Actually I liked Civ 6 Loyalty. I think it worked really well and was logical. Yes it was difficult to settle a colony next to an opposing Civ, but then it should be. However, it was possible to do it, if you had leaned heavily into colonising as a strategy. Some civs are set up to be colonisers, and with the right civic cards and wonders and governors, it was totally a legitimate tactic. That is how it should work. If you think it is advantageous to go and settle far off lands, then you need to be set up to do that, not all civs should be able to do it easily.
I think that is a big step up from the way things work in Civ 7
I don't know if the settlement limit is a more fun implementation. Corruption was always disliked by many people, but the fact that is passive and hard to calculate was part of it's appeal. Settlement limit is such a blunt, simplistic rule, it leads to boring gameplay. You can either stick to your settlement limit (which I mostly do in many games, because it feels better) in which case you just feel pretty restricted. Or you ignore it and min max and do lots of busywork to manage it, which isn't fun either. Either way, it doesn't feel immersive, the blunt nature reminds you that you are playing a video game at all times.
Passive, under the hood mechanics is something the game needs more of, and less obvious arbitrary numbers that make you feel like you are playing a video game.
Yeah, I liked the loyalty in 6. It was best in that it prevented people sneaking in a tiny city in between all your others. At times it sucked (trying to get a colony going), and other times I didn't like it because it was too easy to cheese (basically as soon as you captured your opponent's biggest city, suddenly you never had loyalty issues with their other former cities and could actually culturally flip the rest of their empire).
I don't mind the settlement limit, but I agree that it can use more. I'm definitely in favour of changing it to some sort of "administration" cost, where cities cost you more settlement limit than towns do. And you could even add on more to it - settle a city outside your trade network, it costs an extra point to the settlement limit. Captured/occupied cities should maybe cost you an extra settlement point in their occupation period. And then often you just get so many happiness bonuses from other sources (city parks can give you like 60 happiness in some settlements, pavilions can give you like 10+, etc...) it's just too easy to cover the max -35 penalty you can hit on your biggest cities.
Civ keeps running into problems trying to figure out if they want nice round values, or they want weird calculations. Like in 7, they simplified adjacency so you don't have this whole minor/standard/major, or some of the weirder calculations from earlier versions. But at the same time, the growth formula is some complex cubic equation, and specialists still end up with cases where they give you 0.5 hammers and cost you 1.3 food in some cases. Sure, it sometimes was a struggle to understand why one trade route was 12.6 gold and another was 10.8 in civ 6, but at the same time, do I care? Computers can do math for me, tell me the biggest number and I'm happy.
Even though I didn't like loyalty I agree with both of these takes.
For me, I think the settlement limit is a bad idea, because it's an arbitrary value.
So why is it 5 or 10 or why is it different for some Civs and what about picking up a Civic let's you place down a new city without a happiness penalty?
What's funny is that Civ5 had the same exact system, where the happiness was essentially the settlement limit. So the more happiness you had the more cities you could settle.
And when you played on lower difficulties with higher base happiness, you could literally play wider - if we ignore the culture / science / great person penalties.
I agree with some others about an administration value, but I think we could tie it to the Corruption / Loyalty value, or the Happiness value or something along those lines.
So it's not arbitrary and it makes sense (more cities means more corruption means you lose them more, and acts as a settlement limit).
Also, I think Civ would benefit heavily from under the hood mechanics.
If anyone here plays Rimworld, that's a game with heavy simulation systems and a story based AI events system.
In many ways, Civilisation could benefit from a similar systems based design, where there's many interlocking mechanics to achieve a natural gameplay feel - and the "Game" is built on top of the system rather than the other way around.
Other systems based games, like Streets of Rogue, is one I quite enjoy. In that game, all the different objects and systems interact with each other and the game is built on top.
Breath of the Wild is the same.
Strategy games aren't usually system based, but I think the future for replayable strategy games is definitely system based.