a certain direction and angle 5 seconds after spawn, and have a 80% chance of killing at least one enemy. Or know exactly which paths to take after losing 3 teammates in a match.

Spamming cities is not the optimum strategy, by the way, because of diminishing returns. At least not in single player. I don't play Civ6 pvp, so I wouldn't know.
 
Last edited:
It’s not just you. The AI in civ 5 and civ 6 are not that good. The probable reason is the 1 unit per tile rule (1UPT) that civ 5 and civ 6 adopted. Before Civ 5, the AI could build stacks of doom, and was programmed to do this well. It’s much easier to execute : put 20 units in a stack and head for an enemy city. Civ 5 and 6 require players to build melee units, ranged units, support units, place them properly in relation to one another, and maneuver them through mountains, hills, etc. The AI isn’t good at this.

In addition, the latest patch of Civ 6 overvalues science, and the AI cannot support a big army because it builds too many research buildings.

Those are the usual answers. I also find that the Civ 6 AI is just passive after about turn 50. I’m never really worried about hostile action if I get 4 or 5 cities out there.

This is probably partially true but in my experience, 6 is a lot easier than 5 too, so I don't think 1UPT is the main factor.
 
There is an ideal decision tree, and finding an approximation of it is enough to make the game easy.
I really don't agree with this - the decision tree is different if your neighbour is Macedon or Canada; it's different if you're on the coast or inland; it's different depending on the resources you find; it's different if you have a close neighbour or plenty of land to settle unchallenged; it's different if you use Tech & Civic Shuffle; it's different if a Barb Scout finds you on t3 etc etc.

The beauty of Civ, and the reason programming the computer civs is so hard, is precisely because there is NO ideal decision tree that works in all circumstances!
 
To me, although Civ4 still being the crown jewel of the civ series in my heart, the beauty of games like (SMAC and) civ 6 is that the sheer number of levers you can pull makes turns more interesting. Even if it feels easier because of the AI.
Never agreed more completely with any statement made on the forums. Civ4 is still the crown jewel of the franchise. SMAC was awesome because it was so far ahead of it's time in gameplay and strategy all while adding a unique thematic setting and, while there were few "teams" (leaders/civs/factions/nations/etc.) to choose from, each provided a VASTLY different set of advantages, disadvantages, and experience. And Civ6, I couldn't really pinpoint exactly what keeps me playing this with the other options available, but I think you nailed it on the head - the sheer number of levers you can pull. It's been more-or-less proven that production is the most valuable yield, but I'm drawn to maximizing faith- and (especially) gold-generation as they have the distinct advantage of being able to insta-purchase in any location. And there's many avenues towards increasing your generation of all 3 of these yields. Furthermore, the "number of levers to pull" is probably also the reason I'm addicted to Societies mode (more levers) and why I'm now at the point where I play culture victories almost exclusively: 3 of the victory conditions(space/religious/diplomatic) allow for an ever-so-slightly variance of approach to what ultimately becomes the same journey every time you play. Domination has one of two start-to-finish itineraries, depending on whether or not there's a lot of coastally accessible capitals or not. But culture victory, there's several different sources (levers) of the game-winning tourism yield and going all-in on any one of them will lead to victory, but also partially investing in several makes for varied experiences. I'm sure I'll forget some and don't want to imply that all of these are equally effective, but methods of generating tourism range from wonder-spamming, city-spamming with renaissance walls everywhere, GWAMs and artifacts, national parks and rock bands, ski and seaside resorts, the flight tech "powering" map-culture improvements and colossal heads, the Biosphere approach, relics (including heroic and void relics), entertainment/water park/neighborhood buildings, and Tata/Ibuka with campus/industrial zone spam. Then there's several levers to multiply those effects. And Vietnam, Ethiopia, China, Khmer, Mapuche, Indonesia, America, Maori, Sweden, Canada, Hungary, Georgia, and Magnificence Catherine each provide unique approaches to tourism.

As far as the OP and why Civ6 feels easier than previous iterations, a lot of good points have been made but something that I feel has been overlooked is that this is a 30 year old franchise and games today are easier than they used to be. I don't want to get all "Grandpa" and say how kids have it so easy these days, but games back in the NES days were so challenging compared to what's coming out today that they added replay-value because you couldn't succeed unless you practiced and memorized the layout of all of the levels. I have a guilty-pleasure addiction to reaction videos (I KNOW! it's a complete waste of time) and laugh so hard when kids today attempt those old games: Mega Man 1 with the screen crammed with things each of which are shooting other things, Contra with the same thing except one hit from anything kills you (to which my record is 3 start-to-finish gameplays with a single life), Ghosts N' Goblins and all the haunting nightmares from not the content but the experience, absolute perfection of timing needed for games like Kung Fu, Mike Tyson's Punch Out!! earning the two exclamation points, Ninja Gaiden being funny for having a life-meter when all players firmly believe that the game uses gravity against you for insta-kills, Zelda 2 (and Castlevania 2) with all those WTF moments where without a strat guide you just have to guess what to do, Ikari Warriors, TMNT and Battletoads.
 
Never agreed more completely with any statement made on the forums. Civ4 is still the crown jewel of the franchise. SMAC was awesome because it was so far ahead of it's time in gameplay and strategy all while adding a unique thematic setting and, while there were few "teams" (leaders/civs/factions/nations/etc.) to choose from, each provided a VASTLY different set of advantages, disadvantages, and experience. And Civ6, I couldn't really pinpoint exactly what keeps me playing this with the other options available, but I think you nailed it on the head - the sheer number of levers you can pull. It's been more-or-less proven that production is the most valuable yield, but I'm drawn to maximizing faith- and (especially) gold-generation as they have the distinct advantage of being able to insta-purchase in any location. And there's many avenues towards increasing your generation of all 3 of these yields. Furthermore, the "number of levers to pull" is probably also the reason I'm addicted to Societies mode (more levers) and why I'm now at the point where I play culture victories almost exclusively: 3 of the victory conditions(space/religious/diplomatic) allow for an ever-so-slightly variance of approach to what ultimately becomes the same journey every time you play. Domination has one of two start-to-finish itineraries, depending on whether or not there's a lot of coastally accessible capitals or not. But culture victory, there's several different sources (levers) of the game-winning tourism yield and going all-in on any one of them will lead to victory, but also partially investing in several makes for varied experiences. I'm sure I'll forget some and don't want to imply that all of these are equally effective, but methods of generating tourism range from wonder-spamming, city-spamming with renaissance walls everywhere, GWAMs and artifacts, national parks and rock bands, ski and seaside resorts, the flight tech "powering" map-culture improvements and colossal heads, the Biosphere approach, relics (including heroic and void relics), entertainment/water park/neighborhood buildings, and Tata/Ibuka with campus/industrial zone spam. Then there's several levers to multiply those effects. And Vietnam, Ethiopia, China, Khmer, Mapuche, Indonesia, America, Maori, Sweden, Canada, Hungary, Georgia, and Magnificence Catherine each provide unique approaches to tourism.

As far as the OP and why Civ6 feels easier than previous iterations, a lot of good points have been made but something that I feel has been overlooked is that this is a 30 year old franchise and games today are easier than they used to be. I don't want to get all "Grandpa" and say how kids have it so easy these days, but games back in the NES days were so challenging compared to what's coming out today that they added replay-value because you couldn't succeed unless you practiced and memorized the layout of all of the levels. I have a guilty-pleasure addiction to reaction videos (I KNOW! it's a complete waste of time) and laugh so hard when kids today attempt those old games: Mega Man 1 with the screen crammed with things each of which are shooting other things, Contra with the same thing except one hit from anything kills you (to which my record is 3 start-to-finish gameplays with a single life), Ghosts N' Goblins and all the haunting nightmares from not the content but the experience, absolute perfection of timing needed for games like Kung Fu, Mike Tyson's Punch Out!! earning the two exclamation points, Ninja Gaiden being funny for having a life-meter when all players firmly believe that the game uses gravity against you for insta-kills, Zelda 2 (and Castlevania 2) with all those WTF moments where without a strat guide you just have to guess what to do, Ikari Warriors, TMNT and Battletoads.

I will agree that games are easier, but overall it’s a good thing

Most of the difficulty of the old days was horribly frustrating BS difficulty
 
It's not really 1upt. V Is significantly harder than VI. Maybe it's not as hard as IV but it really isn't as far behind as people make out.

Two things make VI easier:

AI enemies don't crank out units like they did in earlier iterations. IV would be a breeze if it's AI built the puny stacks that would equal the armies you typically see in VI. In fact stacking would make those weaksauce armies even easier to roll. SoDs were only ever scary because they often outnumbered you bigtime. In V they were incompetent but you really did have to weather a zerglike horde when one DoWed you. If you weren't inflicting three or four casualties to every one you sustain you could still be wiped by V's less than competent AI. VI's AI builds a dozen or so units that don't get replaced quickly during a war. All you have to do is slaughter their initial force then you can wipe them out city by city easy peasy. No real attrition at all.

The other thing has been observed by many. They seem to be scripted to actively avoid winning before turn 290-310 (standard speed) or so. So really there's no pressure as long as your plan gets you there before 290. Just no competition. They'll research unnecessary techs, build spaceports and do nothing with them, stop spreading religion or conquering civs around midgame, etc.

These two things kind of make the AI less like opponents and more like bits of the map you may or may not have to interact with. I think Shakakhan nailed it with the games today are just easier analysis. Maybe to make it more approachable, I don't know. It's just the trend. I have one of those little Sega genesis emulators and play games on that with my sons. They're often flabbergasted at the idea that you get game overs when you lose your lives. That's just not something they experience much in the modern games they play.
 
Last edited:
Civ 6 is different - there is an optimum decision tree. There is one main strategy - spamming cities - and all you're doing is optimising the build order. Once you've found something close to the optimum strategy, it's easy. You're just tweaking to shave off the number of turns before you're victory comes. It's more like a science - there is a correct way, and it's just a matter of finding it and then repeating it. There is an ideal decision tree, and finding an approximation of it is enough to make the game easy.

I can't agree to that. There are many random aspects (maps, other civs, etc.) which render the game quite complex. If there were one optimum strategy, then the AI would be unbeatable (if the developers were not using their feet to type code - trying to translate a French expression here!).

What makes it easy for the human player is the incompetent, predictable AI behaviour.
 
It has little to do with 1UPT.

The current model of melee, ranged, siege, whatever unit types sucks because each unit has a role and weakness and – as you all say – if you knock out the key components of the AI army, it's pretty much cleanup time. Even some human generals (thru history) had a sketchy understanding combined arms, let alone AI. There are so many historical examples on how armored brigades got destroyed because infantry support was absent.

Variety offered by Civ6 makes the AI bad because this variety is not self-contained.

A single unit ("an army") with various stats would fare much better in the hands of the AI. And it would also make jumps from tech to tech much smoother.

This is how HOMM3 units worked and the AI was much more competent in combat.

If you had a single unit type, lets say the first one we could call "armed militia" with stats:
Manpower: 100
Ranged shots: 2
Siege equipment: 1

And a city garrison with
Manpower: 100
Ranged shots 3

where attacking a city garrison would substract:
3 manpower for each 1 damage to city manpower (melee)
0.5 manpower for each 1 damage to city manpower (ranged)
0 manpower for each 1 damage to city manpower (with siege equipment used that turn)

Turn 1a: Militia attacks city garrison with "siege equipment" and does 20 damage. New garrison manpower: 80.
Turn 1b: Garrison attacks militia for 20 damage with "ranged". New garrison manpower: 70. New Militia manpower: 80
Turn 2a: Militia attacks city garrison with "ranged" and does 20 damage and receives 10 damage. New garrison manpower: 60. New militia manpower: 70.
Turn 2b: Garrison attacks militia for 20 damage with "ranged". New garrison manpower 50, new militia manpower: 50.
Turn 3a: Militia attacks city garrison with "ranged" and does 20 damage and receives 10 damage. New garrison manpower: 30. New militia manpower: 40.
Turn 3b: Garrison attacks militia for 20 damage with "ranged". New garrison manpower 20, new militia manpower: 20.
Turn 4a: Militia is out of "ranged" and "siege" and can only melee. But since both parties have 20 manpower and the cost of melee is 3 for 1, it would need 70 manpower to succeed in storming the city.

This would mean that the number of siege shots and ranged shots would be extremely useful to have, but you could still win by sheer numbers.

Compared to the above scenario, if you had 6 units just meleeing the city in one turn, it would fall (6*20 damage = over 100 damage). But 6 of your units would also lose 60% of their strenght.

It would also make "spending a turn constructing 1 siege equipment" a worthwile proposition, as long as you have more than 1 unit adjacent to city walls. And it would also make "out of siege equipment, lets get outta here" retreats reasonable.
 
Old World and Gladius are 1upt and better than any Civ A.I in combat so that is a nonsense reason.The reality is Civ A.I is not a priority for the devs as it does not sell copies.The amount of work they did with A.I each patch was small.

With these new devs and the stupid stuff we got in frontier makes me think it will be worse in Civ7.
 
  • Like
Reactions: uhu
I think most of the pining for stacking is less about making the AI better and more about being averse to the big change in the series. Earlier iterations' SoDs were only a problem because the production bonuses allowed AI to massively outnumber you.

Civ VI's AI just does not build armies like IV or V's. It's easy to overcome their early number advantage and out produce them. Especially if you have the chops.
 
Top Bottom