Are science buildings overpowered?

[Grubsnik]

This is to all the people talking about how Science buildings are just fine.

A strategy is to make a choice, where other choices are also viable.

Do you ever look at your pop 12 city and think, I think I'll make this my -city and just ignore science buildings there?

You ever look at a 12 pop -city and think, I'll just focus on barracks and armoury here, no need for a library?

Of course you haven't, not if you are playing seriously at least.
Getting unis up and running is never a real choice, the only potential strategy choice is about how far down the Archery->Construction->Machinery techline you need to go in order to stay alive long enough for you to get your universities.

You can get Education without even researching mining, though most will at least get to construction on the way, but otherwise it seems like a sound choice, the faster you get to education and those lovely universities, the faster everything else will come to pass.

You can also get to Chemistry without researching writing (or even pottery) but I'm fairly sure that going through Education and building libraries, NC and universities will get you there faster than just beelining Chemistry from the start, despite having nearly no overlap between the two techlines.

Education beeline: 2416 beakers
Chemistry beeline: 4623 beakers
Overlap: 264 beakers

The same case can be made for philosophy/NC and machinery
Machinery beeline: 1770 beakers
Philosophy beeline: 368 beakers (this also implies you are skipping pottery and calender in the machinery beeline!)
I think we'll be getting those 368 beakers back, with interest long before we reach machinery.

All in all, it seems to me like the only strategic choices you can make are to try and deviate from the main course in time to gain a temporary military advantage.

You can go Philosophy->Machinery for a Crossbowmen offensive
You can go Education->Dynamite for an artillery offensive(this is where your crossbowmen from the previous gambit go to die)
You can go Scientific theory->Flight for a major bombing campaign (though you will need oil for this, so if you've neglected the nautical tech-line, it may be faster to go directly to plastics first for labs)

I even did a few test runs to compare going straight for Construction vs. getting pottery+writing and building a library first, turns out the detour into writing costs you a grand total of 1-2 turns total, nothing more, but leaves you ahead by 100 beakers of science.

So in conclusion, you pretty much need to prioritize science at all points in the game, meaning it never really becomes a valuable strategic choice, just the default way the game is played.

 

[Matthew.]

Tech costs go up significantly at certain points and science buildings are for bringing you to the next tech "level". Obviously getting to that next level sooner than later is the ideal choice, but there is enough "choices" in between where I don't think it is a problem. I regularly pick up techs like iron working before uni's, muskets/cannon before schools, fertilizer (sometimes railroads) before labs, etc.

Now of course one can say "science buildings are OP for getting the fastest tech rates" but that is just plain stupid. That is like saying "when trying to get the most gpt, gold buildings are OP" or "when trying to get the most hammers, production buildings are OP".

That is what science buildings are designed to do: to increase tech rate.

The real issue goes back to the AI, as always. If the AI was competent and the human attempts to rush science tech while ignoring military (thus able to ignore production/gold), it should be able to crush the player. Yet we have plenty of examples of players winning Deity games with 2 CB's for the entire game.

 

[Grubsnik]

Perhaps it's the whole "a few ranged units will keep me safe" that is messing stuff up, but honestly, for a sampling of 4 "specialized" cities, the production city, the cashcow, the science city and the culture farm. You'd get these values:

The production city:
This is usually a decent food with plenty of hills location, so you stick a workshop in there and later on a factory, build farms and mine some hills, along with at least a library and a university, so you can run 2 scientists

The cashcow:
This is usually a location with few hills but with at least 1 or 2 luxuries and some decent food possibilites. So you improve the luxuries, build farms when there is water and tradepost the rest.
Building is slow, so you buy library, university and public school(when gold permits), while building market, bank and potentially the stock exchange, a workshop will be added to build iron works and a factory might happen if you have spare coal, but neither is a priority.

The science city:
The location includes a mountain, and at least 2 good food spots, so you can support 2 and later on 4 scientists. You build or buy library, university and observatory and cheer as your low pop city outputs ~50-100 science pr. turn, later infrastructure is added on a "what will make it output more science" basis, a city connection ensures it doesn't become a financial drain, any great scientists generated are walked off and settled in the capital. This is the specialized science city.

The culture farm:
Location includes enough food to run a bunch of specialists and potentially a river, so you can build a garden for those extra GP points. The city is where you build all your culture-guilds, in case you don't want them to take up slots/production in your capital. Of course, once you have a pop 12+ city just standing around with a garden, you'll want to add in the science buildings.

Sure, nobody really does specialized cities to a notable extent anymore, but that doesn't change the fact that no matter what kind of city you are making, getting science buildings up and running in them, is not a choice, it's a must have.

Once you have Iron works in place, getting a workshop for a low production city may not really make sense, building markets and banks in a production city is also totally irrelevant, but getting those science buildings up and running is never ever a choice, and that may seem to indicate that they are just a tad more important than they need to be.