Combat Cheats by the AI

[Stapel]

arretium,

No offense, but your story is quite laughable. Check bamspeedy's link and judge again. Some members on this forum are completely lunatic and have analyzed each and every possibility of AI cheating.

The AI does not cheat. If you think it does, please come over with evidence, and not with a handful of unlikely battle results.

We, as civ addicts, have seen it all: loose a tank vs spear. But we never have a winning spear vs a tank, as we upgrade our spears, before they get attacked by tanks. And we have all lost 4 MAs to an infantry, but we all have seen our super MA sending three MIs to heel in one turn.

 

[Padma]

arretium, you may feel quite free to disagree with us; that is certainly within your rights. But as has been pointed out several times already, there are people here who have put an incredible amount of time into testing everything they can about the AI. The results of their tests show the AI does NOT cheat at combat.

We also have the word of the Firaxis programmers (if search was working I would dig it up, but it is buried many pages down by now) that the AI follows the same rules/odds in combat as the human player.

Basically, unless you can provide proof, instead of a handfull of "Hey! That Spearman beat my Tank!" - type anecdotes, we simply aren't going to believe you.

 

[Moulton]

Here is a test you can try yourself: Set Preserve Random Seed off, and re-run your own battles several times... compare the results.

 

[Catt]

Others have already said the same thing, but I'll chime in . . .

Taking isolated incidents, even a series of 30 or 40 battles, is insufficient to come to a conclusion that the AI cheats in combat, particularly in light of developers' statements to the contrary. Others have tested combat involing literally thousands of trials. I have tested many hundreds of trials (in the course of testing radar towers) in a variety of circumstances (bigger empires, technological advancement, etc.) and never seen any anomolies in my testing. Fact is that probability is predictive, and wild sort-term results are to be expected. I've had streaks of incredible bad luck and incredible good luck -- the bad luck is much easier to remember () and is most noticeable when it comes at the wrong time (i.e., "I just need the odds to play out as expected over this short run of trials and I'm golden, otherwise I'm in trouble . . . aarrgh, why am I losing unit after unit" ).

 

[The Last Conformist]

There's also the very good question why the Firaxians should have spent time on implementing some very complex system for giving the AI combat advantages, without showing it in the editor like (most) other AI advantages. If they thought the AI was to soft in war on the higher levels, they could've simply given it even more free units, or simply made extra units materialize in the AI's capital every few turns.

 

[arretium]

Stapel Wrote:

"No offense, but your story is quite laughable. Check bamspeedy's link and judge again. Some members on this forum are completely lunatic and have analyzed each and every possibility of AI cheating."

Well No Offense Stapel, but your arguement is laughable. You critique my argument but you fail to point out its flaws, you provide zero evidence to support your point, and simply engage in an ad homenem attack. Perhaps being a civ addict has reduced your ability to engage in conversation with others?

Where's YOUR evidence?

 

[superslug]

Stapel doesn't need to point out any evidence, it's all over this site if you look. The 'lunatics' he refers to have spent thousands of hours in the editor and the programming analyzing the crap out of the game looking for anything that will help them play better. The AI does not cheat.

 

[Speaker]

Originally posted by arretium
Where's YOUR evidence?

If I may interject on Stapel's behalf. His evidence is in the many posts where helpful CivFanatics have created scenarios with 1000s of warriors and other units attacking each other to show how the odds are even over time. I cannot link to any because the search function is out of operation, but I can assure you that they are there. This same post pops up every month or so and is always shot down. You are not the first and will not be the last. But your attitude here is not a great way to intoduce yourself to these message boards. Arguing with Bamspeedy about the inner workings of Civ3, a field in which he is undoubtedly the expert is pure folly.

 

[arretium]

Padma, Moulton, Catt:

In the spirit of brevity, I'm responding to all three of your statements at once.

As for evidence, I'm compiling it. I do believe a stasticial base should suffice at say 600 battles? Have you all taken statistics? If you have, I believe you'd agree that it would put the likelihood of error at around 3%.

Firaxis Programmer's "Word" Comment

I hope not one of you would actually take the "word" of a programmer who was told by you from another civ friend who heard about it "somewhere." My dear friends they call this latent hearsay. The realiability of sources for any type of information in these circumstanc es is pathetic. If we were trying to "prove" matters in a U.S. or British court of law, such an attempt would be met by an "objection" from the opposing side regarding the use of "hearsay". The judge would of course sustain such an objection because what we are talking about here is utter none sense. In other words, if this is your evidence, you've got to better than that.

 

[Speaker]

There have been chat sessions, where Firaxians who created the game answered this exact questions. If they don't know, who does? And as for your statistical base, there have been people who ran combat tests with tens of thousands of battles.

 

[superslug]

600 battles? The combined community here has probably tested 6 million battles! And as far as the "Word" comment goes, the Firaxians have chats and interviews here regularly. Some of them are even members at this site, while others look around. There's a number of beta-testers for the next expansion pack from this site that are dealing with the programmers on a daily basis about the game's workings.

This thread has been done a thousand times, and since you're not listening to anyone, I suspect the experts are going to give up and let it sink. We've had this argument too many times before...

 

[Speaker]

Hah, you beat me last time Superslug, but I beat you this time. Rubber match upon Arretium's next stubborn comment?

 

[superslug]

I just told another newbie in another thread I was done with deaf ears. I'm not posting anymore in this thread either...

 

[arretium]

Speaker wrote:

"But your attitude here is not a great way to intoduce yourself to these message boards. Arguing with Bamspeedy about the inner workings of Civ3, a field in which he is undoubtedly the expert is pure folly."

I see. So we have reached the point of simply ad homenen attacks now, haven't we? I think if you check back carefully you will see that my response to "bamspeedy" was quite civil.

There's nothing worse than the superior attitude of people who they think they are right and you are simply stupid. Some christians have it about death and hell, of course not all. Some Republicans have (especially those on FOX news) have it about being with the GOP and the evils of being a Democrat. Some Europeans have it about being American. Basically, these attitudes all share one thing in common. No matter what the opinion about the demented il-knowing other side, whether it be stupid democrats, hell-goers/budhists/jews/any other religion that is not the correct one, Americans, they are always wrong. The similiarity of the other side is that they all suffer from a severe case of narrow mindedness. Their mind can't be changed.

I did not know that Bamspeedy was considered the expert at Civ3. Bamspeedy is the expert at Civ3? Is this the consenus opinion of the community?

I'm getting the sneaky suspicion that even if I had the smoking gun of statistical evidence, a video tape of Sid Meir's development team discussing the cheat, and the actual code itself showing how it operates, that you still would be in denial.

 

[Speaker]

I did not say Bamspeedy was the expert at Civ3. I said he was the expert on the "Inner Workings of Civ3." Take a look at some of the links in his signature that show the careful work he has done to show how the game operates. Take a look through the archives and find the exhaustive work that has been done on this subject. You are wrong. Deal with it.

Continued claims to the contrary will result in my resorting to simple, petty namecalling.

(This will be my last post in this thread)

 

[Rhino Ryan]

Originally posted by arretium

I disagree. All of the battles discussed above in my examples occured on desert or grassland terrain OUTSIDE of a city, unless otherwise noted. I have already factored in the terrian, river, and city improvement considerations.

Desert and grassland terrain provides a 10% bonus.

Your river analysis is irrelevant. I am well aware of the 25% bonus, but others might like to know about it. . In any respect, there are no rivers in any of the terrain discussed above.

Your civil defense analysis is irrelevant to open combat considerations. Only the "radar tower" could be a factor but then the AI would have to build the radar tower on my own territory and then remove it before my next turn. Such a tactic would also be cheating, since the AI would have to move their own worker to my terrain thru some sort of "right of passage way activiation" without my consent, build the radar tower, and consequently destroy the radar tower before my own turn. I'd argue such a move by the computer would likewise also be a combat cheat. I suspect we'd agree that the AI's radar tower is not a factor in this discussion.

City bonuses depend on city size. A wall is equivalent in defense to a size 7-12 city. 13 and above receive a 100% bonus based on size of the city.

Your math is incorrect above. You discuss civil defense and then provide an example using mech. infrantry. Specifically, your mathematical computation regarding "29.7 +.15 " does not make any sense. Please provide an explanation for your different considerations.

I stand by my conclusion reached above. I've played this game for many years, in fact, more than a decade when you factor in all of the different civilization versions. [/B]

Okay dude but stop using all the big words... I understand them all but it sounds like a newb wrote it.


On your first Post you said "his one modern Armor took out 3 of mine" First off Modern Armorms are attackers not defenders, I haven't checked recently but I think it is 24.16.3. I know it still seems odd but what you need to do ois when you come across that situation replay that one battle 10 times and see if the same outcome comes up 8-10 times. I know it seems harsh but next time when in battle focus on the enemies life bar, it is a trick that is awsome !!!
-Rhino

 

[arretium]

600 is a fine statistical number. The margin of error is quiet small with that amount of battles.

If I was in Las Vegas right now, I'd be well on my way to "provin' the house was stacked!!" OR, as you guys might say, I'm about to hit one hell of a lucky streak.

Well, we shall see.


Cheers,

Aret

 

[Rhino Ryan]

arretium-
Word to the wise. DOn't swear as Moderators will close the thread and the crazy dilema will be over (for 4 weeks when someone else will ask the same question).
And don't argue with the best. I know you say that we are saying we are taking sides and that we side with people and think that wahtever you say is wrong but it is you that is dong that. All you have done this entire thread is ridicule (sp?) people on there false acusations. With things like I am sorry but I believe you inquirments are completely off, Where is you evedence, and unless you can support you theory you cannot prove anything. Just say it... You think that you are better than anyone on this site and when I wrote this you had what 8 Post and arguing with those of 1000's. I myself with 60 or so am not the most experienced and I n=know that but do not take this site as some chat room, things have ways of biting you in the back. And I don't mean to be offensive here but I can conclude and have most everyone agree with me that you have been a real jerk about this and nobody likes it.
Really sorry but calm down and listen to what these people have to say.

 

[Strider]

Originally posted by arretium
Speaker wrote:

"But your attitude here is not a great way to intoduce yourself to these message boards. Arguing with Bamspeedy about the inner workings of Civ3, a field in which he is undoubtedly the expert is pure folly."

I see. So we have reached the point of simply ad homenen attacks now, haven't we? I think if you check back carefully you will see that my response to "bamspeedy" was quite civil.

There's nothing worse than the superior attitude of people who they think they are right and you are simply stupid. Some christians have it about death and hell, of course not all. Some Republicans have (especially those on FOX news) have it about being with the GOP and the evils of being a Democrat. Some Europeans have it about being American. Basically, these attitudes all share one thing in common. No matter what the opinion about the demented il-knowing other side, whether it be stupid democrats, hell-goers/budhists/jews/any other religion that is not the correct one, Americans, they are always wrong. The similiarity of the other side is that they all suffer from a severe case of narrow mindedness. Their mind can't be changed.

I did not know that Bamspeedy was considered the expert at Civ3. Bamspeedy is the expert at Civ3? Is this the consenus opinion of the community?

I'm getting the sneaky suspicion that even if I had the smoking gun of statistical evidence, a video tape of Sid Meir's development team discussing the cheat, and the actual code itself showing how it operates, that you still would be in denial.

Simply fact of the matter is.... We are right... It's been proven time and time again that the AI DOES NOT cheat in combat. It has been tested by some of the best players this site has ever seen and I'm sure that if it was not true, then all of them would not have gotten the same result.

Originally posted by Rhino Ryan
You think that you are better than anyone on this site and when I wrote this you had what 8 Post and arguing with those of 1000's. I myself with 60 or so am not the most experienced and I n=know that but do not take this site as some chat room, things have ways of biting you in the back.

If we judged everyone's experience on there post count then we would have a VERY messed up site here. DaveMcW is one of the best Civ3 players I know, yet has a post count of 500 or less.

 

[Catt]

Originally posted by arretium
I hope not one of you would actually take the "word" of a programmer who was told by you from another civ friend who heard about it "somewhere." My dear friends they call this latent hearsay.

The Firaxian comments I refer to are from either: (1) a direct post by a Firaxis member here or at Apolyton; or (2) Firaxian comment in one of the numerous chat sessions they have regularly held here and at Apolyton. They're not hearsay - they are direct testimony. Of course, you may choose to attach no credibility to the testimony, but that doesn't change its nature.

The realiability of sources for any type of information in these circumstanc es is pathetic. If we were trying to "prove" matters in a U.S. or British court of law, such an attempt would be met by an "objection" from the opposing side regarding the use of "hearsay". The judge would of course sustain such an objection because what we are talking about here is utter none sense. In other words, if this is your evidence, you've got to better than that.

Reliability has almost nothing to do with the hearsay rules of the common law tradition of US and British courts. Nor does a blanket statement that hearsay is "nonsense." Hearsay is generally excluded not because of its inherent lack of credibility, but because opposing counsel does not have an opportunity to examine the witness whose testimony is the subject of the rule. There are numerous exceptions to the hearsay rule - something like more than a dozen - highlighting the fact that hearsay is excluded not because of a lack of credibility, but because of lack of the opportunity to examine a witness. Just because I say heard Mary say X doesn't mean Mary didn't say X or that the credibility of my testimony as to Mary's statement is suspect -- the hearsay exclusion exists because, in court before the jury, opposing counsel wouldn't have a chance to confirm with Mary that she said "X" or to enquire as to the circumstances surrounding her saying X - the jury would not have an opportunity to examine Mary's credibility surrounding the supposed statement (they have ample opportunity to examine my credibility). But enough with a discussion of common law. We're not in any court, other than the court of public opinion.

As others have said, the "precedent" established here and at other forum sites is that the AI does not cheat except in a few, well-known ways (see TheNiceOne's thread for a good summary, and look at the editor for the difficulty level bonuses if you consider them "cheats"). In the face of the precedent, it is you that must come up with convincing evidence that prevailing opinion is incorrect. As I originally posted, a series of combats in one game is, in my opinion and others', insufficient evidence to prove a point. And though it's been a long while since statistics, 600 trials seems an awfully small sample.

Originally posted by arretium
If I was in Las Vegas right now, I'd be well on my way to "provin' the house was stacked!!" OR, as you guys might say, I'm about to hit one hell of a lucky streak.

To paraphrase what you said of me and others, I hope the above statement is not indicative of your understanding of probability. As I said in my original post, the laws of probability are predictive -- applying them to a historical series of events and attempting to divine the outcome of future events is a fool's errand. Since combat in Civ 3 is a series of independent trials (unless you think it isn't, and want to make that argument too), past events have no impact on future events. If you were in Las Vegas and had just lost 600 bets in a row at the roulette table, betting each time that the number would come up black, you'd not be due for a bunch of red results -- the chances for a red or black outcome would remain exactly the same regardless of the immediate past (but you might be justifiably suspicious of the accuracy of the wheel).

Seems to me you might be trying to prove: (1) that Civ 3 combat is not subject to the law of independent trials; and (2) the AI has advantages that get "turned on" in certain circumstances. It would be very interesting evidence if true, but until you supply something repeatable by other players in a test scenario, something more than lamentations about your luck against the Chinese modern armor over the course of a war, or something more than disparagement of those who don't agree with your hypothesis, your comments will remain not much more than fairly articulate rants (albeit calm rants) about your bad luck.