AGCEEP Portuguese AI

[Twoflower]

Ya I've seen this too. It really bugs me. I've removed West Africa from the areas of expansion for the Portuguese AI at the time that it's doing this and it hasn't gotten into a west African war since. Mind you I've only run it 2 times so far. So I still have my hopes up.

IDLF, if removing West Africa from Portugal's areas of expansion stops or reduces Portuguese dows in West Africa, you are saying that you suspect that the AI's dows are guided by its areas of expansion? If this is really true, we might have a smooth solution... and the Canaries will be colonized anyway by Castile if it owns it I think, so the area of expansion doesn't really serve any purpose.
EDIT: Interestingly enough, the vanilla AI for Castile/Spain does not have Iberia among its areas of expansion...in fact, this is a change that got introduced by the AGCEEP relatively recently, EEP didn't have it either - and strangely enough, when I spent lots of hours playtesting because I wanted to fix the Portuguese AI in EEP, I rarely saw AI Portugal destroyed by AI Castile, which is in fact why I objected to IDLF's observations in the first place.

 

[Norrefeldt]

It's purely coincidental. I have seen it numerous times but more often than not thye will call their allies into wars.

Well i had like ten coincidences on a row where CAS, POR and SPA didn't call allies in their African wars. Might be different in European conflicts, at least I hope so. I play with latest beta.

Upping relations between POR and GRA might do some good. Right now it's 80.
In a hands-off I ran Spain held Tangiers for a long time, a Portuguese core. It didn't make the relations decline. I do think the AI is more likely to DOW if they can get a core though, so CAS might DOW POR for that. But it's only a 5% chance that POR gets the Canaries, so I'm not sure we should alter the sequence for that.

Since the HYW often goes back and forth with a lot of waring, we should try it out carefully before we add POR to the ENG alliance. Not very historical either, is it?

IDLF, if removing West Africa from Portugal's areas of expansion stops or reduces Portuguese dows in West Africa, you are saying that you suspect that the AI's dows are guided by its areas of expansion? If this is really true, we might have a smooth solution... and the Canaries will be colonized anyway by Castile if it owns it I think, so the area of expansion doesn't really serve any purpose.

I never heard of such an interpretation and we should firmly establish it before trying to use it.

 

[Norrefeldt]

EDIT: Interestingly enough, the vanilla AI for Castile/Spain does not have Iberia among its areas of expansion...in fact, this is a change that got introduced by the AGCEEP relatively recently, EEP didn't have it either - and strangely enough, when I spent lots of hours playtesting because I wanted to fix the Portuguese AI in EEP, I rarely saw AI Portugal destroyed by AI Castile, which is in fact why I objected to IDLF's observations in the first place.

Well, I guess it would do no harm removing it.

 

[unmerged(31425)]

Running handsoff tests with Denmark and Holstein, I noticed that one way by which Portugal goes down very often (in fact this sequence happened in all tests except one) is as follows:
1. Portugal allies with Castile early on
2. sometime after ending the war with Morocco, Portugal gets bored and dows Granada, calling Castile in
3. Castile refuses to attack its vassal, which also hits its relations to Portugal badly
4. Castile declares war on Portugal because they now finally got the opportunity to do it

Usually in my tests Casile honors the DOW. What are your warmonger settings for both countries and do they have 125 relations at the start?

The Portuguese AI seems to willingly commit suicide by these silly dows.

Exactly, but it's the same problem for all AIs. Nothing unique about the Portuguese AI here.

Btw, I got more thoughts to throw in:
1. I'd like to know how often Castile dows Portugal because of an ahistorical choice in the Canaries sequence. I mean, even if we reduce the relation hits, Castile still gets a core on a Portuguese province, which means a steady decline in relations and a permanent CB (that encourages the AI a lot to dow). I'm quite sure that Portugal is almost doomed by the moment it refuses to cede back the Canaries....in fact, I have been destroyed by Castile in a game where I played Portugal and was greedy enough to keep the Canaries. I mean, it is only a five percent chance, but if this five percent chance means certain destruction of Portugal in five percent of the games just because of this event, we may have a problem. Have others made the same observations after Portugal has taken the B option and refused to cede the Canaries back? I would not like removing the event sequence, but we maybe either need to prevent Portugal from choosing this or not give Castile a core, at least not yet at that time, if that way we can reduce the likeliness of Portugal being destroyed by five percent.

You're probably right. I haven't payed attention to it so far though, but then again it is a bit rare.

2. IDLF or anybody else, do you know how much the AI takes into account alliances and guarantees of independence when choosing its victims?

AIs compare the economic strength of their alliance vs the economic strength of their opponents alliance. I don't know whether the AIs are weighing an economic figure or whether it's done by province count.

I ask because if this matters to the AI, we might want to try having England guarantee Portugal's independence, or if that does not suffice perhaps even an alliance between Portugal and England - which would both be quite historical (the guarantee of independence completely, the alliance at least somewhat). Of course, an alliance would need to be separate from England's evil HYW alliance since we do not want Portugal dragged into useless wars; the alliance should only be useable by Portugal - which can be done if we have England's other alliance as its primary alliance; then England when going to war will call its normal allies while Portugal when attacked by Castile or anybody else can call England.

This would be disastrous. 15th century England and 15th century France are wrecked right now too. These 2 countries are in wars almost non-stop for the 15th century. They then drag all their poor unfortunate allies into these wars and wreck their economies. This type of alliance will make extensive Portuguese colonization all but impossible.

This of course depends mostly on whether Castile would take that into account, since I doubt that even an alliance with England would suffice to save Portugal against Castilian onslaught.

Castile will probably not attack Portugal if allied to England because it weighs the economic strength of the alliances, but this still won't be preferable as a Portugal that has massive inflation and few colonies and TPs by 1500 isn't much of an improvement than one that gets axed by Castile.

The ideal circumstance for the Portuguese AI is to have to fight as few as wars as possible in the 15th century.

 

[Twoflower]

I never heard of such an interpretation and we should firmly establish it before trying to use it.

Well, it is fact though that Castile does not need Iberia as area of expansion because it will colonize the Canaries anyway and that this got added rather recently....I was however just trying to expand on what IDLF has tried; if removing West Africa from Portugal's areas of expansion stops Portuguese dows in West Africa, removing Iberia from Castile's should stop Castilian dows in Iberia, no? And the area of expansion is one of the differences between AGCEEP and vanilla and also between AGCEEP and EEP, and as such it should at least be examined whether it is indeed a reason for the different behaviour.

Since the HYW often goes back and forth with a lot of waring, we should try it out carefully before we add POR to the ENG alliance. Not very historical either, is it?

As I said, it would be a second alliance for England that should never ever be called upon by England....if this would happen, that would of course be a devastating change. But wait, England does not already start in an alliance, does it? So forget about that idea.
The guarantee of independence might be worth consideration though

 

[Twoflower]

The ideal circumstance for the Portuguese AI is to have to fight as few as wars as possible in the 15th century.

Yes, right. My idea was that England would have another primary alliance (the one with Burgundy etc.) that it uses for its own wars and that only Portugal would be able to call for the alliance with England. This can be done in EU2 and sometimes works well, but in this case I agree it's just too risky, because England will often lose its first alliance and then use the alliance with Portugal for its wars, and more importantly England doesn't start in an alliance right now, so there is nothing to set up as its primary alliance unless we want to wreck the HYW sequence.
A guarantee of independence by England would do no harm and would be historical, though.

 

[unmerged(31425)]

Well i had like ten coincidences on a row where CAS, POR and SPA didn't call allies in their African wars. Might be different in European conflicts, at least I hope so. I play with latest beta.

For African wars they don't usually call on their allies. I thought you just meant wars in general.

Since the HYW often goes back and forth with a lot of waring, we should try it out carefully before we add POR to the ENG alliance. Not very historical either, is it?

Like I said in my last post this will ruin Portugal. It ruins any countries involved in these extensive wars really.

I never heard of such an interpretation and we should firmly establish it before trying to use it.

As a word of caution with areas of expansion, I'm about 85% sure that they need to have a connection by areas in the list to be able to explore correctly. Before I made sure there was a connection I was unable to get explorers to explore with the AIs that lack that connection. Now they will explore albeit it's still too little. In fact I am trying extreme measures because I can't get the Portuguese AI to consistently explore Malacca by 1515. Currently I'm trying giving the AI 3 extra explorers in 1495 or so.

 

[unmerged(31425)]

A guarantee of independence by England would do no harm and would be historical, though.

What's the code for it? I'll start expirementing with it tommorrow.

 

[Twoflower]

As a word of caution with areas of expansion, I'm about 85% sure that they need to have a connection by areas in the list to be able to explore correctly. Before I made sure there was a connection I was unable to get explorers to explore with the AIs that lack that connection. Now they will explore albeit it's still too little. In fact I am trying extreme measures because I can't get the Portuguese AI to consistently explore Malacca by 1515. Currently I'm trying giving the AI 3 extra explorers in 1495 or so.

Do areas of expansion affect dows by the AI or not?

 

[unmerged(31425)]

Do areas of expansion affect dows by the AI or not?

I'm not sure. I suspect they don't. The way the wars happen for Portugal in west Africa is that the AI explores the provinces bordering the coastal ones, then DOWs those countries, usually because they have been bad boys and I'm assuming it's because they have too much BB and/or are at war already(being at war already is factored into AI DOW decisions as well). So I stripped it of the right to explore them and gave the AI the ocean in southwest Africa to explore instead.

It's stopped the wars so far but now explorers just "diddle fart" around there. Their exploring isn't very efficient anyways though so like I said I'm trying extreme measures to get the AI to explore more. I hope like made the 3 explorers will work. It may be yet another AI cheat but man nothing else is cutting it so far. I can't get the AI to explore Malacca in time more often than about 1 out of 12.

 

[ForzaA]

Playing a little game of Portugal myself, I was nice and even gave the Canaries back...

Castille even dowed me at -50 relations WITH a RM... and I didn't do anything particularly nasty (never attacked anyone to get BB or relation hits, didn't compete out that many merchants because I kept forgetting to send them)

It basically looks like Castille thinks that Portugal is an ideal target to attack, so likely we'll have to make them think Portugal isn't much of a target.

btw, would MA help? (atleast forces Castille to take an extra stabhit IF they want to DoW)

 

[unmerged(31425)]

Alright I think I've noticed some important things about AI exploration.

First of all I have noticed that Bartemeo Dias always explores the central Atlantic till the Carribean, the east coast of Africa up to Suez, and the Persian gulf. He does this almost everytime the AI makes him explore.

I've also noticed that Da Gama will go explore India everytime he explores and not other areas unless they are on the way of course. Juan De La Cosa, the Spanish explorer will also explore east Africa almost every game too.

The thing about Dias is the AI doesn't have alot of these areas as areas for expansion. Other explorers don't disobey the areas for expansion. This would indicate that some explorers are hard coded, and not neccesarily in a historical way, to explore certain areas. Dias is the only explorer who disobeys the area for expansion for Portugal and explorers the eastern Carribean. I've noticed it in numerous games in the past too for some time that he explores this region consistently. Because of this I'm going try switching his and Da Gama's ID numbers to see if it will break their pre-determined exploration routes.

Another anomoly I'm almost certain of is that the areas of exploration need to be connected. A break of more than one link seems to stop exploration past that 2nd gap. Therefore it may verywell be neccessary to have complete connection from the capital of a country all the way to every point of expansion. This may need to be a land not jsut sea zone path too, but I'm not 100% sure aboutthat yet. Incidentally Dias and Da Gama still explore their historic routes despite these gaps in the esploration paths. Which adds weight to my conclusion that they are hard coded to explore certain areas.

So bear in mind all that when making areas for expansion you'll need an unbroken path to the farthest points.

 

[unmerged(13277)]

Alright I think I've noticed some important things about AI exploration.

First of all I have noticed that Bartemeo Dias always explores the central Atlantic till the Carribean, the east coast of Africa up to Suez, and the Persian gulf. He does this almost everytime the AI makes him explore.

I've also noticed that Da Gama will go explore India everytime he explores and not other areas unless they are on the way of course. Juan De La Cosa, the Spanish explorer will also explore east Africa almost every game too.

The thing about Dias is the AI doesn't have alot of these areas as areas for expansion. Other explorers don't disobey the areas for expansion. This would indicate that some explorers are hard coded, and not neccesarily in a historical way, to explore certain areas. Dias is the only explorer who disobeys the area for expansion for Portugal and explorers the eastern Carribean. I've noticed it in numerous games in the past too for some time that he explores this region consistently. Because of this I'm going try switching his and Da Gama's ID numbers to see if it will break their pre-determined exploration routes.

Another anomoly I'm almost certain of is that the areas of exploration need to be connected. A break of more than one link seems to stop exploration past that 2nd gap. Therefore it may verywell be neccessary to have complete connection from the capital of a country all the way to every point of expansion. This may need to be a land not jsut sea zone path too, but I'm not 100% sure aboutthat yet. Incidentally Dias and Da Gama still explore their historic routes despite these gaps in the esploration paths. Which adds weight to my conclusion that they are hard coded to explore certain areas.

So bear in mind all that when making areas for expansion you'll need an unbroken path to the farthest points.

Hmm no!

No!

No!

No!

No!


It is hard work and takes enormous amount of time to get it right. It is not just slamming in some values and then it works. Far from it!
It is precision work setting areas of exploration, that involves number of explores, the time they are available, the length you want them to explore and the time you want them to do it in.

Number one golden rule; never ever let the ai run out of places to explore. Cause then they will go on their own.
For European countries that means a trip to the Carribean or Newfoundland or to the Persian Gulf – which means he explores the entire African coastline.
This is btw why Zheng He goes to Europe!

 

[Toio]

My question here would be, should Castile really be DoWing much of anybody pre-1492? If not, could we set Castile to a peaceful AI until they become Spain?

Have you tested this?
what will Aragon do ?
my testing shows that Aragon DOW castile more often. Is it because I had this set Castile to a peaceful AI I do not know.

Someone might know if a peaceful AI is a bait for DOWs.

 

[unmerged(31425)]

Have you tested this?
what will Aragon do ?
my testing shows that Aragon DOW castile more often. Is it because I had this set Castile to a peaceful AI I do not know.

Someone might know if a peaceful AI is a bait for DOWs.

I doubt it. The AIs weigh the economic strength of their and their opponents alliance. This is why smaller countries at times DOW larger ones.

Switching the Castilian AI from it's current warmonger setting of 10 to 0 does curb aggression I've noticed. But since Portugal gets bullied so much already it only puts a dent in the problem but doesn't sufficiently solve it.



By the way I've ran Jinnai's MP lowering proposal. Castile base MP was lowered to 19 from 48. This makes it only 2 higher than Portugal's. And it seems to do nothing. I've ran it about 4-5 times and I can't notice any difference. Castile is just as aggressive to Portugal and eveyone else as it usually is, so I've given up on the idea.

 

[unmerged(31425)]

Hmm no!

No!

No!

No!

No!


It is hard work and takes enormous amount of time to get it right. It is not just slamming in some values and then it works. Far from it!
It is precision work setting areas of exploration, that involves number of explores, the time they are available, the length you want them to explore and the time you want them to do it in.

Number one golden rule; never ever let the ai run out of places to explore. Cause then they will go on their own.
For European countries that means a trip to the Carribean or Newfoundland or to the Persian Gulf – which means he explores the entire African coastline.
This is btw why Zheng He goes to Europe!

Then please explain why Da Gama never explores past India. Why does Dias always explore these locations? Why does De La Cosa show up in east Africa almost every game(every game since I started looking for it). I have given the AI plenty of locations to explore and it is ignoring them and these explorers are doing these things regardless of the areas for expansion all on their own. Please explain why they are doing that.

 

[unmerged(28872)]

A longtime lurker's point of view reguarding the proposed POR/CAS AI cheat:

IF other methods of curbing the CAS aggression can't be found and the cheat MUST be implemented. I personally feel that the event is a lot easier to swallow if the event was framed more as an ahistory event then as an AI cheat.


If CAS has taken POR lands, then history has gone astray. Have the 'cheat' be a 'real' ahistorical event instead of an hidden AI cheat. The event would effect both the AI and the human.

Have the POR event say something vaugely like

"CAS has taken our lands, but through diplomcy... blah, blah, blah... an agreement was reached where POR would regain them"

(a) option "GREAT" Gain control of the lost proviences, +150 relations to CAS, RM to CAS, -200ducats. (The vassilization seems like it could cause unwanted diploannexition or economically cripple POR worse then a one time 200ducats or the missing territory, but ya'll know better how the economics of this would work)

give it a (b) option! "NO! that costs to much"



You could even give CAS a ahistory event to accept (a option) or refuse the treaty (B option).. (as a human playing CAS I'd be mad at POR getting it territory back with no chance to tell them "Screw off"... and sometimes I like to see the game go very ahistorical... currently it does it a little too much.


Basically the event is the same... but by not hiding it and giving it some flowery and hopefully vaugely possible ahistoric text you've changed the event from an AI cheat to a fantasy event. Something the forum members may be more likely to find more 'pleasing'. At least it don't seem nearly as harsh to me if it's framed in this manner....



.......back to the bushes where the lurkers live......

 

[WiSK]

A longtime lurker's point of view reguarding the proposed POR/CAS AI cheat:

Welcome

If CAS has taken POR lands, then history has gone astray. Have the 'cheat' be a 'real' ahistorical event instead of an hidden AI cheat. The event would effect both the AI and the human.

I agree strongly with the first part. History has gone astray. Not only is Castile taking Portugese lands ahistorical, it is also implausible. Castile would have sued for a tribute, promises to aid the reconquista, trade agreements or similar, but not for control of lands. Which is why I still don't consider this a cheat.

We can't make the CAS AI ask for vassilation during the peace process, so we implement this by immediately exchanging the conquered provinces for a vassalisation.

"CAS has taken our lands, but through diplomcy... blah, blah, blah... an agreement was reached where POR would regain them"
(a) option "GREAT" Gain control of the lost proviences, +150 relations to CAS, RM to CAS, -200ducats. (The vassilization seems like it could cause unwanted diploannexition or economically cripple POR worse then a one time 200ducats or the missing territory, but ya'll know better how the economics of this would work)

Okay, in principle, for plausibility's sake I would agree with 200 ducats and RM instead of vassalisation. However, the point of the event - and the reason it works - is that the vassalisation gives Portugal a "force shield" against further aggression from Castile. As far as I understand from IDLF, a royal marriage on its own won't do the trick.

give it a (b) option! "NO! that costs to much"

Yes. I mentioned a (b) option before, but no one seemed to pick up on the idea. I want to know whether it would change anyone's opinion if there was an action_b which didn't concede the provinces.

Basically the event is the same... but by not hiding it and giving it some flowery and hopefully vaugely possible ahistoric text you've changed the event from an AI cheat to a fantasy event. Something the forum members may be more likely to find more 'pleasing'. At least it don't seem nearly as harsh to me if it's framed in this manner....

Well, I'm not convinced the human player should ever see this event. It's an AI solution for an AI problem. Once a human gets involved, they should be able to choose whether to follow history or not.

I know that when I play either POR or CAS, I make up my mind fairly quickly about whether my intent is to conquer or befriend. In that case, I take responsibility for the outcome. However, when playing any other nation I really do want to see POR doing its thing. Especially as the Netherlands, a game without POR is just crap.

 

[unmerged(31425)]

A longtime lurker's point of view reguarding the proposed POR/CAS AI cheat:

IF other methods of curbing the CAS aggression can't be found and the cheat MUST be implemented. I personally feel that the event is a lot easier to swallow if the event was framed more as an ahistory event then as an AI cheat.

This should make no difference. Calling it an AI cheat or an AI event or whatever changes nothing. It is still a technically ahistorical event that kicks in to fix gameplay problems. Call it whatever you want, that changes nothing. If calling it something other than an AI cheat event will get people to accept it than that makes their whole argument complete nonsense.

If CAS has taken POR lands, then history has gone astray. Have the 'cheat' be a 'real' ahistorical event instead of an hidden AI cheat. The event would effect both the AI and the human.

Humans will always pick B unless for some strange reason it benefits them to pick A. Like if they realized they just got their BB too high and need to lower it. Either way it will result in humans getting more advantages over AIs. So what the point?

Have the POR event say something vaugely like

"CAS has taken our lands, but through diplomcy... blah, blah, blah... an agreement was reached where POR would regain them"

I personally am opposed to fantasy events in the game as a default setting. The fact that Italy or Germany can unify etc actually annoys me and this is also the case for some others too who hold fairly deterministic views. But eh, it's a minor point as it rarely ever happens. I haven't seen the AIs do it once yet.

(a) option "GREAT" Gain control of the lost proviences, +150 relations to CAS, RM to CAS, -200ducats. (The vassilization seems like it could cause unwanted diploannexition or economically cripple POR worse then a one time 200ducats or the missing territory, but ya'll know better how the economics of this would work)

I have never seen the event lead to diplo-annexation even once. It even gives a high relation boost, putting the countries in a ideal setting if diplo-annexation was realy possible. I've seen it fire and play out for sometime easily in excess of 20 tiems now and no diplo-annextions have happened yet. Like I've said in previous posts, I suspect it's the BB because Castile can on occasion diplo-annex Portugal by purely peaceful means.

give it a (b) option! "NO! that costs to much"

What's the sense in having it not fire even 5% of the time. The fact is the poor Portuguese AI has so many things that can go wrong for it as is, even with my setup, which is much much more reliable than the one in the 134.3 currently.

You could even give CAS a ahistory event to accept (a option) or refuse the treaty (B option).. (as a human playing CAS I'd be mad at POR getting it territory back with no chance to tell them "Screw off"... and sometimes I like to see the game go very ahistorical... currently it does it a little too much.

Well if we made it into a human as well event then it wouldn't be a "historical event" now would it.

Basically the event is the same... but by not hiding it and giving it some flowery and hopefully vaugely possible ahistoric text you've changed the event from an AI cheat to a fantasy event. Something the forum members may be more likely to find more 'pleasing'. At least it don't seem nearly as harsh to me if it's framed in this manner....

.......back to the bushes where the lurkers live......

Hmm. You've made my day with this post friend. You inadvertedly perhaps gave additional fuel to my argument. The fact is that if German and Italian unification events can be in the game then why can't my AI cheat here be in too? Is it any less historically plausible? This being the case the fact that someone doesn't mind that these 2 fantasy events are in the game but my AI cheat event can't be is rather hypocritical. I mean good grief my event actually helps the game. The Italy and germany unifacation events don't. They are just there for kicks.

Thanx bro! I can't wait to see what Twoflower and Norrefeldt are gonna say about this.

Besides my whole point is that making this mod as challenging as the vanilla without AI cheats is just a pipe dream. I've pointed out why in previous posts and no one has refuted my argument there yet, indeed for those particular posts they haven't even tried. That being the case they ought to either admit I'm right or admit they don't want to believe I'm right but have no real reason as to why I'm wrong. Instead what do they do? They dodge it. Because they have nothing to say about it and they know it. But perhaps the truth is less important to them than winning this argmuent. Or perhaps it just bugs them that I figured out a flaw in their modding philosphy and they are too proud to admit it.

Should Twoflower or Norrefeldt or anyone else for that matter not agree with my analysis here they may feel free to refute my argument in post #240 of this thread. I challenge anyone to prove I am wrong when I say it is impossible to make this mod as challenging as the vanilla without AI cheats. If you can't than apparently I'm correct now ain't I.

 

[unmerged(13277)]

Then please explain why Da Gama never explores past India. Why does Dias always explore these locations? Why does De La Cosa show up in east Africa almost every game(every game since I started looking for it). I have given the AI plenty of locations to explore and it is ignoring them and these explorers are doing these things regardless of the areas for expansion all on their own. Please explain why they are doing that.

You are way way to quick to make generalization.
All you need to do is run a stock v1.08 game with the latest beta, just once and you can first hand see that your theory is not true. Da Gama does not even go to India, and Diaz much less, in stock. They will both happily go everywhere else but India! So how can they possible be hard coded to do a special job in India?
Da Gama and Diaz in stock, has a thing for Newfoundland, Caribbean, Scandinavian, Mexican gulf, Egypt, and the whole South American eastern coastline, none of them gives 5 cents for India.

Look I already told you this is hard work, there is no corner cutting in this. You cannot just slap in a few values and expect it work like you thought it would. It almost never does.
Like I said, you can’t give him to little, but you can also not give him too much. That too will mess it up. If too much he will on his own account select exactly what forfill his ‘need’.

Ex. pick two areas only, like one on the west coast of America and say one in china. Have you ever wondered what makes him pick one over the other? Cause he will do it always, 100% of all games he will do the same thing. What makes him choose one area over another, for first exploration?
Is it location?, is it the area itself?, is it the provinces in the area?, is it distance?, is it the time it takes? Is it explores availability?, is it conquistadors availability?, is it the current attitude (all parameters) of the ai?, or is it something else?

Like I said, this is precision work! And it takes time, time, time, time, time… but you can make any explorer go anywhere you want, any time you want, by supplying the exactly right amount to explore at the right time.

- but remember that exploration setting, is also possible colonization candidates, which makes it even harder to do!