Programmed Irritation

[Buttercup]

Buttercup, when your Galley comes across an enemy Galley, stop and fortify. Let them attack you. See if you get better results.

When it comes to Barbarian ships this is precisely what I do, because Barbarian ships always attack.

Your method is a stroke of genius bar two small flaws:

1) Having my ships wander up to the enemy, fortify, wait, does not really fit my desire to have a navy which 'dominates' the seas.

2) Portugal didn't attack any of my Galleys, fortified or not.

& oh heavens, that ocean fortification thread was aaaages ago. I can't remember if it adds a zero or a 0.1 bonus!

 

[Cyc]

Here'sa simple solution to your problem...

It should help both Buttercup and Spoonwood (along with the rest). Play a game together. Buttercup you start a game and post the 4000BC save here. Spoonwood, you pick up the save and tell Buttercup what to do. Buttercup, you do exactly what Spoonwood tells you to do in the game. Then post that turns save. Each turn you play, Buttercup, gets saved and posted here for Spoonwood to analyze and post instructions for. That way, everyone knows what the deal is and what's going to happen next. No more vague descriptions, no more grandious statements. It will all be on the table for anyone to see.

If you screw up the turn, Buttercup, or don't follow Spoonwood's instructions, then he wins the arguement. OTOH, if Spoonwood's instructions are followed to the letter and your oddities still emerge in his environment, this would lead to your proclamations being accurate and well worth a discussion.

What say you, gentlemen? Have the huevos to put your money where your mouth is? Screenshots may work occasionally, if you both agree, but the save should always be posted by Buttercup to verif the difficulty and other options available for that turn (such as trades, etc).

 

[Buttercup]

I can assure you that wouldn't help in the slightest, lol.

Spoonwood is the Civ Religious Fanatic, whereas I am the Civ Logical Sceptic, and never the twain shall meet.

But seriously, Spoonwood only plays the hardest levels, whereas I only play the mid range levels. During his games the idea of an open fight is almost non-existent whereas on the medium levels it's quite possible to throw your troops around with abandon depending on the scenario.

Spoonwood wont attack anything without 300 Artillery in a stack because Spoonwood plays on levels where that is the only way to win. I wont use 300 Artillery because on the medium levels it gets to a point where they just slow you down and end up good for nothing but the opening exchanges and defensive security.

VMXA pretty much likes to just all-war all the time and is a specialist Army creator, but his games involve virtually zero Diplomacy and very little in the way of role-play or non-violent victories.

Raliuven is the direct opposite of VMXA and specialises in negotiating himself to 4 turns per tech technology based victories.

I like to either build horses and trample my foes non-stop, hard and fast, combined with diplomatic manoeuvring or just go the all-peace route with acres of ass-kissing and unhealthy Wonder addiction.

Other regular posters I'm not so sure about but I have my suspicions.

The reason these 3 posters and myself post so much is precisely because each of our differing techniques provides varied and interesting divergent results. If we all played the same this forum, and the game itself, would be like watching paint dry.

 

[Spoonwood]

At Buttercup,

I can assure you that wouldn't help in the slightest, lol.

Spoonwood is the Civ Religious Fanatic, whereas I am the Civ Logical Sceptic, and never the twain shall meet.

You simply aren't a logical sceptic, especially at this point in time. If you were, then you would have posted in full detail (even your last post didn't have enough information) your unlikely battle results, as well as a log of the rest of your battles and the conditions they happened under especially. As for myself, I still stand waiting for complete details and full battle results.

But seriously, Spoonwood only plays the hardest levels, whereas I only play the mid range levels.

1. You've already posted a game which refutes this claim you've made about yourself.

2. No, I don't only play the hardest levels. As part of the Quartermaster's competition for the HoF, I'm an Octathlete. This means I have one of the fastest finishes at each difficulty level. I also have other games at lower difficulty levels on record. See my HoF profile. Yes, all of those links, and others below, consisted of public information before you started running your mouth Buttercup.

During his games the idea of an open fight is almost non-existent whereas on the medium levels it's quite possible to throw your troops around with abandon depending on the scenario.

I don't know what you mean by "open fight".

Spoonwood wont attack anything without 300 Artillery in a stack because Spoonwood plays on levels where that is the only way to win. I wont use 300 Artillery because on the medium levels it gets to a point where they just slow you down and end up good for nothing but the opening exchanges and defensive security.

First off, I've rarely seen anyone carry that much artillery (I don't doubt some exception exists though). Second, no, I do NOT always bombard with artillery first... if that's what you mean. You can find some of the games in my profile, linked to above, where I played with the Iroquois and won via domination. Those games almost entirely consisted of mounted warrior rushes, and very little to no catapults, military wise. If you consult my histographic thread (you can find saves for the first game here) which contains references to a few different games now, you can also figure out that I simply don't always use artillery units when attacking... especially once I have multiple armies ready to roam around. Or you can look at my Sid conquest games here , and here. I simply conquer too fast in some situations to always use artillery type units.

That said, I generally use fast units, artillery, or armies when attacking or some combination of the three (which might just mean one of them)... only some exceptions exist. When I start a war, my initial stack will contain artillery type units and I'll definitely use them to my advantage as much as a I can. But once I plunge deeper into enemy territory, the artillery type units play less and less a role at least before railroads and artillery proper... though I'll definitely keep using them if say fighting with knights and trebuchets in the middle ages... it just becomes harder for them to keep with my fast units. Artillery type units have actually ended up playing the biggest role in some upper level spaceship games I've played, where I'd rather put money into research enhancing buildings and research, and where my wars tend to start late.

VMXA pretty much likes to just all-war all the time and is a specialist Army creator, but his games involve virtually zero Diplomacy and very little in the way of role-play or non-violent victories.

VMXA doesn't always play always war, though he does seem to like always war a lot, and hardly ever plays peacefully. He has said as much before, as have others.

The reason these 3 posters and myself post so much is precisely because each of our differing techniques provides varied and interesting divergent results. If we all played the same this forum, and the game itself, would be like watching paint dry.

I don't always play with the same technique. My playstyle has varied a bit with the victory condition I've sought to win by.

At Cyc,

Here's a simple solution to your problem...

It should help both Buttercup and Spoonwood (along with the rest). Play a game together. Buttercup you start a game and post the 4000BC save here. Spoonwood, you pick up the save and tell Buttercup what to do. Buttercup, you do exactly what Spoonwood tells you to do in the game. Then post that turns save. Each turn you play, Buttercup, gets saved and posted here for Spoonwood to analyze and post instructions for. That way, everyone knows what the deal is and what's going to happen next. No more vague descriptions, no more grandious statements. It will all be on the table for anyone to see.

If you screw up the turn, Buttercup, or don't follow Spoonwood's instructions, then he wins the arguement. OTOH, if Spoonwood's instructions are followed to the letter and your oddities still emerge in his environment, this would lead to your proclamations being accurate and well worth a discussion.

What say you, gentlemen? Have the huevos to put your money where your mouth is? Screenshots may work occasionally, if you both agree, but the save should always be posted by Buttercup to verif the difficulty and other options available for that turn (such as trades, etc).

I have no problem with this, and actually like this idea. I encourage Buttercup to post a save, and get this started.

 

[Buttercup]

I'd rather chew on a dead cat...

 

[Cyc]

I'd rather chew on a dead cat...

I guess you'll be chewing on a dead cat and eating crow at the same time, Buttercup. But they say a balanced diet is good for you. You might want to throw in a stick of Buttercup to make it go down easier.

 

[Spoonwood]

I'd rather chew on a dead cat...

I honestly don't feel surprised. Don't kid yourself here Buttercup... not many around here take you all that seriously here... and not many have all that much reason to do so.

 

[Raliuven]

How about this for a compromise? Play a game with the preserve random seed off. Save every turn at the end of the turn so all you need to do is press enter to go to the next turn. That's the hard part. Then, whenever you believe the irritation program has occurred, post the save from the turn before and we can analyze the exact conditions under which you are claiming the irritation program kicks in. If the battle is run 100 times and falls within the range of the battle calculator, then nothing odd is occurring and your perception is a matter of selective memory or bias data. Let’s put this to the real test and get it done. You make the save, I’ll run the test – even better, multiple people can run the test and the data can be gathered.

Regarding the settling pattern of the AI: the problem is that you have no idea what the AI was after when it settled in the pattern. You've done no testing or theorized why the AI may have selected those placements over placements that appear more convenient to you. Your default is that they are programmed to irritate you (us). While they have been programmed, the fact that you are irritated by the programming does not mean that there is a lack of logic behind their movements.

The problem I have is that none of what you post is other than observation. There is little that is ‘logical’ in your skepticism. Every time someone makes a logical counterpoint you never actually address the issue, you simply produce more random data. I have explained why this does nothing to support your ‘logical skepticism’ and even demonstrated the point with a simple test. You have not tested anything and have no unbiased data. If you are arguing just to argue, fine. If you are actually trying to convince anyone, you need proof or at least a testable theory.

 

[Spoonwood]

How about this for a compromise? Play a game with the preserve random seed off. Save every turn at the end of the turn so all you need to do is press enter to go to the next turn. That's the hard part. Then, whenever you believe the irritation program has occurred, post the save from the turn before and we can analyze the exact conditions under which you are claiming the irritation program kicks in. If the battle is run 100 times and falls within the range of the battle calculator, then nothing odd is occurring and your perception is a matter of selective memory or bias data. Let’s put this to the real test and get it done. You make the save, I’ll run the test – even better, multiple people can run the test and the data can be gathered.

Regarding the settling pattern of the AI: the problem is that you have no idea what the AI was after when it settled in the pattern. You've done no testing or theorized why the AI may have selected those placements over placements that appear more convenient to you. Your default is that they are programmed to irritate you (us). While they have been programmed, the fact that you are irritated by the programming does not mean that there is a lack of logic behind their movements.

The problem I have is that none of what you post is other than observation. There is little that is ‘logical’ in your skepticism. Every time someone makes a logical counterpoint you never actually address the issue, you simply produce more random data. I have explained why this does nothing to support your ‘logical skepticism’ and even demonstrated the point with a simple test. You have not tested anything and have no unbiased data. If you are arguing just to argue, fine. If you are actually trying to convince anyone, you need proof or at least a testable theory.

Here here!

 

[Lone Wolf]

The AI knows where they all are. The AI knows where to drop Units. It knows every weak spot to the smallest detail.

The AI knows, or cheats, as you prefer, the whole map.

Some player had occasionally made use of it by deliberately exposing their inner cities in order to lure the attacking AI units inside their empire to their deaths. It's not the most reliable tactics, though.

 

[Buttercup]

I feel a lengthy response coming on!

According to the Probability Calculator posted by Spoonwood, the following is a statistical fact:

A Veteran Modern Armour Unit has a 99.8% chance of defeating a Veteran Spearman Fortified on Grassland. The Modern Armour has an attack of 24, the Spearman a defence of 2.7. In fact the Spearman has a 0.3% chance of victory (the fun of rounding bringing the combined total to 100.1%)

Oddly, by making the Modern Armour Elite we can finally achieve a 100% success statistic for the Modern Armour. So now we can see the range of results possible from the supposed Random Number Generator during a battle. The weakest possible Fortified Veteran defender, available within the first 20 turns of a game still has a defensive win value right to the end of the game fighting the strongest ground based Veteran Unit it's possible to make.

Once we have established that this is how the game operates it's battles then no testing is required, all one is doing by testing the battles is discovering what we know already - that *any* result is *possible*.

In fact, average results are likely even more misleading than *extreme* results because an average might only occur 10% of the time! This is basic maths which can be easily understood by calculating the average of the following numbers:

1,2,3,4,5,6,7,8,9,10

The average of these numbers is 5.5, a result with zero chance of happening. This is why the phrase "there are lies, damn lies... and statistics" is so popular, because statistics can often provide 'logical' impossibilities. In effect, by turning off the Random Seed Generator, one is simply exchanging one bizarre result for another over and over again.

Now, for the purposes of my theory, that the AI likes to Irritate the player and that the majority of this is a programmed intention, turning off the Random Seed Generator has no impact, because it's only the first result that is important. It's the first result which will tickle the players irritation cords. With the Random Seed Generator turned on there is nothing the player can do, but with it turned off the player still has the Irritation of having to reload and a general sense that they have *cheated*.

To conclude: without access to the full code of the game, it is impossible to say if, at specific moments in the game, the AI will make certain Units, either theirs or yours, superunits. Some games, when you reload, the game knows you've reloaded and will make the second attempt deliberately easier in order to simplify game progression, there is therefore a possibility that this game might use that method should you encounter a superunit during play, in that reloading simply removes the initial bonus.

Now lets completely chuck out this last idea. Let's assume no such superunits exist and it is all utterly random bizarre results. I am quite happy to do this. I am quite happy to accept that the game is just a non-stop tirade of unpredictable, unplanable, crap shoots. The reason I am more than happy to go this route is because it still fits my overall theory that the programme is operating under Programmed Irritation because having a combat system which is *designed* to give regular and expected bizzare results is, by definition, Irritation Programming for a game that is based on Strategy, Tactics and, most importantly, Planning.

 

[Spoonwood]

Once we have established that this is how the game operates it's battles then no testing is required, all one is doing by testing the battles is discovering what we know already - that *any* result is *possible*.

Not all results come as equally likely.

In fact, average results are likely even more misleading than *extreme* results because an average might only occur 10% of the time! This is basic maths which can be easily understood by calculating the average of the following numbers:

1,2,3,4,5,6,7,8,9,10

The average of these numbers is 5.5, a result with zero chance of happening. This is why the phrase "there are lies, damn lies... and statistics" is so popular, because statistics can often provide 'logical' impossibilities. In effect, by turning off the Random Seed Generator, one is simply exchanging one bizarre result for another over and over again.

That doesn't show such. First off, there exists no such thing as an average result in this example. Second, if there did exist such, then there would exist an extreme result of 9.9.

Now, for the purposes of my theory, that the AI likes to Irritate the player and that the majority of this is a programmed intention, turning off the Random Seed Generator has no impact, because it's only the first result that is important. It's the first result which will tickle the players irritation cords. With the Random Seed Generator turned on there is nothing the player can do, but with it turned off the player still has the Irritation of having to reload and a general sense that they have *cheated*.

Yet again you prefer to leave your ideas speculative at best, and not make them precise enough such that they can get meaningfully examined and tested.

 

[Bucephalus]

Here here!

Hmm, that would be 'hear, hear', Spoons.

 

[Raliuven]

I feel a lengthy response coming on!

According to the Probability Calculator posted by Spoonwood, the following is a statistical fact:

A Veteran Modern Armour Unit has a 99.8% chance of defeating a Veteran Spearman Fortified on Grassland. The Modern Armour has an attack of 24, the Spearman a defence of 2.7. In fact the Spearman has a 0.3% chance of victory (the fun of rounding bringing the combined total to 100.1%)

Oddly, by making the Modern Armour Elite we can finally achieve a 100% success statistic for the Modern Armour. So now we can see the range of results possible from the supposed Random Number Generator during a battle. The weakest possible Fortified Veteran defender, available within the first 20 turns of a game still has a defensive win value right to the end of the game fighting the strongest ground based Veteran Unit it's possible to make.

I don't see how this has any relevance. Since the unit has a defense statistic it has a chance, no matter how small, of winning. The fact that the combat calculator may round to 100% does not mean that the spear winning is impossible, just extremely unlikely.

Once we have established that this is how the game operates it's battles then no testing is required, all one is doing by testing the battles is discovering what we know already - that *any* result is *possible*.

. . .

Now, for the purposes of my theory, that the AI likes to Irritate the player and that the majority of this is a programmed intention, turning off the Random Seed Generator has no impact, because it's only the first result that is important. It's the first result which will tickle the players irritation cords. With the Random Seed Generator turned on there is nothing the player can do, but with it turned off the player still has the Irritation of having to reload and a general sense that they have *cheated*.

The issue with this is now you have two moving parts to your theory. If you are right that the RNG disengages at the Irritation point, it should produce an observable statistical difference. As I pointed out earlier, I would see a divergence of 25% or more from the expected results as suspect based on the 'controlled' test of the coin flip. Out of 1 million flips I never once saw a set of 100 flips outside of the 25% divergence. That is really not that large a margin to overcome if the Irritation program exists, right?

The second piece to this is that somehow the Irritation Program will now know that I have reloaded the scenario and it does not reoccur just to irritate me further. But if we speculate that the irritation factor is completely random in it application and so much so that it can't be identified from a simple deviation, then the answer seems that it is a simple deviation, not an Irritation Program. If it walks like a duck and quacks like a duck it is not an irritation program no matter how much walking quacking ducks irritate me.

In fact, average results are likely even more misleading than *extreme* results because an average might only occur 10% of the time! This is basic maths which can be easily understood by calculating the average of the following numbers:

1,2,3,4,5,6,7,8,9,10

The average of these numbers is 5.5, a result with zero chance of happening. This is why the phrase "there are lies, damn lies... and statistics" is so popular, because statistics can often provide 'logical' impossibilities. In effect, by turning off the Random Seed Generator, one is simply exchanging one bizarre result for another over and over again.

The problem is not the statistics in this example but your application of the math. Either you are tracking the results of the rolls or the combined total of the rolls. If you track the results of the rolls you should expect a graph that is basically a straight line. The longer you roll the d10 the more even the graph will appear in the long term. You have as much chance to roll a 1 as a 10 or any other number. Only in the short term will you see large variations. I feel like we've gone over this before. Limited datasets will lead to bias observations.

Now if you intend to roll the d10 and add the results and then average them, you will uncover a bell-shaped curve at the 5.5. Because you are adding the numbers and dividing them you are not beholden to receiving a number within your specified set. This is like saying I can't possibly roll a 5 and a 6 because there is no number 11 in the set.

But this is my point about throwing out more irrelevant information. Anyone that has a basic grasp of statistics or math can see where you went wrong. So either you don't know or you are purposely being misleading.

To conclude: without access to the full code of the game, it is impossible to say if, at specific moments in the game, the AI will make certain Units, either theirs or yours, superunits. Some games, when you reload, the game knows you've reloaded and will make the second attempt deliberately easier in order to simplify game progression, there is therefore a possibility that this game might use that method should you encounter a superunit during play, in that reloading simply removes the initial bonus.

Reloading does not always make things easier. If you do not have 'preserve random seed' turned on then you could (and eventually will) receive any result possible. I could win flawlessly the first time and then lose flawlessly the second time.

Now lets completely chuck out this last idea. Let's assume no such superunits exist and it is all utterly random bizarre results. I am quite happy to do this. I am quite happy to accept that the game is just a non-stop tirade of unpredictable, unplanable, crap shoots. The reason I am more than happy to go this route is because it still fits my overall theory that the programme is operating under Programmed Irritation because having a combat system which is *designed* to give regular and expected bizzare results is, by definition, Irritation Programming for a game that is based on Strategy, Tactics and, most importantly, Planning.

We are right back to where we began in post #2 with vmxa.

The rng has too great of an impact on CivIII and we tried to get them to tone that down. They wanted obsolete unit to still be viable. I felt that was fine, with small limitations, not massive swings.

By that I mean att 4 kills def 2, repeat in test and get defender winning with no damage. It can go from A wins cleanly to B wins cleanly. That is too much rng.

The answer to your Irritation is that the programmers wanted to make obsolete units viable therefore ridiculous results occur. Perhaps the problem is one of perception? Instead of thinking of it as a spear or a tank, think of it as a 1/2/1 vs a 16/8/1. Each unit has statistics so each unit has a chance of winning. Maybe the spearmen buried the tanks in a hill of rolling timber or dug a tank trap. That's all storyboarding and rationalizing the unit appearances. Otherwise it was a longshot win and it is a mathematical certainty that it will eventually happen.

You can be irritated with the programming but that does not make it Programmed Irritation. The fact is that these things can be predicted; you can plan for them or avoid the situations. That is how players win at high levels on a consistent basis. It may irritate you that you don’t understand them. You seem to become even more irritated when people try to explain it to you. What’s left? You have created a mythical program that seems to exist only because it is so elusive.

 

[Buttercup]

The fact is that these things can be predicted; you can plan for them or avoid the situations.

I'm sorry... what?

Because the player knows that the results are utterly unpredictable they can predict this and plan for it?

Regardless of who won what at what level, I have no idea how you can claim this with a straight face.

 

[Raliuven]

I'm sorry... what?

Because the player knows that the results are utterly unpredictable they can predict this and plan for it?

Regardless of who won what at what level, I have no idea how you can claim this with a straight face.

Ahem . . . artillery will help reduce the odds to a point that you will rarely experience this problem.

Otherwise I make sure to create units with a comfortable margin so that I can absorb the losses and still keep moving. Sometimes I miscalculate and there is one unit left in the captial. This is not an irritation program. This is my lack of planning.

And the results are not utterly unpredictable. Are you a Sith lord or something? This is not an 'all or nothing' scenerio. Most of the time things turn out they way you expect them to. In my test the combat calculator was off by 2.8%. How is that not predictable? While I can't predict every single combat individually, I know how things are going to turn out if I am sending tanks against their spearman. The only question is - how long is this going to take? The closer the odds are to 50%/50%, the more units I should produce to get the job done or I need to increase the odds of winning by, for example . . . ahem . . . using artillery.

Put this another way - on an average day I don't crash my car. Why do I carry insurance? Becaues one day of my life I will get into a crash and I will want insurance. Maybe it never happens. Maybe it happens more than once. The answer: have insurance all the time.

So . . . I plan as if there will be pitfalls along the way. If I stage an attack that could fail and it does I have no one and nothing to blame other than myself. I can accept my faults without needing to blame the program, a divine being, a magic fairy or the CIA. In the heat of the moment I may say that the game cheats but I don't honestly believe it. I know who is to blame. After my 5 year old tantrum I grow up and accept the consequences.

If it happens often enough I need to 1) reevaluate my tactics/playing style or 2) investigate why it is happening. Again, you may be right. But my magic fairy is still as valid as your Irritation Program and unless you prove it otherwise it is just as likely.

 

[Spoonwood]

Ahem . . . artillery will help reduce the odds to a point that you will rarely experience this problem.

Otherwise I make sure to create units with a comfortable margin so that I can absorb the losses and still keep moving. Sometimes I miscalculate and there is one unit left in the captial. This is not an irritation program. This is my lack of planning.

And the results are not utterly unpredictable. Are you a Sith lord or something? This is not an 'all or nothing' scenerio. Most of the time things turn out they way you expect them to. In my test the combat calculator was off by 2.8%. How is that not predictable? While I can't predict every single combat individually, I know how things are going to turn out if I am sending tanks against their spearman. The only question is - how long is this going to take? The closer the odds are to 50%/50%, the more units I should produce to get the job done or I need to increase the odds of winning by, for example . . . ahem . . . using artillery.

Put this another way - on an average day I don't crash my car. Why do I carry insurance? Becaues one day of my life I will get into a crash and I will want insurance. Maybe it never happens. Maybe it happens more than once. The answer: have insurance all the time.

So . . . I plan as if there will be pitfalls along the way. If I stage an attack that could fail and it does I have no one and nothing to blame other than myself. I can accept my faults without needing to blame the program, a divine being, a magic fairy or the CIA. In the heat of the moment I may say that the game cheats but I don't honestly believe it. I know who is to blame. After my 5 year old tantrum I grow up and accept the consequences.

If it happens often enough I need to 1) reevaluate my tactics/playing style or 2) investigate why it is happening. Again, you may be right. But my magic fairy is still as valid as your Irritation Program and unless you prove it otherwise it is just as likely.

I think using fast units here as much as possible can also help due to retreat factor (a retreat feels less irritating than a loss). Also, using armies.

 

[Buttercup]

I think using fast units here as much as possible can also help due to retreat factor (a retreat feels less irritating than a loss). Also, using armies.

Precisely!

A good recent example of this occurred in the game I have posted in the Hall of Fame thread.

Because I had the luxury of starting on a vast isolated island with decent land types for my Civ's Agricultural ability I had the luxury of going virtually the entire game without the slightest bit of Irritation. However, I eventually got to the point where I decided it might be wise to make contact with other Civs in order to acquire some luxuries and prepare for the Irritation of not having a Resource appear on my island.

I took over the closest island (opening the invasion with a healthy stack of combined forces), which turned out to be a small tundra dominated paltry 5 city severely backward Civilisation, then, when that was done and dusted and culturally safe, opened up relations with the rest of the world.

The rest of the world had not even got to Navigation by this point while I was nearly finishing Darwin's Voyage and heading towards Combustion, so I still had plenty of time before any potential invasion. But because this was the first time I had played a Large Archipelago Emperor game, I had no idea how many troops I would need to keep on my new island in order to properly defend it once the other Civs had got past Magnetism (open ocean Frigates and Galleys).

When the surprise attack did come (the only one of the game) I had 6 Cavalry Units on the island, but also a huge reserve from my main island that I could ship over in one turn (and ready to fight the next turn). The surprise attack consisted of 10 Units, quite a lot by AI standards and certainly the largest AI island invasion I had ever seen. Luckily I had sent over 4 additional Units and now had precisely 10 defenders (9 Cavalry and one Mechanised Infantry).

The AI's strongest defensive Unit was a couple of Veteran Musketmen, the rest was a couple of Cavalry, Medieval Infantry, a Knight, Swordsmen, Spearmen. Needless to say, I had the obligatory chuckle and endeavoured to attack their weak 10 with my strong 10 (having already bombed their fleet to the bottom of the ocean with my vast collection of Bombers I had positioned there the previous turn upon noticing a large fleet heading my way).

As per usual I got pretty crappy results considering the enemy was unfortified and had vastly (ahem) weaker defence than my attack. And by the time my 10 had finished shooting there was still 3 enemy Units on my Island with another load looking like it was going to land next turn. I shipped over 4 more Units and fortified my injured (and the two unscathed Units) in the towns closest to the invaders.

Undoubtedly, had I had 10 Artillery or two Armies waiting for them in addition to my Cavalry then the battle would have gone smoother, but I had not gained any Armies due to the fact that all I had invaded the entire game was one titchy island (which I really couldn't be bothered to use as an Army farm, I wanted it all Beaker producing ASAP) and as for Artillery, I was keeping my stock of those securely on my main island in case the AI was smart/Irritating enough to do a two-front invasion.

But as it turned out, just a healthy stack of Cavalry was all that was needed. Having 14 Cavalry produces pretty much the same result as having 10 Cavalry and 4 Artillery for the sake of island defence against weaker foes, and I would probably prefer 14 Cavalry because with the 'retreat if losing' possibility I am still left with another Unit to add numbers to a cities defence whereas with Artillery I am left with an extra stack to defend at the end of my turn (and this is about resource allocation to small island defences, not main island defences).


But on the subject of the Random Number Generator, the really great games, in my and many other gamer's eyes, do not permit irrational Irritating results when advanced units take on out-dated units and a modicum of tactical knowledge is applied. Civilisation II, Medieval Total War, Classic RPGs such as Balder's Gate, all follow the principle that, once one is significantly ahead in the advancement tree, one does indeed become 'invulnerable' to puny attacks/defences (or suffer such light losses it's barely worth registering as an issue). I can't help but think that the designers of this game broke this basic mechanic for a reason and that that reason had nothing to do with 'letting weak Civs have a chance of pulling through' as the Human player is unlikely to ever be in that position, at least not that many of the 'casual' gamers who buy the game and are not of the 'hardcore determined' type who will fight each and every map to the bitter end, ie: that's not where the money is. So, was it just an excuse for a broken mechanic or was it an excuse to permit Irritating battle results...? I guess, for the time being, we'll never know.

 

[Spoonwood]

Because I had the luxury of starting on a vast isolated island with decent land types for my Civ's Agricultural ability I had the luxury of going virtually the entire game without the slightest bit of Irritation.

If playing for the HoF, and you want to minimize the AI making demands/sneak attacking you, turn down the aggression level to "least" aggressive. I don't know how many upper level games, including Sid games, where the AI will happily leave me alone when I've played with the AIs at "least aggressive". Gifting them tech, and/or goods will also help with their attitude. If playing a space/diplomatic game for the HoF with scientific tribes, at age changes you'll want to do this anyways to pick up their free techs. Selling tech for gpt also can help, and even gifting extra luxuries can help to get more gpt/help pick up the AI research rate at higher levels.

But on the subject of the Random Number Generator, the really great games, in my and many other gamer's eyes, do not permit irrational Irritating results when advanced units take on out-dated units and a modicum of tactical knowledge is applied. Civilisation II, Medieval Total War, Classic RPGs such as Balder's Gate, all follow the principle that, once one is significantly ahead in the advancement tree, one does indeed become 'invulnerable' to puny attacks/defences (or suffer such light losses it's barely worth registering as an issue). I can't help but think that the designers of this game broke this basic mechanic for a reason and that that reason had nothing to do with 'letting weak Civs have a chance of pulling through' as the Human player is unlikely to ever be in that position, at least not that many of the 'casual' gamers who buy the game and are not of the 'hardcore determined' type who will fight each and every map to the bitter end, ie: that's not where the money is. So, was it just an excuse for a broken mechanic or was it an excuse to permit Irritating battle results...? I guess, for the time being, we'll never know.

We have no idea how many troops each unit the designers thought it might represent. The people with the superior technology don't always win in "real life". So, one can speculate that the answer to your false dichotomy comes as "neither"... they just wanted to ensure the possibility that results like the Battle of Isandlwana could happen.

On top of this, I do recall talking to someone several years ago, who told me that he could often win multiplayer civ II games online with inferior technology using Fudamentalism. So, I'm not so sure what you claim about civ II holds.

 

[Buttercup]

they just wanted to ensure the possibility that results like the Battle of Isandlwana could happen.

I seriously doubt they would design the entire game's mechanics based on the unusual and occasional moments in history when superior troops have been *heavily outnumbered*.

I can't believe no-one would have taken someone aside during this phase ans simply said "Yes dear, but we don't want to turn *every* battle into Isandlwana now, do we...?"

If anything, such 'excuses' are likely the result of people answering questions post-production and trying to justify a ludicrous mechanic.

Civ II is famous for not suffering from this mechanic, if you take a Tank troop across a collection of Spearman cities you will not lose any Tanks. Someone somewhere may have done something, but I seriously doubt it was Ancient versus Modern.