TheWizardKs TheWizardKs

AI cheats?

AI cheats?

I have been modding the game some and decided it would be nice to have subdivisions of the hulls and moved them around in the tech tree as to when you get them. The tiny hull for instance does not come until you research 'basic logistics'. As a result, the AI's do not have immediate access to them and can not build their 'standard scouts'. What then happens is they do not build any scouts, instead concentrating on colony ships and constructors. This is fine, however it rather bugs me that even with no scouting, the colony ships head *straight for* colonizable planets, and the constructors go on a direct path to resources. In other words they definately cheat. I would like to ask that this part of the AI be fixed, I seem to recall numerous posts I've read that stated they do not cheat, so it's calling out time guys   
19,127 views 32 replies
Reply #26 Top
How convienient. Call a bug a feature.


It is not a bug, it's intentional. During the beta there was a time when survey ships could only autosurvey known anomalies... which made the function pretty much completely useless.
Reply #27 Top
However I did win the colony rush phase in the game I'm playing: I got 77 planets out of about 480, the best AI got about 65.
The game is 1.31, suicidal, gigantic, scattered, 9 opponents, all abundant except anomalies which are common...The AIs do buy colony ships aggressively and probably have some interesting expansion techniques -- So do I.


Maybe you should try the settings as outlined by TheWizardKs? Would be interesting to learn the results...

The bonus's the AIs get at higher difficulty levels are not cheats -- they are there to give better players a game that is challenging and not a walkover.


How do you define cheating then? IMHO, the reason why the code is made to "cheat" is nearly alays the same: to provide challenge.
Reply #28 Top
t is not a bug, it's intentional. During the beta there was a time when survey ships could only autosurvey known anomalies... which made the function pretty much completely useless.


In a perfect world the autosurvey should, besides exploring known anaomalies, move the survey ships in an efficient manner to the great unknown in order to discover hidden anomalies. Unfortunately, we are not living in a perfect world, so the feature is probably a good idea. However, the feaure, like many other features, should be properly documented. Easy for me to say...
Reply #29 Top
How do you define cheating then?


However, AFAIK AI cheating is understood as using information denied to player.


Pretty much is what I feel AI cheating is.

Handicaps/Bonus's are not cheats IMO.

One of the big attractions of GC2 for me and quite a few players I believe -- Is the developers commitment to the AI not having access to any different information than the human players. Also the AIs do not treat the civ controlled by human player any differently than they treat the other AI controlled civs.

I do believe there are quite a few fuzzy areas but I do believe the developers are basically committed to those concepts and try their best -- and I applaud and support them.




Reply #30 Top
Handicaps/Bonus's are not cheats IMO.


I do not want to split hair over this but knowing more/less than a player can be considered bonus/penalty. Ok, I guess you mean that bonuses tied to higher difficulty level than "default" are not cheating. I would not agree to such definition but I do believe that bonuses are necessary for higher difficulty levels.

Ido believe there are quite a few fuzzy areas but I do believe the developers are basically committed to those concepts and try their best.


So do I.
Reply #31 Top
The AI does what it does relatively well and pretty much has the appearance of doing what a human player does. I send out scouts right away. So does the AI. I see a nice fat planet and put a colony ship on a B line for it. So does the AI. I don't see a problem.

Reply #32 Top
Bottom line for me is that I have had this game for 8 monthes now and I still find it challenging, fascinating and most of all fun.

I don't give a rats hindquarters if the aI 'cheats' (a mis-terminology to start with since a computer ai is not a conscious entity, has no conscience, is amoral, and therefor cannot cheat by definition as cheating involves a moral choice; the aI can only only follow the rule set it is programmed to use) to accomplish this - I'd rather have it the way it is than to have an aI that mimicked a human behavior and was forced to act the same as a human without having human intuition - one any 6 year old could beat anytime after figuring out it's flaws.

The devs have made perhaps one of the most challenging 4x aI routines ever; it doesn't take long to see that; well, perhaps it does...for me it has taken 8 monthes and still counting...