• Civilization 7 has been announced. For more info please check the forum here .

SGOTM 15 - One Short Straw

Is the tile at gold-4N a peak? If not, settling there will get the cow and be on the coast where there could be some seafood. :yumyum: Settling at deer-NE would mess up that location...
Gold-4N is a peak and the coast is completely blocked off as far as we can see. We'll have to defog the NE eventually.

On hunting, we don't need the worker to land on the deer till T30 if we decide to send the T30 settler to G1N. That means we do't need to start researching hunting till T24. The borders expand on T23, so we can limit our choices to:

AH
BW
TW
XQ

We'll still need to test with hunting, though, I guess.
 
I just ran a test game to see when the Oracle would likely go. I set up a new test game with 10 AI that don't start with Mysticism. It's possible that Neil had a few that start with Mysticism and forced them not to research Meditation or Polytheism...

It took quite some time T30, before any AI teched Mysticism. Polytheism went on T39. PH was learned by two AI (T84 and T88). The Oracle went on T97 (450 BC).

So it does look like the fact that they all have Buddhism and Hinduism went quite early that the AI avoid the religious techs. It could also have been the fact that I didn't have any religious nuts in the game...

I did notice that the owner of the Buddhist shrine doesn't tech any faster in the early game. He does, however, stockpile a TON of gold. I had fogotten that the shrine gives gold, not commerce, so it can't be converted directly to beakers. :blush: His tech pace will speed up as he expands because he'll be able to keep his science/espionage sliders at 100% for the entire game.

I also noticed that Toku starts with The Wheel. In this test game, he built roads while he was waiting for AH, his first tech, to be completed. So, in our game, hopefully he's improved one AH resource and built a few roads so that he can scoot over to the sheep to meet our warrior in the next 4 turns...
 
Gold-4N is a peak and the coast is completely blocked off as far as we can see. We'll have to defog the NE eventually.

OK, then it does look like settling Gold-1N will mess up the cows, which won't be able to be worked by any city. Do we care? Sure, the cow tile is a good one, but our second, third and fourth cities need to be as strong as possible, right? Based on the fact that we can't raze any cities and the fact that we're going to go after Toku and Hammi soonish, maybe we don't need a lot of filler cities.

On hunting, we don't need the worker to land on the deer till T30 if we decide to send the T30 settler to G1N. That means we do't need to start researching hunting till T24.

I assume that gives us enough turns to complete the camp while the settler walks to Gold-1N.

The borders expand on T23, so we can limit our choices to:

AH
BW
TW
XQ

We'll still need to test with hunting, though, I guess.

If we're sold on settling Gold-1N then yes, we should go Hunting next. However, if we're not sold on settling Gold-1N next, then we could start with something else and switch to Hunting if something good appears in the fog when the borders pop. However, settling by the gold is a strong play as it blocks Toku a bit and Gold-1N, in my opinion, is stronger than Gold-1E because the city gets up and running faster. As you said, testing is in order. I plan to run Gold-1N test today.

I vote for researching XQ next... ;)
 
Mitchum said:
Regarding Gold, the biggest reason I would consider settling it next is to work the early pre-improved deer. Since Gold-1E can't do that, I don't see the benefit. At gold-1E, you'd have to research AH next, improve the sheep on the turn it's settled and then work the gold after growing to 2 pops. This would put gold-1E way behind gems.
I'm a big fan of messing with the AIs. Settling aggressively toward Toku fits in with this paradigm, but since we can't raze Cities, we want to have a good idea of where he will settle so that we can know if settling a location like Gold-1E will actually mess him up or not.

Now, what I would be interested in seeing is a map that matches Toku's location and the rest of the land near him (if the current saved game doesn't already). Then, just end the turns a lot (make sure that it's the test saved game and not the real saved game! ;)) and wait until Toku settles... AI WILL sometimes settle differently due to Barb units and the location of where a neighbour settles, but for the most part, they will relatively stick to their guns.

In particular, if the test saved game has a unit or two of ours spawn-busting the area to Toku's north-west, we'll probably see where he will end up settling in the real game (unless the precense of a Horse Resource or something slightly alters his settling priorities).

Once you've got a good idea where he will settle, go back to the auto-saved saved game (maybe save a copy of it) that is, say, 6 or so turns before he settles (i.e. before he completes his Settler, so that he'll go through the settling decision again). Next, place a City at Gold-1E and then end the turns to see if the location of where Toku settles gets affected.

If his settling location gets affected, then try a third test where you reload your saved game from roughly 6 turns back (it might be less than 6 turns if Toku's new settling location was settled a bit quicker) and this time, wait for his settler party to move out before placing a City at Gold-1E, just to see how us settling while his Settler Party is on the move will affect him.


Only in that way will we get a good idea of what he's going to do... for example, it might be better to just let him settle a good spot for us if, by settling Gold-1E, for example, we force him to settle 1-off from the Fish Resource or something stupid like that.

On the other hand, he might end up settling a City that misses the Gold Resouce, in which case we'd only really have a Deer to pair up with the Gold (making for a yucky situation)... in that case, we'd have a strong argument for settling our own City by the Gold + Sheep + one of Deer or Corn.

Further, if Toku is just going to settle a City by the Corn (that perhaps also grabs the Sheep) but that doesn't mess up where we want to settle our Gold City and doesn't mess up a Fish City, but only if we don't settle a City in the east first, we may just want to let him do that so that he doesn't settle a crappy location for us to have to capture later.


LowtherCastle said:
Dhoom, you can see the peaks in my screenshot above.
Hmmm, are those Peaks at the edge of the map or just partway through the map?

Actually, the more important question is: if we settle a City on our mainland, can our Cultural Borders expand over top of those Peaks or will the Peaks be "invinsible from capture" for Domination Land Area calculations?

Even if they aren't invinsible to being captured, how feasible Culture-production-wise will it really be for us to get them within our Cultural Borders?

I mean, just how hard of a Domination Victory are we looking at here? Could it end up being that Conquest even becomes easier than Domination (assuming that no AI is isolated by a circle of Peaks)? We'd probably end up playing the two Victory Conditions in a similar manner anyway, but we may have to plan for the concept that we'll have to capture every single AI City before we can win the game, unlike last month's game where we were able to avoid warring with some of the AIs.


LowtherCastle said:
We'd have to figure out how to build a library quickly, in Delhi probably, and then work the 2 sci for 8.5 turns. That means a writing beeline. As a back-up, failed gold from GW and/or SH + REX would give us enough beaker-producing capacity to get COnstruction quickly.
As nice as it might be to beeline Writing, I think that we'd want Bronze Working first... Chopping (or even pre-Chopping using Fast Workers) or whipping will likely be required to get that Library built within a reasonable time frame. Writing before Bronze Working would only allow us to put a few manually-built Hammers into a Library first... probably not enough to warrant going Writing before Bronze Working versus getting whipping/Chopping earlier from going Bronze Working before Writing.

If the AIs won't Open Borders with us, then beelining Writing won't give us Open Borders, either (unless we're willing to gift City #3 just to get an AI to Pleased so that we can get Open Borders with him ;)).


ZPV said:
Do we need cats to take down Kyoto? If there are only 2 archers there, then they're costly and excessive. If there's anything more than that, then they're just the ticket.
Unless we Oracle Code of Laws, research Alphabet, gift him Code of Laws, revolt into Caste System, get a bunch of Espionage Points on him, build a Spy, and switch his Civic, he's going to have another Archer, likely another 2 or 3 Archers, from having whipped them.


ZPV said:
There's no point bombarding at 2%/turn for 50 cat-turns though - we'll just have to grin and bear it.
Even without Bombarding, Cats will make a huge difference in the warring. We could, say, throw 10 Chariots at the City and hardly even scratch the defenders. Throw 5 Cats at the City instead and we'll just need simple clean-up units to take down the defenders.


Would we be planning on pre-Chopping Forests for Catapults and then whip + Chop a stack of early Catapults?


Mitchum said:
How many city sites do we really need to be able to break out of said prison?
Four Cities with a Construction beeline and whipped + Chopped Catapults. Five Cities at most but such a rush is generally done with only 4 Cities to keep the Maintenance Costs down and because that 5th City really isn't going to be productive enough to produce units, with the exception of settling near a bunch of Forests and Chopping said units.


Mitchum said:
This applies equally to both Gold-1N and Gold-1E. Gold-1E does claim more new land, but the concept is the same.
Regardless of which Gold City location gets picked, we'll want to be sure to put spawn-busters in place... I really can't stomach the idea of risking a Settler's death to a Barb Animal if we can plan ahead to prevent it.


LowtherCastle said:
He might have connected all land into one landmass. In other words, a city on the tiles to the west of Gems might not give us 2cpt TRs...
So, you're thinking that the line of Peaks not only makes achieving the Domination Land Limit harder, not only blocks us in somewhat, but also reduces the chance of gaining the benefit of an instant +2 Commerce per City Trade Route, particularly in a game with no Colonial Maintenance Costs (due to having no Vassals)? Talk about a sinister setup!


LowtherCastle said:
Previously, I calced we would need 4 axes or 5 chariots per archer garrisoned in Kyoto, plus reserve for bad luck. That was without the -25% crossing river bonus, so it might be 5/6 now.
Yuck, there is a River to cross, too? Yeah, we definitely want Catapults. And Warriors. A Warrior lure or two might convince an Archer or two of Toku's to come out of his City. Certainly, the AI is good at setting up such bait (although the AI will do things like send out an Axeman that you can attack with a Chariot but then have a Spearman hiding right next to the Axeman that will kill your Chariot on the following turn); we should remember to do the same.


LowtherCastle said:
AH
BW
TW
XQ
What is "XQ"?
 
I played a test game and settled Gold-1N, Gems-1N, Marble-1SW and had another settler sitting for several turns. My tech path was Hunting -> BW -> Wheel -> AH -> Masonry -> Writing -> PH -> Math. I was about 7 turns from Math when the Oracle was built in 1160 BC (T71). I built a total of 6 workers, 5 warriors, 4 settlers, 1 Library and 1 Hindu Monastery. Religion spread to 2 of the 3 founded cities, but it took until T60 for the first spread to happen (that city was settled on T34).

I did not like this opening at all for several reasons:

1. It seemed like the only thing to build for a long time were workers, settlers and warriors.
2. It took a long time to hook up the cities, which is why it took so long for religion spread.
3. The library came late.
4. My cities were all small because I spent so much time building workers and settlers. Hence, I didn't whip once...
5. Without horses or copper in our test game, I was only able to build warriors.
6. The workers quickly ran out of things to do.

So, I'm starting to think that skipping Hunting and giving up on the deer for a while may be the way to go. I'm not 100% on this but I'm now leaning that way. We need more testing to decide on this, of course. Skipping Hunting means that our first two cities would be Gold-1E and Gems. Sure, not researching Hunting causes us to lose 20+ beakers on AH, but it also causes us to "waste" 62 precious, early beakers on Hunting which we could skip if we ignore the deer for now...

I think we need The Wheel and Writing sooner so that we can hook up our cities for religion spread and give us something to build (libraries would be useful in the captial, Gold and Gems, no?). Libraries will speed up our research and make the Construction beeline a lot more doable. Would TW -> AH -> Writing make sense? Does delaying BW that long hurt us too much?
 
Anyone know what XQ is, or did some hacker mess with my tech chart?

I thought XQ stood for eXtremely Quick and was the pre-req for Fishing. It's called XQ because you can research it in zero turns, but I don't want to waste that many turns researching it just yet because we have more important techs to learn!
 
Played another test game. Settled Gold-1E, Gems-N and Marble-SW. In gold my improvements were gold mine -> sheep pasture. In Gems my improvements were pig pasture -> gems mine. Tech path was TW -> AH -> Writing -> BW -> Masonry -> PH -> Math (T72). Two chops on T72 and the Oracle / Construction sling was completed on T73 (1080 BC). I had built 7 warriors (6 plus the starting warrior), 5 workers, 3 settlers, 1 GS (born on T73 in Delhi), 3 libraries, 1 rax and the Oracle. I had a TON of wasted worker turns building roads, which I assume will be used up to mine copper and/or pasture horses. :mischief:

I did not use any failed gold. I did not use 100% binary science, nor did I run 0% science slider while libraries were being built, so I think we can shave 3 turns from this easily. I deleted all forest growth. I removed religion when it spread. I whipped 3 libraries for 2 pops each. EDIT: The libraries, although a bit late, provided the needed borer expansions.

I liked this opening a lot better. With TW, workers had things to do. Getting AH early helped set up Gold (sheep) and Gems (pigs) quite nicely. Early Writing allowed early libraries, which made self-researching Math possible. Not having BW didn't seem to hurt too badly since I needed to be at 4 pops before I could whip, which happened right around the time I learned BW anyway. Earlier chops would have sped up REX, I guess, but it sucks to use too many forests when Math is around the corner anyway. I think I chopped a total of 4 forests (1 into Gems and Gold libraries and 2 into the Oracle).

Since I think settling aggressively toward the AI is much better than settling behind our own front line at Gems, I don't plan to test the Gems first option. I'll leave that for someone else.

BTW, 7 warriors seems like a lot but it was nice to use them to fog bust and to explore north toward Hammi... In our game, if we have horses, it may be more like 5 warrior MPs/fog busters and 1 chariot. EDIT: Or no RAX and a few chariots/axes.

Does anyone plan to do any testing? If so, which variant?
 
Nice work, Mitch. I wonder how much of the difference in feel was Hunting first and how much of it was BW-Wtg vs Wtg-BW. I don't have any time to test today, unfortunately.
 
Apologies guys. I completely spaced out uploading the save for T19...

WARNING: THIS IS THE REAL GAME SAVE!!!
 

Attachments

  • Gandhi BC-3240 eot.zip
    68.5 KB · Views: 91
Nice work, Mitch. I wonder how much of the difference in feel was Hunting first and how much of it was BW-Wtg vs Wtg-BW.

It's hard to say. Researching Hunting at all adds about 62 (Hunting) - ~22 (pre-req bonus beakers lost on AH) 40 beakers of total research. That's about 1.5 turns at 100% reasearch IIRC.

Gold-1E wasn't too bad. Obviously, it took longer to grow to 2 pops so I wasn't able to work the gold for ~3 turns after the mine was finished. Better micro may be to work the sheep (3F1C) until the gold mine is done, work the gold for 4 turns until the sheep pasture is done, and then grow to 2 pops working the improved sheep. I worked the unimproved sheep until I grew, and then started working the gold mine. Improving the sheep first could also be better allowing quicker growth to 4 pops = a quicker library whip... :crazyeye: One minor point, settling at Gold-N wastes a forest because you have to settle on top of it.

We really need TW and AH before thinking about Writing. If we slip BW and/or Hunting in there, I'm not sure that we'd have time to self-research Math in time. We could burn a GS on Math, but that seems wasteful to me. I'd rather use him to bulb up the Optics/Astro line or to build an academy.

In my first attempt, I researched Hunting, BW and Masonry all before Writing. I'm not sure what I was thinking there... :confused:

@LC, didn't you post your T20 save yesterday? That's the one I used. Should I be using this T19 save?
 
Great testing, Mitchum!!! I'm glad someone got off their bum. Have you noted when the AIs have been completing SH/GW? Have you noted whether we connect cities when we only have a road at Deer1E and a city at Gold1N? In other words, does the Deer-tile-eastern-river work without deer being roaded?

I've been simply analyzing the save and thinking things through. Got some data to add to the meat grinder.

1. On T23, the Gems lion gets trapped between Gems 1W/1SW/2S, so it will fogbust for us. The warrior will have to approach from the north to fortify a turn or two on the forest before getting attacked by the lion and hopefully surviving.

2. We'll need two fogbusters in the south, 1 to the east and whatever toward Hammy.

3. We could go 0% research for 1-3t till border expansion, yielding 9-27g, giving us 3-13t of 100% research, depending on settling Gems or Gold first.

4. Toku's settler will popT30-33 at the earliest. He's an eternity from researching BW (like 20t or more) and probably doesn't have a commerce tile since he beelined agri-AH, so he'll have to build his settler by hand. So he could settle Corn+Sheep as early as T33 or Corn+Sheep+Gold, T34. So if we settle Gold1E on T33, we're good to go, by a hair. Of course, we don't know what appetizing sites he has out in the Prison Parking Lot...
 
Mitchum, yes, I posted the T20 save. The T19 save is for demo screen analysis. Speaking of which:

Hammy just gained score=6, Toku score=8. That's the T0 land score, meaning Hammy has 6 tiles in his inner cross, Toku has 8. I wouldn't be surprised if Toku has coastal access, but then again, we're in Prison, so capturing Kyoto might keep us landlocked.

Both have completed two techs. I haven't anaylzed what Hammy might have researched, but it's not BW. Only Mahabodhi has that so far.
 
Have you noted when the AIs have been completing SH/GW?

No. :blush: I went back to my autolog and it's not in there for some reason...

Have you noted whether we connect cities when we only have a road at Deer1E and a city at Gold1N? In other words, does the Deer-tile-eastern-river work without deer being roaded?

Yes, that single road segment connects the two cities.

I've been simply analyzing the save and thinking things through. Got some data to add to the meat grinder.

3. We could go 0% research for 1-3t till border expansion, yielding 9-27g, giving us 3-13t of 100% research, depending on settling Gems or Gold first.

I'd consider running 0% research for one turn, but doing it for three turns delays our first tech by two turns, which is too much this early on. Unless we go Hunting next...

4. Toku's settler will popT30-33 at the earliest. He's an eternity from researching BW (like 20t or more) and probably doesn't have a commerce tile since he beelined agri-AH, so he'll have to build his settler by hand. So he could settle Corn+Sheep as early as T33 or Corn+Sheep+Gold, T34. So if we settle Gold1E on T33, we're good to go, by a hair. Of course, we don't know what appetizing sites he has out in the Prison Parking Lot...

I was able to settle Gold-1E on T33. I had built one road segment. I'm not sure if you could do the same without it though, which implies TW next if T33 is some magic number.

I'd love to be able to come to some agreement on what to research next. That would give us the ability to play forward 3 to 4 turns to see if we're able to steal a worker. It will also allow Delhi's borders to pop, revealing more land.
 
Mitchum: Note that the T19 save I posted above is the REAL GAME SAVE!!! ( I Just realized what you were asking in your last post.)
 
I found the wonder dates in the in-game log. Duh!

Stonehenge - 1920 BC and 2160 BC
GW - 1640 BC and 1600 BC
Oracle - 1200 BC and us in 1120 BC
GLH - n/a and 1200 BC
 
Mitchum: Note that the T19 save I posted above is the REAL GAME SAVE!!! ( I Just realized what you were asking in your last post.)

:eek: Oops. I guess we're settling Gems-1N next and researching AH...
Spoiler :
Just kidding.
Spoiler :
How dim do you think I am?
Spoiler :
Don't answer that!
 
Yes, that single road segment connects the two cities.
Okay, I tested this for Gold1E and without culture on Deer1E, we need both tiles roaded. So connecting G1E requires 3 roads, whereas connecting G1N only needs 1 road.
I was able to settle Gold-1E on T33. I had built one road segment. I'm not sure if you could do the same without it though, which implies TW next if T33 is some magic number.
With two fog busters we should be able to make T33 to G1E without any roads, unless we get some forest growth in an inconvenient location. I just tested G1E with AH beeline. We get AH on T32 meaning we know where any horse is before settling. That's a pretty good option, imo.

I'd love to be able to come to some agreement on what to research next. That would give us the flexibility to play forward to see if we're able to steal a worker. It will also allow Delhi's borders to pop, revealing more land.
Me too. :)
 
Okay, I tested this for Gold1E and without culture on Deer1E, we need both tiles roaded. So connecting G1E requires 3 roads, whereas connecting G1N only needs 1 road.

:confused: There should only need to be two road segments since we will have culture on the Deer itself. Seems strange to need 3 segments vs. 1. Testing... :scan: You're right, of course, but I had to proove it to myself.

With two fog busters we should be able to make T33 to G1E without any roads, unless we get some forest growth in an inconvenient location. I just tested G1E with AH beeline. We get AH on T32 meaning we know where any horse is before settling. That's a pretty good option, imo.

Yes, a very good option. It also allows us to pasture the Sheep first should we choose to do so. AH is a good next tech for Gems as well.

My vote for your next turnset (3 to 4 turns) is:

Border pop = knowlege
Research AH = knowledge plus flexibilty (works for gold and gems + horses)
Steal Toku's Worker = knowledge, flexibility plus totally awesome!
 
Top Bottom