So, it's just the AI, then ?

[Iron Weasel]

My big gripe about the AI is the fact that they seem to be afraid of water or something. I've played several Continents maps on Prince difficulty and in each case it normally comes down to the point where I hold one continent and an AI civ holds the other but the AI won't send any units over. I finished a game up last night where I had one continent as Ramsses II and Gandhi (of all leaders) on the other. According to the Demographics chart, his military was almost double the size of mine so just for shiggles, I declared war and waited.

And waited.

And waited.

50 turns (!) later he still hadn't sent one unit my way. Considering our continents were 4 tiles apart at the closest point you would think he would send something across. Being the curious type, I loaded a bomber and 2 fighters onto a carrier and parked it off his coast. To the credit of the AI, he sent one destroyer out to greet my carrier, but I was close enough to see several cities and each turn I saw at least a dozen units just shuffle around the area between his cities.

 

[ButSam]

It appears from a poll I did and this thread that there are two main issues with the AI (possibly three):

(1) Tactical AI sucks, particularly across oceans. (Ironic, since transporting units across oceans is so much easier in Civ 5)
(2) AI does not appear to be good for either Cultural or Diplomatic victory (at least, so far in the poll I am running nobody has said they lost to the AI by Cultural or Diplo, so the AI doesn't really threaten the player with cultural or diplomatic victory; this also makes the cultural/diplo victory easier for the player to attain...particularly the diplo victory...since this combination means the AI most likely undervalues city-states.)
(3) The AI may want to prioritize upgrading units more (perhaps...not completely convinced of this one yet but I've heard anecdotes to that effect on the board)

 

[automator]

Agreed on the AI, but I can see how it's bad.

Previous games, the AI got to be "harder" because it could build a bigger army, and in previous Civs, a bigger army was a better army. SoD! But now the AI can't rely on sheer volume to make up for the fact that it's a computer program. Whereas the old combat system was pile versus pile and the best odds match up would be chosen, tactical combat is more difficult to code in. I certainly hope it gets better, as I really enjoy the combat system and would like it to improve through more challenge.

 

[rschissler]

It isn't the only thing wrong, but it may be the hardest thing to fix.

 

[ButSam]

Agreed on the AI, but I can see how it's bad.

Previous games, the AI got to be "harder" because it could build a bigger army, and in previous Civs, a bigger army was a better army. SoD! But now the AI can't rely on sheer volume to make up for the fact that it's a computer program. Whereas the old combat system was pile versus pile and the best odds match up would be chosen, tactical combat is more difficult to code in. I certainly hope it gets better, as I really enjoy the combat system and would like it to improve through more challenge.

Agree only partially. Why is it hard for the AI to win peacefully thru Culture or Diplomatic Victory, then? 1UPT has minimal impact there. Also, why can't the attacks be somewhat more coordinated? You are right about why tactical combat is tougher to code up now...but not others.

That said, the play 2K Greg had as Japan seemed to have good AI...

 

[jtwood]

The AI is not the *only* problem but, as far as I'm concerned at least, it's the only *serious* one.

There are some minor glitches and bugs but if the AI were more competitive, it would be a good game. It's just not finished.

I think the AI is two-thirds of the problem, but there are some serious imbalances in building/unit/trait cost-to-benefit that need to be ironed out as well.

You fix those two things, you make this a great game with some quirks.

 

[Stalker0]

I think the AI is two-thirds of the problem, but there are some serious imbalances in building/unit/trait cost-to-benefit that need to be ironed out as well.

You fix those two things, you make this a great game with some quirks.

Yeah there are several other issues, but as others have pointed out these can be tweaked with ease. Buildings too slow? Make them cheaper. A trait is too weak, make it stronger.

that can be done in 5 minutes with an XML change. The AI...much harder.

 

[Freesmog]

So, i've read the forum extensively in the past few days, to see what I should expect tomorrow when I finally get the game.

It seems many features are debated, but each change seems to have as many proponents as oponents. It therefore seems to me that none of them are bad in itself, but that it's basically a matter of taste.

The one exception seems to be the AI, which seems dumb dumb dumb. I wonder if it is really worse than previous civ games. i recall the AI has never been very bright.

So, do you think it's safe to conclude that the only real problem in civ v is the AI, and that even that is not really a new one ?

Of course not. There are all kinds of issues, like expecting a major upgrade in overall playability from civ4. However, I do think AI should ALWAYS be the highest priority. Unfortunately modders will have to make a much better game after the current PAID beta test is completed. I recommend continuing with Civ4 mods unless you wish to participate in beta.

 

[SuperJay]

The AI definitely needs work. You sound like a player who could still enjoy the game despite the drawbacks, though. If you have significant doubts, definitely play the demo first - it would have saved all of us a lot of whining if people would have just TRIED THE GAME FIRST.

But yeah, minor issues elsewhere; the lack of challenge provided by the AI is the one thing that it seems both sides can agree on.

If you do buy the full game, I recommend avoiding archipelago maps for now. As others have mentioned, the AI seems really, really hydrophobic.

 

[Guardian_PL]

Agree only partially. Why is it hard for the AI to win peacefully thru Culture or Diplomatic Victory, then? 1UPT has minimal impact there. Also, why can't the attacks be somewhat more coordinated? You are right about why tactical combat is tougher to code up now...but not others.

That said, the play 2K Greg had as Japan seemed to have good AI...

Yeah, if we were to belive "a likely story" that Napoleon did try to assault him from other directions through the ocean, but Greg's archers sniped them all, so Nappy decided to push aaaall its forces through the chokepoint. But hey, that was surely because the AI was fighting three other civs at once

...Did anyone got invaded through the sea so far? I know I haven't...

 

[Lord_Gaspar]

On my first game the AI tried to invade via the sea. They failed badly but it was a smart attempt at first.

Prince diff, large map with continents.

The main continent was sort of like an up side down U. I was on the bend and had a city at a 2 tile choke point. The AI could not get past my fortifications so they sent several units via sea around the choke point. It almost worked because I did not notice them at first.

Over the course of the game I fought 4 wars with this country. In each war they tried this same tactic, but every time they sent the land units unescorted. After the first war I built up some naval units, after which the AI lost at least 10 units to my navy for no reason.

 

[franlato]

The worst AI in a civ game?? lolll

The AI in vanilla Civ IV wasn't able to conquer cities. At least this one sometimes can (in my game napoleon conquered 4 AI)

 

[dannythefool]

Yeah, if we were to belive "a likely story" that Napoleon did try to assault him from other directions through the ocean, but Greg's archers sniped them all, so Nappy decided to push aaaall its forces through the chokepoint. But hey, that was surely because the AI was fighting three other civs at once

That's actually not entirely unlikely, I had a similar situation recently. An AI was attacking me through a narrow land bridge that was only one tile wide (earlier in the game, I blocked it out from my part of the world with a single warrior). It *did* actually try to attack by swimming around the land bridge. I didn't have any navy so it could have worked. Problem is just, he tried that with two or three units while I killed I think about 20 that tried to get through the one tile chokepoint.

 

[ButSam]

Also, in the middle of the walkthrough there is a message that pops up stating that France captured another city *and* that civ was eliminated. How did that happen if France wasn't fighting at least 2 civs (and who knows how many city-states...at least 1 after 2K Greg made friends with 'em)? Also, by the end of the game, didn't you notice Greg was about to lose? So apparently the strategy employed worked. Perhaps it wasn't the best, but it was competitive.