Caster of Magic Release thread : latest version 6.06!

[Seravy]

I don't like the idea of starting with settlers.

[namad]

"Dark Elf +20 "
+20 from where they were before? which was a big negative? or +20 total? which ends up being a buff of like +50?

[Seravy]

"Dark Elf +20 "
+20 from where they were before? which was a big negative? or +20 total? which ends up being a buff of like +50?

All numbers are the new total.

[Nelphine]

I certainly accept disliking starting with settlers. I'd be interested in hearing your thoughts against it though (especially from the standpoint of a test).

My reasoning for it is from early civ games that sometimes had 2 settlers, and from moo2, where you had the option to start at different tech levels.

I found in both these games that the extra starting resources did make a huge difference on the game. I know I tended to play advanced games in moo2, specifically because I like the long term end game better than the brutal rush tactics of an early game. So that's why I'm interested in seeing if such an option could help alleviate some of the concerns you presented.

I'm certainly not overly attached to the idea - but the more I understand, the more I might be able to suggest something else that might answer some of your concerns.

[Seravy]

I rather have something like gold than settlers, as those just make starting at a bad location worse. When everyone else gets 3 good cities and you are stuck on a desert island, that's not helping.
The other problem, this diminishes the effect of not just random factors, but the difficulty level setting as well - The AI having more production means they finish the "mandatory" stuff in a shorter time than the player.
The third problem, it opens up possibly unbalanced rush tactics (not the settlers though, but any other resource), kinda like the "barracks first" tactic in Warcraft where you could use your starting resources on building some basic units while the enemy was busy making their town hall. (except here you would be buying 9x swordsmen instead of your sawmill, or something like that)
(btw for a same reason Mana Focusing, Archmage and Sage Master are powerful - you get resources for free in the early turns.)

If anything, I'd prefer starting with a more advanced game state (3-5 cities per player with some units and buildings, maybe some starting uncommon spells) but the game isn't balanced for that so it'd be a bad idea. Sorcery and Chaos are ultra powerful in the late game (especially Sorcery) for the price of weak early game options - skipping the early game makes there outright superior to anything else, especially as realms strong early can't use their advantage either.

btw I'm playing a random game on hard and I rolled Lizardmen, I admit I had a very good start (2 neutrals, 1 noble, mithril 7 tiles from my capital) but my graph grows like 3 times faster than anyone else's including the Myrran wizard. I didn't have any special retorts or anything either, just 11 books with mana focusing. idk if the new Historian overrates population (which I have lots of), and if not, how much is the fault of lizardmen, and how much is from the starting conditions (and possible bad AI starts). I guess I'll know more after the game, one thing is certain, I won't find out how much the AI changes affect anything this game.

[zitro1987]

I was playing against a powerful lizardmen wizard in my 3.02 'hard' game, but was easy to damage the empire with flying, resist-magic warships. Despite the high charts, especially in military might, this wizard didn't seem that intimidating military-wise (mostly unenchanted lizard stacks, some chimeras). The wizard often attacked my cities with stacks of units, but for the most part, they were defeatable with normal defenders like catapults, magicians, halberdiers, wolf raiders, etc. I expected worse I guess.

Note: an interesting idea to counter the sprawling nature of lizardmen empire is to forbid 'city walls', weakening their defensive capabilities.

[Nelphine]

I don't think AI are remotely overpowered for the AI. Strong, sure, but not overpowered (except maybe that they stengthen the other wizards by preventing the human from expanding much.)

The strengths of lizardmen are in things like colonizing 20 cities with the full knowledge you'll lose 4 or 5 - because any other race would only get 8 cities in that time, which still leaves the lizardman 8+ cities up.

But the AI CAN'T leave his cities undefended. He can't focus on spamming settlers that fast.

Similarly, my lizardman armies are unbeatable because I can choose which to buff, and go crazy on it - and use them on offense early enough that the AI can't stop it. The AI on the other hand, has to spend too much on defense, too much being able to deal with any situation (because the AI can't actually understand the situation), whereas I can hyper specialize.

I would not want lizardmen AI weakened. Other AI are certainly stronger than them. But as a human, I can abuse their strengths and minimize their weaknesses.


@settlers:
Things like starting in a desert island will demolish anyone. It doesn't happen often, but no amount of tweaking will fix it (any more than a life/chaos player who starts on a one city island, not on the shore, and doesn't get chaos channels or archangel).

Yes the AI will get a bog boost from settlers - but I fail to see how they would get a smaller boost from multiple cities. And with settlers, the human can make better choices about the 2 bonus cities, and do a better job of mitigating a bad starting city (except on a desert island). Thus, starting with settlers should (all else being equal) decrease the gap between AI and human - because the human will choose better city spots. Of course the AI will get a huge boost from extra cities, but that will happen regardless, whether someone starts with the current build, settlers, or bonus cities.

I specifically like settlers, instead of gold, extra cities, spells, a starting hero, or any other resource, because the human can use settlers better than the AI (by choosing a spot that is best for them in that came at that time, instead of choosing the best spot based on a formula that gas to cover many scenarios at once); and because in the current game, the decision to build settlers competes with developing your capital, or building military. Due to AI bonuses it doesn't care nearly as much about that decision (especially on higher difficulties) because it simply buys things constantly.

And the early game is often decided because of the choice the human had to make in those early turns - not choosing settlers early enough means you don't die, but usually leads to an agonizing slow death in the late mid game.

[Nelphine]

Alternatively, probably the biggest early game factor that decides a wizards fate is the starting city. Could the map generator be tweaked to give a specific tile for the starting spots? (Must be population 14-17, no 'ores' (including wild game or other special things) on poor, 1 tier 1 ore on averafe, 1 tier 2 ore on rich, always on a coastline)

[Seravy]

Alternatively, probably the biggest early game factor that decides a wizards fate is the starting city. Could the map generator be tweaked to give a specific tile for the starting spots? (Must be population 14-17, no 'ores' (including wild game or other special things) on poor, 1 tier 1 ore on averafe, 1 tier 2 ore on rich, always on a coastline)

No, I don't want to bring back the 75% chance to crash when starting a new game on Tiny or Small landmass due to not having enough of those spots on the map. (and that was with 12 pop required)

The "max 1 ores" restriction sounds safe. Always on coastline is interesting as it also means most very good spots won't get picked. (water doesn't produce that much food)
Restricting population from the other end (no more than 15) should be also doable.

I'm not sure I like losing the chance to have a good capital though - I believe the area around it (such as a pop 25 spot with 2 gems) is equally relevant, and the number of neutral cities far outweights even the best ores so it wouldn't be effective nearly enough to be worth losing the fun of having those resources occasionally.

[Nelphine]

I don't think the area around it has as big an effect. It's very important, but the ai is going to get settlers faster (in general, at least safely), and it already knows where the best city spots are (or at least the standard definition of better - i'm still shocked the system is so tilted towards max pop. I'd always considered the difference between a pop 10 city or a pop 15 city to really be negligible, at least until end game, but there such a huge difference on growth rates, it's astounding.)

Actually speaking of, this is another argument for making growth rates closer together. Max pop is nice, but it should be nice for .. the max pop. It shouldn't also be such a massive contributor towards actually getting TO that population. Currently, growth rate is linear based on max pop of the tile. So it goes from 1 to 25. That's a huge difference, even from 10 to 15. If you changed that to, oh, a bell curve from 5 to 15, that would make a huge difference in how important those secondary cities were, and it would mean a pop 12 city isn't MASSIVELY worse than a pop 18 city.

Ores of course would continue to be a huge influence, but that's one that the players can control via the minerals choice at the start.

Sorry, back to my original point - AI already is going to get good cities (even if not the best), regardless of whether they are close or not. So the area around the capital only really matters for the human player. Standardizing the capital city will normalize that for the AI, and you'll be able to get an expected growth for AI (since they'll have a standard capital, and will always get a city of at least x goodness, including time to actually colonize it, from later settlers). Which will allow you to more easily balance the ai bonuses.


And while it's nice to have the option of a super nice capital, you're still leaving the option of having a super nice neighborhood (which is more important anyway).

(Also as a note: almost all my impossible and extreme games have been without taking neutral cities. you have systems already in place to balance the power of neutrals, at least for the human. And as I've been discussing, AI isn't super good at getting neutrals anyway. So if the player chooses to have a game where they need neutrals to do well, that choice is already on them. So i wouldn't worry about it as a factor, even if it IS a big one. Big risks (wanting lots of neutrals) comes with big punishments when it doesn't pay off.)