What needs to be done?

AI needs to be taught how to fight, period.
This is already second entry in a row where AI can't use naval and air power at all. AI ships just float there kinda menacingly, but they don't ever attack. The only ships I remember seeing to attack were from hostile independent powers. The only tactics AI can use for land forces is mindless swarm of units, power in numbers. Bring some of the wargames logic into this hex based map, I want to be rolled from time to time, so that I could have heroic and tragic tales of valiant but hopeless resistance, instead of so much winning. Winning all the time gets boring.

And do something with the religion system. How can you make such a religion system that managing its units is hands down the most tedious busywork in two installments back to back? Seriously, eliminate those pesky missionaries and leave it to buildings and influence missions, smth.
AI is very good at defense, but I think I remember reading the devs saying that the AI was too good and made playing unfun so they told it to basically be less aggressive.
 
Mathematical proofs are unlikely to be reconciled to something as arbitrary and subjective as "unfun".
Well there are also logical proofs. And I said the game was objectively poorly received not objectively unfun.
Which means your claims remain unfalsifiable, which of course renders them unusable (or at best, heavily assumptive / flawed) as a logical proof. That's the long and short of it.
What was the design intent? We have a few statements. Do they adequately explain the scope of features? If not, then the stated design intent is falsifiable. In this case, some other design intent must be behind those features. Without further primary source data, we have to use other dev statements inferentially, given a knowledge of the industry overall.

The conclusion will be the most likely explanation of the feature set until falsified by more convincing data.
Back in my day, it used to be enough to say "I don't like it". I don't know when this changed to require some kind of ironclad, objective mandate of correctness. That's generally not how games design has ever worked, nor has it generally been how games are typically received.
Video game culture was never like this. Being mad that people don't like a thing you like to the point of insisting they stop talking about it is post-reddit.
 
Well there are also logical proofs. And I said the game was objectively poorly received not objectively unfun.
Oxford commas are fun. But there's no design intent for making something poorly-received, so. Again, assumptive logic.
What was the design intent?
I answered this, in a part of the quote you didn't, uh, quote:
And impossible to answer for us, none of whom (presumably) are Firaxis developers.
You're also misusing "falsifiable". You cannot prove a design principle wrong with feelings. You can say "the game is unpopular", or "the game reviewed poorly". These are at least demonstrable, even if statistics aren't fully-formed and we only have what us, the public, have access to.

But slicing this apart into specific design principles and their relative success is very hard to do. There have been multiple threads that have attempted this! And, presumably, something that developers who want their games to be popular (or more popular) also attempt to triage the same. Obviously, we can't rule out that some developers might want to intentionally create an unpopular product. A mainstream, well-known franchise like Civilisation is probably (but not impossibly) not a candidate there :D
In this case, some other design intent must be behind those features.
Why? Designs can miss. It happens.
Without further primary source data, we have to use other dev statements inferentially, given a knowledge of the industry overall.
Why? There are a near-infinite of inferential possibilities given the possibility space that is "multiple years of games development". How do you choose?

(the answer is: subjectively, each according to our own experiences and worldviews)
Video game culture was never like this. Being mad that people don't like a thing you like to the point of insisting they stop talking about it is post-reddit.
I didn't say anyone shouldn't be talking about anything. Anyone else is not me, and I am not responsible for them, nor do I want to argue for them. But I do think wanting constructive, good faith discussion is a valid goal to strive for. Not really a goal I would think anyone would want to criticise, because of well, the inferences ;)

Regardless, people have been mad at things to the extent they've wanted to shut down discussion about said thing for eons. Drawing the line there isn't our job as posters, either.
 
I can see a classic mode being introduced in an expansion, I just have a hard time imagining how the era system could be made to feel smoother.

But as far as civ choice, a starting point for a classic mode would be to simply put all civs on the initial list of options, have some text informing players that they will have civ specific units and structures only during one of the eras, and maybe add some capital names for the civs to use in other eras if relevant to that civ.

After that, it gets more complicated to make it play smoothly and I don't know how they would do that. You'd have to polish a lot more to make it feel natural. Science and civics trees, era transitions, many systems would need to account for there now being players that remain the same civ. I suppose you'd also have to decide whether to implement it with the ability for the player to change their mind and actually switch when they get to that point, or go all in and say that's not possible in this mode. (Or what if a player wants to play 2 civs over 3 eras? Then you'd want to keep the option to switch but still make non-switching transitions smooth.)

I really do have a hard time imagining things could become that seamless with the way eras are.

As for the "what needs to be done" though, I think classic would be a nice to have because I don't think it's can "save" the game. The game needs to flesh out its own ideas to have a chance of living a fuller cycle, like 4, 5 and 6 and getting more than one expansion. Religion, victory paths, crises, and the modern age are all parts of the game that could use an expansion sized rework, not just a patch. I do wonder what other ideas might have already been planned for a potential expansion that we're not thinking of because there's so much to either improve or outright fix.
 
Last edited:
I think we can all agree Civ7 did not meet the mark for many in here, and even those who like the game, like me, see room for improvement.

The last few months we've got a big patch once a month, changing balance, changing era change mechanics, maps etc.

Now, if this continues, what do you think is the major part(s) of the game that needs improvement, or changes?

  • More and better maps? Or are you content with the recent additions?
  • More civs? Which in that case?
  • Addition of scenarios? Which in that case?
  • Better modding tools? Better modding documentation?
  • Removal of Denuvo? I know some see this as a must, looking at you @Dale , any other takers?
  • Changes to the age system (I think removal is probably not in the cards, but changes definitely could be?), if so, which would you like?
  • Balance? It's getting better, is it satisfactory for you now, or close to it?
  • Pricing? Not a part of the game per se, but the game is costly, too costly for some.

Let's give Firaxis something to chew on. :)

The only way to save Civ VII is with Classic Mode

Nothing else will have a significant impact
 
I think a potential classic mode should be decided when setting up the game and allow no switching. All civ choices available. Additionally I believe all opponents need to be locked into their civ as well.
Civ switching is so absurd in my book.
I won my 2nd full game yesterday, in the modern era as Hatshepsut of the stupid Meiji of Japan.
Now I have started a new game as Benjamin Franklin of the Mississippi people in the ancient era.
It makes no sense whatsoever.
I am sure, with a bit of work. They could give us the choice of staying with one Civ from start to finish.
All Civs should be available in all ages and they could give them bonuses and or units applicable for each age.
 
In Civ7 and Civ6 (even going back to Civ3, which I still play), the AI players understand how to send 2-3 units to gang up on one of my units. Strays, or singletons, get attacked.
I've seen that tactic both on land and on the water. I've not see much use of air power, the same as you.

But the AI players in both Civ6 and Civ7 don't attack a city very well. The best / most threatening city attacks I've seen were in Beyond Earth: Rising Tide, from AI ships. The oceans/coasts are perfectly flat terrain, meaning that their pathing algorithm worked. They would bring both ranged and melee ships, using them in the right order sometimes and in the wrong order sometimes. Wounded cities were especially vulnerable to capture or recapture.
Oh, yes, now that you've mentioned, I remembered those AI wolf packs in BE:RT. Yes, there the AI was more competent in taking cities, it was also helped by how much squishier the cities were there. Could they bring at least that AI aggression logic from over there into Civ 7? I have now played a game on an Archipelago map, and finally I have seen AI ships attacking my units, but they did it so timidly and with scattered units, they stood no chance against a human using admirals and commanders.

AI is very good at defense, but I think I remember reading the devs saying that the AI was too good and made playing unfun so they told it to basically be less aggressive.
In which way is AI very good at defense? Apart from throwing as many units at the frontline in the human wave form, as their bonuses allow? Which gets whittled away anyway.
 
I've just seen devastating AI attack yesterday in Civ7. The thing is that current game balance favors sea units over land ones and air over all else. This makes sense to force players to build more balanced military and usually plays toward player. But yesterday I had was against Axum and while land war was mixed, they launched like 10 of their unique ships and with most of my settlements being on shore, it was really tough. Not great at tactics, sure, but sheer number of firepower was enough to break my defenses.
 
So it seems that the devs are listening to suggestions 👌 Good work Firaxis!

Let's do a quick recap of what’s been discussed (and seen by the devs!) on the Discord feedback list:
  • The Golden Ages of Pirates - adding pirate forces into Exploration Age ✅ Patch 1.3.0
  • Wide & Tall Empires - rebalance population costs so that the really big cities would require feeding towns ✅ Patch 1.2.5 (actually achieved by rebalancing building costs)
  • Towns vs Cities - implement clear visual distinction between these (towns more rural, cities more packed)
  • Government System - Improve the Government UI/System so that it really matters what goverment you run
  • Roads - have some influence on their placement
  • Narrative Events - add more immersion to the game
  • Canals/bridges/dams - add these features as they were in Civ 6
  • Civ-switching - improve civ transition (including transfer of abilities to next civ, keep building style until overbuild etc)
  • Legacy Paths rework (including new victory conditions)
  • Map Improvements - map generation, size, number of players available
  • AI - better handling of airplanes, siege units, treasure fleets
  • Adjustment to ages' length - smoothen transition from Antiquity to Exploration (post-classical/medieval age)
 
In which way is AI very good at defense? Apart from throwing as many units at the frontline in the human wave form, as their bonuses allow? Which gets whittled away anyway.
It will engage you to bait you closer to city walls then retreat behind the walls. It's good and reserving forces behind defenses.
 
AI is very good at defense, but I think I remember reading the devs saying that the AI was too good and made playing unfun so they told it to basically be less aggressive.
Really? So they just dumbed down the AI to the point of being suicidally stupid at war to make the game "fun"?

Back in my early days of playing Civ 6 over three years ago, I harbored the silly delusion that the AI played smarter at the higher difficulty levels. Imagine my disappointment when I learned that, no, all that happens is the AI gets extra settlers, warriors, and yield boosts, but still plays like a moron. To me that is not "fun," and I think making the AI civs morons is precisely what leads to the human player snowballing in Civ 6, with the result being a tedious endgame.

If the devs really know how to make the AI smarter, then make it smarter, at least as an option for the higher levels!
 
Really? So they just dumbed down the AI to the point of being suicidally stupid at war to make the game "fun"?

Back in my early days of playing Civ 6 over three years ago, I harbored the silly delusion that the AI played smarter at the higher difficulty levels. Imagine my disappointment when I learned that, no, all that happens is the AI gets extra settlers, warriors, and yield boosts, but still plays like a moron. To me that is not "fun," and I think making the AI civs morons is precisely what leads to the human player snowballing in Civ 6, with the result being a tedious endgame.

If the devs really know how to make the AI smarter, then make it smarter, at least as an option for the higher levels!
The idea of giving AI players bigger starting bonuses (or other advantages, such as cheaper research or lower maintenance costs) goes back at least as far as Civ3. It was not unique to Civ6.
However, the AI players had different levels of aggression among the games in the franchise. Civ3 had an aggression setting when starting a game; Civ4 had options for "Always War".
Many players have observed that the Civ6 AI players are MUCH less aggressive than their predecessors.

Specific to Civ7, I have found the AI will try to *build* their way to a legacy path or a victory (in Modern). A Sovereign Harriet Tubman beat me to the World Bank. A Sovereign Isabella won by score when I couldn't get my World's Fair built before the age ran out. They regularly build space projects. The human player, having caught up, can usually -- but not inevitably -- reach a victory condition first.
 
Really? So they just dumbed down the AI to the point of being suicidally stupid at war to make the game "fun"?
I don't have the quote source but I do remember a developer video explaining that there was an issue of the AI always winning and that being an issue. However, this could have been a theoretical statement about design philosophy and maybe not what happened with Civ 7.

However, there is an issue in Civ 7 of the AI not pursuing victory conditions in the third age specifically and I do think the quote I have in mind was related to that. To be fair, skilled players can win in Modern in a very few number of turns, so perhaps it's that they calibrated the AI to not be so aggressive in trying to win.
 
I don't have the quote source but I do remember a developer video explaining that there was an issue of the AI always winning and that being an issue. However, this could have been a theoretical statement about design philosophy and maybe not what happened with Civ 7.

However, there is an issue in Civ 7 of the AI not pursuing victory conditions in the third age specifically and I do think the quote I have in mind was related to that. To be fair, skilled players can win in Modern in a very few number of turns, so perhaps it's that they calibrated the AI to not be so aggressive in trying to win.
I believe you're referring to "playing to lose" video by Soren Johnson
 
Moderator Action: Can we please not turn this thread into another "No Civ-Switching" thread! If it continues, thread bans will follow.
 
  • Towns vs Cities - implement clear visual distinction between these (towns more rural, cities more packed)
I don't think they'll ever change this to be as varied as I want but I hope they do. I ultimately would like more town types to choose from (they've started this but to fill some gaps), and this idea that you can go mostly towns only for a large empire (or high population). I see a trade off between investing in specialists for city development or investing in "empire infrastructure" like aqueducts, irrigation systems, roads.
  • Government System - Improve the Government UI/System so that it really matters what goverment you run
They could implement a mini Civ 6 style thing you can choose between having more traditions, a military bonus, etc.
  • Roads - have some influence on their placement
I think we need just a little more micromanagement of food supplies and trade. My ideal system has that the number of settlements a settlement directly connects to reduces its tile distance to all those settlements by 1, down to 1. So if your city connects to four cities, then a 6 tile road connecting to that city becomes a 2 tile road. Then I would tie happiness to distance from capital where cities can mitigate that value, along with other policies (like colonial administration policy removing distance from capital requirement in distant lands or something). This distance calculation also affects the gold value of trade routes, as well as the distance maximum food supplies can be sent. With a system like this you can build a "backbone" of trade cities that allow you to trade more gold further, keep distant towns happier, and allow you to ship food further to support distant growth. Of course, as always, war can disrupt trade routes.

I also think there should be imperial infrastructure like imperial roads that contribute to lowering distance cost or are resilient to war.
  • Narrative Events - add more immersion to the game
This would be good fodder for a full expansion that would have this be something like half of its focus. With scenarios.
  • Canals/bridges/dams - add these features as they were in Civ 6
I would do this but combine with warehouse concept for things like irrigation projects that both control flooding and do something like provide riverside ag bonuses an additional tile past the river.
  • Map Improvements - map generation, size, number of players available
I'd like an official implementation of the much bigger maps. I'd like the "standard size" map to be the size of a "continent" and I want a system with an oceanic distant lands and a steppe separated distant lands (so three core standard sized lands), including the ocean and the steppe and maybe two or three peripheral half-continent lands.
  • Adjustment to ages' length - smoothen transition from Antiquity to Exploration (post-classical/medieval age)
I do believe they will just have to implement a Medieval age in the end, which will begin in late antiquity and include Western Rome and Byzantium, but end before the Renaissance. I would then call Exploration the "Global Age" implying global trade and contact rather than about exploration alone, without pigeonholing it in that it should also include the Enlightenment more.

Exploration civs would have to be relegated to one age or the other (Normans are a good fit for simply the medieval age). Others might need to be split in two. I think we'd need a Tang and Ming China.
 
It will engage you to bait you closer to city walls then retreat behind the walls. It's good and reserving forces behind defenses.
So what? I park and entrench my meatshields just outside the walls while I blow up those walls from safe distance with ranged units and/or siege units supported by commanders, and move in. And all AI can do to counter that is to throw more mass produced meatshields to the line. I have seen no tactical use of AI commanders or combined arms use of the AI ranged/siege units to support the defence. Just a swarm of meatshields until there are none left. Or just plain sacrificing their undefended commanders or ranged units left to die to the melee on the line. Nothing good "at the defence", just the same "power in numbers" thing, if that even can happen.
 
I'm fairly new to the game, but I'd say one of the things that I find the most frustrating at the moment is that there are a lot of game mechanics where you just can't find proper info on how it actually works in the game. I get that Civ is no longer a PC game first and foremost, which explains the overall direction they take with the UI, but even so there is still so much information that is just not available at all. If you want to actually understand the game mechanics and what drives them, then you're currently just out of luck. There are too many headscratchers right now.

There are also many systems, such as setting up trade routes, that just feel very clunky at the moment.

Again, I'm aware that I won't be getting a PC-centric UI overhaul ala Old World for the game, but there really needs to be more clarity about the game mechanics.
 
Back
Top Bottom