Programmed Irritation

You definitely need to visit the Creation and Customization Forum. You would be surprised at some of the units you find there.
+1 :goodjob: I'm still amazed at the number of people I can see here in the General Discussions forum that I have never seen in the C&C forum! Look at the Download Database too!
 
An interesting foray into the challenge of guessing how the AI makes decisions but unfortunately none of it is based on any knowledge or fact, it's all perhaps possibly and maybe. In this game I've been careful to top the tree in all areas possible, land area, resources, culture, military strength, gold, diplomacy, production, population. I must admit there are many cities without any units actually fortified in the city, but if that was a factor in the calculations then surely *every* AI civilisation would attack me, not just one from the opposite side of the globe. Aside from Irritation I can see no reason at all why the only prerequisite for a pointless big stack invasion is that the AI civilisation doing the attack must be as far away from my civ as practically possible, the most hardest AI civilisation for me to counter attack.

Only one problem - your own argument collapses under the same analysis. Basically you are saying your wild guess is as good as my wild guess, only my wild guess has no evil AI super intelligence involved. Which seems more likely? This is a math game. Improbable does not mean impossible.

So why are you irritated? Maybe it is a matter of selective memory. We tend to overemphasize bad things. When everything works, you don't pay much attention. My car runs 364 days out of the year, but the one day it breaks down it is a piece of crap. Selective memory.

In addition, you are battling the 'illusion' of the game matrix. Why does the AI seem to win offensive battles more often than not? Well, the AI made a predetermination that it should win the battle based on the numbers and that triggered the attack. And we are surprised that it wins most of the time why? The reason that army production is emphasized by experienced players is because it removes this calculation - if the AI never attacks then the human initiates all of the attacks, which, if the human calculates properly, should result in a majority of wins. This is how experienced players destroy Sid level game enemies.

We know this calculation takes place because the AI will run back and forth between target cities when you add this spear, that archer or a catapult. Suddenly the entire stack turns around and goes somewhere else to attack. Armies are not even necessary for this. A 6/6 Ancient Cav is virtually immune from attack until MDI or LBs appear and they are even then unlikely to be attacked if they are on high ground and/or have a defensive shot unit with them.

You may argue that the AI has omnipotent view of the map. I've seen arguments for and against this. I don't know what I believe, but I plan as if the AI can see everything I'm doing. What I know for 100% sure is that the AI knows instantly, globe wide, where barbarians appear. You can test this by trading maps with a backward civ and suddenly every barb camp in existence appears on the map even if they have no way of knowing that information. Similarly, leaving cities unprotected, which is a magnet for AI attack, may be 'known' by the AI and trigger the invasion.

To me that seems a more likely hypothesis backed by observable functions of the game than a wild guess at a irritation program.
 
To me that seems a more likely hypothesis backed by observable functions of the game than a wild guess at a irritation program.
Soooo, basically you're backing up what Buttercup is saying.
 
Soooo, basically you're backing up what Buttercup is saying.

What, exactly, do you think I am supporting? I do not believe that an irritation program kicks in at any point. What he has mentioned are predictable reactions by the AI that can be explained, IMO, but the observable functions of the game to date. This events have been explained by players that can exploit this predictability and beat the game at Sid. His evidence is. . . what? Circumstantial based on an small test sample?

So we are generating a conspiracy theory why?
 
What, exactly, do you think I am supporting? I do not believe that an irritation program kicks in at any point. What he has mentioned are predictable reactions by the AI that can be explained, IMO, but the observable functions of the game to date. This events have been explained by players that can exploit this predictability and beat the game at Sid. His evidence is. . . what? Circumstantial based on an small test sample?

So we are generating a conspiracy theory why?
Uh, no. And really, I know better than to argue with the Big Brother type, who calls the posting of any opinion that is contrary to theirs a conspiracy theory. Dash those people who dare threaten the proven (we know, cause we proved it, dammit) conservative theories that we have laid out as fact! Whatever.You've been posting here how long? Try taking a look at your 20K SG, or your all war SG.

I didn't say you were generating (or supporting) a conspiracy theory. That doesn't make any sense. :rolleyes: What YOU quoted Buttercup as saying was ~ An interesting foray into the challenge of guessing how the AI makes decisions but unfortunately none of it is based on any knowledge or fact, it's all perhaps possibly and maybe.

What I quoted you as saying was ~ To me that seems a more likely hypothesis backed by observable functions of the game than a wild guess at a irritation program.

So basically, your agreed with Buttercup that your hypothesis (what you feel is an educated guess) is better than Buttercup's guess. Backing up what Buttercup had said.
 
Okay, how about we change the tone of the exchange? I'm sorry you are offended by 'conspiracy theory'. No insult was intended. A run of bad luck with the RNG is not proof of anything other than that the house always wins when you gamble. Commiserating in regards to bad luck is one thing. Suggesting that there is a hidden program to supplant the RNG is another.

From the OP:
I had two of my 'odd suspicions' confirmed today regarding the Programmed Irritation of AI gameplay, both of which prove that neither luck nor Unit strength are factors when deciding who wins or loses a battle in certain situations.

Emphasis added. Proved it? Did I miss a mountain of evidence? There is no screen shot, no uploaded save and no indication if the city was on a hill, attacking across a river or anything in either scenario. A complete account of units was not included. What was the status of each unit? Veteran? Regular? In the first example he took two of the cities without problem but focused instead on the one he didn’t capture because the AI swooped in and stole it. He invited the AI to the war! How was this some great surprise?

I won’t even touch on the VASTLY superior unit remark, but once more we get an incomplete picture of the event and the surroundings. If a spear is fortified in a city and that city is size 7 or has a wall or is on a hill or attacking across a river, that spear has a defense of a 3.7. Suddenly those swords and AC are actually favored to lose the fight, not win. But we don’t know because the details are not in the OP. Not to mention that even if it was on flat grassland they still have a defense of 2.7, a near match for the sword or AC. If they had three spears there is a pretty good chance of losing the battle. Two go for you and three against and you failed to take the city. He did fail, but then his 4hp AC managed to kill a fully healed elite spear and capture the city. Bet the AI was irritated that its fortified fully healed elite spear couldn’t hold back that last human unit.

If he gives support to his argument with evidence, great. I can claim there is a magic fairy in your computer making decisions but that doesn’t make it true or likely. You’d probably want to see the magic fairy (or at least evidence) and rightly so. PaperBeetle suggested a great way to test his theory. When vmxa responded he was handily blown off. I've come back with logical explanations (as have other players).

At least I was honest and said it was a hypothesis. Programmed irritation proven? I guess those reading the posts will decide which argument is more persuasive.
 
And so they should, Ravi. Before I get started, let me say that I am glad that we are changing the tone of this discussion. Seriously, I never meant to take on a full blown arguement. To me, Buttercup was expressing a personal opinion about the Civ3 game that he purchased. (I say "he" because you have already claimed Buttercup was a he, and for all I know, you may be right.) Also, for all I know, Buttercup may just be trolling for me to give some input so that what I say can be trashed by others. I don't really care. We have a hundred people sitting in this forum and no one posts anything. How boring is that. At the least, I can liven this place up a bit.

You have quoted Buttercup once again, and I have emphasized as well the same quote ~

I had two of my 'odd suspicions' confirmed today regarding the Programmed Irritation of AI gameplay, both of which prove that neither luck nor Unit strength are factors when deciding who wins or loses a battle in certain situations.

Buttercup was just stating a personal observation about his game play. This observation "proved" his own 'odd suspicions'. Not yours, not mine, his. He then went on to describe what he calls Programmed Irritation. I have read and re-read Buttercups posts to try and decide if what I thought he meant from the begining is what he really meant. And it seems it is.

What you have posted in the above post is absolutely true. Except the part about there being a hidden program in the game set to irritate the human player. Buttercup never said there was a hidden program within the game. From the begining, he's claimed this is just a game, one that he believed to be numbers based and having logic problems. But the more he dug into the game, the more irritated he became at the repeated instances he observed of AI habits. Not once did he claim there was a seperate or hidden "Irratation Program".

Everything in Civ3, 3C3, whatever title you please, is a coded program. When a Tribe shows up in a certain color, that's not a hidden program that does that. It's just part of the game as a whole. Sorry for the simple example, but I didn't want to make it too complicated. It's all digital programming. So I hope that we can agree that everything that happens in the game is part of a computer program. There is no human inside the machine, nor fairy god mother making the decisions for the game.

Have you ever noticed that at the begining of every game you've ever played, that the AI neighbors you have tend to build towards your nation? According to what I hear you say, this is just coincidence. It's not part of the program. I think not. I believe the AI is programmed to expand its nation towards yours to try and limit your growth potential. I believe that would be a smart way to program growth options for AI nations. Not only does it limit your growth but initiates contact between the nations. To me, this wouldn't be a hidden program. It's just the way the game's been designed. Ok, let's say you and I agree that this is just normal programming. Nothing hidden, nothing secretive, no conspiracy. Let's also say this 'programmed action' by the AI irritates Buttercup. This irritation wouldn't be happening if the action wasn't being done by the AI. The AI wouldn't be doing the action if it wasn't programmed to do it. Therefore, this is 'programmed irritation' for Buttercup. That's the point he was trying to make.

I become irritated when the AI declares war on a nation that vastly out-numbers it just because my nation is in between the two of them, so that they can trek their troops through my lands. There are certain aspects of AI behavior that is irritating. These forums are filled with testimony. And the AI is following some basic programming when it commits these actions. They are not doing it because the want to. This is the programmed irritation that Buttercup is speaking of.

I also became irritated when posters came on this thread and tried to tell (in several incorrect ways) Buttercup that he was wrong. I'm just trying to help Buttercup express himself, as like you've said, his posting is not as articulate as it could or should have been. Of course, he's not as practiced a writer as you. Just as I am not. I wish you played C3C as well as you write. I wish I played C3C as well as you write. :)
 
\Have you ever noticed that at the begining of every game you've ever played, that the AI neighbors you have tend to build towards your nation? According to what I hear you say, this is just coincidence. It's not part of the program. I think not. I believe the AI is programmed to expand its nation towards yours to try and limit your growth potential. I believe that would be a smart way to program growth options for AI nations. Not only does it limit your growth but initiates contact between the nations.

I don't know that this is always true. In some games it seems like my neighbors are coming after my borders first, but in other games I am relieved to see them build their first few settlements away from me. If there is some sort of programmed preference for expanding toward the human player, I think it is down on the priority list behind settling near a good resource (whether or not the resources have been revealed yet).
 
I don't know that this is always true. In some games it seems like my neighbors are coming after my borders first, but in other games I am relieved to see them build their first few settlements away from me. If there is some sort of programmed preference for expanding toward the human player, I think it is down on the priority list behind settling near a good resource (whether or not the resources have been revealed yet).

Ok, I'll change what I said to 99% of the games. I've seen the AI ignore resources and plant cities in my direction. At least in my games, the human player is an AI magnet. I think that maybe it's something we ignore now (if it was ever noticed by some). Again, I play on different maps than most people.
 
Thank you for signing the peace treaty. It will remain active for 20 posts or until war is declared! :D

I disagree that he did not insinuate that there is a hidden program. I go back to the quote I pulled from the OP.

I had two of my 'odd suspicions' confirmed today regarding the Programmed Irritation of AI gameplay, both of which prove that neither luck nor Unit strength are factors when deciding who wins or loses a battle in certain situations.

So then he has proved that neither luck (RNG) or strength (Unit Stats) are factors in certain situations. So what is the determining factor? Vmxa tried to explain the swings of the RNG and even explained why it does this. Irritating yes, but not mysterious.

He then doubles down on this by saying later in post #4 when his is brushing off vmxa:
This isn't even about the random number generator, I'm actually suggesting, if you read again, that in many circumstances there isn't even a generator in action, the wins and losses are pre-determined based on the level of 'irritation' it generates in the player - a totally different algorithm.

He is suggesting a real program exists and that it somehow trumps the regular programming just to irritate us. I am arguing that it does not exist and, if it does, he has done a botched job of making any sort of argument it is favor. He provided incomplete and circumstantial evidence and did not test at all other than causal observation.

I do have to wonder what Buttercup was looking for in way of response to the OP. This outcome is probably exactly what he intended. I agree with your analysis with regards to Buttercup - he has every right to be irritated with the programming of the game. One would need to wonder why you would keep playing the game if it irritated you so much with its basic programming. I personally don't mod games but it might be a nice remedy for the irritations. But if you are going to make a claim like this, you better to be willing to argue the mechanics of the game.

There are hundreds of players lurking for advice and we want to encourage them to play, enjoy and, better yet, play with us on the SG and GOTM forums. When stuff like this gives out bad information, we feel it necessary to point it out so that less experienced players don’t believe this wholesale without actual proof. That is why I am taking issue with it.

Regarding the AI growing towards you. I think the reason 99% of the time this happens is for a much simpler explanation. They want the same nice land, Bonus food or luxury you do. In PTW the AI will use RCP, so if the city is in a nice spot and fits on their RCP, then no wonder we are in competition. Since the AI can't plan tile improvements beyond its cultural boards it seems odd there would be a special program that would drive the AI towards the human.

This is what I mean by hidden programs. Like I said before, this game is all about math. The AI crunches the numbers and makes a decision. It seems more likely that the AI has counted the tiles and assigned it a higher value and thus selected it for settling. It prioritizes this based on value of the land and nearby luxuries/hidden bonuses and distance. Maybe corruption. The human makes much the same decisions, though we certain do think about expanding towards the AI (in some cases) to purposely block them in. Is there also a value that pushes the settling location to a higher spot if the human is in that direction? Maybe. Maybe that directive exists in regards to all civs and not just the human. Perhaps not less irritating but at least it’s fair! :lol:

Again, it seems this could be tested. Create a game with your civ in one direction and some bad/mediocre land between you. Then give it bonus food, luxury and resources in the other directions with better land. If the AI choses to settle towards you first, then it could be a compelling argument. I might even run this test just to satisfy my own curiosity now. :D

I totally agree the AI does plenty of annoying things - but that's mostly because it’s a AS rather than an AI. :D
 
Ah, well I guess you win then. :goodjob:
 
"If the faith in your own lie is strong enough, people will find ways to believe you."
(I don't know if anyone ever said this - I could be the first)
 
Rali said:
There are hundreds of players lurking for advice and we want to encourage them to play, enjoy and, better yet, play with us on the SG and GOTM forums. When stuff like this gives out bad information, we feel it necessary to point it out so that less experienced players don’t believe this wholesale without actual proof. That is why I am taking issue with it.
Dunno if there's hundreds of lurkers out there, but I +1 this. The whole business seems like a typical conflict between accepted model and novel theory in science. I don't think anyone in the 'orthodox' camp here is against new theories per se, but we'll challenge them until there is evidence which outweighs the evidence built up in favour of the old. Evidence in favour of the accepted combat model extends at least as far as some extensive 'lab' testing, and possibly as far as details from Firaxis (I can't be sure about that, as I haven't been here from the launch of Civ3).
Last time we had this discussion, I said I'd try to do 'field' testing of combat results to build evidence for or against this kind of proposition. I'm sorry, I didn't do that (except for some logging of human-on-AI bombardment results from one of Buttercup's modern age scenarios). I reaffirm my comitment to the principle. Maybe in the next GOTM - we get longer to play them these days.
As for testing "special circumstance" combats, I already suggested one way, but I don't feel qualified to test it myself, as I never came across a situation where I honestly believed the game wasn't playing fair with the RNG.


I'm aggrieved we always come to blows over this kind of stuff. But I'm happy we're all still here posting about it. :)
 
Dunno if there's hundreds of lurkers out there, but I +1 this. The whole business seems like a typical conflict between accepted model and novel theory in science. I don't think anyone in the 'orthodox' camp here is against new theories per se, but we'll challenge them until there is evidence which outweighs the evidence built up in favour of the old. Evidence in favour of the accepted combat model extends at least as far as some extensive 'lab' testing, and possibly as far as details from Firaxis (I can't be sure about that, as I haven't been here from the launch of Civ3).
Last time we had this discussion, I said I'd try to do 'field' testing of combat results to build evidence for or against this kind of proposition. I'm sorry, I didn't do that (except for some logging of human-on-AI bombardment results from one of Buttercup's modern age scenarios). I reaffirm my comitment to the principle. Maybe in the next GOTM - we get longer to play them these days.
As for testing "special circumstance" combats, I already suggested one way, but I don't feel qualified to test it myself, as I never came across a situation where I honestly believed the game wasn't playing fair with the RNG.


I'm aggrieved we always come to blows over this kind of stuff. But I'm happy we're all still here posting about it. :)

Buttercup did this more. Ramble about how the game sucked with some nonsensical idea. Then he responds to the person who gives some response - but he attacks him on his person not his arguments. Then gives some more nonsensical ramblings.
I bet it happened here to. I don't know, he's on my ignore list.
 
Some interesting points have arisen since my last visit!

Thanks Cyc for attempting to help clarify what both yourself and I find pretty obvious but some others prefer to ignore via blinkers. Don't worry too much about Raliuven, I thought he was a troll when I first started posting but it turns out he's not being obstreperous for incitement reasons but rather that he just enjoys deconstructing viewpoints, something I myself enjoy doing. I suspect Raliuven is a fan of political argument, hence his choice to call everything a 'conspiracy theory' instead of 'unproven theory' as PaperBeetle might.

My reply to anyone who uses the tired old 'conspiracy nutjob' technique too many times in general debates is simply the quote "If there's no government conspiracies why do governments employ secret agents?" Ergo, the programme for Civ3 is a secret which suggests steps would have been made to make correct external analysis of the programme more complicated to copy. A small Irritation programme would act as an excellent red-herring to such analysts as it would adhere to no 'logical' rules. Similar to how Map-makers add and remove small insignificant details to their finished product so that anyone copying and pasting their map can be easily identified.

Just to clarify some points which have become hazed:

Firstly, Cyc has a pretty perfect grip on what my post is about and no I'm not trying to get Cyc into trouble, lol.

Secondly, the main problem with forums such as this is that the more one contributes, the more one feels 'ownership' of the information which is presented on a day-to-day basis and if someone 'comes along' and says something which you might feel gives 'your baby' bad press then the knee-jerk reaction is to stamp on it.

However, this post post isn't about being negative about the game. This, I believe, is the big misunderstanding some of the regulars are having with this thread and why some are being more hostile than common sense would dictate. This post is about trying to expose a mechanic of the game which will help all comers to understand that Irritation is something you will have to overcome if you are to achieve Civilisation Game success.

The difficult part of this knowledge and advice is that no-one knows how Programmed Irritation is programmed nor when it is triggered nor whether it is simply the unintended consequence of every other programmed event overlapping with each other.

As PaperBeetle rightly says, if it's a programme then it can be tested for.

However, during a course of a civ game there are literally millions of events taking place and it would take a monumental feat of patience to try and isolate and correctly chart every instance of Programmed Irritation which might only happen 4 or 5 times a game.

But the fact that I can post this thread and have people immediately know what I'm on about does in itself prove proof. This would be the kind of evidence which would initiate a 'deeper' analysis in search of 'conclusive' proof.

The actual programme or programmes that are used to produce the Irritating mechanics might not be specifically titled an Irritation Programme but this is what I choose to call it/them.

Aside from the instances I have mentioned there are many other 'unfathomable' game event scenarios which fit this thread which I have both encountered and seen other people talk about and this thread would be a good repository for such random 'Irritating' events.

Hence I stand by my initial statement that Civ3 is based far more on minor Irritation events than any solid challenge and that this was an intentional decision made by the creators to create a game which would not be 'too easy' but at the same time didn't actually programme the AI Civilisations to 'win'.

By doing so the programmers may not have a specifically titled 'Irritation Programme' but the end result is 'Programmed Irritation' to which each player will, during a game, often come across events which make no game-sense in any regards other than the AI is simply doing it's programmed duty of Irritating the human player.

I think the examples I have provided so far are exemplary examples of this theory.

The AI Civilisations do NOT work together to defeat the human player. If this was the case then they could simply ALL declare war on the human player after the discovery of Iron and Horses and refuse any peace deals.

The AI Civilisations do NOT try and 'win' the game individually because they will unalterably choose to research Music Theory before they have learned Railroad, regardless of level.

So, once all other possibilities have been excluded, what is left, however improbable, must be the truth.

QED, The AI's job is to Irritate the player whenever possible to which, if this is the case and there are no other possibilities to be explored, would mean that a specific 'Irritation Programme' for certain events is by no means beyond the realms of possibility and, probably, most likely. This programme would act in the background and be an illogical calculation similar to the Luck variable in RPGs but with negative modifiers rather than positive modifiers.
 
Yep it happens if you let your wars take a long time then allies will from time to time steal a goal from under your nose thats one of the reasons I like my wars to be swift and devastating, also one of the reasons I tend to wait until the Industial age before embarking on big wars. (Rail crew behind your attack units driving you right through the enemy in less than 5 turns (by which time you have either won or sue for peace)).
 
The difficulty in testing for an aberration that occurs 4-5 times out millions of calculations is exactly what would be expected from a RNG. So let’s try this again.

Statistics 101
Flip a quarter an infinite number of times. As the number of flips approaches infinity you will find the split between heads and tails to be 50/50. However, if you pull a subset of100 flips you will find a varying number of heads vs tails. You can plot the results revealing a bell-shaped curve. Although instances of 0 heads or 100 heads will be rare, it will be expected. This is because the flip of the coin is a true random generation – each flip is independent of the previous flips.

So why do these results seem ‘odd’ to a human? Humans are incapable of producing a truly randomized list of numbers. If you ask a human to start randomly picking numbers they will quickly bring bias into the equation. After picking a few odd numbers you’ll pick an even number to balance it out. You’ll try to use numbers that have not been used before. This is not random. Task a human with making a random list of100 numbers between 1-100 and you will get a list of all 100 numbers scrambled. Task Excel with the same job and it will be rare to find a list that contains every number between 1-100. Repeated numbers will be common.

The game correlation to this is when you believe you just have to win the next battle because the other five went against you. If you don’t win, the game is obviously cheating. In this instance your rational mind is rebelling against the RNG because you have brought a bias into the calculation that does not exist in the program. The absence of this bias in the program is what you consider ‘irritating’. It is not the presence of something that should not be there, it is the exclusion of something that you believe should be there.

In simple terms, you are not irritated because the program exists; you are irritated because it doesn’t exist.

Finding 4-5 incidences out of millions of calculations is expected. It is not even surprising that they caught your attention. Because they are rare by nature you will selectively pick these incidences out as special. Congratulations. You have discovered the short end of the bell curve.

Your application of Sir Arthur Conan Dolye’s famous Sherlock Holmes quote is painful. You have neither created a list of the impossible nor have you examined every possible alternative. While it sounds good in fiction, it is rather hard to apply in real life.

How about we employ Occam instead?

Option 1: The reported events can be explained within the currently understood framework of the game and the RNG. Variances are expected and 4-5 identified occurrences out of millions of calculations fits with the statistical concept of a bell curve. We accept that people will encounter these variances and their frequency will increase as the number of calculations increase. We accept that they may be ‘irritated’ when the variance works against them.

Option 2: There is a secret program, or a collision of programs, hidden with the game to irritate the players. Even though the occurrences of the irritating events can be explained by applying a basic understand of statistics to the current RNG and game mechanics, I know it exists because, well, I’m irritated. Further proof of the existence of this program is that I know that other players are irritated by the game from time to time as well. In addition, it is logical that this program exists because it would be an excellent tool for the programmers to use in counter espionage activities against their competitors.

Now, apply the following:
We are to admit no more causes of natural things than such as are both true and sufficient to explain their appearances.
-and-
When you have two competing theories that make exactly the same predictions, the simpler one is the better.

Lucky for me I don’t need to convince you; that would be impossible. All I need to do is convince the other readers. The only one knee-jerking here is you. Try providing evidence for your theory rather than applying poor deductive reasoning and personal attacks.

P.S. Comparing the makers of Civ3 to the CIA is just plain silly.
 
Yessiree. That adheres to the old addage - "If you can't confuse 'em with the facts, baffle 'em with bullsh1t." :goodjob:
 
I don't get it. Why can't I do both? Or do I have to flip a coin to decide? :lol: :p
 
Back
Top Bottom