The game is too fast and too easy

[Sostratus]

@Sostratus , I’d be particularly interest to hear your views.

A Ball of jargon approaches!
From a system engineering standpoint we are looking at the tech tree representing a quasi fixed amount of research to complete. I shall call this quantity "total tech yield." The question posed is dealing with players finishing the game too quickly by leveraging several methods. Note that when I talk about tech i'm referring to both science and culture trees. Making techs cost more, and thus increasing the requisite total tech yield for a given level of progress, is considered barbaric and taboo by @acluewithout . He has spoken.

We observe that overall tech rate depends on tech rate density (how much science you can extract per area of your empire) and tech rate area (How much area you have making science.)
We also observe that tech rate density depends on tech progress, IE you can achieve more science per area with later game tech than you can earlier.
We also observe that tech rate area is a joint factor of how much area you control and how much of that has a tech rate increasing district/buildings.

What we want is to shift some of the subfactors affecting tech rate area and density, from being a function of tech progress to being more dependent on things which are themselves of a function of time.
The question of whether or not the science and culture buildings give you too much yield is a discussion people can have. I personally think they are abhorrently efficient and should be revisited entirely in their design. But I will avoid that for brevity.
Specialists take pops. Pops require food and growth, which requires time. Shifting balance of yield from buildings to specialists may increase dependence on critical tech factors with time.
Half of rationalism is tied to pops and thus, time. But not the other half. I would almost rather they boost specialists or at the very least come with an increase in maintenance, which ties into...
Increasing the economic cost of operating tech rate increasing districts/buildings/specialists will require more other things to support it than it does now. While this reduces tech rate area by limiting how many campuses you can spam, it also ties tech rate area to a sufficient economic base, which requires time to develop. If you needed 3 cities worth of CHs and trade to support 2 cities worth of Campus+TS, then you have radically changed the paradigm. (Example ratio, of course.)

In short summary:
We can influence tech rate density by just changing the numbers on buildings and districts etc.
We can influence tech rate area by increasing the economic cost of supporting tech rate infrastructure.
We can influence the rate of change of tech rate density and area by tying more things to time based systems like pops and eras.

The last point matters because total tech yield is the integral of tech rate, so of course, reducing the derivative of tech rate will have a stronger upstream impact on tech rate than just cutting it directly. this is because tech rate scales with itself, so ******ing the snowball has an increasing delay on achieving the needed total tech yield.

i could give much more specific example implementations but that's my eye in the orbital layer view of this.

 

[acluewithout]

@Sostratus Cheers!

I'm okay to adjust tech costs in principal, but there are so many mods that do that already that I don't think it's worth touching that - see Take Your Time referenced above, which is adjustable at set up.

I was okay with adjusting maintenance cost, but from some recent discussions I wonder if that's actually contrary to the current design. ie maybe the game is designed to be gold heavy as it is faith heavy so you're always buying things. I think Civ needs some sort of wider economic management challenged, but tweaking amenities etc. might be closer to the intended design.

So, that leaves actual yields to muck around with. And if mucking around with yield balance could also make tall cities more useful, that's great too. The question is - how much can a few numbers here or there fix ?

Basically, I think one way to do that is delaying the scaling of science and culture, hence nerfing T1 buildings more and making rationalism etc escalate more slowly. I think making lots of cities inherently less valuable for science and culture would help too, but that's hard to do without really messing with the game's design.

 

[Sostratus]

The question is - how much can a few numbers here or there fix ?

Even just making something like a university give science per pop would be an enormous shift. Because then you need pops... which require food... which require farms and time to grow... so you need more builders sooner... etc.
It's hard to appreciate because of how minute it is, but then even 2-4 science per city is equal to a science CS with 3-6 envoys, which is widely considered to be a lot.
I'd disagree with this in principle but you could also say something like, okay, you can only place campuses and theater squares in your first district slot if they'd have an adjacency of at least X. Otherwise they have to go into the second or third district slot in the city (presuming the adjacency limitation decays away.)

 

[Victoria]

views

I am not sure how positive I can be here because I feel the fundamental issue is in game design.
I am also a little confused over if the query is about game speed or science or just general ease but I’ll put my general view across.

You play a game on deity and the first issue you face after barbs is your neighbours. Based on terrain and many other factors like resources, taking your neighbour is often not so simple unless you re-roll for flat terrain. You can bait and defend which does work to some degree, using the terrain, your memory and concentrations to whittle a civ down enough to take. You then take the civ and have double the cities and a big advantage.

So a skilled combat player can gain an advantage without specifically talking science. It takes time to get this skill but it is hard for the developers to combat. Bottom line is few people statistically get to this stage but it is not that hard.

Taking the ramp up in science. The AI has the same tools as you and a hearty bonus to compensate for a rather complex set of decisions. Some civs next to you are going for RV and CV and these civs become more vulnerable to you over time because they are actively not concentrating on science. This to me is a key aspect of my argument.
For the civs that do concentrate, you eventually overtake them and this is where I feel people are concentrating the debate but before then you have both skilled warmongers and opportunists (Taking advantage of other civs choice of victory or their bad luck) taking civs not concentrating on science (or being overrun by barbs/ other civ). So warmongers and opportunists have not been considered in this discussion. I consider Re-rollers opportunists of sorts)

So on to the science maxxing. How do we do it? What gives us the advantage over the AI + their bonuses. Decision efficiency, card slotting, projects, GP usage, era manipulation, overflow management, specialists, chopping, pillaging, district placement/discount. These all have their place and with a bit of practice we manage these far better than any AI that does not have a lot of funding behind it. I think adjacency, the adjacency card and chopping are the main issues. Certainly AI city placement can still be poor but it is irrelevant to the long rambling point I am trying to make.

The developers could spend a lot of time refining this science side of things. They could also make all their civs concentrate on science early rather than other victories. But what would be a realistic outcome of this? Maybe the 0.25% who can deity science push better than the AI would become 0.035%. Is this of great value to Firaxis? The game would be duller and it would not fix opportunistic play nor warmonger play.

I said over a year ago I feel the game is complete, warts and all. If you wanted to tweak science to slow the runaway then to me the science adjacency card is the killer, not rationalisation (+14 for some pretty mountains and sulphur is a bit much). Chopping and magnus, Pingala, Kilwa, the way science % stack, off continent bonuses. These are the areas that affect it most. I do struggle with why we get a campus before a commercial hub and this does not help either. The game has been baked a long time, there is a lot in there that contributes that is now just part of the game. The few deity min/max players could be lessened but without dealing with other fundamental baked in issues. A science mod would help a few but feel this is all tilting at windmills. The problem is not the windmills, you could say the problem is you are in Spain.

 

[HiRezAudio]

I think the biggest problem is that we expect too much from AI - it really is not that clever in the stuff available for games, probably about the brain equivalent of an educated earthworm.
For a great discussion of this please - I beg you, as this is so well worthwhile - to go check out this Ted Talks presentation:

 

[Victoria]

For a great discussion of this please -

Yes it has been posted here before and provides kudos to what some of us have been saying for years.
Even at my work, we have a few big AI teams, all who encourage the myth of AI.

I love the Facebook AI for hiring people that turned out to be sexist because the historical data it learnt off was sexist

I was explaining to my teenager boys the other day. Even if you have a self learning thing like us. How many years does it take to learn, will it ask for external feedback like children do, what motivations do you give it and will it finally end up as insane as some of us?

 

[lotrmith]

Currently Civ VI is like, for standard speed deity game on Lakes map, if you tell a good player how many cultural CS and science CS are in the game, and what the notable ones are (Geneva, Nan Madol etc.), and he takes a look of the Future era tech tree on Turn 1 and is like "Hmmm... 4 science CS and 3 cultural CS, this is a good map, and I can skip 3 tech in the future era, so I can win within 155T, and if I try harder I can get it to 140+T".

And some really good ones don't even bother to play the end game any more, they are like, "See I got Industrialization on T95, so I can win by T144 or T145. Time for next game". And you know he's correct. It's a sad state for a strategy game when people can conclude something like that.

I can't help but interject and say this feels like hyperbolic nonsense. Not saying they don't exist, but I haven't seen a sub t150 science win since vanilla with exploits like Scythian infinite horse gold, or R&F Magnus Goddess of the Harvest.

Currently in our own community the best game posted is T167.

Further still, any really good science win is almost assuredly the product of a fantastic map roll and a lot of luck, after many many 'sub par' victories preceeding it on perfectly average maps.

 

[hhhhhh]

I can't help but interject and say this feels like hyperbolic nonsense. Not saying they don't exist, but I haven't seen a sub t150 science win since vanilla with exploits like Scythian infinite horse gold, or R&F Magnus Goddess of the Harvest.

Currently in our own community the best game posted is T167.

Let me quantify. What I'm talking about is the Gathering Storm version as of September 2019 update, before NFP (version number is probably 1.0.0.341). Maps are usually Lakes.

I've seen tons of sub T160 victory, although sub T150 is kinda more precious. But even myself got deity standard Lakes games with Industrialization on T96 (although I didn't bother to finish that one since late game is standard and tedious). Admittedly sub T150 victories do need some luck. However even a Chinese streamer who I don't think is the best among Chinese players constantly get sub T170 victory in his streaming (it's hard to play very carefully in streaming since you have to keep talking).

Here are some videos of sub T160 victories from a very good Chinese player:

Rome T147 (1) (2) (3)
Russia T155

There are more in his channel but I just don't want to copy them all. In a China game he banned himself from using Great Wall and still win on T155.
Bear with that the game is in Chinese but what he said should be completely understandable to you (he didn't say a single word for the entire video...)
Not saying all players are as good as this one but many thing can be learned and copied from him so that you don't need much creativity to win a sub T170.

 

[lotrmith]

I think that, given the lengths you have to dig for them, that these record setting science wins are not nearly as prevalent as to warrant claims that the game is too easy.

 

[hhhhhh]

I think that, given the lengths you have to dig for them, that these record setting science wins are not nearly as prevalent as to warrant claims that the game is too easy.

But I do think the game is kinda "solved" in the sense that if you learn from these players you can turn a standard deity T250 victory into a T170 one, as long as you want to. I wasn't playing before June 2019 and only paid attention to those top players recently but the improvement I got by copying a few strategy is huge (and it soon make the game feel kinda repetitive).

 

[lotrmith]

But I do think the game is kinda "solved" in the sense that if you learn from these players you can turn a standard deity T250 victory into a T170 one, as long as you want to. I wasn't playing before June 2019 and only paid attention to those top players recently but the improvement I got by copying a few strategy is huge (and it soon make the game feel kinda repetitive).

Well sure but isn't that most (all) strategy games? Also this is the vast exception to the norm for playing circumstances. Yes, you can know what the best things to do are to achieve whatever victory in the shortest time, but can you consistently and reliably execute those things with most Civs on most random rolls of most maps?

Just ballparking to demonstrate but say perhaps 10% of players even *know what is best to do*, perhaps 1% are able to do that a couple times, and maybe 0.1% are able to do it consistently? For the record I think those numbers are really, really generous:

Looking at global stats, 42.3% of players have won a game, and 5.4% have won a Deity game so if you normalize to only players that have won a game at all (to approximate people who 'actually play/played the game'), that's about 12.7% of actual players who have even won a single Deity game.

So if roughly only 1 in 10 players who have won a match have won one at the hardest difficulty I think the premise that "the game is too easy" is a stretch. You've got to remember that here at Civfanatics we represent a minority as a portion of the hard-core base of players and even then a smaller portion still of us are even Deity players. (That's also important to extrapolate to other conversations and discussions so as not to alienate even members of our own community, as I sadly see happen now and then.)

 

[Myomoto]

Besides a general rebalancing of civic and tech costs, I think the game needs more empire management mechanics. It's too easy to to just plump down cities all over the map and collect their yields. There is very very little descision making to be done IF to settle another city, it's predominantly WHERE to settle it. After a city is built, it's basically just a matter of plumping down the respective districts in a specific order to get the desired yields.

Instead mechanics like health/disease, corruption, public order/civil war should be needed to be dealt with. The game in general needs more disruptive mechanics to keep the late game interesting. Empire management should be largely reactive: The game does something to cause you a problem - fix it. Few turns breather to work on your victory type, next problem, rinse repeat.

 

[JesseS]

Instead mechanics like health/disease, corruption, public order/civil war should be needed to be dealt with. The game in general needs more disruptive mechanics to keep the late game interesting. Empire management should be largely reactive: The game does something to cause you a problem - fix it. Few turns breather to work on your victory type, next problem, rinse repeat.

There’s a real danger that this sort of random malus event is far more annoying than it is interesting. Ever tried a science game on apocalypse mode? The need to constantly repair stuff makes the end game more tedious, because it just constantly interrupts what you were doing with a couple repair orders. A lazy implementation of the quoted ideas would be similarly tedious and meaningless (i.e. civil war means I need to kill a few units and then repair their pillage damage; disease means a few cities instantly lose a few population).

The tough part about it is that these random impacts almost necessarily need to be limited in scope—a nuisance more than an existential threat. I don’t think most players would actually like it if their games were ruined to the point of not being able to win because a plague killed more than half their citizens, a revolution disbanded their army and completely overthrew them, or a massive barbarian horde razed most of their cities and sacked their Capitol.

The underlying problem to me is the victory conditions themselves. The final 20 or 30 turns of any game get quite dull, once you realize that the AI can’t viably win before you and the only thing that’s going to help you win faster is 1) researching and launching space projects; 2) maximizing tourism, mostly with rock bands; 3) slamming units against capitols; or 4) sending apostles/missionaries out. Once you get to that point, the interesting empire-building parts of the game and associated trade-offs just become irrelevant. Extending this portion of the game by increasing tech costs or whatever is not going to actually improve the game.

 

[hhhhhh]

Well sure but isn't that most (all) strategy games?

Not really. Many strategic game have a better depth. Before I throw out examples like Chess and Go, even Civ V seem to have a better depth.

And to clarify, I won't call Civ VI early game "solved" - even today many good players are still exploring interesting openings of certain Civs. But I do consider late game "solved", meaning that it has become a list of standard procedures that everyone can duplicate. So for good players it's actually less fun unless they choose sub-optimal play (which is painful for them to do so).

So lots of things are just "not happening" for them:

Climate change? Not before I travel 50 light years in space.
Outback Stations giving extra food? I only need Rapid Deployment for Amundsen-Scott Research Station.
Missions giving extra science? Man I'm already making 2000 per turn.

Let alone subtleties like some random bonus like Cahokia Mounds provide extra housing - if you are winning within 20 turns what's housing gonna do for you?

That's why I say late game is kinda in a sad status. (A fun game I just watched is one with Magnificent Catherine - by running lots of project she got a culture victory on Turn 88/500. Although it's very fast it's still a novel way to pursue culture victory. I wonder if this will become a series of standard steps making culture victory boring for Magnificent Catherine. Could be.)

 

[Karmah]

Tbh, as someone who now plays as a casual rhythm, the game last longer after the next patches (I'm taking about a year worth of them).I don't think I fully adapted to the current meta. I think I had time for ten games in a year. I play immortal and always win but none of them were fast wins, like turn 270+.

 

[DanQuayle]

A fun game I just watched is one with Magnificent Catherine - by running lots of project she got a culture victory on Turn 88/500.

That project seemed massively unbalanced indeed since it unlocks so soon. I bet I can cut down on that time too on the proper setup.
Did that game involve some conquest?

 

[hhhhhh]

That project seemed massively unbalanced indeed since it unlocks so soon. I bet I can cut down on that time too on the proper setup.
Did that game involve some conquest?

No conquest, 9 cities. Didn't even get a classical golden age.

 

[Aurelesk]

I don't know about increasing the cost of all techs and civics. I do not think the issue is about being too quick, but about running away and snowballing. Increasing the Science cost will only delay the game further, and make already the ininteresting part of the game even more ininteresting. On the contrary, we should reduce the gaps between the running away civilizations and the others civilizations.

Furthermore, increasing the tech cost will give the civilization even more time to fulfill the Eureka, decreasing even more the odd of having to waste some Science.

I have a couple of solutions about reducing the gaps. I don't know if those ideas are good, but I am listing them anyway:

• Link the power of Eureka to the Era: they could be 50% for all technology of the current era, and be less rewarding when the technology is the era ahead (for example: 10% per era). For example, in the medieval era, the Eureka will be 70%, 60%, 50%, 40%, 30%, 20%, 10% and 0% for the ancient, classical, medieval, renaissance, industrial, modern, atomic, information and later eras respectively.
• Science discount: if a civilization has discovered a technology, that technology could be 20% cheaper for everyone they met (or maybe link that value to diplomatic visibility / cultural domination?). Furthermore, that discount could be even higher (35%?) if at least half of the civilizations in the game has been discovered.

In this case: if you are running away, the other civilization will not be that far behing because they will need less Science to unlock the technologies.


But I think they are still two issues with Science: the adjacencies are too easy to have and some Great Scientists are too powerful. Hopelfully, the envoys tweaks seems to correct a little.

One thing I would like is a change about the adjacency bonuses. It is one of the easiest to get the highly coveted +3, contrary to the Theatre Square that need some work (wonders), or the Commercial Hub that mostly need a Harbor triangle on river to reach that +4. The Campus adjacencies could be linked to how likely is it going to face a natural disaster, like:

• Volcanoes (+2), and their Volcanic eruptions.
• Reefs (+1) and their Hurricanes / Coastal Raids. Maybe the Hurricanes should be more likely on areas with a lot of Reefs?
• Geothermal Fissures (+1). Maybe those Geothermal Fissures should face some danger, like some Gentle eruption, or being really close to really active volcanoes?
• Rainforests (+0.5), and the Wildfires. Maybe those Wildfires should be able to destroy districts adjacent to them?
• Mountains do not yield Science adjacency anymore, as they are facing no danger, even if it is odd as Observatories were kind of linked to high places

Furthermore, the Great Scientist Hypatia, Isaas Newton and Albert Einstein are way too powerful. The "+X Science to all Y building" should disappear for something else.

Maybe one way to prevent the Campus spam is maybe change how the district cost works. Instead of being linked to how far you are, maybe linked to how many we have (like Settlers / Builders?). So it will be harder and harder to have Campuses in your empire if you build only that.

 

[Myomoto]

There’s a real danger that this sort of random malus event is far more annoying than it is interesting. Ever tried a science game on apocalypse mode? The need to constantly repair stuff makes the end game more tedious, because it just constantly interrupts what you were doing with a couple repair orders. A lazy implementation of the quoted ideas would be similarly tedious and meaningless (i.e. civil war means I need to kill a few units and then repair their pillage damage; disease means a few cities instantly lose a few population).

The tough part about it is that these random impacts almost necessarily need to be limited in scope—a nuisance more than an existential threat. I don’t think most players would actually like it if their games were ruined to the point of not being able to win because a plague killed more than half their citizens, a revolution disbanded their army and completely overthrew them, or a massive barbarian horde razed most of their cities and sacked their Capitol.

The underlying problem to me is the victory conditions themselves. The final 20 or 30 turns of any game get quite dull, once you realize that the AI can’t viably win before you and the only thing that’s going to help you win faster is 1) researching and launching space projects; 2) maximizing tourism, mostly with rock bands; 3) slamming units against capitols; or 4) sending apostles/missionaries out. Once you get to that point, the interesting empire-building parts of the game and associated trade-offs just become irrelevant. Extending this portion of the game by increasing tech costs or whatever is not going to actually improve the game.

I think as optional modes it would work quite well. I do indeed want existential threats, by civil war I mean basically several of your cities splitting off on their own. You could have an emergency opportunity to have allies come help restore the cities to you.

But basically I mean semi-large, extinction level threats (which, for the most part could have been negated by proper management - but it is impossible to guard against all of them at the same time), not just nuisance and busy work (*cough* recruit partisans *cough* ). Reality is rarely predictable and straightforward, something that has tested ancient and even modern civilizations beyond their breaking points.

The current game feels too simplistic and predictable once you get settled in and comfortable in the Renaissance era, and it just becomes a kind of solitaire test of ruthless efficiency to achieve victory in the fewest possible moves.

 

[kotpeter]

I want to talk about the subject in a more abstract way, and then go down from there. Maybe my conclusions will line up with someone else's, I don't know yet because I'm just writing out the stream of thoughts that comes from my head.

There's a lot of macro decisions that can be made throughout the game about different things: where to settle, what to build/research, where to move troops, etc. However, only a few of them really matter when it comes to science and culture:

• Where to place your cities to maximize adjacencies and production and how to do it as fast as possible
• How to get to universities/labs faster without getting wiped out
• How to get to rationalism faster without getting wiped out
• Optional against AI: how to grab the most powerful great scientists and engineers.
• Optional #2 but sometimes required: how to wipe out weaker neighbors to amplify yields (only matters if someone else has more land and yields)

At least that's my experience on Deity and in multiplayer when I play peacefully; by focusing on these things better than other players I'm able to win most of the time. There's a lot of micromanagement to this stuff, but it can be muscle-memorized throughout 10 or so games.
And here comes the problem (at least for me): it quickly becomes stale and repetitive unless I start creating challenges for myself (e.g. go for other victory types). Maximizing science and culture is crucial, and you can't really go on without it. But there's one way of maximizing these yields that's much better than the others: spam cities, spam districts, sit on policies, adjacencies and buildings of those districts. And it doesn't really feel like you've done anything; the yields just come by themselves. Even if you try alternative means of getting these yields (Missions, Ziggurats, Fez, Babylon), they're far from what you're able to achieve with districts, so they need to be stacked with the primary method described above.

So more cities -> more districts -> more yields. And nothing to prevent from keeping this spam on and on: amenities don't penalize you well, settlers and districts are cheap enough; so it's all about settling and conquering sweet places and filling them with your districts, and that's all. Wins the one who's faster in doing this.

However, it looks like Firaxis are aware of this issue, since with the latest update they added another way to generate science and culture (through faith while playing Ethiopia or Voidsingers secret society); however, I feel like this approach is flawed by design. That's because the optimal play now is to stack this bonus with the districts' yields, and they become even more cosmic, and the progression is even worse than it has to be.

To be honest, I don't think there has to be such thing as science cost and culture cost. This thing makes your progression linear with the coefficient equal to the amount of yields that you have, and there's no upper limit to that. Instead, I think that techs need to have the fixed amount of turns to complete (how about 8 on standard?), and your science (if enough) is only able to reduce this amount of turns by a fixed value: there's a baseline science value dependent on era, and by having enough science you're able to cut up to 3 turns off of every tech. Eurekas cut another 2 turns off. Having more science than everyone else on the map cuts one more turn, so you're able to research everything of current era in two turns. Also, techs of future eras (later than current era) cost 2 turns more per era, and techs of previous eras cost 2 turns less per era, so beelining into future eras is expensive and needs time to be sacrificed for that.

This way, by setting up a correct yield baseline, you allow people to calculate for the exact amount of science they'll need and plan for that instead of forcing them to maximize science every game. Since there are options to gain science, people would use them and might need fewer cities and campuses. It might even allow tall gameplay if specialists are buffed and amenities are reworked properly.