SGOTM 03 - The Real Ms. Beyond

[Kodii]

I know what binary science rate is, but I probably do not know the details. If someone is able to dig them up, I'd be glad to read into it

 

[greggo]

What I understand about binary science rate is the following:

You run at 0% science until you have enough gold on hand to let you fully research a tech at 100% science. For example, if tech A would take 10 turns to research at 100% science, and you'd be at -10 gold each turn, then you keep science at 0% until you have 100 gold, then research tech A at 100% science.

The benefits are:
1. Tech discounts - since techs cost less if other civilizations have them, letting as many AIs research the tech before we start cuts down on the price.

2. Gold on hand for upgrades/rushing.

3. Flexibility in tech choice. If we suddenly get into a war, we can change to military techs without having "wasted turns" in a half-finished growth tech.

4. Saves rounding errors... Haven't studied the math in detail, but basically all numbers are stored as integers, so running non-binary science makes your libraries, etc slightly less efficient.

I think that's the gist of it, let me know if I've missed anything, or if anything isn't clear.

EDIT: Not that it matters for our game, but warlords apparently negates point 4 by using 2 decimal places for science and commerce.

 

[Sir Bugsy]

At our present rate of posting, Alan is probably going to have to give us another thread because this one will soon bog down the system.

 

[Kodii]

Ok, thats pretty much what I knew already for binary science rate. Its good information for whoever doesn't know what it is though.

As for our post rate.. I've seen faster ... then again, we're quickly approaching the number of posts in our SGOTM2 thread

 

[Iainuki]

The granaries don't do us any good until we grow. In fact, I don't think they kick in until we grow once after they're built.

Not exactly. Granaries begin storing food from the time of their completion, so having them finished sooner does help. However, they don't actually return any food until you grow once after they've been finished, so in that respect, you're correct.

What I understand about binary science rate is the following:

You run at 0% science until you have enough gold on hand to let you fully research a tech at 100% science. For example, if tech A would take 10 turns to research at 100% science, and you'd be at -10 gold each turn, then you keep science at 0% until you have 100 gold, then research tech A at 100% science.

The benefits are:
1. Tech discounts - since techs cost less if other civilizations have them, letting as many AIs research the tech before we start cuts down on the price.

2. Gold on hand for upgrades/rushing.

3. Flexibility in tech choice. If we suddenly get into a war, we can change to military techs without having "wasted turns" in a half-finished growth tech.

4. Saves rounding errors... Haven't studied the math in detail, but basically all numbers are stored as integers, so running non-binary science makes your libraries, etc slightly less efficient.

I think that's the gist of it, let me know if I've missed anything, or if anything isn't clear.

That's a great summary. The article on CivFanatics about it emphasizes the rounding errors, but honestly, that's not the reason I bother with it. The most important factor is the strategic flexibility from having cash on hand and not committing to a tech until you can finish it in minimum time. On this map, the discount from AIs knowing is liable to play a larger role than usual: if the AIs do the usual business of trading around a tech all at once, the research cost for a tech could drop quite dramatically over a single turn once everyone has Alphabet and contacts with each other.

Speaking of Alphabet and tech trading, I've become convinced that not trading Alphabet for as long as possible (i.e., until some AIs have it and have started trading it around) is a powerful strategy for the human player: you not only get to serve as tech broker, but the AIs often ignore important techs, slowing down their growth if they can't trade for them.

 

[Compromise]

Not exactly. Granaries begin storing food from the time of their completion, so having them finished sooner does help. However, they don't actually return any food until you grow once after they've been finished, so in that respect, you're correct.

That's good to know. So I guess regarding whipping as soon as you turn to size 2 vs. right before you grow to size 3 is: is it better to multiply one tile's food by 1.5 or work both tiles. In St. Pete, I'd take the 2 tiles, in Rostov, just 1. Of course, the earlier you whip, the sooner the whip anger abates too.

That's a great summary. The article on CivFanatics about it emphasizes the rounding errors, but honestly, that's not the reason I bother with it. The most important factor is the strategic flexibility from having cash on hand and not committing to a tech until you can finish it in minimum time. On this map, the discount from AIs knowing is liable to play a larger role than usual: if the AIs do the usual business of trading around a tech all at once, the research cost for a tech could drop quite dramatically over a single turn once everyone has Alphabet and contacts with each other.

I agree.

Speaking of Alphabet and tech trading, I've become convinced that not trading Alphabet for as long as possible (i.e., until some AIs have it and have started trading it around) is a powerful strategy for the human player: you not only get to serve as tech broker, but the AIs often ignore important techs, slowing down their growth if they can't trade for them.

I think that's smart. From our lowly tech position, do we research the techs behind alphabet and then start trading them once alphabet is our of the box?

 

[Compromise]

On the game, I suggest we let Moscow complete the granary, then put a turn into a settler and whip it. We can send the St. Pete warrior down as a city garrison for the iron isle city. St. Pete will be small enough for a while (with all that jungle) that it doesn't need an MP. Or, we can move the one from Rostov over there.

That will give our galley time to verify whether that's an island or just a thin peninsula. Regardless, it's an incredible site. We could put workshops on all the flatlands and still feed them, though I'm not sure the game will last long enough for workshops to become truly good tiles.

We should decide whether we want to go for writing at full speed or keep some cash stored up in case we need to upgrade a warrior or two in a hurry.

Edit: Of course, the ultimate irony will be that we rush over with a settler and a warrior and two turns later a barb axe walks in and sets up shop.

 

[Iainuki]

On the game, I suggest we let Moscow complete the granary, then put a turn into a settler and whip it. We can send the St. Pete warrior down as a city garrison for the iron isle city. St. Pete will be small enough for a while (with all that jungle) that it doesn't need an MP. Or, we can move the one from Rostov over there.

That will give our galley time to verify whether that's an island or just a thin peninsula. Regardless, it's an incredible site. We could put workshops on all the flatlands and still feed them, though I'm not sure the game will last long enough for workshops to become truly good tiles.

We should decide whether we want to go for writing at full speed or keep some cash stored up in case we need to upgrade a warrior or two in a hurry.

This sounds like a plan to me. Will being able to upgrade a warrior to an axe actually stop an attack? Two axes should stop most AI invasion forces, but I'm doubtful that one could do it.

My suggested tech path takes us far away from making workshops useful, unfortunately. (Late Guilds, nonexistent Chemistry.) I don't think they'll be relevant enough of the game to bother with.

 

[greggo]

Not exactly. Granaries begin storing food from the time of their completion, so having them finished sooner does help. However, they don't actually return any food until you grow once after they've been finished, so in that respect, you're correct.

Not sure if I'm repeating what's already been said, but just want to clarify for myself on this point.

A size 2 city which is a turn or two away from growing to size 3 whips a granary. The next turn it is size one, and will grow the next turn; however, when it grows, it will have no food from the granary stored?

I believe I have made this mistake in some of my SP games - IIRC in previous versions finishing a granary the turn before growing, or the same turn as growing was ideal.

Going to take a quick peek at the save now to get a better idea of what's happening everywhere...

Okey dokey! Here are my initial thoughts, to play on tuesday night unless we're still discussing at the time.

I checked on a test game, and it costs 110 gold to upgrade a warrior to an axe.

Currently, writing will take us 7 turns at 100%. We have 43 gold on hand right now, and would need 56 to sustain 7 turns of research. One turn at 0% would give us +21 gold, so a total of 8 turns to research writing. It would take 6 turns at our current pace to get enough gold to be able to upgrade while researching writing. By this time, I think we'll have another gems mine and the floodplain online, so a turn _might_ be shaved off there. This would be a total of 12 or 13 turns to research writing. To me, I think the question is will any cities be ready to build libraries in 8 turns? (Is there anything else we need to rush to writing for?)

Onto the cities:

Moscow will produce, as suggested, a granary by hand (3 turns), a whipped settler (2 turns). With the two workers on the copper, it should be online in around 6 or 7 turns, if my math is right (doubtful, but I'll use that number anyway). After the settler I'm thinking some turns into barracks, then start on an axe as soon as possible.

St. Pete's can whip its granary two turns before it's due to grow to 3. After growing back to size two, I think it's next build should be a worker, which can be whipped ASAP (not coming near the happy cap for a while, I don't think the -2 will hurt at all).

Rostov will also whip a granary out ASAP. Currently, the next build is an Obelisk, but I think it could whip out an axe as well when it regrows.

Even if get writing in 8 turns, our copper would be hooked up by then, give or take a turn. With our current military situation, I would much rather be building an axe or two at that time than a library. Since there's no rush for libraries, I'd prefer to be safe and stock up some gold before writing starts.

Our galley can make one move west to further scout iron island and then turn back to pick up the settler and warrior from rostov. To me it looks like it will probably be seperate from the romans, but it's impossible to be 100% certain. Even if it is connected, I think we should still move to settle there.

 

[Iainuki]

A size 2 city which is a turn or two away from growing to size 3 whips a granary. The next turn it is size one, and will grow the next turn; however, when it grows, it will have no food from the granary stored?

I believe I have made this mistake in some of my SP games - IIRC in previous versions finishing a granary the turn before growing, or the same turn as growing was ideal.

That's right. This is a change from Civ3 and earlier, designed to reduce micromanagement.

 

[LKendter]

I checked on a test game, and it costs 110 gold to upgrade a warrior to an axe.

I thought it was more expensive, but I can't argue with the test game.

To me, I think the question is will any cities be ready to build libraries in 8 turns? (Is there anything else we need to rush to writing for?)

At this time Moscow is the only city with enough commerce due to the palace.

Moscow ... After the settler I'm thinking some turns into barracks, then start on an axe as soon as possible.

I agree 100%.

St. Pete's can whip its granary two turns before it's due to grow to 3. After growing back to size two, I think it's next build should be a worker, which can be whipped ASAP (not coming near the happy cap for a while, I don't think the -2 will hurt at all).

I agree with worker, as we really need them. St. Pete doesn't have the tiles right now for growth until some jungle is cleared.

Our galley can make one move west to further scout iron island and then turn back to pick up the settler and warrior from Rostov. To me it looks like it will probably be separate from the Romans, but it's impossible to be 100% certain. Even if it is connected, I think we should still move to settle there.

This is where I disagree. I don't want a stray city on the Roman island.

 

[LKendter]

The other micro technique that I think may be useful from here on out is binary science rate. It takes a bit to explain it, so I'll just pose the question: Does everyone know what this is and how to use it? If not, I'll try to explain it or point to a link that does so.

I somewhat know what it is. I have never used it as I think it is a PITA that adds extra MM to the game.

 

[Compromise]

I think you've got it, greggo. And I agree with LK's comments.

I think you have enough time to make a full move west with Zim (the island-exploring/settler-transporting galley) if necessary, though a half move might reveal the end of the island.

In Moscow, I'd switch to all mines, the quarry and the seafood. That will max our productivity and get us the granary in 2 turns. Then, with a turn into a settler, you can whip it quickly.

You'll want to move the St. Pete warrior down to Moscow in the inherited turn. That Moscow warrior will take longer to get to the coast than the settler, but it may not be critical that he be on the same boatload, though it would be nice.

I really like the idea of having enough cash on hand to upgrade one warrior to an axe. I'd much rather not do that, but as noted, we'll be concentrating on military buildup for a bit here, not library whipping.

From what I can see on the map so far, the spot I marked "Iron Island" is a great place to put the new city. Being able to work the mines will make that an awesomely productive map, especially for archipelago.

Edit: Also, I guess you can send Alan (the worker putting up a cottage on Moscow's floodplain) over to mine the iron. Obelisk first build, then workboats I'm thinking. And, of course, build the iron mine before the road so the city can work that tile asap.

I sure hope we get that city. Any national wonder suggestions for it?

 

[Compromise]

This post is an edited version of Ianuki's great post. I have simply edited it to outline form and am posting it here for reference. I have added no comments here, but I do have some and will post when I can.

Overall goal: Win by (backdoor domination) diplomatic victory. Thus we need the UN which is enabled by Mass Media: Astronomy + Printing Press -> Scientific Method -> Physics -> Electricity -> Radio -> Mass Media

Astronomy prerequisites:
Compass + Machinery -> Optics + Calendar -> Astronomy
Metal Casting -> Machinery
Writing -> Mathematics -> Calendar

Printing Press prerequisites:
Paper + Machinery -> Printing Press
Mathematics -> Currency -> Code of Laws -> Civil Service -> Paper (Preferred route)
Meditation or Polytheism -> Priesthood -> Code of Laws -> Civil Service -> Paper
Meditation or Polytheism -> Priesthood -> Monarchy + Writing -> Feudalism -> Civil Service -> Paper
Polytheism + Masonry -> Monotheism + Writing -> Theology -> Paper

Other Techs We Want:

Literature (req Polytheism): Great Library, Heroic Epic
Polytheism

Education (req Philosophy): cheap universities, stoned Oxford
Writing -> Mathematics -> Currency -> Meditation + Code of Laws -> Philosophy
Writing -> Alphabet -> Meditation + Drama -> Philosophy
Meditation -> Priesthood -> Code of Laws -> Philosophy

Liberalism (req Education): free tech
Education after philosophy above

Cossacks: (req Military Tradition, Gunpowder, and Horseback Riding)

Civil Service + Philosophy -> Nationalism + Music -> Military Tradition
Literature or Drama + Mathematics -> Music
Education -> Gunpowder
Monarchy + Writing -> Feudalism + Machinery -> Guilds -> Gunpowder

Mercantilism: req Banking
Monarchy + Writing -> Feudalism + Machinery -> Guilds + Currency -> Banking

Tech Path Proposals:
1) Iron Working, Pottery, Sailing, Writing, Agriculture
2) Great Library: Alphabet, Polytheism, and Literature
3) Below

Education through Theology:
Monotheism -> Theology -> Paper -> Education
Super-early universities (500AD?), Oxford(100AD?), Oxford science specialists
Great scientists can assist
Shortens AI's route to Liberalism

Philosophy lightbulb:
Drama -> GS pops philosophy (found Taoism)
early switch to pacifism
possible Angkor Wat allows 2H priests
Drama and philosophy have trade value
maybe Music first, burn GA on tech later

Economic Development with Naval Lightbulbs:
Early GS pop Compass and Optics (req hand-researching Metalcasting and Machinery)
Hand-research CS, Paper
Later GS pop Astronomy
Rough tech order: Math -> Currency -> (optional Priesthood) -> CoL -> Metalcasting -> Machinery -> CS -> Paper
gets us economy-improving techs
can trade techs

Combining The Three:

1) Mathematics -> Meditation -> Currency or Priesthood -> Code of Laws -> Philosophy (lightbulb) -> Civil Service -> Paper -> Education (lightbulb)
[vulnerable to AI Philosophy pop]

2) Mathematics -> Meditation -> Drama -> Philosophy (lightbulb) -> Theology -> Paper -> Education (lightbulb)
[very early liberalism]

Putting It Together: the Liberalism Slingshot

Either get astronomy asap or sandbag to get later, more expensive techs
If asap, best paths to get there would probably be the Philosophy lightbulb via Drama followed by Theology-Paper to Education, or Philosophy lightbulb via Code of Laws followed by Civil Service-Paper to Education.

Banking and Mercantilism:
Feudalism -> ... -> Guilds
Let AIs discount this tech line for us

Cossacks:
must choose between sandbagging liberalism and rushing to cossacks
if rushing cossacks, probably best to use liberalism for astronomy
-> Philosophy lightbulb via drama -> Music -> CS -> Optics (pop) -> Education (pop) -> Liberalism (& astronomy) -> Nationalism -> Military Tradition -> (trade for Guilds?) -> Gunpowder -> BOOM

Great People and Wonders:
Great Prophets (undesirable):
Build a shrine for a religion if we found/capture one.
Lightbulb Civil Service w/ Theology but w/o Monarchy.

Great Scientist:
*Build two academies, one in Moscow and one in St. Petersburg.
Lightbulb Physics.
*Lightbulb Education.
Lightbulb Printing Press.
Lightbulb Philosophy.
Lightbulb Optics.
Lightbulb Paper.
Lightbulb Astronomy.
Lightbulb Electricity.
Lightbulb Compass.

Great Engineer:
*Rush the Great Library.
*Rush the UN. (Maybe save 2)
Rush Oxford.
Rush Angkor Wat.
Rush the Taj Mahal.
Rush the Sistine Chapel.
Rush the Forbidden Palace.
Lightbulb Machinery.

Great Artist:
Lightbulb Radio.
Lightbulb Mass Media.

Wonders:
First tier:
Pyramids (done)
Great Library (to be rushed)
The UN (required for our master plan)
Hanging Gardens (More engineer points, more health, more population for whipping and specialists, we have the necessary stone--one of my favorite wonders in any game, and well-suited to this one.)

Second tier:
Angkor Wat
Sistine
Taj Mahal

National wonders:
Forbidden Palace
Oxford
National Epic
Heroic Epic
Globe
West Point, 2nd tier

Wonder Cities:
Moscow (GE focus): Pyramids, Hanging Gardens, "Forge"d engineer
St Pete (GS focus): National Epic, Oxford
Note: Iron Island might be better than St Pete because it has production!
Novgorod: Scrap world wonders
Low income, high production city: Heroic Epic
Food-rich: Globe

 

[Compromise]

Game goal: Backdoor domination (vote ourselves diplomatic victors) asap

Overall game plan: (Tentative)
Phase I: Early expansion with razor-thin defense. Build Pyramids.
Phase II: Consolidate defenses to repel and dissuade AI attacks. Local coast exploration.
Phase III: Quick early wars to eliminate long-term adversaries and build up land power. Secure circumnavigation bonus.
Phase IV: Tech rampup to cossacks. Build infrastructure that can support large empire.
Phase V: Rapid cossack-fueled expansion to acquire necessary population which researching post-cossack UN techs.

While we're taking care of the immediate needs of each turnset, I'd also like to get a handle on our overall strategy. Does this sound right? Or are there some phases out of order?

 

[LKendter]

Hanging Gardens (More engineer points, more health, more population for whipping and specialists, we have the necessary stone--one of my favorite wonders in any game, and well-suited to this one.)

I fully support this one. With stone I feel it would be crazy to pass this one up.

Junk city: Heroic Epic

How about low income city for the description? If "iron city" has NO income besides the water tiles, then how about Heroic Epic there? LK124 is the perfect example why I like better cities. 2 turn maces, 1 turn catapults, and 1 turns spears in the Middle Ages. That one city has been awesome for building military.

 

[LKendter]

Game goal: Backdoor domination (vote ourselves diplomatic victors) ASAP

Overall game plan: (Tentative)
Phase I: Early expansion with razor-thin defense. Build Pyramids.
Phase II: Consolidate defenses to repel and dissuade AI attacks.
Phase III: Quick early wars to eliminate long-term adversaries and build up land power.
Phase IV: Tech ramp up to Cossacks. Build infrastructure that can support large empire.
Phase V: Rapid Cossack-fueled expansion to acquire necessary population which researching post-Cossack UN techs.

I think we are close to starting Phase 2. We will be building real troops soon.

Somewhere in these phases we need locate the enemy. I think we need it soon with at least one "spare" boat exploring. Locating a good wonder, the shrine of the largest religion, etc could be very useful to our plans. Locating more AIs will also help with tech pace, and let us spread out tech trading among more civs avoiding the fracking tech cap.

Don't forget the circumnavigation bonus. Fastest finish could be decided by a couple of turns, and getting the troops there sooner will be the difference.

 

[Compromise]

How about low income city for the description?

I changed it to "low income, high production." Iron Island could be perfect.

Also, I added some exploration and circumnavigation statements to phases II and III respectively.

Good suggestions, LK.

What I'm most concerned about is us over- or under-expanding early on. I don't know which we're most likely to do. Since we have a technologically-motivated endpoint, I'm leaning right now toward very high-quality, high-population core cities to start with. We'll probably need to take a few AI cities (I'm thinking Mao and Alex, maybe Gandhi) early on so that our power ranking will be so high as to dissuade the AI from attacking us. Then tech. Then massive population grab.

 

[Iainuki]

I sure hope we get that city. Any national wonder suggestions for it?

I don't think we'll want to put any national wonders there, at least not any of the ones we've planned for. My intuition says that spot is too good for the Heroic Epic (since chances are it will want to be building things other than military at points), and it doesn't have enough food for the Heroic Epic/Globe combination. However, it's a possibility. How much military will we need?

 

[Iainuki]

I checked on a test game, and it costs 110 gold to upgrade a warrior to an axe.

Currently, writing will take us 7 turns at 100%. We have 43 gold on hand right now, and would need 56 to sustain 7 turns of research. One turn at 0% would give us +21 gold, so a total of 8 turns to research writing. It would take 6 turns at our current pace to get enough gold to be able to upgrade while researching writing. By this time, I think we'll have another gems mine and the floodplain online, so a turn _might_ be shaved off there. This would be a total of 12 or 13 turns to research writing. To me, I think the question is will any cities be ready to build libraries in 8 turns? (Is there anything else we need to rush to writing for?)

If we're whipping Moscow now for a settler, I don't see that rushing Writing is a necessity. If we don't need to upgrade a warrior, we can burn the cash for deficit research later on the Alphabet/Polytheism/Literature track. It would probably be a decent idea to whip Moscow for a library when Writing comes in.

Moscow will produce, as suggested, a granary by hand (3 turns), a whipped settler (2 turns). With the two workers on the copper, it should be online in around 6 or 7 turns, if my math is right (doubtful, but I'll use that number anyway). After the settler I'm thinking some turns into barracks, then start on an axe as soon as possible.

St. Pete's can whip its granary two turns before it's due to grow to 3. After growing back to size two, I think it's next build should be a worker, which can be whipped ASAP (not coming near the happy cap for a while, I don't think the -2 will hurt at all).

Sounds good. The road on the copper will take 2 workers 1 turn, and the mine 2 workers 3 turns, so it should be online in 4 turns if I'm counting correctly.

I'd be careful about overwhipping--it's not the happiness cap that's the immediate problem, it's the lack of ability to whip the city any more for 30 turns, since there are limits to how much unhappiness we can stack. (If, for instance, we wanted to whip out an axe there.) If you're pretty sure it will wear off before we need to whip again, go ahed, though.

Rostov will also whip a granary out ASAP. Currently, the next build is an Obelisk, but I think it could whip out an axe as well when it regrows.

The only problem with not whipping the obelisk is that it takes a long time to get the border pop, and I suspect the second gems mine will be done by then if we wait.

Even if get writing in 8 turns, our copper would be hooked up by then, give or take a turn. With our current military situation, I would much rather be building an axe or two at that time than a library. Since there's no rush for libraries, I'd prefer to be safe and stock up some gold before writing starts.

Good call. If the military situation is secure, we can whip a library in Moscow when Writing comes in, anyways.

Our galley can make one move west to further scout iron island and then turn back to pick up the settler and warrior from rostov. To me it looks like it will probably be seperate from the romans, but it's impossible to be 100% certain. Even if it is connected, I think we should still move to settle there.

Definitely find out if it's an island or a peninsula. If it's an island, we want to settle it now. If it's not, I'm still leaning to settling it: the only land bridge from Gandhi's/Caesar's island is a one-tile-wide forested strip, which would be easy to defend. (Chances are a fortified axe on the land bridge would deter any attack.) However, we should decide as a team what to do if the spot is connected, since there isn't universal agreement.