What happened to the AI?

[Loren]

Ulm, a difficult AI.

They *NEVER* did anything. Their army never increased, they never built anything, never took anything, never researched anything. The army size slowly wore away, they were seiged for a long time and finally vanished.

What in the world could have caused this? The only game mechanic that I can think of that could do this is unrest over 100 but how could that have happened for years?


(I also had a related experience not too long ago. I found unrest over 400 in a province. As far as I can tell it was nowhere near the battle lines--I got it from a dominion kill, not combat.)

[NTJedi]

The AI chooses random scales... and if it chooses bad scales it will flop like a fish on land. Even an AI on impossible will struggle if it chooses bad scales.

Use the map edit commands from the docs folder and edit the .map file to provide the AI some healthy scales.

[dirtywick]

A random event could have called knights to their home province, and they were sieged and couldn't break it. I've seen that before.

[kasnavada]

(I also had a related experience not too long ago. I found unrest over 400 in a province. As far as I can tell it was nowhere near the battle lines--I got it from a dominion kill, not combat.)



A possibility is the villain event where it stays hidden and raise unrests. The AI probably didn't think of patrolling.

[PvK]

Yep if the AI gets besieged by strong independents before it expands, it can get stopped.

[Meglobob]

dirtywick said:
A random event could have called knights to their home province, and they were sieged and couldn't break it. I've seen that before.



I have seen this happen to human players in MP games. http://forum.shrapnelgames.com/images/smilies/tongue.gif

As long as its not me, you after laugh... http://forum.shrapnelgames.com/images/smilies/biggrin.gif

[Aezeal]

Hmm I had 2 unrest events in 5 turns in a MP game.. when the first was patrolled away (which hampered expansion) and the start army was moving a second happened.. I can tell you it doesn't make for a nice start.

[Lingchih]

I mostly do not bother with the AI events. Screw them, I have more important things to attend to, like conquering Indys.

[Aezeal]

hmm well unrest 30 on first turn severly hinders taking indies since my income was ruined I patrolled it out in 2 turns I think and then left. Then my army was gone and I got something like it again.. low income and then I had to lower income even further to get it away.

I think it would be nice if random events didn't happen in the first 5 turns. A huge bonus or a terrible setback in those critical turns makes the game a bit too random.

[Endoperez]

Aezeal said:
I think it would be nice if random events didn't happen in the first 5 turns. A huge bonus or a terrible setback in those critical turns makes the game a bit too random.



Yeah, it sure be nice if we had 120 points more in pretender creation. No one takes Luck, any way, or wants to take it, or needs the good events in the early game, or wants to pay off the small risk of ruined game with design points. http://forum.shrapnelgames.com/images/smilies/rolleyes.gif

I know that you can suffer from bad luck even with Luck scale, but the risk is so small it doesn't matter in the long run, IMO. Many players already consider Order/Misfortune superior to Luck, without Turmoil - more benefit with less points spents, in other words. Unless luck's importance in late game is improved, I don't want to see it's importance in early game made any smaller.

[MaxWilson]

Aezeal said:
I think it would be nice if random events didn't happen in the first 5 turns. A huge bonus or a terrible setback in those critical turns makes the game a bit too random.



I don't see that as terribly more uncommon or more of a setback than moving an expansion army into a province at the same time someone else moves into it from the other side. I'd rather lose a turn of capital income than an army that I spent three turns building.

-Max

[MaxWilson]

Endoperez said:

Aezeal said:
I think it would be nice if random events didn't happen in the first 5 turns. A huge bonus or a terrible setback in those critical turns makes the game a bit too random.



Yeah, it sure be nice if we had 120 points more in pretender creation. No one takes Luck, any way, or wants to take it, or needs the good events in the early game, or wants to pay off the small risk of ruined game with design points. http://forum.shrapnelgames.com/images/smilies/rolleyes.gif

I know that you can suffer from bad luck even with Luck scale, but the risk is so small it doesn't matter in the long run, IMO.



I agree with both. If Luck had a greater impact on the early game, e.g. Luck-3 means there is ZERO probability of getting bad events in the first 5-10 turns, I might be at least somewhat tempted to take it. Or at least if I did take it, I wouldn't have my lab burn down anyway and feel cheated on my 120 design points.

-Max

[Endoperez]

MaxWilson said:
I agree with both. If Luck had a greater impact on the early game, e.g. Luck-3 means there is ZERO probability of getting bad events in the first 5-10 turns, I might be at least somewhat tempted to take it. Or at least if I did take it, I wouldn't have my lab burn down anyway and feel cheated on my 120 design points.




Has that actually happened to you? Wow, that's hard.

[MaxWilson]

I don't think so, I believe I heard it on the forums somewhere. I got turned off Luck pretty quickly when I first started and haven't played much with it since.

-Max

[Loren]

dirtywick said:
A random event could have called knights to their home province, and they were sieged and couldn't break it. I've seen that before.



Except they owned a province for a long time. Something of this sort was my first thought but it's not consistent with the graphs--they *DID* own their province for years.

[Endoperez]

They owned the province because they had the castle, I think. You can't win wars without storming all the forts you are sieging.

[JimMorrison]

I was under the impression that the score graphs will show you as losing a province, when you are besieged (or gaining one when you initiate a siege. Could be wrong, but I'd swear that's the behavior I observed.

[Endoperez]

JimMorrison said:
I was under the impression that the score graphs will show you as losing a province, when you are besieged (or gaining one when you initiate a siege. Could be wrong, but I'd swear that's the behavior I observed.



Could be. I'm not that clear on it, to tell the truth. I mainly play SP, and rarely long enough to conquer an AI's castle.

[Edratman]

I suspect that the AI nation got an invasion on the first or second turn. I've seen it two or three times to the AI and had it happen to me once. When it happened to me I had bad settings. From that occurance, I guess that the same thing happened to the AI. I remember one time when the AI was stuck at one province they were under the knight siege when I got to the province.

Both the knight and barbarian events are too strong for starting armies to relieve the siege.

I like the chance that bad things can happen on the first turn. Unpredictability is one of the charms of the game. Pretty much everyone has learned to recruit a mage the first turn with bad scales because of the lab destroyed event.

[VedalkenBear]

I'll note that I quite often take Luck-3. I also play with CBM routinely, which IIRC increases the probability of events being 'good' with Luck-3 to 95% (from 90%). And I _still_ have gotten labs burned down or temples hit by earthquakes in the first 3 turns.

However, since I play SP, I can simply start a new game with the same parameters if I don't feel like suffering from that handicap.

I also play nations with fortunetellers a lot to more or less guarantee good events in those provinces.

Yeah, I don't like bad events. However, barring an option to out-and-out remove all random events from the game (with the attendant nulling of the Luck scale), I have no issue with the current system.

[Loren]

Edratman said:
I suspect that the AI nation got an invasion on the first or second turn. I've seen it two or three times to the AI and had it happen to me once. When it happened to me I had bad settings. From that occurance, I guess that the same thing happened to the AI. I remember one time when the AI was stuck at one province they were under the knight siege when I got to the province.

Both the knight and barbarian events are too strong for starting armies to relieve the siege.

I like the chance that bad things can happen on the first turn. Unpredictability is one of the charms of the game. Pretty much everyone has learned to recruit a mage the first turn with bad scales because of the lab destroyed event.



But that would have shown as the loss of the province early on. That's not what happened, though.

[Edratman]

I do not remember exactly what the score graph showed when I encountered it. I know the nation was low, but the graphs are not large enough to read much better than that.

And when a castle is besieged, ownership is not really decided in some game terms. Although the beseiger is nominally the owner, neither side can recruit in the province. The castle still owns the lab and can research and summon. I don't know who gets the gems, but I think it goes to the castle. The besieger gets the taxes. So it is possible for the graph not to reflect besieging, specially when the siege drops the number of provinces to zero, but the god is still alive.

[PvK]

Ok, so I ran a test game as LA C'tis with Turmoil-3, Unluck-3, Growth-2, Heat-3, Prod-3, Magic-3, just staying in my home capital, and here's what happened.

All went well enough until brigands started hiding out in the province, which turned up the unrest, and then lots of events started wiping out or driving away my population, cursing my pretender and troops, causing my income to vanish, etc. Two barbarian hordes tried to take my province but were defeated by my provincial defense force. I tried lowering taxes and patrolling, but my economy collapsed and my population plummeted.

Finally, in the last month of year 2, an earthquake hit and trogs swarmed out and seiged my castle.

The provinces graph DOES go down when a province is seiged.

However, I had no income and no lab and no way to pay to rebuild it, or to recruit troops, for many months before that happened, due to having only one province and unluck/turmoil/unrest/brigands in my capital. I have 4910 population left in my capital, no temple, no lab. Unrest is 94, gold income is 2 with tax at 50%. I do happen to have a large undead army which I could use to break out with, but not every nation would have any such thing.

[Aezeal]

Endo: I have nothing against bad luck and it's events in most of the game, but I do think the impact it can have on the first turns is too much so I think it should not happen the first turns or at least not happen to the starting prov the first 3-5 turns.. after that luck will still be worth it's points without the option of getting your game ruined by misfortune (or even with luck as mentioned) and in MP some events would ruin your game in most cases

[Loren]

Edratman said:
I do not remember exactly what the score graph showed when I encountered it. I know the nation was low, but the graphs are not large enough to read much better than that.

And when a castle is besieged, ownership is not really decided in some game terms. Although the beseiger is nominally the owner, neither side can recruit in the province. The castle still owns the lab and can research and summon. I don't know who gets the gems, but I think it goes to the castle. The besieger gets the taxes. So it is possible for the graph not to reflect besieging, specially when the siege drops the number of provinces to zero, but the god is still alive.



But going from 1 to zero should have been clear in the graph. It did happen but long afterwards.

[silhouette]

Loren said:
But going from 1 to zero should have been clear in the graph. It did happen but long afterwards.


Probably just a coincidence of actions that makes it look like that. For example, the AI sent its initial army and took a capital-adjacent province with very heavy losses, and on the same turn a indep. event sieged their capital. That could leave them with 1 province still, but very low income and low ability to recruit for a looong time, making them easy pickings later. Did they finally die from another AI, or from dominion death, or ?

Sill

[Aezeal]

does the besieger get the taxes?

[Sombre]

Yes.

[MaxWilson]

Mostly. The besieger gets the base taxes, and the besieged gets the admin bonus taxes. So in a 200 gold province with an Admin 40 fort, you'd normally get 240 gold per turn. If you're besieged, you'll get 40 gold and he'll get 200. So Saith KO.

-Max

[Edratman]

Pvk has proved that the graph goes down when besieged so I have no realistic explanation for your observations.

[Loren]

silhouette said:

Loren said:
But going from 1 to zero should have been clear in the graph. It did happen but long afterwards.


Probably just a coincidence of actions that makes it look like that. For example, the AI sent its initial army and took a capital-adjacent province with very heavy losses, and on the same turn a indep. event sieged their capital. That could leave them with 1 province still, but very low income and low ability to recruit for a looong time, making them easy pickings later. Did they finally die from another AI, or from dominion death, or ?

Sill



Yeah, that scenario is possible.

I have no idea what they died of, I never saw them.

[PvK]

My example also showed that you can become impotent due to bad luck + turmoil + unrest via hiding independent brigands, if you don't attack and your lab is destroyed. No income plus the AI being unwilling to attack with a weak army could have him stuck with no income.