AI "cheating" is necessary and welcomed!

[Grumbold]

The goal should be to make an AI opponent capable of putting up a significant challenge without cheating of any kind. Then if you have to weight a few factors in its favour, whether it be research, citizen happiness or accellerated construction, the cheating will not so obvious as to be infuriating. Capturing cities to find they do not contain the basic improvements necessary for their existance is not acceptable in 2001. If the Civs can fight an offensive war in a vaguely competent fashion and ally with each other to oppose the front runners then even a good player shoud have problems when opposed by multiple AI nations sharing their tech research and allowing troops through each others territory.

[Roman]

quote: Originally posted by Ralf on 05-04-2001 03:45 PM Mark Everson quote:[*] Most importantly is of course that Firaxis from the very start, tries to design and implement ideas in as "AI-friendly" ways as possible. Many added new ideas works for example like "double-edged swords", in the sense that they act equally as much on the human player, as on the AI-civs. But, because the human player always is prone to be the more active and expansionistic one - the burden often lies much more on the HP and then one the AI-civs. This is something the game-designer can (and should) exploit.

I disagree on this. Features should not be eliminated from the game just because the AI wouldn't "know" how to use them. Such features should merely be bypassed by the AI and should be assumed to be done at high (100%?) efficiency or simulated statistically.

[polymths]

quote: Originally posted by Roman on 05-04-2001 08:42 PM I disagree on this. Features should not be eliminated from the game just because the AI wouldn't "know" how to use them. Such features should merely be bypassed by the AI and should be assumed to be done at high (100%?) efficiency or simulated statistically.

I believe that this really hits the nail on the head!

Just because the human player is asked to manage and keep track of lots of units and lots of detail doesn't mean the AI Civs should.

In essence the AI should really be playing a much more simplified version of Civ than the human player. Certain things should definitely be simulated statistically such as tile/city development, wonders, tech research, etc. I think that one reason why past Civ AIs have stunk is that they were asked to manage way too much detail. For a strong Civ AI, every feature and aspect that can be modelled statistically and simulated statistically should be done so. This will leave the AIs actual analysis and decision-making to be much simplified.

[Cyclotron]

Cheating? Well, I don't like it, but I think it may be neccessary.

Bottom line: Make the AI very, VERY good... if that can't happen, then make it cheat.

------------------
- Cyclotron7, "that supplementary resource fanatic"

[Ralf]

quote: Originally posted by polymths on 05-03-2001 09:55 PM Tiles will "age" slowly through time at a realistic rate. Some token workers will roam around to give the illusion that actual workers are working the land but in reality the tiles are simply "maturing" through time by themselves automatically. If these token workers are killed, then the maturing rate will slow until new token workers start roaming the land. But these token workers will not need to be managed at all but will just roam around aimlessly for illusion purposes. As for city improvements and wonders, these will be awarded to the Civ at certain intervals of time based on the theoretical productive capability of the Civ, assuming that no other Civ already has that wonder.

I think the maturing AI city-area tiles sounds OK. But I cant say the same about the "awarded" CI:s and Wonders idea. Yes, the AI shouldnt have to "evaluate" what to build - it can just follow player-editable template-lists most of the time (until something triggers it to temporarily take over direct command). But the AI-civs must sure as hell work for a living!

By "work for a living!" I mean that there should always is a clear and 100% straight connection between exactly how many harvested city-area tiles and production-improving CI:s/Wonders an AI-empire currently have, on one hand - and what that same AI-empire is capable of producing in terms of raw shields, lightbulbes, coins; leading to new CI:s, units and Wonders, advancement in tech-tree and so on, on the other hand.

Above connection-ratio can be a player-favourable 0.9:1 ratio, a fair 1:1 ratio, or even AI-favourable 1:1.1, depending on what pre-game level you choose to play on. But that connection must be fair and crystal-clear - so for the love of God: NO free CI/Wonder give-aways. MOST civers is bound to feel heavily cheated by "free give-aways".

[This message has been edited by Ralf (edited May 05, 2001).]

[Ralf]

quote: Originally posted by Mark_Everson on 05-03-2001 08:49 PM I just don't think that you can make it appear realistic. Especially in the military area. As I understand your 'attack patterns' description above, the game would actually Need good AI to execute them! There is no such thing as a limited set of preprogrammed attack patterns that will suffice for most circumstances. And if the military AI is not sufficient, then having the AI armies mature at the same rate as a good players' would do no good! The AI will still lose horribly.

I certainly agree with above.

--------------------

quote: Originally posted by Urban Ranger on 05-04-2001 03:04 AM "By comparison, the game AI (or any silicon-based intelligence for that matter) is 100% dead and non-experiencing (of course). It lives in a 2-dimensional "flat" world, figuratively speaking - by that i mean it cannot possible "overview" anything." This is definitely not true. There are various ways to program some sort of overview into the AI, for example, using overlayed weighted fields of influence.

Urban Ranger; If you dont accept that artificial intelligence just cannot "experence" and "see" things in real-time, the same way as living intelligence can - then its no point for me to continue arguing. Besides, the problems surrounding artificial intelligence in general, is not just about how to evaluate and respond to constantly shifting world-data - its also about how to retrieve and digest that massive amount of data in the first place - you seems to believe that the latter is already given for free.

--------------------

quote: Originally posted by Roman on 05-04-2001 08:42 PM I disagree on this. Features should not be eliminated from the game just because the AI wouldn't "know" how to use them. Such features should merely be bypassed by the AI and should be assumed to be done at high (100%?) efficiency or simulated statistically.

Well, Im not much for statistically simulated-, or percentage probability AI-responds - or so called "fuzzy AI-rules". Sounds good in theory, but the results often stinks. Some percentage-probabilities is needed here and there, of course - but in general, they should be restrictive about it.

[This message has been edited by Ralf (edited May 05, 2001).]