The Let's Play Archive

Civilization 2

by Melth

Part 9: Mechanics: Pros and cons of making military units early-game

Bloodly posted:

You're going into extreme detail, but at the same time, your reasoning is also not clear at all, because I think I'm missing the initial basis for the conclusions made.

I mean, I don't even understand why or even how the hell you're capable of spamming cities and settlers without any military. This is clearly because I am dumb and no reflection on you. I couldn't play on Deity at all. I would never say 'Oh no' to a new city.

Keep going.

Well I'm glad you're liking it anyway, but I definitely want my reasoning to be clear to people and for even newcomers to Civ 2 to understand what's going on. Do you have any particular questions I can answer? Are there any important aspects of the game I haven't explained well enough?

So you asked how I can spam cities and Settlers without a military. Good question. But the real question is this: Why would I want to have or not want to have military units?

Let me start with the reasons that building or owning a military unit can be bad and that I therefore want to have as few military units as possible:

1) Construction time. It takes several turns for a city to produce a military unit. During those turns the city cannot produce anything else which i might rather have (like a Settler). This is a serious problem because even a slight delay before producing my first Settlers can have a compounding effect that hurts me throughout the game. See, the sooner I build more Settlers, the sooner those Settlers can turn into new cities. Once I have more cities, I can produce even more Settlers faster. Which lets me found new cities even faster. And so on. The more cities I have, the better off I am in every way. Each city gives me Tax money, Science progress, and can build me more units. Therefore, spending time building military units hurts my whole economy- and it can even hurt my military in the long term.

2) Upkeep cost. Every unit has an upkeep cost which I must pay every turn. This cost is paid in "shields", which are the units of Production. For example, a city might start off with 2 shields worth of Production every turn. This means it would take 5 turns to build a unit which costs 10 shields- like the basic "Warriors" military unit. However, that newly built Warrior has an upkeep cost of 1 shield per turn. Therefore my city now only has 1 shield worth of Production to spare- the other shield is automatically spent maintaining the Warrior I already built. So if I wanted to build a second Warrior, it would take 10 turns instead of 5. And since that Warrior would also cost 1 shield, my city would no longer be capable of producing anything at all. Both of its shields would be spent on upkeep every turn instead of making anything new. Thus, building too many military units can prevent you from building anything else ever. Now my government starts off as a "Despotism" and I eventually turned it into a "Monarchy" and then a "Republic" Each of those governments has different special perks and problems. One of the perks of Despotism and Monarchy is that the upkeep of several units per city are paid for free. However, Republics do NOT have that advantage and it was part of my strategy to change my government into a Republic as soon as I could. Therefore, I knew that I would need to pay the upkeep of every single unit in my military eventually, with none of it for free, so I had an incentive to keep my military very small.

3) Unhappiness. One of the special disadvantages of a Republic is that my citizens become unhappy (which can cause "Civil Disorder", a major problem) if I have military units out in the field. This means I have a major incentive not to send my military units out to fight somewhere else in the first place. And if I'm not going to use them to fight, they're obviously not that useful.

So with the reasons to not build military units out of the way, what are the purposes military units serve and the reasons I might want to build them afterall?

1) Scouting. Several military units (like Horsemen) are capable of moving very quickly. This makes them very good for revealing the map and thus helping me figure out where I should build my cities in the first place. HOWEVER, Horsemen were not available to build at the start of the game. The only military unit I COULD build doesn't move faster than a Settler, so it can't actually explore better. By the time I could build Horsemen, I actually had done most of the exploring I needed to. Thus I didn't build any scouts.

2) Preventing Civil Disorder. As I've mentioned a few times, "Civil Disorder" is a status a city can be in. While in Civil Disorder the city doesn't produce any units and doesn't contribute any Science or gold. Essentially, the city stops functioning entirely. What causes Civil Disorder? Well every city has a size, starting at 1. That size increases as the city stores up Good (the food gauge to increase size is always in the top-right of the city menu). A city is said to have a number of citizens equal to its size. Those citizens can be "happy", "content", or "unhappy". Think of "content" as the normal situation. If there are more unhappy citizens than happy citizens in a city, the city will go into Civil Disorder. It will stay in Civil Disorder until some action the player takes causes there to be as many happy citizens as unhappy citizens. So if you wanted to end Civil Disorder, you would need to either turn some content citizens happy or turn some unhappy citizens content. There are MANY ways to do that and I outlined a bunch of them in this post:

One of those many ways to cause unhappy citizens to instead be content (and thus end Civil Disorder) is what's called "martial law". This is an option under Despotism (the default government) and Monarchy, but NOT as a Republic. The way martial law works is that for every military unit you have in a city (up to 3), one unhappy citizen there becomes content. This is a decent way to end Civil Disorder. In fact, I did a bit of this. I DID build a pair of Warriors and also a single Horseman whose main job was to sit in cities that had been in Civil Disorder so that the disorder would end and the city would function again. HOWEVER, I knew that I was planning on becoming a Republic as quickly as possible. You CANNOT use "martial law" as a Republic. Therefore, any military units which I built just to enforce martial law would become useless for that purpose. Therefore, I decided not to build many military units for that purpose in the first place. Instead I used some of the other methods for preventing or ending Civil Disorder, such as building Temples or preventing the cities from growing past size 1 (cities above size 1 have more unhappy citizens in this difficulty mode)

3) Defeating barbarians. Barbarian units can appear on any tile at any time and will begin trying to kill your units and conquer your cities. They are a major problem. You can build military units to try to defeat the barbarians when they appear. HOWEVER, barbarian units are just too strong and too numerous for me to defeat them in battle easily anyway. In particular, the military units I could build at the very start of the game (Warriors) would be completely worthless against any barbarians. Therefore, there wasn't any point in building them to do that. In general, the sheer number of units I would have to build all over my country to be safe at this point is just too high to be feasible. Fortunately, the barbarians will allow me to pay them money to go away and not attack my cities. This is a much better deal, so that's what I often do until it becomes more feasible for me to maintain a strong military.

4) Defeating other civilizations. Sooner or later, fighting with another civilization is inevitable. Military units are essential for that. HOWEVER, it's easy to avoid war at the beginning of the game and a war then would not be good for me. Since I didn't want an early war and was capable of avoiding one, there was no need to build a pile of military units to fight it.

Thus what you can see is that military units would actually not help me much in any way at all. None of their 4 basic purposes would be useful with the strategy I'm using. And, in fact, because there are disadvantages and penalties for owning or making them in the first place, they might well be less than useless to build. So that's why I didn't make many.

It also helped that I happened to get a few free military units as random prizes at the goody huts early on. By using those 2 free units (plus a single one that I built) skillfully, I did the work of an entire army

Carbolic Smokeball posted:

If you're going to spam cities early game, Monarchy's martial law benefit is a great way to control happiness. It's for this reason I rarely b-line for republic anymore

As I think I demonstrated in yesterday's update, Republic has its own great way to control happiness in numerous cities. Furthermore, Republics experience slightly less unhappiness from owning a large number of cities in the first place.

Making a pit stop to Monarchy can definitely be a tempting option, but I believe that it ultimately does more harm than good if you're definitely set on playing as a Republic/Democracy eventually. All of those martial law units suddenly become not only useless for that job but also a major drain on your Production. And of course you've set your Science back by dozens of turns. On the other hand, pure Monarchy all the way is the key to at least one extremely powerful Deity strategy. That's the one I typically use to win by 1500 AD.

Fister Roboto posted:

I just need to point out that the plural of metropolis is metropolises if you're speaking English, or metropoles if you're speaking Latin.

Melth the immortal God-Emperor of Romankind speaks only Roman. Any resemblance whatsoever to English is purely coincidental.

ajkalan posted:

One interesting quirk with the Barbarian leader units is that they become Diplomats when your own Diplomat or Spy bribes them to join you, like some kind of Diplomat zombie-bite victim. On lower difficulties where I was flush with gold, I enjoyed bribing them and using my new Diplomat duo to buy an army from the now-leaderless Barbarians. Not really practical or useful, but bloodlessly destroying your enemies like that feels like its own kind of accomplishment.

Ah, I'd forgotten about that! Someone needs to step up and be the OrangeFluffySheep of this LP; Civ 2 is a game that demands drawings and cartoons of its zany action!

Using Diplomats on barbarians actually can be a good strategy. They're often fairly solid units and they can be bought dirt cheap. Better yet, it pays for itself if you can use one new barbarian unit to capture the leader. Takes a bit of tactics, but it can work.

Fionordequester posted:

Man, you're really getting violent now aren't you Melth? I mean, Xerxes looked like he was kind of a mean guy anyway, but are you sure you're comfortable bullying people so much? War can be an awful thing you know...

In all seriousness though, it looks like you've just about gotten the hang of your screenshots now! Only nitpick though. With screenshots like this...

It's a bit hard to tell what's going one. I mean, it SAYS there's an archer about to attack the Elephant, and the text below that shot said there was, but I couldn't actually SEE it at all. I spent like 1-2 minutes going "ok, wait, there's an archer? Where is it?". I think with situations like that, you'll want to be careful to make things visually clear too, and not just try to explain the situation with text.

War is hell but peace sucks!

Unfortunately there's actually not much I can do about that screenshot and others like it. The Archer actually WAS invisible when that announcement came. Units appear and disappear a lot in this game at odd times. I guess I could have doodled in a stick figure Archer or something to show where he was supposed to be?

Redmark posted:

I usually find exposition about game mechanics and tactics kind of dry, but something about the way you do it is so irreverent and fun It's probably the slightly arrogant tone.

Also your attention to detail is pretty impressive, both in this and in the FE LPs.

Glad you like it!

sheep-dodger posted:

Wait, all those combat bonuses are multiplicative with one another? Good lord, who thought this was a good idea?

Me! It's a marvelous idea. Along with the unit stack mechanics, it's one of the only two reasons why combat is tactical in this game.

To understand why, you first need to notice that in every stage of the game most units have much more Attack than Defense. For example, in the ancient era there's 5 Attack/1 Defense Crusaders or 6/1 Catapults vs 1/2 Phalanxes. In the modern era there's 12/1 Bombers and 12/2 Howitzers vs 10/6 Tanks.

However, there are basically no ways to increase your Attack power other than to be a veteran, while there are a half-dozen ways to boost your Defense.

Thus in every era the defender deployed carelessly is easy prey but the maximally fortified defender is borderline invincible. For example, in the ancient era the toughest possible unit would be a veteran Phalanx in a city with City Walls on a Mountain that has a River. 2 (base) x 1.5 (veteran) x3.5 (Mountain w/ River) x3 (City Walls) for a total of 31.5. Sidenote, ordering the Phalanx to "fortify" does NOT stack with City Walls. The 50% fortify bonus is superseded by the 100% being in a "fortress" bonus or 200% in City Walls bonus. This means that, contrary to popular belief, there is actually no benefit to ordering units in such places to fortify. You'd be better off ordering them to sentry if you just want to not bother with them.

Anyway, the principle that defending units are naturally easy to kill but are extremely tough if you stack the multipliers is key. It forces people on defense to play smart, plan their moves carefully, etc. And note that EVERYONE is on defense at least half the time. Even if you're invading someone, you're still playing defense on their turn, so you had darned well better think tactically. A gigantic attack force that could crush any defense can be slaughtered in a few moves if the invading player is dumb. But if there were units with high base Defense, that wouldn't be true. All the invader would need to do is stack his invasion force up with the high Defense guy and move in. Or if the Defense bonuses were just additive and one of them (like terrain) was just particularly large, the attacker could focus on maintaining that one Defense bonus and be too tough to attack. Multiplicative stacking provides a big incentive to make sure you take advantage of several bonuses at once instead of just homing in on one big one.

This principle also means you need to think very carefully about military matters when founding your cities. If you put a city on a Hill, it may well be impossible for the enemy to take. But its economy will suck forever. If you build next to Forests or Hills, you can use them for Production. But an invader could stand on that Hill or Forest and be too tough to dislodge while they stack up Catapults or Cannons to crush your defenders.

So the low natural Defense but massive power of stacked Defense multipliers make tactics very important whether you're invading or trying to stop an invasion. If there were a lot of stackable Attack multipliers too, that would cease to be true. And that's why there's only being a veteran- which is also a Defense bonus.

But it gets even better.

Besides requiring good tactics, the multiplicative Defense bonus phenomenon also forces good STRATEGY. See, it's quite feasible to put together a defense which can crush any conventional attack. But instead of making invasion impossible against a competent player, what that means is that the attacker needs to get smart. Don't just send your force of 50 Crusaders at the uber Phalanx and hope to chip him down, find a way to walk around him. There aren't a lot of Mountains in the world. And there's like .1 per game with a River. So set up a colony or use transports and attack from an unexpected direction. Or use a Diplomat to buy the city. Or to explode the City Walls. Or attack with a ship- City Walls don't work against ships.

So this mechanic is the fundamental reason that you must think carefully about how to launch your attacks and use combined arms rather than one-unit masses to be effective. This is especially true in the modern era. The enemy can easily be invincible against Diplomats OR land assault OR nukes OR ships OR planes, but he sure can't be invincible against all of them everywhere. Find what he's weak against and where he's weak against it and smash him there. Have Battleships shell coastal cities with City Walls but no Coastal Fortress. Have marines take cities with Coastal Fortresses but no City Walls. Surround big cities and stand on their good terrain so they have to disband their own troops and starve to death. Use Nukes + Alpine Troops or Paratroopers to snatch up cities that don't have SDI Defenses. Buy cities that don't belong to Democracies using Diplomats and Spies. Blow up the City Walls of Democracies using Spies. Or hit them with Bombers or Howitzers. Have Spies plant nuclear devices.

Or my favorite tactic of all: attack the planet itself by causing global warming and then sit back and prosper while they starve or disband their armies because I was smart and relied on Oceans while they relied on land terrain.

Warfare in the modern era is cool because the defender can still maintain the edge against what he's trying to defend against, but there are dozens of different types of attacks you could launch. Of course, there's no way to have the resources and time to employ them all at once. So you've got to make choices. And that's what strategy IS: choosing how to use your limited resources to accomplish your goals.

Carbolic Smokeball posted:

Also some comments on what you said about wonders.

I disagree with your assessment of Colossus. It is an awesome early game wonder, especially where trade can be at a premium. If you're going for a super science city it's nigh essential. It also expires fairly late game depending on how soon you want to get flight so those 200 shields will pay themselves back many times over in that time, ESPECIALLY if you build caravans in that city (or take part in the morally dubious strategy of re--homing all your caravans to that city) That all being said, it's not game-ending to miss it.or really any ancient era wonders.

The game's description of Copernicus' Observatory is incorrect. It actually doubles the science output of the city it's built in. There are actually a few wonders in the game that have incorrect descriptions. The game claims Isaac Newton's College doubles the science output of a city (what Copernicus' actually does), but in actuality it only doubles the extra science produced by the Library/Uni/Lab. As a result, building Newton in a city with no science improvements will have no effect. The last one IIRC is JS Bach. The game says its bonus is applied to all the cities on the continent, but actually applies to all cities regardless of where it's built.

If well planned, Darwin's Voyage can be give a very useful research bump if you have the production to spare. But it's merely about timing: completing DV when you're a turn or two away from completing a tech is incredibly wasteful. But getting it close to completion with a fresh tech being researched and you can turn off your science and pile up some gold for a few turns, or just complete it and keep ahead in the research game. By that point in the game I'm usually far enough ahead to spare the production. Again, not an essential tech, but not useless either. I think Eiffel Tower is unanimously considered the worst wonder, especially in MGE where diplomacy is broken as hell. But even in regular Civ 2 it was useless.

Thanks for the tip-off about wrong wonder descriptions. I just did a whole bunch of experimenting and confirmed that the in-game description of Copernicus's Observatory, Isaac Newton's College, J.S. Bach's Cathedral, SETI Program, Apollo Program, and Hanging Gardens are all false or misleading or incomplete:

-J.S. Bach's Cathedral does indeed work on every continent (same as in Freeciv).

-So does Hanging Gardens (actually, the description of that one doesn't specify that it only works on one continent but I know I read that somewhere). Hanging Gardens also makes 3 people happy in the city it's built in, but the game doesn't mention that.

-Copernicus's Observatory does indeed grant +100% Science instead of +50%. And this is MULTIPLICATIVE with Libraries, Universities, etc (those buildings stack additively with each other). That makes it more than twice as good as I thought it was.

-Isaac Newton's College doubles the effects of Libraries, Universities, and Research Centers but does nothing on its own. This is always strictly worse than Copernicus's Observatory for the same number of research buildings in the city. AND it costs more to build and is available later. Fail. This is definitely on my worst wonders list. Copernicus's Observatory moves up to just being low-tier.

-Apollo Program not only allows construction of spaceships by all players, it also reveals the entire map to the player who completes it (the game doesn't mention this in the description, but you get a message that astronauts bring back photos from space and then all is revealed).

-And SETI Program DOES count as a Research Center in all cities as it claims, but the description that this "effectively doubles your Science output" is completely false. It's a 50% increase over your baseline. It is a smaller % of your total if you have ANY Libraries or Universities or Research labs already.

This is about to get very long and detailed. The TL;DR version is that the Eiffel Tower is actually pretty good, Darwin's Voyage can be mathematically shown to suck, and so can Colossus.

Now you're obviously free to have your favorite wonders, but the math is just not on your side for Darwin's Voyage or Colossus really.

Before I get deep into those, let me talk about the Eiffel Tower. I actually consider it a mid-low tier wonder and I never heard anyone say they thought it was the worst before. The Eiffel Tower's effect is not a huge one, but it's available early and it's cheap to build and its effect is unique: there is no other way in the game to boost your diplomatic reputation.

And diplomatic reputation is a very important resource. First of all, it's worth a lot of cash. People will pay you tons of money to not fight them if you have a spotless reputation and you know how to put them in a corner. With a very good reputation it's also much, much easier to make AI civs friendly to you and thus trade techs with you or to successfully demand tribute for your patience or to get them to share maps. And it makes other people much less likely to betray or sneak attack you or refuse to sign a peace treaty when you want one. Oh and forget about having allies without it. Iirc is actually impossible to EVER get someone to make an alliance with you if your reputation is not at least Excellent (one step down from Spotless, and bear in mind that a single breach of treaty drops you 2 steps).

I said earlier that almost nothing is worth breaking a treaty over, and the near-impossibility of ever repairing your reputation is why. The Eiffel Tower is your one and only chance to fix your reputation if you do need to get up to some skullduggery. Does that make it a great wonder? Definitely not. It's fairly low tier in my eyes. But at the very least it can be situationally good, and unlike most other wonders there is nothing else that can do its job.

What about Darwin's Voyage and Colossus though? Well, Darwin's Voyage only looks worse now that I know Copernicus's Observatory is more than twice as good as advertised. Darwin's Voyage has this huge problem that there are a ton of other wonders that do its job better. The following wonders directly increase your Science or grant you free techs: Colossus, Copernicus's Observatory, Darwin's Voyage, Great Library, Isaac Newton's College, and SETI Program. Numerous others indirectly boost your Science by letting your cities grow faster or letting you devote more citizens to high-Trade tiles and so on. But let's just talk about the directly science-y ones right now.

With all the science-y wonders there are only 3 questions that matter: 1) how many techs will this actually get me? 2) When will it get me them? 3) How expensive/difficult to access is the wonder?

First of all, the SETI Program can't really be compared to the others. For one thing, it's like 20x more powerful than all the others combined. A 50% increase to Science in every city in your late-game civilization is mind-blowing. However, it's unlocked by one of the latest techs in the game. By the time you could get it, the tech race is basically over. There is nearly nothing left to research. And you're probably so far in the lead anyway that it doesn't matter. There's just no way to really compare their usefulness.

Second, the Great Library dominates all the others in all three categories. It's available almost immediately, gives you its benefits in the early-mid game when you need them most, and the sheer number of techs it can give you on Deity difficulty is amazing. Especially since it works BETTER the cruddier your starting situation is. Oh and it's THE science Wonder for people who aren't playing Republic/Democracy.

So the Wonders I actually want to compare are Darwin's Voyage, Colossus, Copernicus's Observatory, and Isaac Newton's College

Now Darwin's Voyage grants you 2 techs. Actually, as you touched upon, it grants you 1 and a fraction techs unless you time it properly and know what you're doing. Let's assume the hypothetical best case scenario where it grants you exactly 2 techs. Now the issue is that these are mid-late game techs. By the time you get Railroad (to unlock Darwin's Voyage), you should already be way in the lead in the tech race. And, as I mentioned, you already must have or could easily have all the techs there's actually a rush to get to: Republic, Monotheism, Democracy, and Railroad itself. So Darwin's Voyage doesn't look so great with regard to question 2; it gives you the techs at a less useful part of the game. It also looks bad in 3; it costs as much as or more than every one of those wonders and is also available later than every single one. So Darwin's Voyage sucks with regard to issues 2 and 3. The only thing that can save it is being better than the others at 1), granting lots of techs.

Let's compare how many techs Darwin's Voyage and Copernicus's Observatory can give.

Now that's a hard question because there are a ton of variables. I'm going to outline 2 different scenarios and compare the numbers in each case. The first scenario will be extremely biased in favor of Darwin's Voyage. I'll assume a lot of poor decisions on the part of the guy considering building Copernicus's Observatory and also that the game ends before he can get much use out of it, among other things. I'll call that scenario "Pro-DV". The second scenario will be a fairer comparison with much more reasonable numbers, but still not using Copernicus's Observatory to its full potential as it might be used by an expert. This will be a more normal game and I'll call that scenario "Standard"

So let's talk "Pro-DV" first. Now Copernicus's Observatory is complete trash for smallpox, so the only people who will ever build it are either Bigpox or Celebration players. By the time they get it, they will have at least a size 8 city. To be absurdly generous to Darwin's Voyage and keep the calculations simple, let's say it's size 8 and never grows larger than size 8. But it's a solid Science city. A size 8 Science city should have 26 Trade (8 Oceans + the city tile with a Road as a Republic) with maybe a point or two of Corruption subtracted out. Let's pretend he never goes Democracy or does anything else to eliminate this Corruption. Call it 24 Trade.

Let's be even more generous to Darwin's Voyage and say the Copernicus guy doesn't average 80% Science rate; he does more like 60%. Ok. So 24 Trade x 60% to Science makes 14 beakers (rounding down). That's the Science output of that city every turn before the wonder or anything else.

Now he will definitely already have a Library and let's say he gets a University 50 turns after building the Copernicus's Observatory. And let's say he never, ever makes a Research Lab. So in the first 50 turns, his Science output would be 21 per turn without Copernicus's Observatory and is instead 42 per turn with it. Thus the wonder itself gets him 21 beakers per turn x 50 turns = 1050 beakers in the pre-University phase.

Now we need to know how long the University phase lasts before the end of the game. Let's say it's a Deity game (which means the fewest turns and is thus best for Darwin's Voyage). Let's say Copernicus's Observatory is built in 1 AD (turn 100) and the University thus comes out at 1000 AD (turn 150). And let's say the game ends fairly early, 100 turns later in 1850 AD. So he has 100 turns with a University on top of the Library. With those 2 buildings but no Observatory he would get 28 Science per turn. With the Observatory he gets 56 Science per turn, so the Observatory is profiting him 28 per turn for 100 turns. 2800 beakers from the University phase.

Add that and the Library phase and we have a total of 3850 beakers created across the whole game by Copernicus's Observatory in this scenario where it's at just about its lowest feasible effectiveness.

Now how many techs is that? Well that's hard to answer since the tech price depends on a ton of stuff. Most importantly, early techs are cheaper than late techs. The most relevant question is how much the techs just after Railroad (when Darwin's Voyage is unlocked) cost since those are the ones Darwin's Voyage will probably be used to acquire. I believe Railroad takes a hypothetical minimum of 25 techs to acquire. But to be VERY generous to Darwin's Voyage once again, let's say the guy actually messes around a TON and acquires a whole 25 other techs either before getting Railroad or before finishing Darwin's Voyage. Thus Darwin's Voyage gets him techs 51 and 52 for free. The question is whether 3850 beakers could purchase both of those techs.

How many beakers do techs 51 and 52 cost those cost? Well that's still unanswerable without a few more assumptions. Like most things in this game, the mechanics of how much each technology costs are bizarre and convoluted and secret and depend on loads of things. This site has most of the details for you if you want a look at the details:

The main thing we need to know is how many more techs the guy has than the "Key Civ" or "Reference Civ". You can manipulate who that is by exploiting a lot of the secret mechanics and underpinnings of the game, but no normal player has any idea how to do that and it's kind of cheating anyway. Therefore, let's assume the Reference Civ is a random AI opponent. In that case, we need to know how many more techs this guy has than the average AI opponent. Let's be generous once again and say that he has 18 more techs than the average AI civ despite playing so poorly that none of his cities went above size 8 and he never switched to Democracy or anything. This will greatly increase the price and thus help Darwin's Expedition in the comparison yet again.

The price of tech 51 is 1632 beakers under these assumptions. Tech 2 is 1664. Their combined cost is 3296, which is less than 3850. So Copernicus's Observatory can buy the techs that Darwin's Voyage granted with spare change left over. And this was under absolutely absurdly favorable conditions for Darwin's Voyage! I mean, I assumed that the Copernicus guy NEVER built his city over size 8, NEVER put his Science rate above 60%, NEVER eliminated Corruption in the city, ended the game VERY early, didn't build his University till late, NEVER built a Research Center, and worked hard to make the cost of the techs Darwin's Voyage gets for free extremely high. And Darwin's Voyage was STILL worse. And that's not even considering the fact that it costs more shields to build and doesn't start helping you till much later!

Now let's see how horribly Darwin's Voyage gets crushed under the "Standard" scenario. Again, nothing that favors Copernicus's Observatory tremendously, just something that looks like how an actual game involving an actual player who's decently skilled and trying to get good use out of Copernicus's Observatory would go.

So first of all, his city should be at least size 12 on average. Let's still just call it 12. Maybe he only has 12 Oceans because this isn't actually a great Science location or maybe he's a Bigpox player rather than Celebration and he finishes higher than 12 but takes a long time getting there. Whatever, call it 12 for simplicity and to still be fairly nice to Darwin's Voyage. A size 12 Science city should have about 38 base Trade.

He's going to be smart enough to switch to Democracy thus get rid of the Corruption penalty, so he will actually have the full 38. And he's actually serious about Science and knows how to play with Luxuries and Taxes down around 10% or so most of the time and still turn a profit, so he has 80% Science.

He also knows that getting Universities is key for good Science gain, especially when he's trying to have one big Science city like this, so he gets a University in that city only 30 turns after building the Copernicus's Observatory instead of 50.

Furthermore, 1850 was an unreasonably early finish time for most players. A typical game would go on well into the 1900s, but let's just call it 50 more turns because we're still trying to be reasonably nice to Darwin's Voyage.

So how much Science does Copernicus's Observatory actually get him over the course of the game? Well in the Library-only phase it's like this:
38 Trade x 80% to Science (round down) is 30 Science. The Library multiplies that to 45. Then Copernicus's Observatory doubles it and thus creates 45 Science per turn. This goes on for 30 turns, so a total of 1350 Science in the Library phase. Then in the University phase it's 30 Science turned into 60 by the Library and the University, so Copernicus's Observatory gives 60 per turn for 170 turns. That's 10200 beakers of Science, for a grand total of 11550 beakers over the whole game.

How many Darwin's Voyage-era techs is that? Well in a real game, the player is going to be facing fairly stiff AI competition on Deity difficulty and also is unlikely to mess around a ton before getting to a critical tech like Railroad. Let's say he has 35 techs upon completing Darwin's Voyage, so it gives him techs 36 and 37 for free. Let's also say he's maybe 12 techs ahead on average instead of 18. In that case, techs 35 and 36 cost 1080 and 1110. A total of 2190. So guess what? Copernicus's Observatory is more than FIVE TIMES BETTER. And it's cheaper. And it starts helping you earlier!

What about Newton's College?

Isaac Newton's College is terrible and strictly worse than Copernicus's Observatory, but is it still better than Darwin's Voyage? Let's keep scenarios "Pro-DV" and "Standard" more or less intact, but update them for the later unlock time of Isaac Newton's College and the different setup it requires.

So back to the "Pro-DV" numbers. 24 Trade and only 14 Science in the size 8 city forever, 60% Science rate, game ends in 1850, the player has 50 techs before Darwin's Voyage finishes and is 18 ahead of the average AI opponent at the time, etc.

However, Isaac Newton's College has University as a prereq. And (assuming a player knows that the in-game description is a lie and thus knows how to actually use it), it would never be built unless the player already had a University. So let's say it's built in the year 1000, with a University already present, since that was the year that the "Pro-DV" player finished his University. There are 100 turns till the end of the game.

So each turn his city has 14 base beakers of Science and his Library and University would make that 28, but Isaac Newton's College makes it 42. So Isaac Newton's College gives him a profit of 14 per turn. For 100 turns. That makes a total of 1400 beakers, which is less than half the value of Darwin's Voyage under these crazy conditions.

What about under "Standard"? Well under "Standard" for Copernicus's Observatory the University was built 20 turns earlier, so let's say Isaac Newton's College is built at that time in this incarnation of "Standard". And the game lasts 50 turns longer, so he has 170 total turns to work with instead of 100.

And his profit each turn is 31 beakers. Remember, 38 Trade -> 31 Science -> 62 with Library and University -> 93 with Isaac Newton's College too so the wonder is giving him 31. Over 170 turns that's a total of 5270 beakers.

And the "Standard" tech price is about 1080 beakers, so Isaac Newton's College buys 5 techs to Darwin's Voyage's 2. That's right. Under even halfway reasonable circumstances for a guy trying to use Isaac Newton's College, even that inferior wonder is more than TWICE as good as DV. And it still builds earlier!

What about Colossus?

Ok, so this one requires a bit more consideration since the Colossus actually expires. Let's start off by taking it back to the basics of the "Pro-DV" scenario though: size 8 city, 60% Science all the time, game ends in 1850, etc.

But the Colossus would feasibly be built earlier than Copernicus's Observatory since it unlocks much earlier and is cheaper. Let's say it's built 30 turns early, with a Library, but that the University is still built in the year 1000. Then let's say Flight is developed about 20 turns before the end of the game since the Colossus guy would put it off for a while and clearly isn't winning by space race with his strategies.

So the Colossus generates +8 Trade in a size 8 Science city. At a 60% Science rate, that's +5 Science. The Library phase is now 80 turns long, and during that phase the +5 Colossus Science becomes 7 with the Library bonus. Then for 80 turns it's turned into 10 Science by the Library + University. Then the Colossus is gone. So over the course of the whole game that's 1360 beakers. Which at the inflated prices we're imagining is not enough to buy even one tech in the DV completed phase. Man, the Colossus sucks! But it's still better than Isaac Newton's College since it's at least generating some cash and happiness too

How does the Colossus look under the "Standard" circumstances? Well now it's a bit infeasible that the city is at 12 size when the Colossus is built. Let's say it's 8 for the first 30 turns then instantly becomes 12 the same time it was 12 when Copernicus's Observatory was built. Now how much further is Flight put off in this 50-turn-longer game? It's unreasonable to assume it's the full 50 turns; this player is researching lots of techs fast and will probably need a good one like that, though he IS trying to get good use out of the Colossus until he needs to start building Bombers for the final wars or tech to space flight. Let's say Flight is pushed back 20 turns.

So for 30 turns the Colossus is generating +8 Trade with 80% Science rate and a Library. That makes 9 extra Science per turn. Total of 270 beakers over the 30 turns of the size 8, Library-only phase. Then for 50 turns it generates +12 Trade with 80% Science rate and a Library. That should be 15 extra Science per turn. So 750 Science. Then there are 100 turns with +12 Trade with 80% Science and a Library and a University before Flight is discovered. So 1900 for that phase. Total Science produced by the Colossus over the course of the whole game? 2920. At the new, more reasonable price for techs it will buy 3 DV era techs. Thus even the Colossus is better than Darwin's Voyage under reasonable assumptions even WITHOUT getting into the fact that it's generating gold and Luxuries too.

Quod erat demonstrandrum: Darwin's Voyage is vastly inferior to all the other Science wonders under even remotely normal conditions for people trying to use those wonders even halfway decently. Funnier yet, it doesn't even shape up THAT well under ridiculous conditions cooked up to favor it. Copernicus's Observatory in particular still flattens it.

Thought Experiment About Super-Science Cities:

So you said the Colossus is good because you can use it as part of a super-Science city. That's an interesting concept and one I think most Civ players have thought about. Partly as a thought experiment and partly to analyze how good or bad the whole strategy is, I'm going to try to figure out what the best possible Science output one city can have is. Now technically this can be arbitrarily high because as far as I'm aware there is no limit to the number of Food Caravans you could bring in. Thus given unlimited time, you could make the population as high as you want and make everyone a Scientist. But this is an obviously terrible strategy. So let's ignore Food Caravans and look at the highest "natural" city size.

Now the apex Science city actually DOESN'T rely on Oceans. You can get more Trade in the endgame with land tiles because of Superhighways. Since you want a high population too, what you want is a city in almost-pure Grassland, every single tile of which has a River. However, you want there to be 4 (the maximum number of resources the generator will let any spot have in range) Mountains with Gold (NOT Hills with Wine as you might think). These must also have Rivers. There are 21 tiles. Each tile has Irrigation, Farmland (the city has a Supermarket), Roads, and Railroads (the city has Superhighways). 17 Grasslands each grant 4 Food, the 4 Mountains each grant 1, That makes 72 Food total, so you can support a population of precisely 36.

The government must be Democracy and the city must have the Colossus. Each Grassland with River and Road grants 2 Trade. This is boosted to 4 by Colossus and Democracy. Then to 6 by Superhighways. The Mountains with gold and Rivers grant 7. Boosted to 9. Then to 13 by Superhighways. So a total of 154 Trade.

Now the complicated bit is Caravans. You'd better have maximally good Trade boosting caravans to make this truly the best. The way to achieve that is actually for another player to have 3 cities just like this one. Which is obviously out of the question. Let's assume YOU have them instead. But no Colossus there of course. Oh and they're connected to this one by both Road and Railroad, both have Airports, the "Trade Route" was established with a Freight unit, and both of course have Superhighways. Since the Trade of the other 3 cities is 116 instead of 154, the Caravan formula you can find here ( says we should be looking at 357 Trade from the sum of the 3 trade routes.

So combining Caravan Trade and natural Trade, the city experiences a total Trade of 511. Assume a 100% Science rate, so that becomes 511 base Science (This city better have Shakespeare's theater). It has 20 citizens working tiles and 16 specialists who are all Scientists and thus grant +3 base Science each. Total base Science of 559. Assume it has a Library, University, and Research Lab of course. And Isaac Newton's College. And Copernicus's Observatory. Well the boosted Library, University, and Research Lab turn 559 into 1677. And Copernicus turns that into 3354. So there we have it! The hypothetical maximum Science that one city could generate without infinite Food Caravans and without relying on enemy civs also having perfect Trade metropoli is 3354. That's roughly enough to learn 1 lategame tech per turn on its own.

Know what the crazy thing is? That's just not that much. Even after assuming multiple 1 in a quadrillion chance terrain formations, infinite time to setup, and everything. Even a halfway decent Celebration strat guy at a similar phase in the game is going to be packing much more total over his giant civ. In my recent 1800s space race win, I had about 2500 beakers per turn before I even got serious about building Libraries- let alone constructing SETI. And in that game I didn't even build Sewer Systems and only controlled part of one continent.

Also, this super city will cripple itself. Once it forces you to learn Flight, it's going to take like a 1000 bulb hit. And actually so will all the cities it's Caravaning with.

Let's take a quick look at a non-absurd super-science city. The kind of thing I've actually nearly made. Ok maybe just a little insane. 1x1 island with a Rivered Grassland. Surrounded by Ocean with 4 Fish (slightly better than Whales for maxing Population). That makes natural size of 26. Democracy, 100% Science rate. Library, University, Research lab, Colossus, Isaac Newton's College, Copernicus's Observatory. Harbor and Offshore Platform of course. No Caravans since it's an island and they're hard to get out there. And you don't have any other comparably spectacular cities to make the boosts huge.

Grassland with Colossus and Democracy and River and Road (and let's just say Superhighways!) is 6 Trade. All 20 Ocean tiles are 4 Trade, so total of 86 Trade. So 86 base Science. Multiply by the Science buildings boosted by Isaac Newton's College and you get 258. Doubled by Copernicus to 516. That's all. That is the most you can get with an absolutely spectacular but conceivably findable super science city site without getting into Caravans. 516 is decent at that stage in the game. But it's going to be like 1/6 of your total science output if you have any kind of decent civilization.

This demonstrates quite handily that super-science cities are just not actually viable. Even stacking up every possible wonder and building on ideal terrain in the endgame only makes them pretty good, and a drop in the bucket compared to what Caravan cheese or celebration strategies will land you in total.

And that's why I regard all of the Science wonders except the Great Library as low or fail tier. Because they're actually just not impressive as a proportion of the overall Science your civ should produce.

Phew! Delenda est Carthago!