Oh, that explains a lot, thank you! Does this mean I'll never get this tasty wheat tile upper the London?
[BTS] Deity One City Challenge on a Huge map with 18 civilizations
- Thread starter valergrad
- Start date
Oh, that explains a lot, thank you! Does this mean I'll never get this tasty wheat tile upper the London?
T4.
Paris border expands and we met Caesar because of that.
We have signed off peace treaty with Frederick - hopefully nobody objects against it.
Worker starts it's job on wheat's tile.
Warrior is now where in reality is Slovenia.
Also this move I am starting project for understanding real peaceWeights of all AI's. This is hidden parameters that influence the AI's attitude against each other ( but not with human player ). May be this is not important to know, but may be it will help later if we will try to win some apostolic or diplomatic voting.
How this works? Each AI gets iBasePeaceWeight from XML file and at the beginning of the game each AI also add to this value random number - 0, 1 or 2.
For example, Lisa has 9 and Frederick 8. This means that in concrete game it could be 11 for Lisa and 8 for Frederick. Or 8 for Lisa and 10 for Frederick.
This hidden attitude they get will be 4 in case of this parameter equal, and -1 for each difference between them.
Another hidden parameters are:
-1 - it is just start for all AI's at Deity level.
IbaseAttitude from XML - it is 1 for both Lisa and Freddy.
iWarmongerRespect from XML - in attitude it will be max for both of them. It is 0 for both Lisa and Freddy, so it will be 0.
+1 if they both in the bottom half of the scoreboard - this probably doesn't apply to them for now as Lisa is first in visible part of scoreboard, I don't believe she can be in the bottom half in the overall scoreboard.
And iWorseRankDiffrenceChange/iBetterRankDifferenceChange - parameters from XML that can get some change if one leader is much higher/lower in the scoreboard than the another. But for Frederick and Lisa this parameters are not big, so this is also should be 0.
So, now we see that Lisa is pleased with Freddi and Freddi is pleased with Lisa ( >= 3). Visible relations between them is 0, hidden is 0 as shows my calculations before. This means that this peaceWeight-based parameter give them at least +3 in attitude. So we can say that baseWeight between them is equal or differ not more than 1. We can't say what is it exactly for now, but we definitely can exclude some combinations ( to be more accurate - 3 combinations out of 9 ). I hope that if I'll analyse all pairs every move, I'll be able to calculate or baseWeights for all leaders, and hence will be able to calculate relations between them with 100% accuracy.
It will be just an interesting puzzle to get all this values.
( the same analysis of Alexander - Caesar relationships gives nothing for now)
Paris border expands and we met Caesar because of that.
We have signed off peace treaty with Frederick - hopefully nobody objects against it.
Worker starts it's job on wheat's tile.
Warrior is now where in reality is Slovenia.
Also this move I am starting project for understanding real peaceWeights of all AI's. This is hidden parameters that influence the AI's attitude against each other ( but not with human player ). May be this is not important to know, but may be it will help later if we will try to win some apostolic or diplomatic voting.
How this works? Each AI gets iBasePeaceWeight from XML file and at the beginning of the game each AI also add to this value random number - 0, 1 or 2.
For example, Lisa has 9 and Frederick 8. This means that in concrete game it could be 11 for Lisa and 8 for Frederick. Or 8 for Lisa and 10 for Frederick.
This hidden attitude they get will be 4 in case of this parameter equal, and -1 for each difference between them.
Another hidden parameters are:
-1 - it is just start for all AI's at Deity level.
IbaseAttitude from XML - it is 1 for both Lisa and Freddy.
iWarmongerRespect from XML - in attitude it will be max for both of them. It is 0 for both Lisa and Freddy, so it will be 0.
+1 if they both in the bottom half of the scoreboard - this probably doesn't apply to them for now as Lisa is first in visible part of scoreboard, I don't believe she can be in the bottom half in the overall scoreboard.
And iWorseRankDiffrenceChange/iBetterRankDifferenceChange - parameters from XML that can get some change if one leader is much higher/lower in the scoreboard than the another. But for Frederick and Lisa this parameters are not big, so this is also should be 0.
So, now we see that Lisa is pleased with Freddi and Freddi is pleased with Lisa ( >= 3). Visible relations between them is 0, hidden is 0 as shows my calculations before. This means that this peaceWeight-based parameter give them at least +3 in attitude. So we can say that baseWeight between them is equal or differ not more than 1. We can't say what is it exactly for now, but we definitely can exclude some combinations ( to be more accurate - 3 combinations out of 9 ). I hope that if I'll analyse all pairs every move, I'll be able to calculate or baseWeights for all leaders, and hence will be able to calculate relations between them with 100% accuracy.
It will be just an interesting puzzle to get all this values.
( the same analysis of Alexander - Caesar relationships gives nothing for now)
NothingBesideRemains
Warlord
- Joined
- May 14, 2021
- Messages
- 113
@CarpoolKaraoke
I have tried to learn vassal mechanics a few times but the numbers havn't stuck in my brain yet.
But I think you need twice the number of land tiles compared to the AI you want to vassal.
I was helped with having alot of land tiles with Paris, but in reality you need to smash the AI down into a little coastal city to have a chance.
Not an easy task! And for subsequent vassals your first vassal counts as half, so it's very frustrating when you are clearing a enemy city turn after turn and the moron you are trying to help are just moving his doomstack back and forth, back and forth.
I think the key in a OCC military victory like this is you need to vassal at least one civ super early - early enough that they have a chance to actually expand after you vassal them.
Needing 2x the land / pop as your first target means likely pinning them to 1 city before they cap. If you could Oracle feudalism that might be possible (this can work on a hand picked normal map, but probably not here - doubtful even if you get some nice techs from huts! Maybe if an AI alsp got early alpha and you were able to trade for Monarchy or even writing?). If you declare early war against that civ to choke them before you get feudalism, that could also help.
Only chance I see is something like going France, choking both Fred and Izzy (need to declare war on her before the world turns Buddhist) Oracling Feud (if that is absolutely impossible, oracling probably like metal casting and trading it for Monarchy, getting in a cheeky forge), vassling Fred, then vassling Liz (can't vassel her first as her single holy city will have more land than yours), then quickly vassling London and any other Civ with limited land.
Note that such a quick push likely means the only early wonder you can get is Oracle (maybe SH if you want the relevant quick culture and early prophet).
If those first few vassals have the ability to do any teching, then you might be able to do a slow crawl through Europe with catapults then trebs then cannons (gotta limit your losses as a OCC!).
No matter how you slice it, at some point the world gets rifles, and hopefully by that point you have enough vassals to press on regardless.
Never tried this on this map, but I doubt it's possible.
But if they expand on more than 100% of your land - aren't they will immediately free themselves? Sorry if this is a stupid question, I haven't a lot of experience in this vassal games. I just have got impression that if civs land become larger than it was during vassalization AND more than 100% of master civs land - it will free itself.
Last edited:
As there is no change in strategy or not yet something to think about or discuss, I'll continue and make one more move.
T5.
No events for this move, worker continue it's march to Middle East ( where he will meet Saladin/Cyrus and probably Hatshepsut ).
Freddi met Caesar and Alexander. Analyzing their relationship gives me not a lot of information, just understanding that Freddi peaceWeight - Alexander peaceWeight <= 8.
Right now we can predict only that Alexander will be probably the enemy of all civs in this area, because of this peaceWeight parameter.
Espionage: Lisa 21/0, Freddi 28/16, Caesar 10/0, Alexander 12/0.
Took screenshot of demo just in case. May be somebody can get some information from this.
T5.
No events for this move, worker continue it's march to Middle East ( where he will meet Saladin/Cyrus and probably Hatshepsut ).
Freddi met Caesar and Alexander. Analyzing their relationship gives me not a lot of information, just understanding that Freddi peaceWeight - Alexander peaceWeight <= 8.
Right now we can predict only that Alexander will be probably the enemy of all civs in this area, because of this peaceWeight parameter.
Espionage: Lisa 21/0, Freddi 28/16, Caesar 10/0, Alexander 12/0.
Took screenshot of demo just in case. May be somebody can get some information from this.
Attachments
NothingBesideRemains
Warlord
- Joined
- May 14, 2021
- Messages
- 113
To my understanding, there are 2 types of vassals - war vassals (they capitulate as part of a peace treaty to end the war) and peace vassals (they offer to be your vassal when you are at peace amd if they really like you). These types have different attributes. In particular, peace vassals can break away if they become too large relative to your size, whereas war vassals can break away if they become too small relative to their original size when they capitulated.
(If Im wrong on any of this please correct me, Im no expert).
At least I can't figure out how it could even be theoretically possible.
It should be possible to get conquest on marathon speed I think.
CarpoolKaraoke
Not James Corden
- Joined
- May 2, 2021
- Messages
- 257
@CarpoolKaraoke
I have tried to learn vassal mechanics a few times but the numbers havn't stuck in my brain yet.
But I think you need twice the number of land tiles compared to the AI you want to vassal.
I was helped with having alot of land tiles with Paris, but in reality you need to smash the AI down into a little coastal city to have a chance.
Not an easy task! And for subsequent vassals your first vassal counts as half, so it's very frustrating when you are clearing a enemy city turn after turn and the moron you are trying to help are just moving his doomstack back and forth, back and forth.
More land? Hmm......
Spoiler T84 revised :
Could also turn off tech-trading with the strategy above. Marathon speed would probably help. I don't have that much time at the moment, maybe someone else is willing to give it a shot?
@CarpoolKaraoke Hah, I did start a Julius wheat plant late last night and it looks somewhat promising for me too.
In a rush now, have to write more later.
In a rush now, have to write more later.
@CarpoolKaraoke Hah, I did start a Julius wheat plant late last night and it looks somewhat promising for me too.
In a rush now, have to write more later.
Are you playing this with save-loads abuse?
Last edited:
Marathon speed make some things easier... But even in marathon you need to understand what to do with Montezuma's hordes in America.
Not save-load really, but I do some worldbuilder trickery. Far from honest clean play.Are you playing this with save-loads abuse?
For example, to ensure that I raze Fredrich city and steal his settler as a second worker, something thats about 5% odds or so.
Instead of save-loading a gadzillion times, I just super-promote the warrior for that turn.
Couldn't be bothered by huts though, so just ignored them, so no BW/masonry huts for me.
Spoiler :
@CarpoolKaraoke Hah, I did start a Julius wheat plant late last night and it looks somewhat promising for me too.
In a rush now, have to write more later.
Scrap that, it doesn't look promising!
T6.
Met Isabella at last. Now I know everybody in Europe Except Russia of course.
Frederick had last move 6 production ( in espionage info), this move 7. This means that it is not settler or worker, so probably a warrior, this is the only option.
As it has now 31 score, it does mean probably Berlin grown to size 2.
Met Isabella at last. Now I know everybody in Europe
Except Russia of course.Frederick had last move 6 production ( in espionage info), this move 7. This means that it is not settler or worker, so probably a warrior, this is the only option.
As it has now 31 score, it does mean probably Berlin grown to size 2.
Attachments
At t7 I see some strange thing:
Frederick got his worst enemy. It is Alexander. But if we check his attitude to Alexander - it is cautious.
This doesn't have sense at all!
First of all - he is annoyed with me, so I should be his worst enemy.
Second - based on code that we see on:
only persons with attitude less than ATTITUDE_CAUTIOUS should be considered as worst enemy.
The only hypotesis that I can imagine: that something with my interface and it shows me wrong information. For example, Alexander is annoyed with Frederick, so based on the code, it actually Alexander should has his worst enemy as Frederick, not counterwise!
If somebody can check this save - can he see if he gets the same worst_enemy behaviour?
Frederick got his worst enemy. It is Alexander. But if we check his attitude to Alexander - it is cautious.
This doesn't have sense at all!
First of all - he is annoyed with me, so I should be his worst enemy.
Second - based on code that we see on:
Code:
void CvTeamAI::AI_updateWorstEnemy()
{
PROFILE_FUNC();
TeamTypes eBestTeam = NO_TEAM;
int iBestValue = MAX_INT;
for (int iI = 0; iI < MAX_CIV_TEAMS; iI++)
{
TeamTypes eLoopTeam = (TeamTypes) iI;
CvTeam& kLoopTeam = GET_TEAM(eLoopTeam);
if (kLoopTeam.isAlive())
{
if (iI != getID() && !kLoopTeam.isVassal(getID()))
{
if (isHasMet(eLoopTeam))
{
if (AI_getAttitude(eLoopTeam) < ATTITUDE_CAUTIOUS)
{
int iValue = AI_getAttitudeVal(eLoopTeam);
if (iValue < iBestValue)
{
iBestValue = iValue;
eBestTeam = eLoopTeam;
}
}
}
}
}
}
m_eWorstEnemy = eBestTeam;
}only persons with attitude less than ATTITUDE_CAUTIOUS should be considered as worst enemy.
The only hypotesis that I can imagine: that something with my interface and it shows me wrong information. For example, Alexander is annoyed with Frederick, so based on the code, it actually Alexander should has his worst enemy as Frederick, not counterwise!
If somebody can check this save - can he see if he gets the same worst_enemy behaviour?
Attachments
CarpoolKaraoke
Not James Corden
- Joined
- May 2, 2021
- Messages
- 257
Edit: Removing from thread.
Last edited:
T7.
Warrior met Cathy. Oh, Cathy, whimsy russian girl, I hope you will not be too demanding to me!
Continue to fill my excel spreadsheet with possible peaceWeights for leaders, don't have much information for now.
The thing that i don't know how to interpret - it is this "worst enemy", i'll create a separate topic for it in case if somebody here knows how it actually works ( I am confused that it doesn't work as it should be looking at the code).
Frederick has 8/9 at warrior - next move he will finish it. Don't forget to check when he will build a new worker - probably I am going to steal it too. Not decided yet, I definitely want 2 workers, but I don't want -6 in relationship.
Saving espionage points for understanding who is spending them to me: Lisa - 27, Freddi - 36, Cathy - 0, Isabella - 2, Caesar - 21, Alexander - 12.
Warrior met Cathy. Oh, Cathy, whimsy russian girl, I hope you will not be too demanding to me!
Continue to fill my excel spreadsheet with possible peaceWeights for leaders, don't have much information for now.
The thing that i don't know how to interpret - it is this "worst enemy", i'll create a separate topic for it in case if somebody here knows how it actually works ( I am confused that it doesn't work as it should be looking at the code).
Frederick has 8/9 at warrior - next move he will finish it. Don't forget to check when he will build a new worker - probably I am going to steal it too. Not decided yet, I definitely want 2 workers, but I don't want -6 in relationship.
Saving espionage points for understanding who is spending them to me: Lisa - 27, Freddi - 36, Cathy - 0, Isabella - 2, Caesar - 21, Alexander - 12.
T8.
Warrior now reached Bosphorus. Next move he will in Asia.
Worker finished wheat farm, switched to this tile. I want to grow as fast as possible before masonry, and then I need to find a balance between grow, hammers and commerce.
Frederick changed his opinion about his worst enemy. Now it is me again. May be this "worst enemy logic" doesn't work at the first move after two civs meet each other? Let's see.
Who is spending espionage points on me? Lisa - 3, Frederick - 4, Isabella - 2, Caesar - 4. Others don't spend, probably they have a better targets.
Warrior now reached Bosphorus. Next move he will in Asia.
Worker finished wheat farm, switched to this tile. I want to grow as fast as possible before masonry, and then I need to find a balance between grow, hammers and commerce.
Frederick changed his opinion about his worst enemy. Now it is me again. May be this "worst enemy logic" doesn't work at the first move after two civs meet each other? Let's see.
Who is spending espionage points on me? Lisa - 3, Frederick - 4, Isabella - 2, Caesar - 4. Others don't spend, probably they have a better targets.
