I think there is a very very strong core of ideas and concepts in Civ6, it just needs a fair amount of polish and reworking, so I hope Civ7 is more of an evolution than revolution.
I like the lower unit density (less micromanagement), but the strict 1 UPT makes you have to solve a sliding tile puzzle every time you move your units, which is frustrating totally uneccessary busywork, since the lower unit density basically makes the need for stacking limits superflous anyway.
Builder charges are a way simpler more elegant system than workers. The only way to improve this is the option to buy tile improvements. Honestly I’d get rid of workers/builders completely and use the district system for everything; all you really need is a Farming District, a Mining District, etc. Less units cluttering the map. If you are producing builders to then move around the map to then say make a farm you are essentially doing this anyways, just with far far more micromanagement involved.
I like the way the district system makes you “play the map”. If this game had more actual warfare in it it would make strategy a LOT more important than just “turtle in city” like earlier titles, since doing so leaves your districts open to pillaging.
I would honestly merge culture and religion. The way religion works in this game feels weirdly tacked on, as amusing as the Apostle Fights are. I’d bring back the ability to steal tiles with enough culture that earlier titles had too.
The concept of “Loyalty” is a good one, the implementation is the typical Civ6 misfire. It should be based on culture, not population. Basing it on population turns it into Yet Another Snowball Mechanic.
If you are going to keep the concept of Gold and Dark Ages, it had to be out of the player’s controls. It’s way way too easy to game. It’s a natural anti-snowball mechanic, but you’ll have to have the balls to actually punish the player in the lead
The AI being a pinata needs to be addressed first, last, and foremost. If the AI can’t or won’t do anything other than passively filling yield buckets to hit an arbitrary win condition, then the game becomes incredibly boring to anyone not a passive sandbox builder. I mean this is the franchise that made Ghandi a symbol of nuclear violence and it might as well be a Sim game.
Yeah, everything you mention has some good and bad. 1upt: good because less unit management. Bad because more unit micromanagement. Charges good because easy and convenient. Bad because too much hassle (especially late game when you get a flood or tornado and have to go through tile by tile repairing). Etc...
Some stuff like religion, as dumb as the apostle fights are, it's a big step up from previous games where you literally had no mechanism to defend religious spread. But then again, it's such a pain when you see a bunch of AI apostles walk all over the map and now you have figure out where to place your guys, block them in, worry about zone of control, still worry about fighting over rivers, deal with the weird way that religious units are embarked and you can attack into water, but you can't attack out of it, except sometimes when maybe you can because I still don't understand all the movement rules, etc...
Loyalty I'm a lot less pessimistic on than some people. I think it's actually one of the mechanisms that has the least downside. Yes, it's too dependent on population, and I absolutely hate how when I capture Paris suddenly Orleans becomes loyal to me because those Parisian citizens are immediately 100% mine. But loyalty at least has nuance, you can control its influence based on policy cards, governor placement, and even other stuff like religion and happiness, and it prevents the AI settling right on your doorstep and applies a natural DMZ that can be hard to encroach into.
But overall, figuring out a good way to make it tougher when you're in the lead will be the big key. I still want a challenge even if I am winning, but I also definitely don't want to invest 20 hours into a game and then be hit by a Mario Kart Blue Shell and be knocked back an era in tech. I do think the golden age system could have been that, except that it definitely hurts by being too easy to rack up score when you're ahead. Like if I just finished back to back golden ages and am coasting through the game, it's too easy to be the first to the next era, the first to build a national park, the first solar plant, the first size 20 city, piling up great people, etc... that it's almost impossible to miss the next golden age. There's a part of me that wonders how different the game would play with the simple change of dramatic ages mode being even more dramatic, where anytime you hit a golden age, you 100% are guaranteed a dark age the next era. Or maybe that's a little too mean, but if Dramatic Ages mode was basically that after a golden age, the next era either you hit the marker and go into a normal age, or you drop into a dark age, so that you can't chain back to back golden ages together. At least if there was something that forced a pause in the system the next era, you might still have to fight a little bit.