Spatzimaus
Mad Scientist
All of the bonuses the AIs get are multiplicative. So this sort of change won't affect the relative balance of AI-vs.-player; the AI might only need 80% as much production to build a building, but both the AI and the human will be getting the same +X added to the city's base values.
AI confusion is not a problem; you can't demolish buildings that have zero maintenance. Since you can't build them (cost -1), the AI basically pretends they don't exist and just benefits from the effects of the buildings without caring about where the effects come from.
And in case you didn't know, I've actually been using this mechanism for several versions now; there's a hidden +1 Happiness building that gets added and removed to cities whenever you use an Empath specialist slot. Since specialists can have Flavor values, the AI even knows that using those specialists results in happiness, even if it doesn't know HOW.
But that'd still cost you one turn embarking, and one turn disembarking. That's what I'm trying to avoid here; right now, if you've got a 1-hex channel of water between two landmasses, it takes several turns to get across. This way, that water would only cost you one movement point.
I've got other ways to improve on this, but that's the basic idea. One issue is pathfinding; the AI would still think going around by land would be faster, because it wouldn't know about the movement refunding. I'm working on that.
In my original design, launching the ship would also create a "First Nest" Natural Wonder near your capital, a permanent wonder that'd give extra incentive for being the first to launch, but that would (at least initially) give that civ heavier mindworm spawns. In the end I dropped that as too problematic, both mechanically and graphically. But the idea of having the launching civ have the heaviest initial infestation appealed to me.
That's basically what they are already. Every turn they're alive, they have a chance of spawning various types of psi units. But I had to do them as units (which can be killed) instead of Improvements (like the barb camp) for practical reasons. What I'm trying to do is make them a bit tougher to kill.
Unfortunately, you can't do that; the combat event for nukes has no information about who or what is being affected by the blast. There IS a Lua function to see if a specific unit has been nuked, but the overhead necessary for this makes it impractical to use for these sorts of triggered effects. I'm seeing if there's a way around that limitation, but no luck so far.
Only the Fusion Lab (1 unit per) and Supercollider (10 units, but it's a Wonder).
And that's not enough to avoid the issue I'm describing. Right now, it's possible for a map to have no Uranium whatsoever. On smaller maps, this is even fairly likely. The reason is that resource distribution is probabilistic; the game sees a Hills tile and says "this has a 1 in 12 chance of having a resource on it." On a huge map, where there might be a hundred hills, no problem, but on a small map that might only have a handful? And of those 1/12, only a small fraction will have a given resource. The result is that it's easily possible for a map to have no deposits of Aluminum, Uranium, et cetera. depending on the exact geography.
This was absolutely crippling with Coal; a civ without Coal simply could not keep up, because of how powerful Factories are, and in a game where you were expected to go into future eras it was just untenable. So, I put in an override, where no matter what the randomization gave, the game would auto-place an additional N/4 small deposits of Coal on the map, where N is the number of civs (including city-states). I reduced a few probabilities to compensate, but the result is that if before you'd see 0-10 deposits on the map, it's now more like 3-7; less randomization, same average. If you've played my mod in the past three or four months, you've benefited from this change. This still doesn't guarantee the coal deposits are in your territory, so there's still motivation to go to war, but it was enough to ensure that there'd at least be SOME out there to go after.
The question is whether I should do the same for the other strategics. Reduce randomness across the board, to prevent a map from crippling all players other than the one lucky enough to get a local deposit. I've done this from the start for my custom resources, and Iron and Horses are common enough that they don't need it, but the others?
First of all, the term SDI does not automatically mean orbital weapons. It includes those railguns, long-range interceptor missiles, airborne lasers, et cetera; the name has changed over the last couple decades, but it's the same programs as before. (Besides which, railguns wouldn't come any earlier in the tech tree than the early Digital era.)
Second, those other things, like AAA, are the reason why I added that minimum chance of nuke interception (30% for A-bombs, 20% for missiles, 10% for PBs). That's what the minimum chance represents, the possibility of using existing non-dedicated defensive systems to prevent a nuke from impacting or having its full effect. Maybe you got lucky and shot down the bomber, maybe you got everyone into bunkers in time.
Originally, I'd handled this through giving the Manhattan Project a 15/10/5% chance and reducing the SDI project accordingly (45/30/15, instead of its current 60/40/20); I talked about this in a couple posts made before the final version was released. In practice, this didn't work well; I needed to add the minima anyway to protect city-states, since they can't build projects, so there just wasn't much point in adding anything with that small of a percentage.
You just can't do that, at least with the XML/Lua we have at present. It's pretty much all or nothing. Hence the question.
