Concerns about difficulty vs. AI.

I collected bits of unit stats in preview and +8 combat strength from difficulty seems to be a really big bonus.

Antiquity era unit strength:

(Tier 1) Warrior: 20
(Tier 2) Spearman: 25
(Tier 3) Phalanx: ??
(Tier 2) Chariot: 30
(Tier 3) Horseman: 35
(Tier 1) Slinger: 15 (ranged strength)
(Tier 2) Archer: 20 (ranged strength)
(Tier 2) Ballista: 10 (ranged strength)

It looks like unit upgrades generally comes with +5 strength, and 8 is bigger than that!
 
Last edited:
This is a big worry for me. In Civ VI, the combat bonus was +4. I'm worried they don't feel comfortable with their combat AI so are compensating it by providing a bigger raw difficulty bonus. Hope I'm wrong.
 
Several of the streams I saw were on Deity, and aside from the AI getting pretty insane yield bonuses, it didn't seem much harder or to go much slower. The game is way too fast at present in my opinion.
Which streams were deity other than Potato? I haven't gone through them all yet.
 
I'm usually more of a lurker, but here goes...

I think the game will be hard enough on Immortal and Deity, but this is primarily because the AI has a huge advantage in some specific areas of the game, such as war. They not only get a +8 combat strength bonus on Deity, they seem to have a much easier time getting war support (I'm not sure if they have a bonus towards spending influence on it, or just get more influence in general), which leads to you getting war weariness and significant combat strength penalties. If you end up fighting multiple wars, with multiple AI's boosting their war support, this can get out of hand really fast (your combat penalty can reach -10 or even -20, while they have a +8 bonus). But I'm sure people will find a way around that. It does make influence a pretty important resource for war though.
With that said, I've seen the AI do plenty of really stupid things, like leaving their army commander undefended and easily killed (and army commanders, especially experienced ones, are super valuable).

Speaking of bonuses, one thing I did like was that the AI didn't seem to start with extra settlers on Deity. At least in the Deity games I tried, the AI's I met right at the start, didn't get their first town / second city until like, turn 20 on standard speed (you can tell from their military milestone points), which is roughly when you can get yours as well. So that's a big positive, at least from my point of view. But that's not to say they don't get major bonuses to other things on Deity, they do (like in the example above).

As for befriending the "barbarians", I did focus pretty heavily on getting more influence, specifically with that goal in mind, and I picked up some attribute bonuses that make it easier as well (and as mentioned above, influence can be a very important war resource due to the fact you can use it to boost war support at the start of the war, which gives a direct combat advantage). That's just one strategy, I'm sure there will be many others.

The aggressive editing was sadly necessary, because we were limited to one hour only. I definitely had to cut out more than I wanted to in order to fit in that one hour limit...

Edit: Oh, and the city cap is more of a soft cap than a hard cap, you just get -5 happiness penalty for exceeding it (for each extra town/city over the cap, up to -35). So exceeding it by one or two is not that big of a deal, there are many ways to boost happiness. The biggest pitfall of doing that is the crisis that tanks your happiness (there's more than one crisis type, so there's some randomness in which one you get) - if that happens and you fail to get it back into the positives quickly enough, the city will flip to the the AI. Speaking of the crisis, it can also be turned off during game setup.

Edit 2: One more thing I want to say is that I played through the whole antiquity age on Deity and standard speed, and it felt easier than Deity in Civ 4 or Civ 5. I found the happiness crisis to be more challenging. BUT this is a very small sample size, so take it with a grain of salt. (and I haven't tried deity beyond antiquity age yet)
Thx a lot Marbz ! I'm one of your biggest fan, you know how many links I have in my "YouTube" bookmarks ? Two : your welcome page and the series (of you) I'm actually watching. For now it's "Civilization 5 Brave New World [Part 17] Deity Let's Play as Hungary", that makes for a good digestive for me. (sorry but it's the truth :D, maybe it's your voice, your character but you make me feel good) And by the way you are the only English-speaking folk that I can fully understand, thx to your accent I suppose. ;) "Good ol' Marbz. :)
Thx for the details btw, appreciated.
 
Overall game looks absolutely fantastic. Can't wait to actually get to play it!

I'm very worried Deity is too easy based upon the PotatoMcWhiskey stream where he seemed to do so well on the toughest level.

While all the AI had higher science and culture output, the AI were generally behind in gold/influence/happiness; at T86 he's ahead in the economic legacy path and tied for lead in the cultural legacy path with a likely economic golden age coming. He easily held off attacks. Obviously no idea how he actually finishes the age and what happens after the transition, but at least as shown appears the highest level may not be tough enough for the release to the influencers. It does appear the AI seems to maintain a continuous edge for science/culture, which is a huge improvement.

Hopefully the AI will be slightly tougher (for the toughest level) for the actual release. It should take more than a few hours of gameplay to learn how to win at the toughest level. Maybe they need to add a Deity Plus level.
 
Overall game looks absolutely fantastic. Can't wait to actually get to play it!

I'm very worried Deity is too easy based upon the PotatoMcWhiskey stream where he seemed to do so well on the toughest level.

While all the AI had higher science and culture output, the AI were generally behind in gold/influence/happiness; at T86 he's ahead in the economic legacy path and tied for lead in the cultural legacy path with a likely economic golden age coming. He easily held off attacks. Obviously no idea how he actually finishes the age and what happens after the transition, but at least as shown appears the highest level may not be tough enough for the release to the influencers. It does appear the AI seems to maintain a continuous edge for science/culture, which is a huge improvement.

Hopefully the AI will be slightly tougher (for the toughest level) for the actual release. It should take more than a few hours of gameplay to learn how to win at the toughest level. Maybe they need to add a Deity Plus level.
I've mentioned elsewhere, but I think it comes down to the fact influence/city states seem overturned, he went all in on them with mementos and leader choice, and he had 0 competition for city states as his main competition for them Machiavelli was steamrolled. Not to mention very defensible terrain against Rome where he barely managed to get his army over in time
 
Marbozir has an interesting video on deity level. He says that deity AI gets a base +8 combat strength which is significant. The AI also spends a lot of influence on war support which will give the human player a combat penalty. Additionally, if you are at war with multiple AIs, that war support bonus will stack. So if the human player is at war with say 3-4 AIs, they could face as much as -20 combat strength that stacks on top of the +8 deity combat bonus that the AI gets. As a result, he says attacking the AI can be challenging especially since the AI will have units defending their cities. But he says that is still easy for the human player to win a defensive war on deity because the AI is poor with offensive tactics and the defensive player gets combat bonuses from wall and fortifications that can offset the combat bonuses the AI gets. The human player can also use terrain defensively, set up choke points, to fend off the attacking AI. He says the AI is also a bit bad at protecting their army commanders so the human player can try to snipe the AI army commander to get an advantage that way.

Deity AI gets big yield bonuses. In one of his games, by turn 20, the AI had a city producing +40 science and +40 culture when his city was only producing around+10 to +15. It is impossible for the human player to compete with the AI in terms of yields.

He says thar AI is not necessarly smarter but Civ7 has designed certain game mechanics to be easier for the AI to handle. However, he notes that the AI is poor at picking attribute points. A human player is better than the AI at seeing the "big picture" so they can plan ahead to get attribute points that stack very powerfully, thus giving the player a long term advantage over the AI.

He says the AI is also not very good at handling some of the crisis.

Overall, he feels that deity is easier in civ7 than in civ4 and civ5. He says civ7 deity is comparable to civ6 deity (different in some ways due to different mechanics).

 
Marbozir has an interesting video on deity level. He says that deity AI gets a base +8 combat strength which is significant. The AI also spends a lot of influence on war support which will give the human player a combat penalty. Additionally, if you are at war with multiple AIs, that war support bonus will stack. So if the human player is at war with say 3-4 AIs, they could face as much as -20 combat strength that stacks on top of the +8 deity combat bonus that the AI gets. As a result, he says attacking the AI can be challenging especially since the AI will have units defending their cities. But he says that is still easy for the human player to win a defensive war on deity because the AI is poor with offensive tactics and the defensive player gets combat bonuses from wall and fortifications that can offset the combat bonuses the AI gets. The human player can also use terrain defensively, set up choke points, to fend off the attacking AI. He says the AI is also a bit bad at protecting their army commanders so the human player can try to snipe the AI army commander to get an advantage that way.

Deity AI gets big yield bonuses. In one of his games, by turn 20, the AI had a city producing +40 science and +40 culture when his city was only producing around+10 to +15. It is impossible for the human player to compete with the AI in terms of yields.

He says thar AI is not necessarly smarter but Civ7 has designed certain game mechanics to be easier for the AI to handle. However, he notes that the AI is poor at picking attribute points. A human player is better than the AI at seeing the "big picture" so they can plan ahead to get attribute points that stack very powerfully, thus giving the player a long term advantage over the AI.

He says the AI is also not very good at handling some of the crisis.

Overall, he feels that deity is easier in civ7 than in civ4 and civ5. He says civ7 deity is comparable to civ6 deity (different in some ways due to different mechanics).

Yes I saw that. On the one hand, not encouraging given how easy Civ VI is on deity, on the other hand it sounds like there are some easy things to fix that will make a significant difference (e.g. use of Commanders and interactions with IPs).

Also worth noting that he does not mention anything related to the full campaign (not sure whether the preview build excludes Modern or if they're just not allowed to talk about / show it?). One of the issues with VI was how badly the AI would tail off in the mid to late game, hopefully the Ages and Legacy Paths make it easier for Deity to stay competitive.
 
Yes I saw that. On the one hand, not encouraging given how easy Civ VI is on deity, on the other hand it sounds like there are some easy things to fix that will make a significant difference (e.g. use of Commanders and interactions with IPs).

Yeah, many of the weaknesses of deity AI should be fixable like making the AI better with army commanders, better at combat tactics, better at handling crisis and maybe giving them an extra starting settler. I would bet that Firaxis will likely buff deity soon after release.

Also worth noting that he does not mention anything related to the full campaign (not sure whether the preview build excludes Modern or if they're just not allowed to talk about / show it?). One of the issues with VI was how badly the AI would tail off in the mid to late game, hopefully the Ages and Legacy Paths make it easier for Deity to stay competitive.

My guess is that the AI will probably be more even accross Ages. So yes, I think the Ages and Legacy Paths will help deity AI stay more competitive. The human may still be better but at least the human won't snowball.
 
The changes around Commanders mean that it's easier for the AI to get units into the fight, and in the streams we're definitely seeing the AI bring a lot of units into fights, which is encouraging. Unfortunately, Commanders also bring a lot of fiddly micromanagement to combat itself (as well as the necessity of protecting the Commander itself) which the AI is terrible at.
 
I'm not sure what I can glean from Potato's repelling of that first Roman invasion. On one hand, he was able to handle it well with minimal losses and the AI took bad terrain, but he had also built up a sizable army to fight off early aggressors that Augustus couldn't see, and so his Commander rushing in to fortify the settlement before the Legatus could reach the high ground was a surprise for the AI. If his army was smaller or more scattered and Rome was able to move a turn or two more, as the AI might have planned, part of me thinks he could have lost the Settlement... but I can't say that for certain.
 
Overall, he feels that deity is easier in civ7 than in civ4 and civ5. He says civ7 deity is comparable to civ6 deity (different in some ways due to different mechanics).
Hopefully they can make adjustments to modify AI behavior or legacy focus so it is tougher than previous versions, not as easy or easier.

One thing I really liked in the video is the AI seemed to be more balanced throughout the Age instead of OP early and then weak after you get ahead in 50-100 turns. That is a great accomplishment!
 
Typically while I think it shouldn't be the case, the more "hardcore" part of the civ community will very probably again have to rely on mods to challenge them.
And with the base game it won't be about whether a player can beat the game but how it's achieved (like as fast as possible or with some conditions).

I think it's healthier to be ready for that.

Also regarding civ5 deity, it's important to note that the deity difficulty jump was mostly due to the AI starting with 2 settlers compared to immortal and below. A bonus that they don't seem to have given to civ7's.
 
Great videos!

I’m shocked to hear that war support lowers your opponents CS globally, not just against your units, in a way that stacks with multiple wars. If this makes it into the release version, a way of reworking this will be the first mod I look for. FXS must be seeing this happen in their games, and just thinking “working as intended”.
 
I don't think Potato's game can be considered as a showcase of how competitive deity AI will be. He played a good game but was also quite lucky (such as receiving a free warrior from a CS right when Rome showed-up at his door so he could fortify while waiting for his main army. Boesitus who was playing at immortal seem to have had a much rougher game against Charlemagne (and i don't think he's a worse player). Some amount of luck (regarding "goody huts", map terrain, independents ...) is bound to influence how a game turns out and if you combine a skilled player with good luck ... well then it's not a surprise the AI didn't pose much of a threat.

What i find concerning is the reliance on such a huge strength bonus. +8 is almost like they are fielding end-of-age units when we'll just have tier 1! This doesn't bode well for how competent they will be, and unfortunately it will make a "fair" combat so unbalanced that we'll have to rely, once again, on exploiting the AIs stupidity, which will only showcase just how incompetent it is 🙄

Typically while I think it shouldn't be the case, the more "hardcore" part of the civ community will very probably again have to rely on mods to challenge them.
Unfortunately, without access to the code and the ability to release dll mods (something that was, sadly, not possible in Civ6), modders will have limited tools to help the AI on the military front. Playing with priorities can help the AI economy but other than producing more units or being more aggressive, it won't change how it plays militarily (or will it, i've never delved that far into modding). With the AI already receiving such a large strength bonus, adding more bonuses is also virtually impossible (there's a point when combat will simply be impossible if you just add more str). What we really need is for the AI to stop being stupid when moving their units, and sadly that's not really something modders can do (at least, not without changing the code) unless combat AI was coded with something similar to neural networks and we can change the weights to affect it's behavior.

I guess we'll see soon enough. Hopefully leaving commanders undefended is something FXS themselves patch sooner rather than later, like they did for the settlers in Civ6 (you could still find unprotected settlers after they changed the AI behavior of course, and maybe it was a design choice, but at least they would no longer send all of their settlers without escort even when they were already at war like they did on release day)
 
It feels that the +8 combat bonus is meant to account for how a player can also raise their own combat bonus through war support, fortification, commanders, city-state abilities, etc. That is not to say that +8 cs is the "better" way of designing AI, but we the players certainly have more tools than before (especially in the early game), and the AI needs to stay competitive.

In Potato's video, we can see that he planned ahead and quickly narrowed down the +8 advantage using various means, from military aid to war support to city-state bonuses, fortifications, etc., and in almost every battle he managed to have at least only minor disadvantages.
 
One notable thing for me was that in Potato's playthrough, an AI was reduced to a single city without any intervention by him, purely the AIs interacting with each other - this would've never happened in release Civ VI and still hardly happens even now. What's more, in someone else's playthrough (I think Ursa Ryan? Not sure), one AI actually got eliminated entirely, again without player intervention. So at the very least AI vs AI seems more volatile. Which probably means the AI can also create more of a threat against an unprepared player. Also, while Potato did hold off the wars Augustus declared on him, he certainly had quite a few more units than I'm used to at least from my own games (and I don't think he usually built that many more than me in Civ VI), so it seems that an army isn't as optional as it was. I also haven't seen cities performing ranged attacks, which is another mechanic that often made it much easier for a player to defend.

Regarding game speed: I agree that the age seems to go by too quickly, but with there being a long age duration that's completely separate from the rest of the game, which I assume is simply that more points need to be gathered before the age ends, I'm not as worried.
 
It feels that the +8 combat bonus is meant to account for how a player can also raise their own combat bonus through war support, fortification, commanders, city-state abilities, etc. That is not to say that +8 cs is the "better" way of designing AI, but we the players certainly have more tools than before (especially in the early game), and the AI needs to stay competitive.

In Potato's video, we can see that he planned ahead and quickly narrowed down the +8 advantage using various means, from military aid to war support to city-state bonuses, fortifications, etc., and in almost every battle he managed to have at least only minor disadvantages.
Exactly this. +8 seems like a lot when you compare it to starting unit CS (20 for warriors). But it is so easy to get at least circumstantial +8 yourself through attributes, commander promotions, improvements, leaders (some of which are +10). We need to see how good the AI is in stacking up their bonuses, but it will hardly be as good as a skilled player. Thus, the +8 seems fair.
 
One notable thing for me was that in Potato's playthrough, an AI was reduced to a single city without any intervention by him, purely the AIs interacting with each other - this would've never happened in release Civ VI and still hardly happens even now. What's more, in someone else's playthrough (I think Ursa Ryan? Not sure), one AI actually got eliminated entirely, again without player intervention. So at the very least AI vs AI seems more volatile. Which probably means the AI can also create more of a threat against an unprepared player. Also, while Potato did hold off the wars Augustus declared on him, he certainly had quite a few more units than I'm used to at least from my own games (and I don't think he usually built that many more than me in Civ VI), so it seems that an army isn't as optional as it was. I also haven't seen cities performing ranged attacks, which is another mechanic that often made it much easier for a player to defend.

Regarding game speed: I agree that the age seems to go by too quickly, but with there being a long age duration that's completely separate from the rest of the game, which I assume is simply that more points need to be gathered before the age ends, I'm not as worried.
There is no ranged attack for cities and unwalled settlements are very easy to conquer. Defence now actually requires a noticeable army, while in civ VI you often had 1-3 turn grace period before a city was actually in a real threat.
 
Its been like this for a while now. I remember winning my first game on deity in civ V, while I never won deity game in IV. VI isnt too tough to win on deity either.
Winning on deity probably gets harder after some expansions.
 
Back
Top Bottom