Science progression seems slow?

[Castar]

This is a thought I have had in general, but was especially apparent in my last game, where I tried for a science victory with Korea and almost accidentally got a Culture victory around the 1980's (I did not accidentally build the Utopia, obviously, but it was apparent I was unlikely to get a science victory).

I have the impression that science progression is slow, in general it takes too long to research most techs, especially from the Renaissance and onwards. In a game like my last one (v150.2, Korea, Warlord, Small Map, Continents+) where I focused on science and great people from turn one, I just barely managed to hit the era's (Renaissance, Industrial etc...) by around the year you'd expect them to hit and then only by making use of free techs from Oxford, Enlightenment finisher, or a crapload of GS's to beeline to the easiest to reach tech. I then usually had quite a while to go before I was actually good and well in the era the game said I was in, because all the other "branches" had to "catch up" with the one I had beelined. I was vastly ahead of history to reach the Medieval period as Korea's science from farms bonus is huge early game, but after that things slowed down tremendously, and only a complete focus on science allowed me to somewhat keep up with actual history.

More details on the specific game: I played tall, with only two cities that did however grow massive very fast and where completely stocked with science specialists (and artists after Cristo Redentor). I was de facto alone on my continent (the only other civ was seperated from me by a mountain range) and didn't build any military units or buildings until the industrial age. By the end of the game the populations where around 40 in both cities, with all of the science buildings and wonders and science-adding policies there are (it's Warlord, after all ), and still all "current-era" techs were something like 20+ turns to research.

With previous games where I focused on conquest more, with wider empires, it was even more ridiculous as I really had to struggle to get into Industrial by, say, the 1910's or so, with lots of renaissance techs still to be researched. This just seems wrong. I would expect that a non-science focused game manages to steadily get along with science, and a science focused game should be somewhat comfortably ahead of the historical curve.

The other Civ's were completely worthless. Agreed, this is Warlord, I like to play an easy game after a hard day/week of work, but even then it's somewhat ridiculous to see other Civs hit the Renaissance around 1900.

So, focusing completely on science, I easily got a culture victory, could have steamrolled all other civs for a conquest victory and just before I won I bought all the city-states because my income was through the roof. And I had built nearly every wonder, so a Time victory was also well within reach. Only a Science victory seemed implausible to achieve. Again, this is Warlord, but the point is not the performance of the other civs (I didn't expect to lose ), but the apparent "difficulty" of Science versus other aspects (production, income, culture).

 

[Sanguinus]

@Castar
I would share your view but I haven't played very much (only prince, marathon) so my opinion is not very representative.
Perhaps at a higher difficulty level the AI will build more wonders.

I also have the impression that culture develops way to fast. Currently it takes only around 30 turns in my game until the next free policy and the same goes for the next science step.

 

[Thalassicus]

There's two distinct questions to consider:

1. Did you run out of things to build in cities?
2. How close to eras match to real-world years?

The first is the gameplay side; the second is purely aesthetic.

 

[Zaldron]

I actually have found that science moves at about the right pace. Recall that the rate at which years pass was recently changed and that it's almost impossible to make years seem correct in more than a handful of games. You need to watch which turn you attain different techs to get a real feel for how fast/slow you're going.

All that said I do believe that culture is attained a bit too fast: I've been routinely picking up cultural wins late industrial or very early modern (sometimes I haven't even picked up broadcast towers yet). Since science victories are attained in the future era I'd wonder if perhaps mid-to-late modern era would be a better target.

 

[Thalassicus]

Players focusing on a science victory research faster than other victory types. An early-modern culture (or conquest) victory probably occurs around the same turn as a future-era science victory. Also... cultural empires tend to be smaller, and #policies is not factored into score, so I try to give culture victories a slight time advantage.

 

[Seek]

@Castar: There's a bug in the AI which reduces many yields to somewhere around 1/5 the actual rate - if you're playing recent versions, you might have run into it, thus the poor AI performance.

 

[Castar]

First of all, again, I'm not concerned about the AI performance, more about the effectiveness of going for science, than going for anything else (money, production, culture).

There's two distinct questions to consider:

1. Did you run out of things to build in cities?
2. How close to eras match to real-world years?

The first is the gameplay side; the second is purely aesthetic.

To the first: yes, in my first city. And late game, quite frequently, as my production exploded with workshops, factories and the like. I did not actually run out of things to build, but only because I chose to switch to wealth, or at one point just build way too big of an army for a peace-game, just for the heck of it. But the two or three buildings that were left cost 1 turn to build, and the few wonders maybe 3 or 4, so I could very easily have built everything before switching to wealth/science.

That's with one mine, one sheep on a hill, two forests (admittedly both got the +3 production opportunity) and the rest farmed plains and a good chunk of towned desert. The second town was less productive, due to lots of grassland instead of plains, but nearing the end it was also running out of things to build.


To the second: the era's actually matched the actual dates quite well, but only because I beelined them. So I hit the renaissance mid-1400's, but then spent well until the somewhere near the 1600-1700s still catching up Medieval techs, then in the mid-1800s beeline to industrial and again spending a good while still researching Renaissance techs etc. As the good "beeliners" (oxford, enlightenment finisher, lots of GS) where gone or slowing down, it actually took me quite a while to get to Modern, around 1950s-1960s I believe.

That's all not very bad, and nearly historically accurate. Point is two-fold: firstly, I only hit the era's from a game-perspective with the game telling me "welcome to <insert era here>". De facto I was still firmly researching in the previous era. Secondly, this was with a pure science-focused game, in a situation where I did not need to care about opponents. Only with these circumstances was I able to somewhat keep up with history, while I'd expect to be well ahead of the historical curve. Other playstyles can't seem to keep up with history in any way.


An added thought: while science generally seems to drag on (science generation explodes, but so do the science costs), production, culture and to a lesser extent wealth seem to completely skyrocket at certain points. Production gets a massive boost the moment you start getting factories, Order, etc., while the end-game graph showing my accumulated culture basically went vertical at the point I completed the Eiffel tower (I had the +2 culture per surplus happy face policy from the piety branch)...

 

[Gamewizard]

If you had such an excess of happiness then maybe you should have expanded some more. 1 pop = 2 base science after all.

 

[Thalassicus]

How many cities did you have?

We run out of things to build in the late game because the industrial & modern eras don't have as many new buildings as early eras. It was this way in earlier versions of the Civilization series as well. Once we reach that point, we can "build" the Research process to speed things along.

 

[Sanguinus]

I've just reached the modern era in my marathon game and have only 10 cities, most of them have nothing more to build and I have built nearly all wonders.
New social policies require only around 30 turns, while a new tech requires 40+ turns. However there is no AI player with more advanced techs, although they might have much more cities.

Am I running towards cultural victory because of my few cities?

 

[Thalassicus]

Ah no, it might be because of game speed. Movement and combat on marathon are basically three times as fast as normal, which changes the balance of things. Even so, it sounds like things are somewhat close for you. What policy trees did you pick?

 

[Sanguinus]

@Thalassicus
I've finished Liberty (1st), Honour (2nd), Enlightment.
I did not finished Order because of the "Factory give-away" I don't want (right now).
I've started Tradition and Piety long ago but stopped due to more important policies.
Now I've just started Commerce which I will fnish first.

 

[Castar]

If you had such an excess of happiness then maybe you should have expanded some more. 1 pop = 2 base science after all.

First of all, I was somewhat landlocked. Secondly, I chose to play tall, originally going for an OCC, but then after seeing my surroundings decided upon two megacities. My pop total was well over 80 end-game, so in a wider strategy that would've been quite a few size ~10 cities. I wouldn't have had as much pop actually, due to the static unhappiness cost of a city. Science is, as you say, almost purely pop-based (indirectly so with scientist specialists and science bonuses on tiles), so expansion doesn't really matter as long as your pop grows swiftly. The large amount of excess happiness was also largely due to Eiffel Tower, a late-game wonder. Mid-game I actually had to keep up with colloseums and theaters due to the speed my population was growing at.

And even if you still think I should have expanded, that's, again, not really the point. The point is that, given all the circumstances, and me going nearly a 100% for science, I could have gotten an easy culture, a somewhat easy conquest and an easy diplomatic victory, but science would have been a very close call if I'd have managed it at all.

We run out of things to build in the late game because the industrial & modern eras don't have as many new buildings as early eras. This is an inevitable part of the game. Once we reach that point we can "build" the Research process to speed things along. It was this way in earlier versions of the Civilization series as well.

I agree with your point, but building research seemed quite irrelevant, actually. I generated hundreds of science in each city, so the (i'm guessing here) 50-70 science added by 1/4th of my production was actually largely irrelevant, maybe cutting a turn or two of the ~20 turn techs.

 

[Thalassicus]

And even if you still think I should have expanded

Not at all! My goal is the other way around: I want to figure out how to make your playstyle more interesting. To do that I need to know about your playstyle, which leads to the questions. A conquest-victory pangaea player on normal speed likes different things than a player going for peaceful tall victories on marathon speed continents. The details are important!

It sounds like there's two important factors 1) nearly a one-city-challenge game 2) marathon speed. It sounds like you weren't involved in much conquest so we can probably rule out marathon speed as an influencing factor (as mentioned before, marathon triples combat and movement speed). This leaves the nearly-OCC style. About half a year ago I added some checks to citystate exports to limit them in OCC games. I did this because some players had much better results in a Featured Game with 1 city than players with many cities. OCC games are typically less complex, so I added some reductions in citystate exports for those types of empires.

With all that in mind, my next question is would higher maritime food have helped you? Or was your population limited by happiness, and more food wouldn't have changed anything? The flip side to this is raising citystate exports to OCC (or nearly OCC) games would also raise culture, and your feedback seems to indicate culture and science were relatively balanced to one another. It's the balance of science (getting new things) -vs- gold/production (building those things) we need to look at.

 

[gdwitt]

You can quickly change science accumulation by increasing the SCIENCE_PER_POPULATION" modifier in vem from 2 to 2.5 or so.
I actually think science is still too fast and prefer to make it at 1.5 or even 1.

I dislike having to fight my enemies with a massive tech advantage either way.

Culture is the best part of the game. There should be more cultural options and trees.

With only 6 trees, we may need to slow things down once museums are available.

 

[Thalassicus]

In v152.1 I raised the policy cost exponent from 1.2 to 1.3, which will have a big difference in late game (probably about +25% total cumulative cost for a culture victory). It's somewhat counterbalanced by the stronger Great Artists.

 

[Jaybe]

In v152.1 I raised the policy cost exponent from 1.2 to 1.3, which will have a big difference in late game (probably about +25% total cumulative cost for a culture victory). It's somewhat counterbalanced by the stronger Great Artists.

Are you referring to VEP_General.xml Policy_Cost_Exponent (which is 2.3)?
The only 1.3 I could find was CULTURE_COST_LATER_PLOT_EXPONENT in VEC_General.xml.

 

[Castar]

It sounds like there's two important factors 1) nearly a one-city-challenge game 2) marathon speed. It sounds like you weren't involved in much conquest so we can probably rule out marathon speed as an influencing factor (as mentioned before, marathon triples combat and movement speed). This leaves the nearly-OCC style. About half a year ago I added some checks to citystate exports to limit them in OCC games. I did this because some players had much better results in a Featured Game with 1 city than players with many cities. OCC games are typically less complex, so I added some reductions in citystate exports for those types of empires.

With all that in mind, my next question is would higher maritime food have helped you? Or was your population limited by happiness, and more food wouldn't have changed anything? The flip side to this is raising citystate exports to OCC (or nearly OCC) games would also raise culture, and your feedback seems to indicate culture and science were relatively balanced to one another. It's the balance of science (getting new things) -vs- gold/production (building those things) we need to look at.

Several things: my game was Standard speed, another player said something about marathon, but that wasn't me. As you did not really count that as a factor it's not really important, however.

Secondly, culture was, in my view, really in the production/gold category, not really balanced with science in terms of progress. Thing is that in order to go for science, you generally pass all the culture techs quite rapidly as well, so you build those buildings and generate quite a bit of culture.

Thirdly, more maritime food... I don't know for sure, but I don't really think so. This might be Korea-specific, however, but I just started farming the bejeezus out of everything, and by the time I had the economy to "buy" more than one citystate (Banks in my massive-pop cities), I was already full on scientist specialists and could easily have sped up pop growth by moving one to a food production tile. Even then, my growth was fast enough that I was frequently balancing around the 0 hapiness point, so even more food from maritime would have been irrelevant.

I think the problem is actually rather simple (although the solution might not be): science costs increase to fast towards the end of the game. Production, maintenance and culture costs increase over the course of the game, towards somewhere in the thousands for wonders or the next policy, while science costs in the industrial-modern age go into the tens of thousands. This is somewhat strange, since the means to generate science are not much different from generating money (pop-based due to working tiles, and Banks), culture and production. The means to generate those three seem to "outrun" the increase in costs as a city/empire gets more developed (or in the case of culture, by building the Eiffel Tower), while the reverse is true for science, the costs outruns the means to produce them. Think about it, IIRC culture costs for social policies (in a empire without too many cities) goes from into the tens for early policies to maybe several thousands for late-game policies. Science goes from around 75 for the first techs to >10000 for late game techs. Science production is generally greater than culture production, yes, but not 5 times, let alone 10 times greater. And I'd say during the course of a game, you need much more techs than you need policies.

 

[Txurce]

How many gross beakers were you using each turn at the end? And the end was turn what?

 

[Seek]

One way to help this would be to move the late-game culture wonders off the science line. We could also buff the Research Lab by population and/or %.

That being said, I wouldn't mind increasing the culture exponent to 2.4 and decreasing Modern Era tech costs a bit.