Quick Answers / 'Newbie' Questions

Yep that seems really strange. Especially since the French did agree to an MA, I would assume this doesn't have to do with broken Trade Reps - otherwise the French shouldn't have agreed to it.
It could be something about my trade rep that broke without me understanding why. Gandhi was upset with me, but not for any reason I could figure out.
Your game shows again why, it becomes really complicated, and when you really count on an MA for fighting a stronger AI and you then can't sign it after declaring war, you might get ripped of the planet. And probably can't even pay gpt for peace :)
According to the F3 screen, I was "strong" compared to Hiawatha, but in my estimation not by much since we were mostly tech-matched - I just had finally managed to build up more tanks (I'd been average to weak in comparison most of the game). But if I was going to go after a Domination win after all, I wouldn't have been looking for an MA anyway.
Side note on selling techs: I never ever sell Electricity. Not for any price. Once they have Electricity, they all research like "instantly" for Replacable Parts, and you want them to have Infantry and faster workers as late as possible anyway. The AI's tend to research Electricity only after they got the Nationalism, 2 Gov's, Steam, Industrialization, Coorporation and Medicine. So not selling them Electricity slows them considerably down on defense, production&development (2x worker speed) and the lower branch of the tech tree. Plus they should't be able to contest you for Hoover - unless on high diff.
Electricity, or Electronics? (Lesson learned on selling Electronics ... won't do that again.)
 
Electricity, or Electronics? (Lesson learned on selling Electronics ... won't do that again.)
Electricity. Yes, I know you sold Electronics, which might be still a good idea if you're behind technologically. Just keep track of prebuilds from the AI's (F7) and time selling it correctly.

That's why I'm referencing Electricity as a side note, it comes early in the industrial tech tree and keeping it to yourself has so many advantages. At first glance, you might simply not want to sell Replacable Parts, since its this tech that is so powerful. But as I learned the hard way (and with an increadible flat learning curve) that in order to keep them away from RP and possible early wonder contest, do not sell Electricity.

As already pointed out, the AI then researches almost immediately Replacable Parts after it. Not only I ended up fighting against Infantry and fully railroaded&productive civ's very soon when I sold Electricity, I also often lost Theory of Evolution and sometimes even Hoover, because the AI's used some prebuilds via Universal Suffrage (which I couldn't or forgot).

EDIT: I usually consider selling it only after the AI's have the other techs I mentioned (Commie, Nationalism etc.), and I'm reasonably sure my two parallel prebuilds for ToE and Hoover are very likely to work out. They gona research it anyway after those techs, and if you intend to switch to Commie (without researching it), or need a push in your gpt, then it's worth considering.
 
Last edited:
Side note on selling techs: I never ever sell Electricity. Not for any price. Once they have Electricity, they all research like "instantly" for Replacable Parts, and you want them to have Infantry and faster workers as late as possible anyway.
OTOH this means that they are paying you money and/or luxuries for Electricity and then that they're using the added production for paying you for whichever tech you've researched next. In continent or archipelago maps, since they cannot invade you anyway, this effectively makes you richer.
Hence why I decided to go for JS Bach's Cathedral instead of Smith's Trading Co. (as well as the aforementioned emotional benefit of seeing a lot of AIs choke on their shields): I was already beginning to be able to max out research and still get a financial surplus. Treasury: OVER NINE THOUSAND and counting.
 
f3 military adviser basically only talks about numbers of units .
 
f3 military adviser basically only talks about numbers of units .
No. For F8 this is almost true, but F3 weights attack value and current HP strongly.

 
only thing ı witness is a given number of turns when ı kill enough numbers of units so that "stronger" falls to "equality" and so on . Won't argue otherwise , now that an aggreate of hitpoints (naturally) would be involved .
 
Hence why I decided to go for JS Bach's Cathedral instead of Smith's Trading Co. (as well as the aforementioned emotional benefit of seeing a lot of AIs choke on their shields): I was already beginning to be able to max out research and still get a financial surplus. Treasury: OVER NINE THOUSAND and counting.
It truly is getting ridiculous. AIs can put up over a thousand gold per turn even when completely mismanaging their resources to win limited-scope wars against various enemies. This actually renders some Emperor games easier than Monarch. The real change is in Demigod and higher where the AI starts with two settlers instead of one and a bunch of warriors.
 
It truly is getting ridiculous. AIs can put up over a thousand gold per turn even when completely mismanaging their resources to win limited-scope wars against various enemies. This actually renders some Emperor games easier than Monarch. The real change is in Demigod and higher where the AI starts with two settlers instead of one and a bunch of warriors.
Another advantage of Emperor is that AI's territory is usually better and earlier developed. So once you conquer them, you dont have to work hundreds of new tiles, except from building some mines instead of irrigation in not so corrupt areas. That might not matter much on standard maps, but does on large or huge maps which I like to play. So less work for my lazy self :)
 
The AI certainly does like to build roads to everywhere. I applied to the English a policy of surrender and replant that proved that you could fit in 3-4 more cities in the same space without leaving so many tiles outside any city's BFC.
 
My thanks to those who responded about Armies and when the AI will attack them.
I do wonder whether the actual criteria used in the program is simpler than making a calculation of combat outcome and attacking if a certain %age is exceeded. Most such calculations involving armies would need to cover more than ten combat rounds and could reach 25. Even with use of combination formulae this is a lot of computation. Could one criterion (or part of one) be the number of health points possessed by the defending army?

Roads everywhere, yes I've found the AI does this even when squares lie outside any city's BFC and are without any resource/luxury/strategic relevance. Probably down to the workers having been "built" and will not be disbanded or merged then having nothing else to do except improve tiles.
 
Roads aren't ‘without any strategic relevance’. It might be an example of redundancy, but the AI building roads everywhere does help it with mobility when transporting its troops. It's only when you reach railroads that you can afford tos end them the long way around.
 
Thanks @Takhisis. My poor language. I regularly duplicate roads and rails for strategic reasons. It's just that there are some tiles that it's a waste of time for the AI to link up. It happens to me every now and then when I make a mistake or am obliged to use less than optimum city placing.

When it comes to rails I'm aiming to find the shortest links for the initial network then work on the least corrupted tiles. I also try to remember that cities are already railed and to take advantage of the diagonal being just as quick as the straight route (cf king moves in chess)
 
The AI's a bit erratic because it also builds roads to resources which you cannot see, OR it just works the highest-yielding tiles and leaves the rest as they are. In my current game I'm seeing how the Iroquois next door carve strange paths through the jungle, including sending a worker into jungle to build a road and then clear the tile, or just properly work random tiles but doesn't connect them so then spends ages moving its workers when just the one or two turns spent on roads per tile would result in an impressive saving already.
 
I do wonder whether the actual criteria used in the program is simpler than making a calculation of combat outcome and attacking if a certain %age is exceeded. Most such calculations involving armies would need to cover more than ten combat rounds and could reach 25. Even with use of combination formulae this is a lot of computation. Could one criterion (or part of one) be the number of health points possessed by the defending army?

I'm pretty sure the same outcome calculations are done as with standard units (the latter is referenced in the war academy iirc). This ofc also involves the health points left of the unit/army. Also I don't believe that it's a lot of computation. For each round you have the probability of winning/losing this round, which you have only to calculate once. Then there is some summing up and (1-x)^y involved iirc, but thats done in no time. Also this might not even be done for every unit. Civ3 could calculate the minimum attack value that a unit should have in order to win against the army with a defined % chance (lets say 25%). If this minimum attack value is 10 and the AI has only units with 6 attack, there will be no more calculations for this turn.

Another pointer that roughly same calculations are used I noticed a long time ago, when I went pillaging with a Horsemen army and he attacked them with Medieveal Infantry (even though it was at full health). There are some notable exceptions to this:

- Invading another continent, taking the first city and fortifing this city with an army (we assume it has the highest chance of winning when defending). Then the AI will attack it, even at full HP, no matter the expected losses. They might even go with Archers against it. This should never happen though, as protecting captured Cities with Armies is risky anyway.
- If you somehow got on red HP on your invading army, it might still be the AI doesn't attack it. Assume you can't defend your army with strong defenders until next turn. As long you make sure you place units with lower chance of winning (i.e. low on defense) in reach of their units the AI would attack those units first. This applies even to workers (also referenced in the War Academy, other uses for workers).
- If you have an Army sitting on a continental choke point it might also get attacked when an AI declares war on you "over it", no matter combat caluclations. Haven't seen this yet though (and I don't intend to find out :) )
 
Last edited:
My mod has a 6/2/2 Hidden Nationality Spy unit.
I pillaged a neighboring irrigation (at peace) and then the Spy was attacked by a worker!! I have never seen that before.

How is that possible, since workers cant attack.
Now, while typing, all I can think of is that the Spy could be invisible. I forgot.
Could that be it? I had never seen a worker attck before.
 
Is that a variant of the "sub bug", but on land? The worker stumbled onto your invisible spy?
 
Is that a variant of the "sub bug", but on land? The worker stumbled onto your invisible spy?
Think so. It was so weird, the worker hitting him with his shovel.
Will check the biq tmr.
I forgot if I set them to invisible. Shouldnt be, actually.
 
Maybe it's related to the galley bug which I experienced not long ago?
 
I'm not sure. A galley attacking my land units is still a galley that has HP and an attack and a defence stat. A worker in unmodded civ has neither HP nor attack or defence.
 
Top Bottom