Liberty is very hard...

[steveg700]

I'd like the finisher on Liberty to be more enticing. Yeah, I know that a GE at the right moments nets your Desert Folklore civ a Petra, and a timely Prophet can net you first pick of enhancer beliefs, and so on. However, all the other finishers grant a long-term benny. With Tradition, you get free aqueducts and then later can buy GS's. With Liberty, there's not even a faith-buy option.

 

[Cromagnus]

Disagree with science penalty. What if I just choose not to build more than 4 cities? Wouldn't liberty be out teching Tradition 4 cities by alot? Even now, with 7 liberty cities properly developed they are already out teching tradition.

Sent from my Galaxy Nexus using Tapatalk

GhostSalsa, re: Liberty Finisher, An academy (w/ NC) is 12 beakers. On t85 (which I usually finish Liberty by) that's usually boosting my science by 25%. (9,7,5,4 pop + NC = 48)

The growth advantage of Tradition+aqueducts isn't all that significant until you finish Tradition, which is usually around that same time, so for a brief moment you can be farther ahead in tech with Liberty, right before Civil Service + Aqueducts leaves you in the dust.

Anyway, regarding the science penalty, and why it IS significant:

A brand new city generates 1 beaker. Let's take a very typical Liberty example prior to turn 100. Your size 10 capital generates ~40 beakers. (10 from pop, 5 from library, 8 from academy, 3 from NC = 26. + 50% = 39, let's pretend 40 for simplicity)

That's how you have to measure your "base" beaker cost. With just this city, if a tech is 200 beakers, it will take 5 turns to research.

If you have one expo, size 6, w/ a library, it generates 9 beakers.

This boosts the tech cost to 210 beakers, but reduces the time to 4.3 turns. (still 5)

Let's say that city started out at size 1, and when it was size 1, you didn't have NC up, and your main city was only outputting 20 beakers, so 20 + 1 is a break-even (+5%)

Now. Your second expo, size 4, w/ a library, generates 6 beakers.

This boosts the tech cost to 220 beakers, but reduces the research time to an even 4 turns. You're generating the *equivalent of 25% more beakers*, (after penalty) but you've *doubled your population to do it*... This is bad, yes?

Your third expo, size 4, w/ a library, generates 6 beakers.

This boosts the tech cost to 230 beakers, and now we're down to 3.8 turns.

I'm intentionally ignoring some factors along the way (mainly that those expos after the first weren't pulling their own weight, tech-wise, at first)

Now, we plant another city. (#5)

It starts out generating 1 beaker. We can't afford to rush-buy the library, because we need that money for other things. (Say, perhaps a university)

Now our cost is 240 beakers, and we're back up to 3.9 turns. Unfortunately, we can't build Oxford until we get 5 universities now, and Oxford will take longer to build. This was the same issue earlier with National College when we had 3 cities. So, we throw a trade route at this city, try to get the library built fast, and rush-buy a university.

We can't place another city now if we build Oxford, but we're going to wait on that anyway for timing reasons.

Sometime around t110 (because we're kicking ass) we have all 5 universities done, and we place our 6th city.

Now let's say our beakers are:
(capital size 14) (14 + 7 + 3 + 8)*1.83 = 59.
(expo size 10)(10 + 5)*1.83=27.
(expo size 8)(12*1.83)=22.
(expo size 6)(9*1.83)=16.
(expo size 4)(6*1.83)=11.

Grand total is 135. If a tech is base 400 beakers, it costs us 500 and takes 3.7 turns.

We drop another city. Total is 136. Now we're at 520 beakers, and it takes us 3.8 turns. We'll break even at 140, when the city hits 3 pop w/ a library, but that costs us 6 happiness.

We drop another city, six now. Total is 141. 540 beakers, and it takes us 3.8 turns again. We'll break even at 4 pop w/ a library, but that costs us 7 happiness. Otherwise, it takes us 6 pop to break even. (9 happiness)

From here on out, it's pretty much the same thing. Each new city breaks even at the cost of 6 happiness and a library or 9 happiness. Until we need to build Oxford, now we need all those extra universities. And it takes longer to build Oxford. And after we build Oxford, our capital is generating even more science.

Now we build public schools. Our more recent expos (unless they have insane production) are kept busy building those schools for a very very long time, or we have to rush-buy them. At which point, the capital is generating an even greater percentage or our total science.

So, it may seem like we broke even, and maybe even eventually generated more science total... but at what cost? Well, for each city, we lost ground before breaking even. With the science penalty, we paid 25% more for technologies than before. Over 125 hundred turns, we research as much as we would have in G&K in 100 turns. Over the course of a typical game, that adds 50 turns.

FIFTY TURNS. And that's only with 6 total cities. The AI usually builds more. This is why the AI in G&K could win by t270, and now doesn't win until t330 or later. It's that simple.

The happiness cost USED TO BE made up for by the faster tech rate. It's now only slightly better than a break-even, and it requires a ton of money and hammers be dedicated to those new cities.

The final factor is internal food trade routes. You can only have so many. There's a fixed-size pool of bonus food, enough to feed a few expos. (or boost a mega-capital)

And you're sacrificing money to use all those trade routes on food in the first place. OCC is impractical because you can't use Trade Routes, and wide is impractical because there aren't enough. So, we're stuck at roughly 3-5 cities being the sweet spot.

Now, Liberty gives us +1 happiness per city, (20% of the breakeven on happiness) but it doesn't help at all with those 50+ turns of extra research. My proposal, droppping it from 5% to 3%, would reduce the effect to 30 turns (over a game) from 50, which would help (somewhat but not completely) compensate. Tradition would still be superior at science because of Monarchy and the limited number of trade routes in a game, but at least early conquest wouldn't *cripple* your science rate... as much. Because you can't afford to annex puppets to build libraries... This is why late-game conquest is so much more profitable than early-game. You're capturing cities with universities and wonders vs adding +15 unhappiness for no benefit whatsoever.

Anyway, I digress. Point is, going "wide", even 7 cities, is nerfed by the science penalty. And my suggestion of reducing it is really not going far enough, but removing it entirely would make Liberty even MORE superior at conquest, and Honor is even more nerfed than Liberty right now. Maybe the Honor closer should remove the science penalty in cities with a courthouse?

 

[Cromagnus]

I really really agree with this. The current mode which is default - if your tech is strong you never have any threats - makes the game fundamentally broken. When you learn how to beeline for Education it's not a strategy game, it's Sim City (and on Deity the AI functions as your "funds" button, I have to laugh a bit at the idea that Deity affirms whether strats are efficient when it introduces so many free resources for the human player). The game needs some historically implausible dues ex machina to throttle tech speed, like barbarians popping up in borders if you're too far ahead. This is not satire; the game is toothless right now. This device could be turned in Advanced Settings off for the huge "I don't want challenges I want a sandbox" contingent. Anyway.

Bet I'm confused how you could say Liberty is better for science early due to the academy. It comes when a Tradition civ would already be near education even given a 4 city before NC start, and the Liberty civ has 10-20 turns to go. An academy at that point isn't enough to make up for getting universities later, usually. I mean as I keep saying I like Liberty but don't see it as a path to early tech lead.

Also, I forgot to mention... bulbing the GS at that point will often finish a critical tech like Chivalry (for arabia/genghis) or Machinery (everyone else) faster than you possibly could with Tradition. It comes at just the right time for a rush. The academy is useful for those multi-tech rushes like Machinery->Navigation, and Tradition gives much less benefit when you're intentionally keeping your capital locked on production and smaller to give happiness headroom, so that academy has a bigger benefit. /shrug

 

[GhostSalsa]

GhostSalsa, re: Liberty Finisher, An academy (w/ NC) is 12 beakers. On t85 (which I usually finish Liberty by) that's usually boosting my science by 25%. (9,7,5,4 pop + NC = 48)

Yeah I was mis-remembering how long it takes to finish Liberty - just closed it on turn 81 in an Austria game thanks to Fuji (a great wonder when you're wide and need to grow slow at first). (However an Academy is going to hurt my growth and Petra just got snagged so I haven't decided what I want yet.)

I feel your math is all more or less correct, 50 turns slower science by end of game is a good way of looking at it. But that's compared to a wide empire without the penalty. Since there's no such thing as "a wide empire without the penalty" in BNW, that's not the relevant comparison. We only need to compare wide empires with the penalty vs tall empires (which also have the penalty, at 15%): wide empires are clearly better eventually. Virtually always by turn 250.

Then, if we want to see how a Liberty penalty reduction would fare, compare wide empires without the penalty to tall empires with the penalty: wide empires are clearly overwhelmingly better eventually. But still will usually lag before turn 200-250 because of population and resource concentration factors. "Resource" meaning science buildings, mainly the University.

I mean as you were working your math step by step it was pretty clear that each new city in a liberty empire isn't harmful for very long, and is beneficial in the long run. The only worrisome times for new city creation RE the science penalty are turn 100-200 when your total empire bpt is over 200 but new cities take a long time to get running (no Order benefits yet).

As you were working your math it was clear that the bigger early game science "penalty" for wide empires is still National Wonders. Potential NC and Oxford delays are still vastly more significant than the science penalty.

Since the comparison in BNW is wide-with-penalty vs tall-with-penalty, the only real question for Liberty is should it boost wide empire science in early game somehow. Or should it remove National Wonder restrictions. Or should those just be patched out altogether (replaced with "must have such and such building in 4 cities or all cities if you have less than 4") (I very very strongly vote for the last one, National Wonders are the whole reason you can't profitably settle after turn 110). Should National College just be flipped to a flat benefit instead of percentage. Should libraries yield a flat beaker count instead of population. Should the science penalty scale down after turn 100 to incentivize mid-game settling and conquering (probably not, that sounds too wonky). Etc etc.

 

[Cromagnus]

Interesting points. I do agree that national wonders are a bigger issue, and that the science penalty isn't as significant in the context of tall vs wide within BNW, generally, but it's more a matter of it no longer being a good tradeoff. Especially re: domination. Prior to BNW, if you could handle the happiness issues, conquest actually improved your science. Puppets didn't generate much science because of the -25% and because they usually lack science buildings. but they generate enough beakers (almost) to compensate for the happiness issues they created. It was a small net gain, but more importantly it didn't slow you down. Now it *really* slows you down.

The same applies to going wide in general. Until Order, you spend a ton of resources propping up new cities, especially on happiness. Faith buildings, gold for coliseums or libraries, etc.

You can break even eventually and eventually get ahead, but the happiness trade off is no longer worth it. A wider empire has more unhappiness per population and less science per population than a tall empire, and there's limited resources, so going wide slows you down. You could go Tradition and use those caravans to boost the population in your capital, getting way more science and gold and way less unhappiness out of the deal. So the end result is that wide can't keep up, because you've used resources to prop up happiness and science buildings instead of boosting growth in your most profitable city. Without the science penalty you'd get enough benefit from each new city to keep pace. IMHO.

I think tweaking the national buildings really would help, but I also like that mechanic. It forces you to make decisions about the timing of new cities and such, which is good. Not that they're going to re-balance. But if they did, I don't think reducing the science penalty for liberty would be excessive. You'd still have the disadvantage of a fixed number of trade routes, which really gives the upper hand to going tall. However I think there are lots of different ways to solve that problem. Maybe in civ 6 they will.

 

[Vitruvius]

Wide does eventually get ahead of tall in the later game and liberties bonuses becomes more pronouced later. If you arnt obssessed with win times, liberty is always stronger. It gives more flexibility in the late game and you can comfortably aim for multiple victory conditons at once.

Sent from my Galaxy Nexus using Tapatalk

 

[anthropophage]

A couple of nitpicking points about Cromagnus' calculations above. The capital has the palace which adds three beakers. Expos with a university get modifier of *1.33, not *1.83.

Doesn't really detract from his point, but anyway...

 

[Cromagnus]

Oh oops right. Kinds strengthens the point about satellites taking a while to break even

 

[Ninakoru]

I think the science penalty is fine as it is. Just remember how you or any civ could out-tech everyone with a good conquering spree back on G&K. Those were REAL runaways and the game was done easily by turn 100 or so. I prefer this way on the science side, tall can compete wide, and going wider doesn't grant a science lead.

I find tall vs wide is quite balanced right now, but still tradition is overpowered, and tradition while offer tall benefits can be used anyways to to wide later on, and better tan with liberty as the game stands now. After all the main restriction to going wide is happiness, and tradition is king at that with the huge monarchy bonus plus the aristocracy extra.

 

[klaskeren]

I agree with cromagnus, and i think that the science penalty is the biggest fun-ruiner in BNW, i dont like that i can easilty win every game without ever having to worry about AI finishing before me, there are many other consequences, mostly nerf to warmongering.

 

[Gori the Grey]

I've found this discussion fascinating. I'm one of those people who always goes Tradition, but just because I'm comfortable with it. But I'm getting bored, so I'd like to try Liberty just to change things up.

Do people who like Liberty think the following start with Portugal would lend itself to that SP? I've met Venice , probably off to the north east by just a little (no trouble obviously) and that's China off to the east. I have no idea what is further east or northeast.

I'll say more about my playstyle in a future post. Regardless of SP, I'm going to start building soldiers to keep China from settling into my future lands. There's incense under the unfortified barb to the east.

Sorry for the bad pic. Gems to the SW. Marble 3 rings out from Lisbon. Citrus beneath that wheat tag. Furs everywhere. Horse placement to follow.

 

[adwcta]

In my typical Tradition science game, I finish turn 230-270.
In my typical Liberty science game, I finish turn 250-270. I've hit sub 250 a couple of times, but only with science-civs.

If you're vigilant about keeping the AI in check (even without using any direct military action), you can keep them from completing the first spaceship part until ~turn 300, and the rest of the spaceship not until turn 350. If you just leave the AI alone, the fastest I've seen anyone with a spaceship part was ~turn 260, and the spaceship doesn't finish until turn ~290. And, liberty snowballs, so even if you "miss" 270, your accumulation will actually (imo) make it safer to actually hit the end of the spaceship tree before game over than Tradition.

The science penalty shifts the wide vs tall balance, but it really screws the AI, because they insist on settling 10 one tile islands in late game (meaning the cities never go past 3 pop). 5% is a nice number, but I think it's pretty clear that a more competitive game would have a % closer to 3% (maybe 3.33%).

 

[Vitruvius]

Before we attempt at balancing a game we should always think about what difficulty level are we talking about or whether if it is single player or multi player.

Sent from my Galaxy Nexus using Tapatalk

 

[adwcta]

Before we attempt at balancing a game we should always think about what difficulty level are we talking about or whether if it is single player or multi player.

Sent from my Galaxy Nexus using Tapatalk

Good point. Let me change that.

Liberty should give -20% science penalty bonus (to go from 5% penalty to 4% penalty).
Deity AI should get an extra -20% science penalty bonus (so expansive civs are down to 3%).

Also, puppets should not count toward the science penalty. They don't count toward the culture penalty, not sure why science would be different. Their science production should also be nerfed to be 50% instead of 75%. This allows warmongers to not be "locked in" to their era due to high resistance times (as it currently is, it's a raw penalty for X turns with no production). This change would nerf Venice, but that's fine.

 

[magicq99]

The science penalty shifts the wide vs tall balance, but it really screws the AI, because they insist on settling 10 one tile islands in late game (meaning the cities never go past 3 pop). 5% is a nice number, but I think it's pretty clear that a more competitive game would have a % closer to 3% (maybe 3.33%).

Well, if I recall correctly, the percentage scales with map size. On a large map, the penalty is 3%. (Please correct me if I'm wrong here)

And I think this shows the main challenge Firaxis faces in the current game design. The way it is right now, Liberty seems to be more powerful the larger the map is. A "Standard" size map is probably the one where they want it to be the most balanced. This suggests that Tradition, by its nature, is more powerful on smaller maps.

So the way I see it, no matter how they buff or nerf either Liberty or Tradition, it will have the same effect on both ends. Buff Liberty to be more competitive with Tradition on Standard maps and it will be overpowered on large maps. Nerfing Tradition will have the same effect.

They probably tried to address this with the scaling science penalty. But Tradition does not have a comparable scaling mechanic. It benefits up to 4 cites, which is a lot on a small map, but less on a huge one.

I wonder if it would have been a better idea to scale the 4 cities number from Tradition, instead (or in addition?) of introducing the science penalty. This way Tradition would steer you to basically a kind of OCC on a tiny or small map, while benefiting more than 4 cities on large and huge maps. Similar to the way the Representation civic scaled with map size in Civ IV.

Anyway, I do not expect that Firaxis will change anything now. The most they will do is adjust some numbers, if they change anything at all. But they will not fundamentally change or move any policies any more, I think.

 

[Laurwin]

warmongering is such un-fun because of science penalty IMO...

war is just so un-viable in mid game often times...

-Capture capital, suffer happiness and science penalty...

-capture puppets, not really worth it anymore in bnw, suffer culture generation, suffer science penalty, suffer happiness.

The point about civ4 commerce-to-science is not relevant to civ5.

Science penalty is another fixed maintenance mechanic, which cannot even be overcome. (maintenance mechanic together with happiness system)

In CIV 4 there is CITY MAINTENANCE COST (gold upkeep per city)

When you grow city, you overcome the penalties, when you build all the necessary buildings in each city, you overcome the penalties. Remember guys, buildings cost NO UPKEEP in Civ 4.

This is why the oft-used phrase about civ4 strategy became commonplace: "land is power"
In civ 4, when you get good land, you defend and improve the land, then you build buildings, then your empire simply GROWS in economy and science.

Land is no longer power in civ 5.

 

[GAGA Extrem]

Land is no longer power in civ 5.

I think it still is. But it just kicks in a bit too late to make a viable difference.
Whenever I play wide, my first 200 turns are just abyssmal. I feel inferior to my tall empires each and every time.

But once you reach the industrial era, a wide empires starts kickin'. You will easily overpower tall nations in terms of science (probably normalized by the number of city penalty) and hammers. You will also have a much bigger supply of important strategic ressources, esp. oil and uranium which is - imho - a significant advantage for late game wars. You will also have more archiological dig sites for munoments, which is quite nice to improve your tourism output.

I must admit, I would be much happier if wide as a bit more powerful early on and a bit less powerful later (to compensate for the better early game). As it is, tall seems to be a much safer bet 95% of the time...

 

[Laurwin]

The key advantage for wide empire, was nerfed heavily in BNW.

This advantage was major factor in vanilla and GnK, (IIRC)

Advantage of wide empire was, GOLD ECONOMY. This was because of land was more "valuable" in vanilla, especially the riverlines (for extra gold trade posts)

Gold income was heavily nerfed in BNW, such as removing +1 gold from riverline. AND moving away the trade posts into medieval guilds. Also, removing +1 gold from trade posts in normal fashion, and moving this extra gold into commerce finisher.

 

[Cromagnus]

I think it still is. But it just kicks in a bit too late to make a viable difference.
Whenever I play wide, my first 200 turns are just abyssmal. I feel inferior to my tall empires each and every time.

But once you reach the industrial era, a wide empires starts kickin'. You will easily overpower tall nations in terms of science (probably normalized by the number of city penalty) and hammers. You will also have a much bigger supply of important strategic ressources, esp. oil and uranium which is - imho - a significant advantage for late game wars. You will also have more archiological dig sites for munoments, which is quite nice to improve your tourism output.

I must admit, I would be much happier if wide as a bit more powerful early on and a bit less powerful later (to compensate for the better early game). As it is, tall seems to be a much safer bet 95% of the time...

I think this is the key. The advantage to going wide kicks in too late. You really want the happiness policies (and others) of Order first. But, the best policies in Order are not the ones that help with new cities. So you really don't want to take those until maybe your 7th Order policy. By then, if you're playing efficiently, you're almost done with the game, and unless you're playing on a Huge map or a Terra variant, there isn't anywhere left to put a city good enough to justify it anyway.

(By wide I mean more than 5 cities)

There's wiggle room when it comes to domination-assisted victories. Sure, if you've captured a size 10 city with Notre Dame and a university, annex away! But it isn't profitable in the general case. The zero-output rebellion + the money and faith you'd need to spend to make keep it happy after annexing while you built a courthouse is not a net profit. But, for a cultural VC, the advantage to being able to build 8 archaeologists at once and have 8 museums is undeniable. (Not to mention the advantage of capturing the cultural powerhouse's capitals to stunt their growth)

But, will it result in faster win than a 5-city empire? Unlikely, since tech is the limiting factor on finish time for science, cultural and diplomatic victory. Since the tech advantage of going wide kicks in too late, it can't keep up.

It's a simple equation:
1) Tech = Growth
2) Growth is limited by happiness and number of internal food trade routes available
3) The capital has the best science output.
4) The capital (With Tradition) has the least amount of unhappiness.

Going with 5 cities (or less) allows you to send all early-to-midgame trade routes to the capital, maximizing growth in the city that has the least amount of unhappiness. This is the maximal approach to science, and going wide goes against this. If you want a new city to contribute, you have to send food to it, not to the capital. Yes, that new city will grow faster with only one trade route than the capital would, but at a base cost of 3 + population in unhappiness and a significantly smaller beaker output until that new city has all the science buildings... at which point it's still producing roughly 20% less science. (100% + 50% (NC) + 50% (Obs) + 33% (Uni) + 50% (Lab) = 283%. Satellites without NC but with ALL other buildings = 233%, or roughly 82%, and that's AFTER all science buildings are built, assuming you found a spot with an observatory!

So, the bottom line is, wide can't keep up on science until long after you get Order, and by then, the game is almost over. The science penalty is absolutely a big factor here, because if it weren't for the penalty, then each new city would *immediately* contribute. Sure the unhappiness would be an issue, but at least it wouldn't be slowing you down, if say, you found a unique resource to camp it on. However, even without the science penalty, the fact that # of trade routes is a fixed size gives the advantage to small empires... even on Huge maps.

 

[adwcta]

You keep assuming NC is in capital. If you're going wide, and one of your first two city spots has a mountain, the NC should be there. The only thing that doesn't synergize with this is CS, but none of your cities are growing as hard as tradition anyway.

Wide liberty plants more observatories, so freedom is actually IMO better than order (with capitalism and the specialist one as your happy boosters, the specialist one has anti synergy with tradition's happiness mechanic anyway, so freedom is very good for wide).

Anyway, I share your issues with the game's resistance mechanic, and how they apply to science and not culture. This (and hammers and gold) is why IMO, peaceful wide is a better route for science. It's harder to pull off though. I have to think this is not intended and if the game was actually receiving patch support that it'd get tweaked.