A New Dawn Bug Reports and Feedback

Status
Not open for further replies.
Civ4ScreenShot0056.JPG
Incorrect name in event: Organized Religion
I also wonder, if the event choice is available, if the required building (The Oracle) is obsolete...
 
Another thing I noted in that game (so still rev 774):
The combat AI has become HORRENDOUS!
It will recklessly attack my units, even at the worst possible odds, and without anything to gain (Examples: A javeliner attacking a stack of axmen in a forest. A archer at 10% life still attacking. The best defender (archer with 4 promotions) leaving the city to chase some units. And so on.)
I was able to defeat egypt easily, because it kept attacking my axmen with their archers, until there were none left.
I am aware that the AI before was too defensive and would not attack in situation where it would be the right thing to do. That could be exploited (I know I did), but it took effort.
Now the AI just slits it`s own throat for you.
 
My test.

One starting save - Huge map, Blitz speed, 38 civs. Vassal states off, AI Autoplay.
After 420 turns ( from 600):

rev765: 33 civs alive
rev777: 14 civs alive

My conclusion: now smaller civs have no chance to survive. If you have not get many cities on start - you doomed.

Upd: have read previous post by Zorn. May be the reason is combat AI, yes.
 
There hasn't been any changes to the combat AI...

If you see a situation where you believe the AI acts oddly, upload a save. I can't debug complex cases like the AI from reports alone.
 
There hasn't been any changes to the combat AI...

If you see a situation where you believe the AI acts oddly, upload a save. I can't debug complex cases like the AI from reports alone.
I have taken a note that different AI operate their units differently when at war.

For the most part when the AI declare's war, they (might) come up to you with an initial (relatively small) stack, usually intent on capturing a city. At that point I am usually able to back up said city with enough defending troops. This results in the enemy stack wondering around your territory for a bit and eventually fortifying itself somewhere in your territory for durations that can extend beyond an entire era. No backup units will ever be sent during this entire period. This despite the AI not being at war with anyone else but you and and bordering your civ directly (thus no reason to not send extra troops). Thus their stack will just sit there until you accumulate enough forces to destroy it. The latest AI this happened to me with was Washington. I will upload a save of that game if I find it lying around.

On the other hand certain (fewer) AI behave noticeably different when at war. I played a game where I was on an island with a single neighbor, Ghengis Khan. I replayed that game about 5-6 times and each and every time we declared war on each other and each and every time, he would send formidable stacks that would overcome even my best defenses. More importantly he would send BACKUP stacks to keep the pressure on my defenses - it got to a point where I had to have one stack to attack and another just to defend. My economy went to hell and even though I was victorious the war took its toll. Again this kind of AI warfare behavior manifested itself every one of those 5-6 games with Ghengis Khan.

My conclusion is therefore that if the AI have differences in how they are programmed to behave as far as managing wars are concerned, that those differences should either be reduced or removed completely and that every AI when at war, operates his troops like Ghengis Khan does. That would give me a good reason to fear enemies that have stronger armies than myself rather then laugh at their inability to manage them. In effect it would make the entire game more challenging.
 
@Arkatakor,

Leader Traits are a part of this.

I don't recall if 45* incorporated any of C2C's LT's programming or if we still use stock BtS. But Traits and Flavors are a part of this observation. Every time I get Shaka in a game I can expect the same behaviour as you got from GK. Montezuma and to a lesser degree Hyanc and Pacal will do the same. So will Peter of Russia. There are a couple of more of the War Monger types, and I would dare say they all have a leader trait in common.

JosEPh
 
@Arkatakor,

Leader Traits are a part of this.

I don't recall if 45* incorporated any of C2C's LT's programming or if we still use stock BtS. But Traits and Flavors are a part of this observation. Every time I get Shaka in a game I can expect the same behaviour as you got from GK. Montezuma and to a lesser degree Hyanc and Pacal will do the same. So will Peter of Russia. There are a couple of more of the War Monger types, and I would dare say they all have a leader trait in common.

JosEPh
The way I see it is that if a method could be found to make all AI's aggressive when fighting a war (which they should be since they are at war) whilst maintaining their individual personalities whilst not at war that would be ideal.
 
The way I see it is that if a method could be found to make all AI's aggressive when fighting a war (which they should be since they are at war) whilst maintaining their individual personalities whilst not at war that would be ideal.

I don't think most players want this, I get the impression that AI personalities are a feature that is rather popular. (I mean otherwise we might as well be "Red Player", "Blue Player", etc...)

Most of the XML that differentiates leaders is related to war or peace. If they all were the same, most of the AI would be identical.
 
45°38'N-13°47'E;13376378 said:
And no, I've left out every change regarding leaders trait when I imported code from C2C.

As someone who pays no attention to C2C development, what have they changed?
 
As someone who pays no attention to C2C development, what have they changed?
There was an entire revolution about leaders traits, as far as I recall, it was something about leaders gaining or changing traits during the game. I've discussed a similar feature with you long before C2C even started but at the time you convinced me it was kind of cheating. :)
 
45°38'N-13°47'E;13376475 said:
There was an entire revolution about leaders traits, as far as I recall, it was something about leaders gaining or changing traits during the game. I've discussed a similar feature with you long before C2C even started but at the time you convinced me it was kind of cheating. :)

I vaguely recall a mod idea about dynamic leader traits, I didn't realize that is what you were referring to. I do agree with my previous conclusion though. :p
 
FYI,
Sgtslick is the author of the Leader Traits of C2C. He's proposed many of the changes from that change there for here in the past, repeatedly in fact.

Not sure you would call them Dynamic or not. Possibly?

JosEPh
 
FYI,
Sgtslick is the author of the Leader Traits of C2C. He's proposed many of the changes from that change there for here in the past, repeatedly in fact.

Care to link me to any of his proposal posts?

Not sure you would call them Dynamic or not. Possibly?

Welll..... Dynamic is opposed to "static" mechanics that do not change throughout the game. With traits, once you pick a leader with a trait, it never changes, making it static throughout the game. Dynamic would mean that it has the possibility to change.
 
Well, I would welcome the idea of "leaders gaining Xp and going up levels" :)
...but I guess you are not going to add that kind of feature :sad:

I think scoreboard is good as any for showing LH "XP"? However, if you mean something else, I agree that it's doubtful this will be implemented.
 
I don't think most players want this, I get the impression that AI personalities are a feature that is rather popular. (I mean otherwise we might as well be "Red Player", "Blue Player", etc...)

Most of the XML that differentiates leaders is related to war or peace. If they all were the same, most of the AI would be identical.
What I was getting at was that different leaders should continue to have different behavioral personalities when it comes to diplomacy (like Tokugawa never trading world map, Mansa Musa being a tech whore trading techs with anyone, etc). What I was trying to emphasize is that AI should behave more or less the same when managing troops (ala Ghengis Khan) in a war situation once war is declared.

Not making a change in how most of the AI manage their troops makes at least 50% of them (if not more) virtually non viable in a war situation (as exemplified in my previous post) which is a huge flaw in the game if you think about it.
 
Care to link me to any of his proposal posts?



Welll..... Dynamic is opposed to "static" mechanics that do not change throughout the game. With traits, once you pick a leader with a trait, it never changes, making it static throughout the game. Dynamic would mean that it has the possibility to change.

Here's 2 3...so far. http://forums.civfanatics.com/showpost.php?p=11784586&postcount=172 http://forums.civfanatics.com/showpost.php?p=11974973&postcount=264 http://forums.civfanatics.com/showpost.php?p=12003863&postcount=337 all in same thread.

Problem is his ideas and then work on them was spread all over in various threads. Then ls612 and him had some difference of opinions and ls612 built his own set of Traits and made it an Option in game setup. And then T-brd did another slightly different one so that C2C has 3 or 4 Options for changing how traits for leaders work. And this was all the way back in v 23 and 24 of C2C. A lot longer than I thought.

I'll link ls612's thread too. When I find it again. :p

JosEPh
 
Status
Not open for further replies.
Back
Top Bottom