Programmed Irritation

Frankly, none of what you have posted would irritate me at all

So losing a Temple one turn from completion is not Irritating? Ok, each to their own...

That is just the way the game happens to be.

Yes, I know, that's the point of the thread...

OK, I can understand if somebody is a little anxious about straying AI units attacking, but really there are several reasons why they are moving over the map, such as exploration and hunting barbarians. Your irritation is just a by-product and not the point. (As an aside, the AI does not launch surprise attacks against units in the open. So, actually there is no, absolutely no need to be scared if your lone settlers and workers etc happen to end up besides AI military units. )

This is interesting information. So, if this is true, then we can discount Wandering AI Warriors from the Irritation Programme! Unless, of course, this is information that is not 'supposed' to be known by 'normal' players. Exploration is not part of their programme as the AI has no 'fog' to explore, but, if you meant explore as in 'make contact', then that makes sense.

And if you don't like the barbs -- this is if I am not mistaken the second time they show up in this thread -- you can turn that off. It's been a long time since I played a game with the barb setting not to either "Off" or "Sedentary".

Dude, if I didn't *like* Irritation then I wouldn't be playing Civ at all. You've got to get out of the mindset of reading Irritation as meaning 'I hate'. I'm exploring the mechanics of the game, not ranting. I think I've made that quite clear on numerous occasions. I think you'll find that your response and subsequent removal of Barbs from your game suggests more about you than it does me.

As an aside, is there any point in constantly spelling "irritating" and "irritation" with a capital "i", i.e. treating it as if it were a proper noun or something like that? And make no mistake I am not complaining about spelling or grammar. You seem to be trying to make a point with it.

Yes. I have made clear several times that Programmed Irritation is the 'name' by which I am grouping the underlying 'objective' of the AI programme. Ergo, if it is a scenario which I think is associated with Programmed Irritation then I shall use a capital letter, even when I shorten it to simply Irritation. The irritation you are giving me by missing the point of the thread would not be capitalised as it's not Programmed Irritation, just standard forum irritation.
 
Buttercup, I have certainly experienced the barbarian and wandering AI behavior that you have seen. The key difference between our experiences is the *motivation*, or reason that we ascribe to those experiences.

AI warriors: Clearly, the AI sends out scouts to perform early exploration. If the tribe does not have the ability to build scouts, it sends out warriors. None of that is Programmed Irritation.

Barbarians: Their purpose in the game is to provoke (or provide) early military conflicts. Their units harrass and destroy undefended units or cities, and (as has been documented elsewhere) they are generated in areas which are covered by the fog of war. Specifically, in tiles that are not in the line of sight of a unit or city.
As has been also documented elsewhere, there are periodic "barbarian uprisings" that happen after a certain number of turns. A larger number of barb units, including barb horsemen are generated. The purpose could be called "Irritation"; I prefer to think of it as danger, that can be dealt with. Barbs were present in Civ 1 and Civ 2, you know; they are a game mechanic designed to discourage a human (or AI) player from under-producing military units.

The random scripts: Ah, here is/are the central point(s). You believe that the barb warriors, and AI warriors, are choosing their actions to force the human player to alter build orders in cities, or thwart the human's plans in other ways. I believe that the barb warriors, and AI warriors, are doing a very simple-minded threat assessment. Is that undefended settler close to me? If yes, I move toward it. If not, I move away toward another target. I have seen the AI civs sending troops 6, 7 tiles distant to stomp a barb camp that I didn't know was there; clearly, they saw the camp as a threat.
It's been well established that the AI can see a human player's undefended cities. Using the logic of the F3 military advisor, the AI makes an assessment of the human's military strength, and whether it should declare war by invading. It does not surprise me *at all* that those assessments of risk/threat are very fluid in the early game.
The AI may be deciding to attack ... then not attack ... then to attack, on successive turns. To me, that is more believable than the idea that the AI looks at what the human is building in a city, counts the turns to completion, and deliberately makes a feint towards the city in order to make the human flinch, by altering his/her build.

Sending out unescorted settlers is a risk; it is a risk that the AI seldom take in the early game. If I have a fog-buster warrior nearby (as you have in your screen shots), then I will sometimes take that risk. If I have found a source of horses (using my scouting warrior) that is 7 squares from the city that produced my settler, I will sometimes send the settler out unescorted and send the recon warrior back to meet halfway. Sometimes it works, sometimes I get ambushed from the fog of war.
But it's not proof of nefarious programming.
 
But it's not proof of nefarious programming.

I have never mentioned the word 'nefarious'. I have only ever used the word Irritation. Please don't confuse the thread by insinuating I mean anything other than 'Irritating'. I'm proving Irritation and Irritation alone, something which you admit is the general purpose of Barbarians. So, essentially, I agree with everything you have said, but, please, don't put words into my mouth that others might be implying but for which I have not stated :)

MysteryX - Yes, I agree with everything you have said, but please remember I am 'testing' for Programmed Irritation, I have said this a number of times. Of course I'm leaving cities empty and leaving fogs for Barbs to appear - that's the point of a 'testing' scenario. In my 'normal' games I am not bothered by them. Playing as Greece with Raging Barbarians, for example, is an awesome tactic for slowing down/distracting the AI while at the same time requiring only one regular unit (the Hoplite) to defend a town, even when the hordes get really quite big.

As for the AI calculating whether it's a good idea to walk next to a human player, that's an interesting idea and one that might be in place, but it still doesn't stop the player being Irritated, the Human Unit is still either missing a turn or being forced into the enemy borders.
 
The main point is that I am not sure that a strategy that is successful for the AI can be characterized as illogical.

I am guessing that you would have a good reason not to just capture an unprotected enemy settler that blocks your warrior. It would start a war, you are not yet prepared to fight a war, and you do not want that war to distract you from your primary goal of expansion.

It is not illogical for the AI to use a settler to block a warrior, if the AI calculates that you may leave the settler alone rather than starting a war.

I do not know exactly what calculations are going into what moves. But I do know that the AI has at least some method of calculating relative power, and if it determines it is in a position of strength compared to another opponent, it will bully the weaker opponent.

It would be illogical for the AI to use unprotected setters to block your path or even just move them up to your military units. I have seen the behaviour a few times. Lone AI settlers do exist, you can even observe their behaviour with scouts, but as soon as a warrior or so comes near they'll run away.

So yeah, it would be illogical, or at least odd, for an AI to block your path with lone settlers.

But ...

... we are not even talking about lone settlers here. In the screenshot I see a stack with a Numidian Mercenary and another unit, presumably a settler. Plus, I have looked at the save game. They weren't trying to block anybody's path, they were on the road to the tip of that land togue to found Cadiz:
attachment.php

I have marked the spot. The save from 110AD shows that city.

Why does the AI chose the route over the hills and mountains? Simple, it offers better defense values. (Unfortunatel for the AI, there seems to be an annoying human warrior blocking the 'best' route.
 

Attachments

  • 370BC.png
    370BC.png
    447.3 KB · Views: 216
(Unfortunatel for the AI, there seems to be an annoying human warrior blocking the 'best' route.

Er, no there isn't. The 'best' route would have been to use the town's road system to jump a square. And I am not blocking the quickest route.
 
So losing a Temple one turn from completion is not Irritating? Ok, each to their own...

You decided to play with barbarians on. Plus, you decided to build a useless temple when you cannot even properly defend yourself against the barbarians that you chose to play with. You just asked for it. :D

Additionally, they are barbarians. As such, they plunder cities, kill workers, settlers, warriors and tear out your terrain improvements. I am really not sure what you imagine to have spotted here. There is nothing out of the ordinary.

If such events are irritating to you, well, they are not to me. They are just what I would expect.

Yes, I know, that's the point of the thread...



This is interesting information. So, if this is true, then we can discount Wandering AI Warriors from the Irritation Programme! Unless, of course, this is information that is not 'supposed' to be known by 'normal' players. Exploration is not part of their programme as the AI has no 'fog' to explore, but, if you meant explore as in 'make contact', then that makes sense.


I know the AI knows the map already. However, they still explore. Or at least, they exhibit behaviour that is indistinguishable from exploration. Those warriors are one indicator. You see them early in the game. Later on, when the AI officially knows the maps this wandering stops. Another indicator for AI exploration are maps that you buy from them later on in the game. I have bought maps that had each and every bit of shroud uncovered, including of course the oceans.

Why the AI explores, I don't know. Maybe it is because map information is worth something. Maybe because the AI is just programmed to do that.

(I don't know if this is not supposed to be known. I think you'll have to find out a lot about the game for yourself.)


Dude, if I didn't *like* Irritation then I wouldn't be playing Civ at all. You've got to get out of the mindset of reading Irritation as meaning 'I hate'. I'm exploring the mechanics of the game, not ranting. I think I've made that quite clear on numerous occasions. I think you'll find that your response and subsequent removal of Barbs from your game suggests more about you than it does me.


Heh. I generally have a very clear vision of what kind of game I want to play. And barbarians rarely play a role. I am not sure that is what you can deduce from my posts here. :p (OK, point taken about that irritation not necessarily means dislike.)
 
I apologise if you're a new player, but if you see two AI Units stacked at the start of the game, it means one is a Settler.
Fine, but it doesn't actually block you from doing anything besides landing on this specific tile (you can still go north or east, Hannibal will complain one time, but he won't kick your warrior out of his territory, and you'll be away from it on third turn). Furthermore, you're implied that the settler is in "danger" when it actually isn't is much danger (Numidian on a Hill against a Warrior). This confused me.

Am I the only one? Does it matter if I am? is 'isolating me' in your argument some form of 'proof' that the purpose of these wandering AI'S is not to be as Irritating as they can?
It is a kind of proof. If no one else finds any irritation in the AI's actions, then clearly the developers, at least, failed to do their job of making irritating AI's properly - only one player among millions finds them to be so! But if the developers failed at their job at making irritating AI's, then these AI's aren't that irritating.

It was? You mean it could, if it wanted, have chosen to perform actions which would have been less Irritating than disrupting Temple production when the said Temple was 1 turn from completion? Couldn't it have simply continued it's path to the Capitol?
Are you playing with NoAIPatrol = 1? Adding NoAIPatrol=1 to your ini file in an attempt to speed the turns up by reducing the AI movement of units affects the barbarians as well. With this flag set, they just fortify where they are and won't move unless an enemy unit (or city, like in your case) is on the NW/SE axis somewhere on the map (barbies see the whole map). This is a bug which the lead programmer has confessed to not understand. It sometimes can give an illusion of bizarre, but smart unpredictability on the barbies' part.

Yes. I said at the start of the post that I was trying, in this scenario, to encourage Programmed Irritation. I think you do not understand the concept of a *TEST*

*FACEPALMS*
Okay, I take back the "sending unescorted settlers out" part. "Worrying too much about random AI units next to the settlers" is still your own making.

This all relates to 'what is the *purpose* of Barbarians in the game.
Then you were ambiguous in your original argument, since it seemed that you take issue with a specific behaviour on part of the barbies. I actually agree with you that they're irritating, but I don't think that was their purpose. They, imho, were designed to force the player in a strategic conundrum (these are not irritating, per se) - do I leave my cities relatively undefended, not wasting the shields and the upkeep cost, but having the risk of barbies invading, or do I build an army - even when there're no Civilizations nearby. However, the developers, probably overly cautious of players complaining about barbarians being too strong, made the barbarians too weak to be anything but an irritation.

That's why in Civ4 they had the barbarians wield a variety of strong units and had only the three lowest levels offer you a combat bonus versus them. Furthermore, in Civ4 they actually had cities, and capturing a barbie city is the opposite of irritation. (They did add a ridiculous rare, but random event in the latest Civ4 expansion that can early kill you with randomly spawning barbs. That ended up to be irritating in the worst sense of the world).

Dude, if I didn't *like* Irritation then I wouldn't be playing Civ at all. You've got to get out of the mindset of reading Irritation as meaning 'I hate'. I'm exploring the mechanics of the game, not ranting.
On second thought, maybe that is a reason why people are discontent with your claims? Because the word has a generally negative meaning. A better way to say what you're seemingly trying to do is to state that the AI's are designed to be an obstacle, not an opponent. You're still, however, overstating your case.

Er, no there isn't. The 'best' route would have been to use the town's road system to jump a square.
Disagree here. From Theveste, it always take three turns minimum. The seemingly "shorter" way is a visual illusion from the "chess king effect".
 

Attachments

  • case.JPG
    case.JPG
    65.2 KB · Views: 58
Er, no there isn't. The 'best' route would have been to use the town's road system to jump a square. And I am not blocking the quickest route.

Well, I don't know where the settler/NM pair came from. Presuming they came from Theveste, there are no faster routes than what they picked. There would however have been a more safe route, i.e. over the mountain that your warrior is on. But in the end, I can only guess what went on there.
 
Well, I've been very happy with what's been discussed about my little examples of early game Irritation.

I think (or hope) I'm correct in concluding that:

1) The Barbarian's job is pretty much Programmed Irritation.
2) Wandering AI Warriors may appear to want to Irritate, but that is not what they are programmed to do.
3) Path blocking is something which might be Programmed Irritation but also might be coincidental depending on the scenario and is probably the hardest area (aside from RNG influenced battles) to specifically prove as a specifically Programmed duty of the AI.
4) The issue of Iron placement is likely Programmed Irritation as no-one seems to be arguing that it's likely random or coincidental.

I enjoyed watching the Youtube video scratchthepitch directed us to, I think everyone should watch that, it covers a lot of the ground this thread is about.

I am now going to (when I get the time) go back to my various saves and see what happens to the events I have listed in this savegame when I tweak them with alterations to see if I can triangulate the motives and movements of the various Irritating pieces on the board. If I find anything of interest in terms of additional proof of anything I shall post them.

In the mean time, I'd be delighted if anyone encounters any instances of Programmed Irritation and cares to share so we can further understand and therefore account for when playing the various Irritation objectives of the AI.
 
As a simple calculation, I make the probability of your not having iron in the 30 hills/mountains closest to your start position about 38% (I think there's 14 sources of iron on the map, and about 440 hills/mountains). Bad luck, but not outrageous.
That said, distribution of strategic resources is obviously not as simple as every tile having a x% chance of getting a resource. For example, from a quick play with the editor, I make the following observations:

1. strategic resources do not naturally occur in adjacent tiles (unlike bonus and lux resources, of course),
2. strategic resources of the same type have a greater radius of exclusivity (I suspect 4 tiles), and
3. rule #2 does not apply to resources on different landmasses.

There might then be further algorithms which generate an uneven distribution of resources. I do vaguely remember this being discussed at some point as knowledge handed down from the devs, in the case of lux resource distribution if not for strats. Or it might have been said in realtion to Civ4 or 5, but anyway... the purpose of this system was not quoted as "to irritate" but to promote interaction between the civs, i.e. to encourage juggling between trading and conquest. Because they thought this might be more interesting than just building your civ in isolation with everything you need in easy reach. You may or may not agree with their judgement (for sure I don't agree so much with Firaxis' idea of interesting gameplay post-2005), and you might consider such algorithms, if they are indeed present, part of the Programmed Irritation. I shall choose to call them Programmed Fun.
 
Buttercup

I've modded the movement of land units in my games to be around 2-3 times faster depending on the unit. Since I usually played large maps (160x160 or so) with 31 civs and raging barbarians, I would see AI units running all over the place fairly quickly. If I didn't go for goody huts right away, the AI got them. They cris-crossed the map and seemed to be exploring everything. They seemed to know where barbarians were and would also hunt them down.

One thing the AI does do that I didn't see mentioned is they'll move a unit next to someone else's territory, and then fortify and just leave the unit there. But they do it to other AI, as well as the player, so I don't think it is something against the player only. I've found fortified AI warriors in another AI's territory as late as 1000AD. The warrior was probably just sitting there for 1000s of years. That may simply be a bug in the "go to" program that was never corrected from Civ2. "Go-to" was poorly done in Civ1-2.
 
^ Well I think this is the other side of the "patrolling AI" dichotomy that Emsworth mentioned a couple of posts back, isn't it? As I understand it, in the early days of Civ3, the behaviour of units that the AI had earmarked as "patrolling" was pretty much to randomly stumble around, in a pretence at exploring the map. But after a while (and after we learned that they weren't really fog-busting) the community started to get irritated by the constant pointless scurrying, kinda like Buttercup is now. So in a patch, Firaxis added the "noaipatrol" option to the .ini file, which made these patrollers all just sit down quietly unless they happened to see something specific to do, like popping huts or whacking barbs. And I guess that's the state all these fortified warriors are in; just minding their own business!
 
Buttercup

I've modded the movement of land units in my games to be around 2-3 times faster depending on the unit. Since I usually played large maps (160x160 or so) with 31 civs and raging barbarians, I would see AI units running all over the place fairly quickly. If I didn't go for goody huts right away, the AI got them. They cris-crossed the map and seemed to be exploring everything. They seemed to know where barbarians were and would also hunt them down.

I have not boosted the land movement of units, but I am thinking on the very large water maps that I play on, it would be worthwhile to boost the movement of ships.

As for much of the AI behavior, clearly it is all programmed, but it is hard to say if it is designed to irritate the player. Until such time as we have access to the source code, or have an open source clone to work with, we will not be sure.
 
^ Well I think this is the other side of the "patrolling AI" dichotomy that Emsworth mentioned a couple of posts back, isn't it?....

That sounds about right - thanks.

I have not boosted the land movement of units, but I am thinking on the very large water maps that I play on, it would be worthwhile to boost the movement of ships.

I limited sail and ancient ship move to 3-5, same as the stock game, but I boosted the steam and nuclear power ship moves to 5 for the earliest (ironclad) up to 11 for the last modern ships. I originally had the later ships moving faster, but that made things a little bit too wild. It was too difficult to use screens and pickets, and too easy to attack and then disappear.

For land move I completely changed the move system from the stock game and the faster moves has greatly enhanced the AI's tactical abilities. It also better simulates the abilities and differences between the kinds of land units.
 
I have done the same in boosting movement for the later period ships, with base movement rates before any bonuses of between 9 and 11. In the Age of Discovery (World Map) scenario, it looks like to get reasonable discovery rates, we are going to have to increase the movement rates of the sailing ships, otherwise, sections of the map are never going to be reached.
 
I have done the same in boosting movement for the later period ships, with base movement rates before any bonuses of between 9 and 11. In the Age of Discovery (World Map) scenario, it looks like to get reasonable discovery rates, we are going to have to increase the movement rates of the sailing ships, otherwise, sections of the map are never going to be reached.
Are you speaking of the standard scenario in 3C3? Or a new .biq or map?
 
I have done the same in boosting movement for the later period ships, with base movement rates before any bonuses of between 9 and 11.

That is interesting that we both ended up using the same move rate.

In the Age of Discovery (World Map) scenario, it looks like to get reasonable discovery rates, we are going to have to increase the movement rates of the sailing ships, otherwise, sections of the map are never going to be reached.

With very large maps, you don't have much choice, or the game will plod along way too slowly. The fact it is a period scenario gives you a lot more freedom because you don't have to scale really early ships to those of modern times.
 
I cant see any cause for irritation in the screenies, simply people exploring and you not cleaning out barbs when you come accross them, all perfectly normal.
 
Are you speaking of the standard scenario in 3C3? Or a new .biq or map?

I am working on the new Age of Discovery (World Map) scenario, which uses Teturkhan's 256 X 256 world map. The larger map has a lot of ocean to cover, especially if you want to try and reach Australia to avoid the competition with the AI in North America.

That is interesting that we both ended up using the same move rate.

My experience in working with war game design companies is that if you have two or more knowledgeable individuals looking at the same issue within the confines of a specific game system, you normally come up with very similar ideas and results.

With very large maps, you don't have much choice, or the game will plod along way too slowly. The fact it is a period scenario gives you a lot more freedom because you don't have to scale really early ships to those of modern times.

Yes, working within a limited time period, in this case, the Age of Sail, makes it much easier to have a greater range of ship speeds for sailing ships than if mechanically-propelled vessels are included. A larger sailing ship has two speed advantages over a smaller ship.

The first is the benefit of a greater speed-length ratio over a shorter vessel. Wave making resistance for a ship increases drastically once the speed in knots exceeds the square root of the ship's length in feet. For example, a ship that is 100 feet long traveling at 10 knots has a speed-length ratio of 1. To increase the speed of that vessel to say 12 knots, or 20%, will likely take an increase in power on the order of 44%, the square of the speed increase. Unless steps are taken to reduce the friction resistance of the hull, which is roughly proportionate to the speed of the ship, the actual increase in power needed can easily reached the cube of the increase, so for a 25% increase in speed, the additional power required can be 200% or double the power for 10 knots. A World War 1 US 4-piper destroyer required 3,000 horsepower to reach 20 knots in speed, while it took 26,000 horsepower, nearly 9 times the power, to push the ship to 36 knots, a speed increase of 1.8.

The second advantage of the larger ship is being able to carry a heavier spread of sail under any given wind condition, so that as the winds grow stronger, the larger ship can maintain a greater sail area than a smaller ship without becoming unstable. There are many cases of a ship-of-the-line running down what would normally be a smaller and faster ship, in lighter winds, when a gale or near-gale was blowing, as she did not have to take in sail. There are also a fair number of instances where the smaller ship ended up capsizing from carrying too much sail while attempting to escape.

When you combine these two factors, it is fairly easy to justify giving an East Indiaman nearly twice the movement factor of a caravel, say having the caravel move at 6 and the much larger East Indiaman move at 9 or 10. Then a Seafaring nation gets a +1 bonus from having trained seamen able to get that much more performance out of a ship, with the boosts from a Navigation School and Magellan's voyage being viewed as coming from more efficient rigs and better hull design and also a much better understanding of the wind and current patterns to gain shorter voyage times.
 
timerover51

Excellent analysis. I agree 100%. But I think we might be straying a little too far from Buttercup's thread subject, so it might be a good idea to continue this on the Age of Sail questions thread in the Creation & Customization section? I've seen the mangled remains by the side of the highway of those who had annoyed Buttercup before. ;)
 
Back
Top Bottom