Keep getting beat on building wonders

[Helmling]

Cool, Derpy. Maybe I should go up. I just remember the one time I tried it, I got licked so bad in the race to the GL that I crawled back to Emperor with my tail between my legs.

Nevyn, you don't have to have marble and stone, but boy does it help.

Speaking of which! Last night I came THIS close. I got every Greek wonder except the Temple of Artemis. I had a sweet start that had stone and marble by my capital, but then more stone and marble just to the south. I went liberty first, grabbed the 2nd marble city site and set it working on the Mausoleum right away. I actually took 8 turns of zero growth in my capital to pump out another settler, grabbed a 3rd city nearby with another stone that was good enough overall to build Zues. Meanwhile, my uber-capital with two stone, one marble, and a mined hill was enough to build the Colossus, Great Library, Great Lighthouse and the Oracle.

Very sweet game...I think I'll go continue it now.

 

[apocalypse105]

One thing I used to do is save before starting each wonder and reloading if I miss it. It's really frustrating if you 1 or 2 turns from getting it and you lose out. I've even had it before where it was on the last turn and the A.I still got the wonder, I don't know what decides that but figure it must be they had more production towards it i.e mine was 190/200 with +11Prod/trn and they had 195/200 with +11. Or maybe even matched like both of us had 190/200 but they had +12/trn and I only had +11 or something like that.

My strategy now, because its a personal goal not to reload turns, is as soon as you choose your wonder focus completely on Production. Plummet your city into stagnation if you have to, don't starve them obviously, but get as much prod as you can. Also, If i go for the early wonders I don't settle until I have which wonders I want and try not to start building them until I have a worker out and at least 4 or 5 population. And if you have hills nearby then build as many mines as you can to up the production, or possibly even chop forests but I don't usually do that myself. I'm sure on some games that early mines were what got me the wonder instead of the A.I.

Hanging Garden is one of the toughest to beat the A.I to I think. I don't know why they get such a love for the Gardens but they've beat me to it so many times.

I'm starting tho thinx the AI gets free wonders


Because I actually started on a wonder and I got beaten

So I relouded and decided not to build it then 10 turns later then normal the wonder goes

 

[Derpy Hooves]

Yes, I've had situations as well where an AI would complete a wonder at different turns, either just a bit quicker then mine or far slower if you didn't build the thing at all.

What I think is happening is that the AI cheats and knows how much time you have left for the wonder, and if necessary it will switch its city over from food to production if it means getting the wonder just before you.
Cheating and pretty lame, but at least I guess it gives the AI a much needed edge.

 

[headcase]

I'm starting tho thinx the AI gets free wonders

They might cheat like Derpy said in seeing how far you are, but building wonders is the one thing they absolutely never cheat on according the XML. However, the boost in apples on King+ and happiness on Warlord+ helps them grow their cities faster.

 

[Camikaze]

The AI doesn't cheat. That's simply baseless suspicion. If the AI never beat you to wonders, there'd be complaints about bad AI. When they actually do beat you to them, the logical implication would be that that is conversely good AI. But the contention is that they're cheating? If an AI is going to beat you in getting to a wonder, then of course it's likely to be within a few turns. If there are 10 players in the game, you'd think there should be an even chance for each to get the wonder. So you're only likely to get a wonder 10% of the time. The AI should, in evenly matched games, be beating you to 90% of wonders.

 

[Derpy Hooves]

Wrong Cami.
Perhaps I could have been a bit clearer but I'm talking about a single game, I reloaded savegames where I and the AI where already building them but he would complete it sooner compared to another save where I would have finished mine later as well. So unless there's a RGN involved somewhere the AI has to be cheating. (This was a really early wonder as well, so no GE if the AI even knows how to use em)

 

[Camikaze]

Wrong Cami.
Perhaps I could have been a bit clearer but I'm talking about a single game, I reloaded savegames where I and the AI where already building them but he would complete it sooner compared to another save where I would have finished mine later as well. So unless there's a RGN involved somewhere the AI has to be cheating. (This was a really early wonder as well, so no GE if the AI even knows how to use em)

You probably should tell the developers where these cheats are hidden in the game files, so they can fire whoever snuck them in there. The AI bonuses have been pretty transparently laid out, and the suggestion that there are cheats worked in has been categorically denied.

If you're just talking about a single game, then that is coincidence. You started building later, and so did the AI. Maybe you built a unit earlier instead, which could have made the AI perceive you as more of a threat. This would potentially make them pump out a couple more units before they worried about going for the wonder.

Anyway, supposing for a second the AI did cheat, so what? If that 'cheating' is defined in certain ways to give the AI specific information (and that's the only way a programmed AI can cheat, unless you were to think it's capable of reading your mind), in what way would that negatively impact on your Civ-playing experience? If you want the AI to play by the exact same rules as you, then of course you're going to beat them. It's unrealistic to expect anything different. There are a lot of different areas of the game in which the Civ5 AI must be competent, and although AI can certainly be improved and tweaked, one that can actually match you in a game that requires the combination of such a set of mechanics and factors is simply does not exist. If you want a challenge, then the AI needs bonuses, as is the case currently. But another way of giving bonuses is to let the AI cheat a little. Would it really be so bad if they did?

 

[caketastydelish]

Noob question: When you invade another Civ's city that had a wonder in it, do you gain the benefit of the wonder?

Because if you do, you may want to just invade a bunch of other civs so you get their wonders.

 

[Helmling]

Wrong Cami.
Perhaps I could have been a bit clearer but I'm talking about a single game, I reloaded savegames where I and the AI where already building them but he would complete it sooner compared to another save where I would have finished mine later as well. So unless there's a RGN involved somewhere the AI has to be cheating. (This was a really early wonder as well, so no GE if the AI even knows how to use em)

I've noticed similar things happening, but the reload was always far enough back that there could have been a change in the AI's decision tree that led to a different completion time. There are random nudges one way or another for the AI's behavior, so I think that's reasonable.

 

[Helmling]

To my surprise it works quite well in immortal as well, started off with the GL and the Oracle, got Chichen Itza, PT(GE rushed), ND and I expect to be able to get the Sistine Chapel ans Big Ben pretty easily as well, bastard Ghandi made the HG, 'henge and the DLC wonders, he won't be long for this world though.

I'm a bit behind in tech but I expect to be able to fix that with the vigorous conquering I'm doing right now, might post a replay once the whole debacle is over, mind you I'm also playing Monty, so nothing really to help me with the wonder building or science apart from the food boost from floating gardens.

I had a gorgeous wonder start on Immortal, so I thought I'd try the Mausoleum-->GL-->Oracle slingshot that works so well on Emperor.

Somebody beat me to the GL while I was still researching Writing.

 

[Hammer Rabbi]

i just finished my Immortal Science vic with Babylon. with no reloads/restarts, i got GL and Oracle and PT and HS, but i tried for the Mausoleum. took 14 turns to build, lost it with 3 turns left. I felt okay with it since it and the Kremlin (later on) were the only wonders I was beat to (really wanted the Kremlin, too). Great Wall is always gone quickly in my games. Stonehenge didnt get completed until the Renaissance era. I could have gone for it but it wasnt in my plan.

 

[Helmling]

i just finished my Immortal Science vic with Babylon. with no reloads/restarts, i got GL and Oracle and PT and HS, but i tried for the Mausoleum. took 14 turns to build, lost it with 3 turns left. I felt okay with it since it and the Kremlin (later on) were the only wonders I was beat to (really wanted the Kremlin, too). Great Wall is always gone quickly in my games. Stonehenge didnt get completed until the Renaissance era. I could have gone for it but it wasnt in my plan.

I'm going to try that game again from turn 1 and try to hit the GL first, then divert to masonry and see if I can snag the Mausoleum before Oracle. Wish me luck. (I don't feel too badly about starting over because I had very little luck from Ruins anyway so it shouldn't be a factor.)

 

[Lance of Llanwy]

In my experience, some AI always builds the Great Lighthouse very early on, which I find perplexing, because that branch of the tech tree isn't very (immediately) important unless you have whales in your capital and the wonder itself is not as powerful as other wonders that could be unlocked with the same amount of beakers(SH and GL).
It just seems odd to me, but it might just be my game.

 

[Hammer Rabbi]

some civs with good water units or a good water UA go for it out of habit i think. If im in a game with England, they almost always get it.

 

[Helmling]

I've been dabbling in Immortal the last two days. I find I can get the GL and the Oracle pretty easily, then speed toward Hagia Sophia. I can reliably nab one of the University wonders, if not both. I feel really constrained, though. The only kind of game it leads to is a long waiting game for the Utopia project. The happiness cap drives me nuts. I feel really limited on growth.

 

[zevel]

Chop chop! micro your tiles, know your enemy.

Uproar east, strike west!

 

[Helmling]

Gawd...more so-so games. I had Songhai come at me before turn 30 with 6 warriors. I was on a hill, but he bankrupted me and killed my workers. In other games, I kept missing one of my target wonders by a few turns. I've been fiddling with this all weekend. Why?!? Why must I keep at this!

 

[Derpy Hooves]

Cami, as far as I recall the devs deliberately said very little about th AI to make it more unpredictable, it was usera messing around with XML files that eventually found out all the relevant valueas for leaders, difficulties and such.
Adding to that the fairly well documented occurance of an AI that stacks combat units (I've even seen this happen myself), it is very likely that it cheats in 'unofficial' ways, be it intentional or not.
If you want to tell te devs go ahead, the chances they'll listen are remote and I for one, have no more faith in them.

And as you could have concluded from my previous message, I don't hate it cheating as I just find it incredibly cheap considering the masses of other things it already cheats at. Does it really need to know when your finishing a wonder ON TOP of its production bonusses? Couldn't they have done some minor behavioral programming?
The only reason I can think of not to is laziness.

And good luck Helmling, if your conquering get autocricies police state ASAP for mad happiness, otherwise just get the wonders, policies and occasionaly rushbuy a happiness building, I shoul note that more conquering can also fix this via lux.

 

[apocalypse105]

Cami, as far as I recall the devs deliberately said very little about th AI to make it more unpredictable, it was usera messing around with XML files that eventually found out all the relevant valueas for leaders, difficulties and such.
Adding to that the fairly well documented occurance of an AI that stacks combat units (I've even seen this happen myself), it is very likely that it cheats in 'unofficial' ways, be it intentional or not.
If you want to tell te devs go ahead, the chances they'll listen are remote and I for one, have no more faith in them.

And as you could have concluded from my previous message, I don't hate it cheating as I just find it incredibly cheap considering the masses of other things it already cheats at. Does it really need to know when your finishing a wonder ON TOP of its production bonusses? Couldn't they have done some minor behavioral programming?
The only reason I can think of not to is laziness.

And good luck Helmling, if your conquering get autocricies police state ASAP for mad happiness, otherwise just get the wonders, policies and occasionaly rushbuy a happiness building, I shoul note that more conquering can also fix this via lux.

The problem with wonder building at higher diffulty is that the AI starts with pottery and other technologies so they can easy pick up early wonders Wich I find a little bit stupid

 

[Helmling]

And good luck Helmling, if your conquering get autocricies police state ASAP for mad happiness, otherwise just get the wonders, policies and occasionaly rushbuy a happiness building, I shoul note that more conquering can also fix this via lux.

What I really want is to pull the coup of getting GL, Mausoleum and Oracle. I was ONE FREAKIN' TURN away from following up the GL with the Mausoleum (and presumably being able to snag the Oracle, which is the easiest of the three) when I got a pop up that somebody else had finished it.

ONE TURN!

So I keep going back and trying...like a moron...or a junkie.

Take your pick of moniker there.