Now, as I actually thought about the whole thing on the bus as I came to my work, I'll throw my two cents on the matter.
On cost progression on buildings and units
- Production scales up with tech and civics, as those two things unlocks buildings, improvements and policies which rise your empire's general production.
- As such, the cost of almost everything rises alike. Look at military units: the cost of a warrior may remain the same, but as soon as you unlock the swordsman (via tech) the warrior is replaced by the swordsman, whose production cost is higher, on the building list.
- The unit cost-progression is, thus, equally defined by tech. The exceptions being settlers, builders and religious, only because they are not replaceable by newer units (tough builders surely could be, and maybe settlers too).
- The major motive for that is to mantain the flow of the game. As your empire grows and gains production power, the costs of things grow along. If not, by mid-to-late, every newborn city would be capable of spamming units and districts.
On production progression
- When you reach industrialization every well-designed older city can easily be producing 30-50 hammers per turn. With the wise planning of trade routes even newborn cities can produce 20+ hammers per turn. (And I've not even played with Gear-Happy Barbarossa yet... it's my next choice. I've played only with Brazil and Rome so far).
- Even on cities with poor locations hammer-wise, on mid-to-late game, you can boost it quickly enough with trade routes. On my actual run with the Romans a caravan for Rome gives +4 food and +5 hammers for the city, and the other 4 bigger cities give +4 and +4. So, five trade routes getting out from that city means a boost of +20 food and +21 hammers each turn. You leave them there for 15-20 turns and the city will be ready to take on all alone from then on.
- So, a newborn city having 20+ hammers each turn, without cost scaling, would mean that it could easily build districts within 3 turns! And I'm not even taking policies and wonders in count. THAT would be game-breaking.
-
So, my conclusion there is that it really needs to be a scaling up on costs of districts, or they would be really messed up by post-mid-game.
-
On how to scale
- Scaling by techs and civics have it's own reason to be. It's them which unlocks the things that boost production, both directly (like industrial district and some policies) and indirectly (things that boost the growth of the cities, for example). So scaling the costs by the same thing that makes the hammers grow looks like a fine tradeoff. Also, it gives some strategic decisions and flavours on how to play the game - like lagging behind in tech to focus in district production, as the op proposed. Or boosting science to increase production to have a fine edge on Tech AND Production, as I do. Superior edge on Tech + Production win wars before the war weariness starts to creep.
- Scaling by time is punishing for those who lags behind in science - as they lag behind they lack the means to increase production, but the costs keep on rising, making it even harder for them to catch up - and makes for a "run for the districts" madness. Just imagine the AI having to deal with that. It really looks like a terrible idea.
- Scaling by districts on same city means that it will be cheaper to have a lot of bad cities with just one district on it, vulgo ICS. No, thanks, I pass that.
- Scaling by districts of the same type already built. Now, that'd be interesting to see, as it punishes ICS., but it also punishes guys like me, who always build districts on the following order: Industrial, The one relative to the victory I'm trying to achieve, Comercial. Nah, thanks.
- Scaling not in production, but in maintenance. That made a lot of sense on civs 1-4, because of the sheer quantity of buildings there was. As you progressed to the modern era there were dozens and dozens of things to build, and every tech unlocked some more. I rarely achieved a point in which there weren't more things to build anymore. And then it make a lot of sense not pushing the costs up, leaving them to be built on 3-5 turns, and then pushing up on maintenance. But, seriously, look at the quantity of buildings we have now. And the way districts are locked by pop cap. If you could build them on 3-5 turns, what would be you doing when there were nothing more to build? Carpets of Doom???
-
I guess that's it. My conclusion is: we need that cost scaling, if we want districts (and sure I want, I loved them). And on the options presented, doing it by tech is the best way to balance things. Maybe that scaling by districts of the same tipe could be something to be tested on a mod, to see what actually happens. But hey, dropping Tech to produce more districts is not THE way to go, just ONE way to go. It is one strategy... maybe something I would even use as I beeline to Apprenticeship, but surely not on a whole game. Unlocking things by tech still beats producing districts a little bit quicker, for me. And I'm glad that Civ VI puts those tradeoff choices in front of us.