| General | Hosted Sites | Civ5 | CivRev | Civ4Col | Civ4 | Civ3 | Civ2 | Civ1 | Misc | Marketplace |
![]() |
|
|
Welcome to Civilization Fanatics' Center. You are currently viewing our site as a guest which gives you limited access to our site features. By joining our free community, you will be able to participate in the discussions, search the forum, send private messages, vote in polls, upload your own screenshots to the gallery, and access many other special features. Registration is fast, simple and absolutely free, so sign up today! If you have any problems with the registration process or your account login, please contact support. |
|
|||||||
![]() |
|
|
Thread Tools |
|
|
#1 |
|
King
Join Date: Oct 2001
Location: Burlington, VT
Posts: 735
|
Alright, I've been playing a game on Emporer difficulty, and I found the early game terribly difficult. The computer was able to expand much faster than I was, and they still had all of their trees at the end of it. However, playing Emporer finally gave me a chance to do something I always wanted, have a decently challenging late game against the computer.
In monarch and below, I was always in a dominate position at the point tanks were being introduced. I was first in tech, land, population, you name it. In my current Emporer game, against 6 other civilizations, 3 of them are ahead of me in technology, two have much better armies, and one even controls a larger landmass. The biggest difference however, is that they all have oil as well, something that never seemed to happen on Monarch. The late game promises to be so much interesting, with espinage and alliances and naval blockades, rather than the standard air raids and Modern Armor massacres that occured in Monarch. So that gave me an idea, perhapes they could modify the difficulty settings to be a little bit harder on the computers in the beginning, but much easier later on. As a possibility, maybe the game could have somewhat of a sliding difficulty scale, giving the computer more production bonuses once you are comfortably ahead. Any thoughts |
|
|
|
|
|
#2 |
|
Emperor
Join Date: Jan 2005
Location: somewhere up north
Posts: 1,539
|
im trying to ajust to civ4 but i have not been chop rushing. I should try that
good luck
|
|
|
|
|
|
#3 |
|
Warlord
Join Date: Oct 2005
Posts: 242
|
Been playing on Emp level for awhile. Won it couple times via Space. Still haven't found a way to win by Conquest.
|
|
|
|
|
|
#4 |
|
Chieftain
Join Date: Feb 2004
Posts: 38
|
Having a sliding difficulty handy would make it too tempting to slide it back to easier setting if u find yourself falling behind
Takes a lot of the challenge out of the game
|
|
|
|
|
|
#5 |
|
Warlord
Join Date: Oct 2005
Posts: 164
|
It doesn't have to be a manual slider. Games as early as NBA Jam for SNES had a modifier built in whereby a game in which one team losing by x amount of points would get bonuses to keep the game from getting ridiculously out of hand.
I could see how if it were done right it would help balance out some of the exponential gains that seem to come from a game like this (tech lead = free GPs = extra turns to build units/buildings/wonders, etc). I could also see how it might be exploited though |
|
|
|
|
|
#6 |
|
Deity
Join Date: Nov 2005
Location: Texas
Posts: 10,263
|
I just wish the difficulty scaled better, it's been like this with every single Civ or SMAC game. It's insanely hard at the beginning compared to how much easier it is towards the end. A tiny mistake at the beginning can cost your entire game (or at least make it much, much harder), whereas individual choices are much less significant later on.
For example, choosing where to place your second city: I've found this can be the single most important decision in Civ IV, and requires lots of long-term planning. Where to place your 22nd city isn’t all that challenging. I wish there was just some way to adjust the AI difficulty from front to back to compensate for this, like a Fade control for the computer handicap to make it harder at the end, or at the beginning, according to your preferences.
Last edited by Thalassicus; Dec 10, 2005 at 09:52 PM. |
|
|
|
|
|
#7 |
|
Warlord
Join Date: Dec 2005
Posts: 122
|
I think it gets easier over time because the human has a fixed strategy in mind and will build tons of power units whereas the AI has to build a fairly even (randomly?) number of units, a lot of which a human player would find questionable for the overall game at the time. Just look at the build suggestions the AI gives you, how many times do we disagree?
What would work for the programmers and to make their life easier (as AI programming is difficult!) is a "strategy editor" for the different ages, per civ. For example, you could force the AI to try a praetorian rush for the romans, now wouldn't that be fun to play against? And to make late game more difficult, a fixed strategy for AI eg. all out war concentrating on certain units only. Another "side" is that there are just a few programmers dedicated to the AI of this game, now if this game had the same number of people dedicated to chess, we would have an unbeatable AI level just like chess! |
|
|
|
|
|
#8 |
|
King
Join Date: Oct 2001
Location: Burlington, VT
Posts: 735
|
Yeah, I definately wasn't suggesting a manual slider, something automatic like the sliders they use in some basketball sports games to keep the game at least a little close near the end. It could even be a set of sliders in different areas of the game, once you are out teching your opponents (which hasn't happened to me yet in Emporer, but I'm sure it could) then your opponents could get a larger research advantage. Once you are outproducing them (which happens pretty much every game once you have a larger landmass) they could get more of a production bonus.
Either that, or just more bonuses in later Eras. I was wondering, but I wonder if that is what the 'Era Modifier' in the XML file is for. I suppose I could manually adjust the settings so the computer doesn't get so many free units/techs at the beginning, but I think the main problem is just that the production binus for the comp is unbeatable at the early game. You just don't have enough choices at that point... |
|
|
|
|
|
#9 |
|
Chieftain
Join Date: Nov 2005
Posts: 67
|
Yep, agreed. Improving the AI civs production and research gradually the longer the game runs is probably the best way to have a more real feel to the difficulty level.
I also find this a bit annyoing, since if you don't take big risks early on at higher difficulty levels you will likely lose. Barbarians and animals are often the biggest threat in CivIV, not the AI civs. |
|
|
|
|
|
#10 |
|
Chieftain
Join Date: Dec 2005
Location: Smak dab USA
Posts: 36
|
Barbarians and animals are often the biggest threat in CivIV, not the AI civs.
by Luhh. Amen. I have kinda, sorta, mastered Noble; harummph. I want to have barbarians if for no other reason to keep me frosty during the first couple thou years. But at Prince level them thar barbs are just that. I lose just about ALL of my mines ,farms, cottages, etc before 1 AD. I also lose my cities. If I dont find copper by about 2400 BC Im in deep doodoo. I have had 9 axemen/archers in rapid advancement at once on a single humble city. I play Fexfxgrand-roosevelt-aggresive-highland-huge-ridged-largelakes-w/barbarians ;"NOT raging". If the AI does consider my strength before attacking, isnt 2archers and 2 warriors per city with walls fair defense by 1500BC? BTW is adding trees at the vary start considered cheating? I do like to chop rush early. |
|
|
|
|
|
#11 |
|
Prince
Join Date: Feb 2007
Posts: 401
|
I just started empororer, the very beginning was very hard simply due to hordes of barbs, so I decided i wanted the great wall, unfortunately 3 civs had to die for me to get it. 200 bc and i already conquered 3 rivals, mehmen elizabeth, and napoleon all with 20-25 axemen and bw only, doesnt seem to bad really although I havent won the game yet. It was surprising to see a 18 population city defended only by 3 archers and a spearmen though??
Maybe next game ill try immortal and i hope the ai is somehow better at war although doubt they will be. |
|
|
|
|
|
#12 |
|
Warlord
Join Date: Dec 2006
Location: Finland
Posts: 294
|
I have the same problem. I seem to win emperor regularly, and after early rush I know I am going to win. But on immortal, not a change to keep up on expansion.
It's hard to even grab the copper, or the needed 3-4 cities to keep up without early rush.I wish someone made a mod which boosts AI bonuses when game progress. Maybe monarch bonuses on the ancient era, emperor bonuses on classical, immortal bonuses on medieval and deity since then. That would keep the game interesting until the end. |
|
|
|
|
|
#13 |
|
Deity
Join Date: Mar 2007
Location: Univ. Florida
Posts: 15,372
|
Adding trees!
You are probably not helping yourself. I find that starting positions always have advantages and disadvantages - very few are actually "bad starts" and this perception is probably only due to not figuring out how it was supposed to be used. Part of the fun is realizing how best to use the starting position. For example: if you get corn, but do not start with agriculture, should you change your 'standard' tech route? If it has marble, could you consider wonder building even as a non-industrial civ? If it is lacking trees, there is a reason - figure that out and leverage. I'm quite sure the computer provides starting cities on a basis of 'minimum usefulness', and everyone gets a fair opportunity. Most restarts are because the player did not see how the start could fit in with their preconcieved notion of how the civ 'must' be played. I think the idea of different civs getting boosts in different eras would be interesting. A praet rush from an AI, wow that would be cool - that's one of the reasons I play MP. Lastly, if you are having problems with barb control and not using the great wall, you might try dropping a city ON a military resource. The time saved building a mine/pasture + road(s), and making the resource unpillagable (except via loss of city) often proves to be a powerful boost. I like to send a settler, worker, and minimum defense. Instead of the worker building a road and improving a tile, I go with the "drop and chop" method for a quick chariot or axeman. Last edited by Ecofarm; Mar 22, 2007 at 10:11 AM. |
|
|
|
|
|
#14 |
|
Warlord
Join Date: Nov 2001
Posts: 168
|
I'd really like to see a truly open ended AI system in the next Civ. Inevitably the game designers and testers are tiny compared to the number of people that will play the game. Removing as much of the hardcoded AI as possible to something modifiable (this would be a challenge, don't get me wrong--I think my dream job would be trying to figure it out) so that as players improved, patches could be released to upgrade the AI with more and more intelligence.
I believe the technologies are there to make this a reality in the next Civ, and I hope whomever makes it attempts to do so. I would find it utterly amazing to be playing on Emperor and lose because the AI out-strategized me. I'm seriously salivating at this...Firaxis or Take Two, hire me! I'd probably work damn near for free on this...give me some food once in a while. Someday I think it will get there.
__________________
DozenLong War is an extension of politics. |
|
|
|
|
|
#15 | |
|
Immortal
Join Date: Aug 2004
Posts: 527
|
Quote:
|
|
|
|
|
|
|
#16 |
|
Warlord
Join Date: Nov 2001
Posts: 168
|
Um...who is Blake? If it's secret advertising for anything it's for Sid Meier to come give me job!
![]() I tried to search for a blake computer game and found Blake Stone. Is that what you mean?
__________________
DozenLong War is an extension of politics. |
|
|
|
|
|
#17 | |
|
staring at the clock
Join Date: Jan 2007
Location: where mise
Posts: 13,290
|
Quote:
__________________
this week in #fiftychat: special guest |
|
|
|
|
|
|
#18 |
|
King
Join Date: Aug 2006
Posts: 943
|
Blake and lustus have significantly improved the AI (their city placement and governor modifications were incorporated into the 2.08 patch. Since then, they've been working on combat so that the AI fights like a semi-moron instead of a complete moron. But I don't follow it closely...). Which in itself should shame Firaxis. Two guys with no pay can significantly upgrade the AI in their spare time?
But it's still, as I understand it, "AI wakes up in the morning, looks around for something to do, and does it. Next turn, repeat." I'm paraphrasing someone there, but I don't remember who. The point is, I'd like to see an AI that's capable of making long-term plans, and the hierarchical decision/value structure to support that. "Given my current position, I think Space Race is my best victory condition. That means (1) I need to expand my lands, (2) I need to maintain peace with at least some of my neighbors, (3) I need to keep my science rate high. Within (1), I identify the best target as (1a) the barbarians who have a worthwhile city on my borders, and (1b) Frederick, who is nearby and not too far ahead in tech. However, upon evaluating the prospects for war with Fred, I decide that a sub-goal is in order: (1b1) research Engineering so I can support trebs. Within (2) (keep neighbors happy), I'll choose to (2a) adopt Isabella's religion, and (2b) gift a tech to Monty." That's the level a human thinks on. And it's hierarchical. I'm not gifting a tech to Monty because I like him, but because I've decided I need good relations with him. And I want good relations not for their own sake, but because they fit my larger goal of a space race victory. Now, that's a major challenge for AI programming. Civ makes chess look easy. To actually accomplish a hierarchical goal-setting AI like this would be bleeding-edge technology, and it would probably take quite a few years, and work embarassingly badly in the meantime. So I'm not surprised that we've got a "wake up in the morning" style of AI. But I can't help dreaming about more. My other dream is a genetic algorithm approach. Mutate the AI in several different ways, let it play against itself, keep the mutations that improved outcome. Repeat. peace, lilnev |
|
|
|
|
|
#19 |
|
War Dancer
Join Date: Nov 2005
Location: Inbetween Shangri-La and Valhalla
Posts: 1,050
|
@lilnev
Nobody would like to play a game, in which you lose every single time. There are commercial reasons why the AI is weak. BTW. I have not noticed that AI improved in 2.08 ![]() edit: I mean wartime AI, but maybe I should read about Blake's AI first ...
__________________
Vanitas vanitatum et omnia vanitas - we are f..... |
|
|
|
|
|
#20 | |
|
Big mouth
Join Date: Dec 2005
Location: France
Posts: 7,710
|
Quote:
. Now the AIs settle on resources all the time. Including food resources .They still get significantly better - population - techs than they did before the patch. They also try to heal their troops now (by retreating badly wounded units into friendly territory, not with medics )
|
|
|
|
|
![]() |
| Bookmarks |
|
| Thread Tools | |
|
|
Similar Threads
|
||||
| Thread | Thread Starter | Forum | Replies | Last Post |
| First Emporer Game: Start | madscientist | Civ4 - Strategy & Tips | 21 | Sep 20, 2007 09:32 AM |
| My First Emporer Game | madviking | Civ3 - Stories & Tales | 18 | Aug 12, 2005 04:17 PM |
| Late Game Workers or Late Game Slackers? | CommandoBob | Civ3 - Strategy & Tips | 6 | Aug 07, 2005 09:16 AM |
| First Emporer game | Damburger | Civ3 - Strategy & Tips | 2 | Nov 11, 2004 02:18 PM |
| First Emporer Game | predesad | Civ3 - Stories & Tales | 30 | Jun 29, 2004 08:34 PM |