I did some quick tests with the 2) mod and it's really causing every city to look the same. The improvements don't cost enough to demand that the player specializes. Since you are required to build housing room in order to level the city, by the time you have to choose what to build you always have the population to build everything. I'm thinking much higher limits would be in order after level 3. It is probably better to tweak the tile limits for cities, coupled with the 1) mod (and a +1 food cost for new cities) that should make it very hard to build low-level towns effectively.
I also tried increasing the Zoc, but frankly, I don't like it. While there wasn't a hard-cap on the range to the nearest settlement, it caused the land to become segmented way too fast. The game isn't balanced for a ZoC war, and the Ai REALLY isn't up to the task.
Current 1.3 beta tile limits for non-capital cities:
level 1 = 4
level 2 = 12
level 3 = 24
level 4 = 40
level 5 = 60
I guess I would change this to
level 1 = 2 (1 food = 2 tile per food)
level 2 = 6 (2 food = 3 tile per food)
level 3 = 12 (4 food = 3 tile per food)
level 4 = 33 (11 food = 3 tile per food)
level 5 = 62 (32 food = 2 tile per food)
However... this is really just doing what the 1) mod is doing anyway, but in a less elegant way. In addition, the tiles for improvements really aren't balanced for their efficiency. Some will cost 4 and do little, and some will cost 1 and do much.
I guess the beauty of the 2) mod drafted in the previous post is that it sets a pretty high limit on how many studies you can effectively build per food. But the downside is that it leaves room for a lot of other buildings to be built instead, which leads away from specialization.
Re-thinking the 2) mod from a perspective of food cost vs study cost, assuming the player always chooses the city to improve on tech research
Level 1 city can hold 2 of each building (2 per food) (max research: 2) (research per food = 2)
Level 2 city can hold 4 of each building (2 per food) (max research: 4*1.2=4.8) (research per food = 2.8)
Level 3 city can hold 8 of each building (2 per food) (max research: 8*1.5=12) (research per food = 3) (libraries can double)
Level 4 city can hold 22 of each building (2 per food) (max research: 22*1.9=41.8) (research per food = 3.8)
Level 5 city can hold xx of each building
This way, a specialized city can outproduce a non-specialized city in food per research cost, especially if secondary buildings come into play that boost % further. But... already at a level 3 city we have a city that uses a massive amount of tiles. Cities will grow huge and cover the map again.
Maybe the solution is to have each city level natively increase the efficiency of the food vs research cost of each built study. I will have to experiment some, brb.
Alright, please comment on this solution:
Citylevels automatically increase the efficiency of pop vs output of the repeatable buildings (through level-ups). But, consider that a level 2 city would automatically employ MORE men in a study and produce a higher output. For example
Level 1 city = 1 study costs 5 citizens, produces 1 tech research
Level 2 city = 1 study automatically costs 6 citizens (20%), produces 1.2 tech research
Level 2 city = 1 study automatically costs 7 citizens (40%), produces 1.4 tech research
Level 3 city = 1 study automatically costs 8 citizens (60%), produces 1.6 tech research
Level 4 city xxx
What does this solve? The cities will be reduced in size on the world map. Before, you needed 30 tiles to produce 1 base research. Now, at level 3 cities, you'll need only 19. So, a massive amount of studies won't cover the entire map anymore. This way, we won't end up in a situation where cities grow too big. Indeed, it is imaginable to further increase the city-levelup modifiers. The repeatable building limit per city would be tweaked for this, so that there is a situation where high-level cities automatically (without choosing levelup bonus) gives you a better FOOD per TECH cost, but with this solution the result is not huge megacity spam.