Crazy Spatz's Alpha Centauri Mod

[dinobot386]

Alright, I'll get to testing this in the morning and I'll change my senario to see if the AI will still choose policies.

It's an extremely large earth map that ive been tweaking TSL and all. Im not sure exactly how much bigger it is than the huge earth map percentage wise, but you can fit 3-4 in italy alone, and its all to scale (except for hawaii which is the size of japan so that polynesia has some room to start with.

And get this... It actually WORKS unlike the map I cannibalized it from

 

[Darsnan]

Tested against the 0.22 patch this morning in one game. Nothing significant to note other than it seemed to take longer to build things (i.e. I was doing a little more of the continuous press end turn than usual), however that just may have been because of my cities locations and the availability of hammers/ worked tiles.

I'm also thinking the KGB is too good and/ or too cheap in the environment I play in (i.e. Industrial era start): my original build queue for Wonders was Louvre-Cristo Redentor, but I am finding I can easily accomodate building the KGB in between the two. I don't know the KGB's importance versus other era starts, but have you thought about changing the cost of the KGB, or toning down its benefits (i.e. possibly less likelihood of "stealing" techs)?

Also, through all my playtesting I've noted the Siamese, Greeks, and French consistently seem to do quite well on an Industrial start. Conversely, the Mongols and Ottomans are consistent push-overs. Question: have you given any thought to tweaking the Civs as well, in order to try and improve them?

----------------
On an unrelated note, there's something I'm working on for the next one. In my experience, the Pentagon is a problem; 50% discount on upgrades is just too much. A player will hold off on upgrading units until he gets it, and then get the massive discount.
I want it to be possible to get cheap upgrade costs, but not in one shot like that. So here's what I'm trying out now:

Military Academy (building): the Elite promotion given to units trained in this city now decreases the cost of upgrading that unit by 20%, in addition to its other effects.
Pentagon: -20% to all upgrade costs, and you spawn Great Generals 25% faster.
Skunkworks (national wonder): -20% to all upgrade costs, +10XP to units trained in this city, and another +5 to units trained in ALL cities.
Bioenhancement Center: the Bioenhancement promotion given to units trained in this city now decreases the cost by 20%, in addition to its other effects.
Nano Factory: -100% to all upgrade costs (so yes, they're all free at that point. They actually cost a token cost, I think 10gp.), plus free Aluminum and a production bonus in this city.

I'm flexible on this whole thing, but it's like with the Food Storage system: instead of one or two big buildings, I wanted a gradual progression. Thoughts?

I like the whole idea behind the "gradual progression" approach, as it makes a player have to think/ plan ahead, and prioritize what the player wants to do - i.e. what their long-term goals are going to be.

D

 

[dinobot386]

Ive notice that too, but also with Washington who colonizes a few areas and then just sits back to be conquered, also the Monguls arn't nearly as agressive as they should be

 

[Spatzimaus]

however that just may have been because of my cities locations and the availability of hammers/ worked tiles.

That's what I'd guess. None of the changes I made in the last two versions should reduce your production.

I don't know the KGB's importance versus other era starts, but have you thought about changing the cost of the KGB, or toning down its benefits (i.e. possibly less likelihood of "stealing" techs)?

Both, which is why I need the feedback on this. Let's talk numbers.
Steal Chance: Originally, I'd given the Planetary Datalinks a base chance of "10". That means that if every civ other than you has the tech, INCLUDING city-states AND barbarians, then you have a 10% chance of stealing the tech on a given turn on Standard speed. Slower speeds lower the chance proportionately, so it's 3.333% on Marathon. Note the "every civ" part; if I'm on a Small map (6 civs, 12 city-states), then there are 18 civs other than me. So if only one civ has the tech on that map size, it'd be a 0.555% chance, if two civs have it it's a 1.111% chance, and so on.
When I added the KGB, I simply split the bonus in half, so both national wonders now have that value set to "5". I could easily change this, then, to be a 4/6 split with the Datalinks having a larger bonus. Or, make it 4/4, and give the last 2 to the Nethack Terminus. The point is that it's adjustable, other than requiring an integer value.

One catch is that the city-states and Barbarians have a custom logic in the core game code that gives them techs automatically based on what the major civs have, in addition to the normal research process. It seems to be a chance-based system, because the city-states don't all gain a tech at once, but it's definitely there.
I could change my logic to simply not count barbs or minor civs, but that'd mean a loop to count up how many have a tech instead of simply using the
Game.CountKnownTechNumTeams(tech.ID)
function.

As for the cost, well, I'll admit that when I add new Wonders I just eyeball the costs and hope for the best. I set the cost equal to 400 hammers (and no maintenance cost) based on the buildings around it. The Seaport (same tech) is 180, the Windmill (tech at same X) is 250, Big Ben (wonder at same X) is 700. Taj Mahal (previous X, Wonder) is 600; so are the Forbidden Palace, Kremlin, and Sistine Chapel. The last National Wonder in the vanilla game, the Hermitage, is 300 and is just a couple tiers earlier than the KGB.

Given that pattern, 400 seemed like a reasonable value. National Wonders fall between the world wonders and normal buildings in cost, so I figured 400-450 would work.
I could up it as high as 500, but I wouldn't go higher than that; if it's too cheap for its effect at that point, then it needs a toned-down effect.

Question: have you given any thought to tweaking the Civs as well, in order to try and improve them?

No. The civs that are good in a later-era start are VERY different from those good in an early start, and I'm still sticking to my original design where an Ancient start is the intended one. I know, plenty of people will start in the Industrial. (I will, too, about half the time.) And I'm trying to make a Digital start viable for people who want a really SMAC-like experience. But I don't want to do anything that'd screw up an Ancient start.

Basically, I see it like the Wonders. Some Wonders are good in the earlier eras, some are good in the later ones, and some only have a single triggered effect (a golden age, a free great person) and are useless after that (meaning capturing a city with them isn't helpful). Some boost a single city tremendously (see Colossus), others give an empire-wide effect. So it's up to the player to know which Wonders are stronger or weaker in the current situation.
I see this the same way. Having France as an opponent in an Industrial start is dangerous. Having them as an opponent in an Ancient start, not so much. And in your Industrial start, once you get into the future Eras, what advantages does France have over any other civs?

As for the Siamese and Greeks, the root cause of their imbalance for them is something I've known to be an issue for a long time: city state bribe rates.

If you start a game, then on the first turn, a 250gp bribe gives 30, a 500gp gives 65, and a 1000gp gives 140. As time goes on, these rates will decrease, so a hundred or so turns later it might be 20/45/95.
But here's the catch: This progression is the same regardless of what era you started in.
So if you started in the Ancient Era, then it'd be 30/65/140 on turn 1, but by the time you'd built up ANY money it'd already be down to 20/45/95, and so on. So Greece and Siam wouldn't be able to afford to bribe a bunch of city-states, because by the time they had enough income to do so the return on the investment would be too low.
But if you start in the Industrial Era, then these civs are making a substantial profit while the returns are still at their highest rates, and it becomes much easier to bribe multiple states.

This is bad design. The game should calculate the progression from year 0, and if you start a game on turn 340 it'd just scale the values down appropriately and START you at the reduced progression.

 

[Darsnan]

Oh yeah, this morning I made the Nuclear era in turn 449, was at +38 Gold/ turn, and had +4 Happiness. The game ended with me being nuked to death by my many enemies.

This afternoon I reached the Nuclear era in turn 400, with +21 G and + 7 H at the time. Siam spawned between me and the Greeks, and the Greeks approached me about removing Siam from this dimension, and they are now down to one city. I've got oil, but no aluminum or uranium in this game: we'll see if that becomes an issue later....


One thing which caught my attention this afternoon were the associated costs regarding Theatres versus Opera Houses, as follows:

Theatres cost 187H and 4G turn in maintenance, and give +3 Happiness and +2 Culture per turn.

Opera Houses cost 187H and 4G turn in maintenace, and give +4 Culture a turn.

Should Theatres cost more, or are the benefits balanced in your oppinion?

D

 

[Spatzimaus]

Just to be clear: I'm not saying that I would never consider adjusting the balance of those civs, just that I HAVEN'T considered adjusting them.

But really, this ties back to an earlier discussion: the elusive Third Mod.

Right now, I've got two mods, the Balance and the Content. But the "balance" mod is a bit of a misnomer, which is why I'd originally called it the Long mod: the balance it creates is ONLY related to whether a game is still competitive long enough for you to reach the future eras. Nearly every change I made in it is along these lines, at least indirectly.
Obviously, the Balance mod has grown over time, to where it's now adjusting things in a wide variety of gameplay elements. But it's still not a full attempt at game balance.

So for a while now, I've considered creating a third mod, creatively referred to as the Third Mod. (More likely, I'd start referring to THIS mod as the Balance mod and go back to calling the other one the Long mod.) It'd be for the kinds of things you described, balance changes that have nothing to do with prolonging the competitive game phase. Things like my new custom non-future national wonders, the "head start" logic for late-era starts, a lot of the tech yield increases, and so on. I'd also probably place my variant of the Tech Diffusion in here (I use that mod every time, but I didn't want to include it in the SMAC mods because I didn't want to force anyone else to use it).
If I did that, then tweaking the civ traits and policies would be more reasonable. The question then becomes whether it's a good idea for me to try my hand at wide-scale game balance in general, especially since Thalassicus has been doing that for quite a while now and has a set of large-scale balance mods in place.

 

[Spatzimaus]

The game ended with me being nuked to death by my many enemies.

The question, then, is whether I need to put more nuclear defenses into the game. For instance, I could make the Manhattan Project add a 20% Interception rate. (No, you wouldn't ACTUALLY be "intercepting" the nukes. I'd think of it more like "nuclear attacks are a bit harder to pull off against a civ that knows what nukes are capable of".) And while the only pre-future building with nuke damage reduction is the Military Base, I could tweak a few others this way. For instance, I've thought for a while of giving the Palace a massive nuke defense, so that nuking someone's capital wouldn't be nearly as effective. But I'd also want it to be tied to some other building, something that could be built in more cities. Arsenal, Military Academy, that sort of thing.

I've got oil, but no aluminum or uranium in this game: we'll see if that becomes an issue later...

Resource distribution is obviously something I've been trying to keep an eye on, but I think the real problem has been that the resource distribution is just so RANDOM. In the vanilla game it's a purely random process. I've tried to overcome this, somewhat, by hard-distributing certain resources. Coal, for instance, is hard-set to place N/2 small deposits on hills, and I've reduced the randomly-distributed coal a bit to compensate. (Others set this way: N/3 small neutronium on any land tiles, N/3 small omnicytes on flood plains, N/4 small iron on any land tiles, and N/3 LARGE neutronium on non-forested hills.)
So no matter what type of terrain you have on your map, or how bad the randomization is, there'll be a minimum number of coal deposits on it in the hills. There's no guarantee that any of these would be near your empire, but at least they'd be on the map somewhere.

I've been tempted to do the same for other resources, adding in a bunch of N/4 or N/3 small distributions for Oil, Aluminum, Uranium, etc. to ensure that a minimal number are placed on the map somewhere, and reducing their chances of random distributions. I'd then reduce the total number of random Small deposits placed to compensate.

One thing which caught my attention this afternoon were the associated costs regarding Theatres versus Opera Houses, as follows:

Theatres cost 187H and 4G turn in maintenance, and give +3 Happiness and +2 Culture per turn.

Opera Houses cost 187H and 4G turn in maintenace, and give +4 Culture a turn.

Should Theatres cost more, or are the benefits balanced in your oppinion?

Yes, this is something I've been unhappy with for a while now. No, they're not balanced IMNSHO; Happiness is better than Culture, it's not a 1:1 ratio to begin with, and obviously the two totals aren't the same even if you DO go 1:1.

In the core game, the Opera House and Museum were both +5 culture for 3 gpt, although the Opera House was reduced to +4 in a patch. In my opinion those were just too cheap, so I changed them to the 4/5 gpt costs paralleling the Theater and Stadium. And I reduced them to 4 and 5 culture (with the museum later being changed to 4 culture and 1 research). But yes, they still feel weak, and I think part of that is my inherent distaste for single-effect buildings.
You notice how often I've changed buildings to do something besides the primary function? A research building that gives a bit of gold, a food storage building that gives a bit of research, a happiness building that now gives a little culture and a little gold? It's not just that I prefer this sort of design, it's that I wanted a smooth transition to the SMAC-like future eras, where EVERY building does multiple things. So these two culture buildings are throwbacks to the vanilla design. Which leads to the discussion:

How would you change the Opera House and Museum so that they do multiple things, while still being primarily about culture?

The Museum's already like this, although I could boost its science output a little (to +2?). But what can I do to the opera house? I could make it be +1 gold, but there's already too much gold floating around. What I REALLY need are some more +food buildings to boost city growth, but I can't really rationalize the Opera House getting that.

The problem with this is, there's just too much Culture floating around as it is. The Colosseum and Library each generate +1 culture, but that's to compensate for the reduction in the Temple (+3 to +1), so that's excusable. But the University now generates +1 (to make up for its reduced research), the Observatory generates +3 (to make up for its VERY reduced research), and you've already noticed the Theater and Stadium. Okay, the Monastery now gets +1 instead of +3 (to make up for its +1 happiness), but you still come out way ahead. I'd originally wanted this, because the need for an additional five policies to make a Cultural victory was getting prohibitive if I didn't boost culture output a bit, but it's throwing off a lot of the balance.
This is the reason why I'd like to find some way to reduce the culture of these buildings, and give them some other secondary effect to compensate.

So what I'm thinking of trying is this:
Opera House: +3 culture, +10% Great People birthrate in this city
Museum: +4 culture, +2 research

10% isn't a huge amount, especially for cities that aren't generating great people, but I've always wanted a way to boost that rate in cities that can't build a Garden. And this'd balance out the total Culture nicely, making up for the boost to the University.

I was also thinking about replacing the University's culture bonus with a similar Great Person boost. (Or doing this to the Public School, but I'm not sure if that joke is worth it.)

 

[Faerillis]

Spatz, I've noticed something with your balancing that you might want to consider. Your mod is fantastic, but I think you may have over-raised the science needed for Nuclear Era techs. Obviously the vanilla versions are too short, but the pacing seems a bit slow for an era where technology increases so rapidly real-world... that and in vanilla it was practically the end-all era so it was mostly reiterating all the same things with higher numbers. I think it would be better to slow things down when Centauri stuff starts to take hold. (Just opinion)

Anyways keep up the great work, you've made one hell of a mod here

 

[Spatzimaus]

Your mod is fantastic, but I think you may have over-raised the science needed for Nuclear Era techs.

Thanks for commenting on this. Let me try and explain what happened and what I'm aiming for, so that you can see the problem.

Back in Ye Olden Dayz, research buildings gave +50% bonuses and techs cost a certain amount. In the Modern Era, a fully-stocked city would generate 7.5 science per population, and most techs would only take ~3 turns. This was bad.

I lowered the research buildings, such that a full set would only generate 3.33 science per population (a factor of 2.25 less than the vanilla game). My goal was and is this:
A player who plays in a normal way, with the occasional expansion war, will average 8-10 turns per tech throughout the game without taking the Rationalism policy branch. (Likewise, I try to balance Happiness around the idea of staying at +10-20 without taking the Piety branch, assuming you're not playing Persia or Egypt, whose UBs add happiness.)

So far so good, right? Reducing the research rate by a factor of ~2.25 put the timing right about where I wanted it.

Then, in the March 1st patch, the devs rebalanced the costs of techs, such that a Modern Era tech now costs about twice what it did before. That ratio should sound familiar; the scaling they applied to tech costs is basically identical to what my building changes provided, over all eras. So in theory, I could completely undo all of my research-related changes and be done with it.
Again note: I did nothing to change the tech costs, in beakers. That was purely the devs' doing in the vanilla game. I changed the multipliers for buildings. This isn't semantics; lowered multipliers means that there's not as big of a difference between an unimproved city and an improved one, which helps the AI quite a bit.

Thing is, it's still going a bit fast in my experience once you really get going, because of the lack of a "corruption" system. I just played a game where, in the Fusion Era with an empire controlling half the planet, I was getting new techs every ~7 turns. Granted, I use a Tech Diffusion mod, but by that point I was the tech leader by a pretty good margin and wasn't gaining anything from that, and besides, that bonus doesn't affect the listed number of turns before you start researching a tech. But more importantly, I've been trying to get away from the population-plus-multipliers basis of the vanilla game's research in general. That why I:
> Added the KGB, which lets you steal techs outright.
> Added the Three Gorges Dam, which gives all Engineer specialists +1 research
> Added the Red Cross, which gives +50% to its host city and gives all Scientist specialists +1 food. (This makes it more likely that a city in Default Focus would use them, which would in turn increase research.)
> Added tech yield increases to the Academy to make its science output impressively large as the game goes on.
> Notice the Centauri Ecology tech, the one you get from building a spaceship. It gives +1 research to all freshwater farms (and because of the gap between Civil Service and Fertilizer, most of your farms will have fresh water). That can add a BIG amount, added over your whole empire.
> Lots of other little things. The Museum gives +4 culture and +2 research, the Monolith gives a little research, and so on. I've done things to make great people more common, and more Great Scientists means more free techs or Academies.
(Okay, on the negative side I cut the effects of the Secularism policy in half, from +2 science per specialist to +1, because it was just WAY too good.)

I'm not saying these are significant enough to take the pace back to where it was before, but these are the kinds of changes I'd prefer to put in, instead of just putting all of the multipliers back up to +50%. So I'm open to suggestions on what could be done to speed the pace up, but I'm NOT going to go back to massive building multipliers. (And the Public School's "+1 science per pop" is WAY out.) Think more like the University's "+2 science per jungle" type of thing...

What I'd like to do, really, is have Golden Ages make all specialists give +1 research, sort of like how gold gets +1 per tile that generates it and production gets a flat +20%. Something like that could make all the difference. I'm just not sure I can DO it; I mean sure, I can do a Lua end-of-turn event that does it easily enough, but that's not the same thing.

 

[Darsnan]

The question, then, is whether I need to put more nuclear defenses into the game.

How about increasing the cost to build nukes? This would then theoretically push out the first instance of nuking, as well as increasing the gap between rounds of bombing. Say increase the cost of Atomic Bombs by 33%, and Nuclear Missiles by 50%?

I've thought for a while of giving the Palace a massive nuke defense, so that nuking someone's capital wouldn't be nearly as effective. But I'd also want it to be tied to some other building, something that could be built in more cities. Arsenal, Military Academy, that sort of thing.

So would the AIs be able to understand that nuking someone’s Palace has a lower probability rate of success? If not then I think this might be considered exploitable by humans.
Also, what are the parameters regarding nuke defenses for detonations on tiles near cities – do the buildings your describing still have the chance of intercepting a nuke one, two, or three tiles away from the city itself (i.e. within the blast radius), and if so does the chance of interception scale versus distance from the city center?

I've been tempted to do the same for other resources, adding in a bunch of N/4 or N/3 small distributions for Oil, Aluminum, Uranium, etc. to ensure that a minimal number are placed on the map somewhere, and reducing their chances of random distributions. I'd then reduce the total number of random Small deposits placed to compensate.

Dumb question on my part but what was the whole point behind Firaxis implementing the strategic resources game mechanic in the Civ series?

How would you change the Opera House and Museum so that they do multiple things, while still being primarily about culture?

So what I'm thinking of trying is this:
Opera House: +3 culture, +10% Great People birthrate in this city
Museum: +4 culture, +2 research

10% isn't a huge amount, especially for cities that aren't generating great people, but I've always wanted a way to boost that rate in cities that can't build a Garden. And this'd balance out the total Culture nicely, making up for the boost to the University.

I like your idea of boosting the Great People generation. Question: how well do the AIs utilize Great People? I know that during wars I sometimes see random Great People units floating around the battlefields, so I know the AIs can get confused on what to do with them, but what about for popping techs, building Manufactories (and other Great People buildings) – would the AIs be able to accommodate/ utilize the generation of more Great People within their domains?

D

 

[Spatzimaus]

How about increasing the cost to build nukes? This would then theoretically push out the first instance of nuking, as well as increasing the gap between rounds of bombing. Say increase the cost of Atomic Bombs by 33%, and Nuclear Missiles by 50%?

The problem with doing this is that the AI's decision-making is done on the basis of Flavor, so if you raise the cost but don't adjust the flavors, then the AI will still build the things just as often as before, which'll suck up even more turns, and therefore cripple it even worse in terms of infrastructure. (And if I do adjust the flavors, then what did I need the increased cost for?) Small upward tweaks are good, and I'll look into that, but I can't take it far enough to REALLY make a difference.

The bigger issue, IMO, is that you've got two nuke units within one tech level of each other. Besides just being kinda pointless in general, this basically doubles up the Flavor values, making it more likely an AI will build one of the two.

So would the AIs be able to understand that nuking someone’s Palace has a lower probability rate of success?

I just don't know. If it was a Lua change then it'd be definitely no, but this IS an already-implemented function in the Buildings table. I just don't know if the AI can tell this defensive bonus exists; similarly, there's an AirDefense modifier to improve a city's defense against air units, which I've already given to the Palace, and I have no idea whether the AI can evaluate its effects. (Since the player can't really tell until after he's made an air attack, I'd lean towards no on this one.)

Even if it deals less damage, nuking someone's capital is STILL worthwhile. It's their capital, after all, it'll be the hardest to conquer normally (thanks to the palace's defense boost), and knocking down its population makes a big difference to the economy (since capital population counts towards EVERY city's trade route income).

Also, what are the parameters regarding nuke defenses for detonations on tiles near cities

None. The defenses only affect how much damage a city takes when being hit by a nuke. They do nothing for anyone NEAR the city, they don't stop improvements from being destroyed, and they don't prevent fallout. SDI does help these, by adding a flat percentage chance that a nuke will be destroyed BEFORE exploding, but that's an empire-wide thing.

I'd tried doing something related to this in my mod. My original goal was that SDI would have a 75% chance of intercepting an Atomic Bomb, a 50% chance against a Nuclear Missile, and a 25% chance against a Planet Buster. This percentage would then be modified by how many Orbital Defense Pods the defender had and how many Geosynchronous Survey Pods the attacker had. I could then go from there and adjust things based on the contents of nearby cities; this is where your comment comes in, since it'd be easier to then expand this to providing nearby units with temporary nuke protection.

Unfortunately, while RunCombatSim triggers for nukes like any other combat, the defender's info (both ID and unit number) are ALWAYS -1. There's no indication of who's being attacked, because nukes attack a map hex and not a unit/city. To make it worse, there's no X/Y being passed, so you can't even figure out the owner of the target hex THAT way, and since nukes are a bombardment unit their X and Y remain the coordinates of their launching city. So, I had to scrap this and just use their existing interception chance stub.

It'd be easy enough to do an after-the-fact cleanup, where if a city with building X has nearby fallout, then there's a X chance each turn that each hex's fallout is cleaned up and the improvement repaired. But this does nothing for the units caught in the open, which would be destroyed.

Dumb question on my part but what was the whole point behind Firaxis implementing the strategic resources game mechanic in the Civ series?

Obviously, I'm not a Firaxis developer. (You can tell, because I comment my code.) But the way I've always understood it, there are three main reasons:
1> Strategic resources limit the number of "offense" military units you can produce. Without them, a player could just keep churning out tanks until he had an unstoppable army, and then roll over everyone else. (This is exactly what'd happen in the earlier civ games.)
2> Most strategic resources are placed in less hospitable areas. Without them, there'd be little reason to settle near deserts, marsh, or tundra. But these are where the most Oil is found, along with many other resources. Hills, too; they're okay production normally, but hills are also where many of the best resources are found.
3> Rarity drives action. You hit the industrial era, where you were planning on putting factories in all of your cities. But wait, you don't have any coal. So, since you just can't afford to NOT have factories (the lack of production cripples you in a wonder race), this forces you to go get some instead of just sitting back in your comfortable little starting area for six millenia. To go get some, you might be able to find some small island to colonize, but more likely, you have to go to war.

Throw enough randomness into the system, and statistically speaking it's practically guaranteed that at least one resource will be like this. The first problem is that going from 6 strategic resources to 9 means that it's even more likely that one resource will suffer from rarity problems, even though I increased the number of spawns by a corresponding amount.

But the second problem is endemic to the core game: the randomness is PURELY random. The game doesn't say "okay, place 10 coal on the map, and they might all be out of your reach", which'd allow situation #3 above. It says "okay, go through each hill hex, and do a 1 in X chance that it has a Large coal on it." You have no way of knowing how many it'll end up with. Then the game places 35 Small deposits (modified for map size and abundance) around the map, but there's no control over how many of those would be on the types of terrains that could have coal, and even if they do then there's no guarantee the coal would be what's placed.

Think of it this way. Go get 30 six-sided dice; each represents one hill in the game. Roll them all; if you get a 1, it has a Coal on it. The math will tell you that you're most likely to have 5 coals on the map as a result. But each individual die is most likely to not be a 1, so the odds are actually not bad of getting significantly less. There's a 0.4% chance of NO coal (1 in 250), a 2.5% chance of 1 coal, and a 7.4% chance of 2 coal. That's more than a 10% chance of having far too little on the map to allow for option 3 above; you can't go get more if there isn't any more to get. And even if there IS an average number, then there's the chance that the five deposits might all be placed on the hills that aren't near you (assuming there are any hills near you in the first place); the game has a small amount of clustering logic, to discourage placing a new deposit too close to an old one, but that only checks within a radius of 3-4 hexes and won't affect empire-wide issues.
So what I did was, instead of rolling 30 dice, I roll 18 and then add 2 automatic successes to the result. Same average number at the end, but less variation, and there's now ALWAYS some on the map somewhere for you to go get if you come up short in your starting area. It still might be distributed badly, to where you wouldn't have any nearby, but you'd at least have the option of trading for it (or bribing a city-state for it, if they're placed right).

I did this for coal, neutronium, omnicytes, and iron in varying amounts. Neutronium, for instance, is almost entirely placed through this flat amount system, while iron just gets a slight boost but is mostly still random. Coal is in between those extremes. The question is whether I should do the same for the other strategics. (Note that water deposits are handled through a separate logic, I'm just talking about land resources here.)

Question: how well do the AIs utilize Great People?

Pretty well, actually. They build the Manufactories and stuff just fine, and I think the confusion you've seen was caused by their activated abilities (Culture Bomb and such). This is part of why I boosted the GP improvement yields; the player was rarely building those, and was mostly using activated abilities, while the AI would build the improvements. I wanted to make sure there wasn't a power disparity as a result.

One thing I've been toying with for a while is to have it so that if you work a tile containing a Great Person Improvement, you gain one Great Person point of the appropriate type. It's actually not hard to do this in Lua; if a city has an Academy within its area, then check to see if the tile is being worked, and if so, add one Scientist point to that city. Repeat for the other types.
But as usual, the AI would have no clue. So I've held off for now.

 

[dinobot386]

Uh, there seems to be an issue with Representation

It's making the costs of future policies go up, not down!

Got it at 90 culture mark, waiting for it to found city, the cost of next policy shot up from 160 to 175 culture

 

[Spatzimaus]

I just tested that out, and no, it's working correctly. (And since my mod does nothing to that policy, I'd have been very surprised to see otherwise.)

So I tried to repeat your experiment. Prince difficulty, grabbed Liberty and Citizenship. Cost of next policy: 90.
(If I settled a city right then, the cost would jump to 120; but, I didn't.)

So then I followed two paths, by saving the game and reloading to the decision point:
PATH A:
Then I grabbed Collective Rule. Cost of next policy: 160.
If I settled a city then, the cost of next policy jumped to 205.

PATH A:
Then I grabbed Representation. Cost of next policy: 160. (Since I only have the one starting city, it SHOULD be the same.)
If I settled a city then, the cost would jump to 190. (An increase of 30 instead of 45. In other words, representation did the 33% discount correctly.)

So it's working as designed. I don't know what you did wrong in your test.

 

[Darsnan]

SDI does help these, by adding a flat percentage chance that a nuke will be destroyed BEFORE exploding, but that's an empire-wide thing.

That was my concern. AIs from my experience always nuke cities, whereas a human can discern better options, like targeting a unit which is centrally located between two cities (thus inflicting damage to both cities)

I'd tried doing something related to this in my mod. My original goal was that SDI would have a 75% chance of intercepting an Atomic Bomb, a 50% chance against a Nuclear Missile, and a 25% chance against a Planet Buster. This percentage would then be modified by how many Orbital Defense Pods the defender had and how many Geosynchronous Survey Pods the attacker had. I could then go from there and adjust things based on the contents of nearby cities; this is where your comment comes in, since it'd be easier to then expand this to providing nearby units with temporary nuke protection.

Unless the AIs can be taught discretion regarding nukes, then I think the best option is to up the flavor of the SDI and/ or place it immediately after the availability of Atomic Bombs (it might be there already – I’m at work right now so can’t check). That way hopefully only one round of nukes occur before there is an option for some sort of defense.

the way I've always understood it, there are three main reasons:
1> Strategic resources limit the number of "offense" military units you can produce. Without them, a player could just keep churning out tanks until he had an unstoppable army, and then roll over everyone else. (This is exactly what'd happen in the earlier civ games.)
2> Most strategic resources are placed in less hospitable areas. Without them, there'd be little reason to settle near deserts, marsh, or tundra. But these are where the most Oil is found, along with many other resources. Hills, too; they're okay production normally, but hills are also where many of the best resources are found.
3> Rarity drives action. You hit the industrial era, where you were planning on putting factories in all of your cities. But wait, you don't have any coal. So, since you just can't afford to NOT have factories (the lack of production cripples you in a wonder race), this forces you to go get some instead of just sitting back in your comfortable little starting area for six millenia. To go get some, you might be able to find some small island to colonize, but more likely, you have to go to war.

Several observations:

1) I think partially I need to adjust my playing style so that I place myself nearer where likely deposits of oil are likely to occur, as usually I find myself not having oil in my games.
2) In regards to Culture Bombs from hostile entities, the AIs don’t place any more significance on winning/ losing tiles containing strategic resources than they do any other tiles. I’ve Culture Bombed Uranium Mines away from AIs, and the AIs merely sent strong rebukes. I then nuked the AIs to death with the bombs I built from the mines they originally owned. I understand that the AIs really can’t put two-and-two together in this regards, and that this is a rare event (not very often a uranium mine ends up in a turf war like I described above), but it is an exploitable option.
3) For the aircraft which can be built from the Oil and Aluminum strategic resources there is the counter of the Air Defense units, and there is also an ersatz offensive unit in the Guided Missile (i.e. if you don’t have oil or aluminum, you still have the option where you can still reach out and touch someone at a distance with missiles). For Uranium there is the SDI as a defensive counter, however currently there is no similar ersatz unit available to simulate a nuke’s offensive capability. Do you plan to put some sort of offensive capability in for those civs who don’t end up with uranium?

Pretty well, actually. They build the Manufactories and stuff just fine.

That’s good to hear! I’ll have to file that away for if I ever get into scenario development for ciV.

D

 

[Spatzimaus]

AIs from my experience always nuke cities, whereas a human can discern better options, like targeting a unit which is centrally located between two cities (thus inflicting damage to both cities)

Except I think that if you do it that way, it does less damage to the cities. Also, if the cities have to be at least 4 hexes apart, then it'd be VERY hard to catch two cities in the blast radius of a single nuke. So it's not too bad for them to do it that way. My only complaint, as noted before, is that there's no information being passed through the Lua events to show where the nuke is heading, which makes it impossible to add any positional defense logic. And that means that units out in the open are basically defenseless other than SDI's flat rate.

The only way, then, to prevent field units from being vulnerable would be to make a way to be ENTIRELY immune to nukes. That is, two SDI-type projects that each add 50% nuke interception, to where you can get to 100% interception by the late Fusion Era. (In other words, turn the Orbital Defense Pod into a one-per-player Project that adds the other 50%, instead of it being just another variant of the Perimeter Defense and Gravity Shield.) I don't know if the AI can handle that, though; it wouldn't know that its nukes would be useless against that player, so I'd have to handle it crudely by making nukes go obsolete at that tech.

Unless the AIs can be taught discretion regarding nukes, then I think the best option is to up the flavor of the SDI and/ or place it immediately after the availability of Atomic Bombs (it might be there already – I’m at work right now so can’t check).

It'd make no sense for SDI to come before Nuclear Missiles. Right now, Atomic Bombs are at T13, Nuclear Missiles at T14, and SDI is at T15 (at a tech that is NOT in line with the nuke techs, so you could in theory unlock it before getting any nukes). The buildings that reduce nuke damage to cities are at T11, T15, T19, and T20, and all but that first one (obviously) are in line with the nuke techs.

As I said before, though, I COULD just make the Manhattan Project give 20% Interception. (Then have SDI give 40%, and if I make the Orbital Defense Pod give the last 40%, you're at 100.) This'd just reflect the idea that a civ that knows how to make nukes would have taken basic precautions against nuclear attack, like underground command centers or Looking Glass, which'd make some nuclear attacks ineffective. This'd give you at least a little defense against nukes pre-SDI, and more importantly, it'd give players who DON'T intend to use nukes (like me) more of a reason to go up that research path.

1) I think partially I need to adjust my playing style so that I place myself nearer where likely deposits of oil are likely to occur, as usually I find myself not having oil in my games.

I've done the same. In my current game (Small Continents / Standard), my island is a nice temperate area, but with one small all-desert peninsula and a small tundra peninsula up north. I didn't settle near either of those, but I just KNOW that one of those areas is going to have oil in it when the time comes, so I'm prepared to drop a settler eventually.

2) In regards to Culture Bombs from hostile entities, the AIs don’t place any more significance on winning/ losing tiles containing strategic resources than they do any other tiles.

No, they do. It's tied to the same value system that the AI uses for placing settlers, as far as I can tell. The thing is, the amount that a strategic resource adds to that value is not huge, and there's no adjustment for rarity, so I'm pretty sure the AI would value grabbing its only Uranium the same as grabbing its tenth now-obsolete Iron deposit. So it doesn't think it's enough to immediately go to war over, although it WOULD then trigger the "need resource" logic and make it more likely the AI would start a war with you soon (the same as if you'd had that deposit in the first place). But the difference isn't huge, because the devs obviously didn't want that type of behavior to dominate.

For Uranium there is the SDI as a defensive counter, however currently there is no similar ersatz unit available to simulate a nuke’s offensive capability. Do you plan to put some sort of offensive capability in for those civs who don’t end up with uranium?

There are plenty of offensive weapons for people without uranium, but no, nothing in the "consumable" weapon category. There IS one nuke-type weapon that doesn't require Uranium, but that's the Subspace Generator (T23), which doesn't really count. (I still have to check, though, because SDI MIGHT have a 50% chance of destroying the Subspace Generator, which'd be bad.)

This was a deliberate choice by me: in the future eras, EVERYTHING requires a resource. You just have to tailor your army (and therefore your style of offense) to the resources you have. Laser Infantry are the last combat unit to not require a resource, so they're always an option to cover your basic defense. (Geosynch Survey Pods are resourceless, but they can't attack and so don't count.) But if you don't have Uranium, you're only missing out on the nuke units, Plasma Artillery, Needlejet, Vertol, and Leviathan. You can still build the Psi units, the Skimmer, the Gravtank and Mobile Shield, the "organic" units (Ranger/Troll/Doppelganger), the orbital weapons, and all of the Titans.
It's still doable. I almost never build nuke units, myself, so you don't NEED them to still win.

Now, once upon a time I'd wanted to add Singularity Missiles, sort of a non-nuclear Guided Missile analogue that could one-shot a Bolo. The idea was that, instead of building them directly, you'd build the Singularity Inductor (which'd be a national wonder in the Nanotech Era) and it'd generate these units every few turns (like Civ4 had). I eventually dropped that idea for practical reasons.

 

[dinobot386]

I checked again, its still doing it

90 culture spent on Representation, next turn 8/160 culture until next policy, found a city, next turn 17/175, continue, after next city it shoots up to 295 culture (i had spent another on mediocraty for a great scientist)

 

[Spatzimaus]

90 culture spent on Representation, next turn 8/160 culture until next policy, found a city, next turn 17/175, continue, after next city it shoots up to 295 culture (i had spent another on mediocraty for a great scientist)

Yeah, I'm not seeing the problem. That's what it's SUPPOSED to do. Try the exact chain of events again, but taking something else instead of Representation, and the cost would be even higher. (Even better, go into FireTuner and turn Representation off and something else on, and you'll see the effect directly.) That's why I gave the numbers in my earlier posts; I held as much as possible constant while comparing the "with Representation" to "without".

Effectively, it goes like this:
> There's a base cost. It's a semi-exponential function (25 + 6*(N-1)^1.7, I think, but there's some screwiness involved with how it applies the multiplier for difficulty level).
> Each city adds +33% to this, normally. With Representation it's only +22%.
> Round to a multiple of 5.

That means that if you've got ~3 cities and take Representation right around the time you'd settle a fourth, then there'll be a strange transition period where the cost of the next policy doesn't appear to go up much at all. (Wait even longer to take it, and there's a chance it'll actually go DOWN.) But that's just a function of switching from one progression curve to the other.

It's not broken.

 

[dinobot386]

Ok, sorry about that

 

[Spatzimaus]

Ok, sorry about that

No biggie. The devs didn't really document that one very well when they implemented it, and I don't think they paid much attention to the balance side of it either. (Really, a big cost discount and a 10-turn Golden Age, when policies in some of the other trees do a 4-turn golden age and nothing else?)

Just to verify, I tried modding it to flip the sign to +33 instead of -33, and things got WAY more expensive. This makes me think that I could put my "SMAC Start" policy back in and get the progression to still follow a more-or-less comparable slope just by playing with these values a bit. I'll look into that for the next version, because I'd really love to get the negative-happiness buildings working again.

 

[Darsnan]

It'd make no sense for SDI to come before Nuclear Missiles. Right now, Atomic Bombs are at T13, Nuclear Missiles at T14, and SDI is at T15 (at a tech that is NOT in line with the nuke techs, so you could in theory unlock it before getting any nukes). The buildings that reduce nuke damage to cities are at T11, T15, T19, and T20, and all but that first one (obviously) are in line with the nuke techs.

One of the things I said when I started helping out on this project was that I wouldn’t fixate on one thing, yet it seems I’ve been doing just that recently, so I do appologize for this. And I do agree that the SDI is in the right position relative to the nucleart weapons. I also believe you’ve done everything you can from your end regarding trying to make this system of nuclear warfare more balanced. However right now I am still seeing the situation as one where you HAVE to beeline for and thru Atomic Bombs, missiles, and SDI, and immediately start slinging the Atomic Bombs as soon as you manufacture them, lest you lose them when the AIs initiate their own nuclear warfare against you. Or to put it another way, the game should be about choices, and here there really isn’t any choice to be made (at least from my experience).

This'd just reflect the idea that a civ that knows how to make nukes would have taken basic precautions against nuclear attack, like underground command centers or Looking Glass, which'd make some nuclear attacks ineffective. This'd give you at least a little defense against nukes pre-SDI, and more importantly, it'd give players who DON'T intend to use nukes (like me) more of a reason to go up that research path

So what game settings do you use? I’m curious how you can get away with not building nukes, and if this revolves around set-up options?

D