SGOTM 11 - One Short Straw

[shyuhe]

That's what double wet corn will do for your start.

 

[mdy]

If we went with the grow to size 4 plan we could get two more warriors and 2 settlers by T51 if we put 2 chops into the first settler and do a two pop whip on the second. This could be bought forward to T48 if we did a 2 pop whip on the first and a 1 pop whip on the second combined with two forest chops. (No roads are built in either of these plans though)

Rexing so fast combined with delaying the silver mine could be a problem if we try to get something really expensive with the Oracle.

 

[babybluepants]

Rexing so fast combined with delaying the silver mine could be a problem if we try to get something really expensive with the Oracle.

Yup. Getting to Writing-Lib-Acad asap is essential (and working silver), if we want an Oracle shot. A third city on this schedule hurts tech in the Pottery-Writing span quite a bit. Probably delays Writing by 5-10t - don't exactly know now.

I think this third city and excessive chop-whip talk is contingent on what we find out by scouting in the next TS or two. I'm not crazy about putting down a crappy city in the face of a CRE neighbor purely for block value, if it kills early tech. If there's a good city or two (maybe essential resources, as well), that's different, ofc.

 

[LowtherCastle]

If we went with the grow to size 4 plan we could get two more warriors and 2 settlers by T51 if we put 2 chops into the first settler and do a two pop whip on the second. This could be bought forward to T48 if we did a 2 pop whip on the first and a 1 pop whip on the second combined with two forest chops. (No roads are built in either of these plans though)

Rexing so fast combined with delaying the silver mine could be a problem if we try to get something really expensive with the Oracle.

Very interesting. When do you get the first settler in these two variants? Edit: Okay, I think I see. 2pop on T43, done on T44.

 

[shyuhe]

If we went with the grow to size 4 plan we could get two more warriors and 2 settlers by T51 if we put 2 chops into the first settler and do a two pop whip on the second. This could be bought forward to T48 if we did a 2 pop whip on the first and a 1 pop whip on the second combined with two forest chops. (No roads are built in either of these plans though)

Rexing so fast combined with delaying the silver mine could be a problem if we try to get something really expensive with the Oracle.

That is my concern as well, since a fast rex without roads (2 +1 coin trade routes) will slow our research necessary for a good oracle. Plus, settling two additional cities without workers maybe be questionable as well, unless there are really good city sites out there.

So I think the issue we need to figure out is whether we want to tech wheel or BW here. Once we choose the tech, we can then send our scouting warrior east to try scoping out the land between us and Zara, as that's the biggest factor in our equation right now.

If it's great land, then we may need to consider spamming out two settlers quickly. I vote for BW here for the flexibility. Wheel may allow us to plant city #2 faster and give us road access, but without BW, we have no chance of spamming out a second settler (if needed). With silver, we should be able to get to wheel--pottery--writing reasonably fast even if we spam out two cities ni quick succession. While we might not get CS from oracle, we should still be able to get CoL or philosophy (my new favorite ).

 

[LowtherCastle]

@Shyuhe and all: Yes, of course, the entire purpose of the second settler is to grab a second excellent site. That's all. For that matter, it carries with it a risk, because barbs will come from the south, west and north and building the third city may trigger their spawning.

Here is my comparison study, as best I can determine. All of these are BW first (hence I called it Mutin/LC, though it's really just Mutineer's idea with BW first). TW first gets the first settler out on T43 but the second only comes around T58 with a 2pop.

      Mutin/LC   bbp        mdy1       mdy2
      --------   --------   --------   --------

T29              warrior2
T30   warrior2              warrior2   warrior2

T38                         warrior3   warrior3

-----------------------------------------------

T42   settler1   settler1
T43
T44                         settler1   settler1
T44
T45
T46
T47
T48              work mine             settler2
T49
T50
T51                         settler2
T52   settler2
T53
T54   work mine           
T55              settler2   work mine
T56                                    work mine
T57
T58
T59   pop4       pop4+pott  pop4
T60
T61   pottery               pottery    pop4
T62                                    pottery

My conclusions:
1. BW-TW
2. Send our current warrior toward Zara scouting as safely as possible.
3. Seeing Aksum's espionage asap might make a big difference, but probably not because we'll need to decide between mdy's and bbp's at pop2, ~T29.
4. There are two decision points.
4a. At pop2, we start building a settler or grow to pop4.
4b. When the worker is done, bbp's variant works better if we work a forest for 2 turns. (Please don't shoot the MMessenger. )
5. bbp's variant gets us pottery 2 turns sooner in exchange for mdy's extra warrior and earlier settler2.

 

[babybluepants]

Edit: nvm, had a comment, but can't remember the test sequence exactly...

 

[mushroomshirt]

OK lots of posts for what will be a pretty boring set! I've played a few of bbp's test saves and if I play to corn irrigation here's how it will go:

turn 15: still 4turns to worker build - Delhi continues to work corn
turn 19: worker is built, start irrigating corn, build warrior
turn 26: corn irrigated, still 2 turns to pop2, 4 turns to warrior

warrior can scout (safely) while this is going on generally in the direction of Zara. Scouting in that direction seems like the most prudent course of action to me.

Only real debate now I think is whether we tech Wheel or BW. It looks to me from LC's analysis that BW gives us the most flexibility. Comments on this updated plan?

 

[shyuhe]

^

I'm fine the with plan - just trying to see where people stand on BW/wheel. I'd give people another day to weigh-in but I think the majority prefer BW right now.

 

[babybluepants]

Plan looks good, mushroom. I like LC's scouting map from earlier today - I'd say we want the initial warrior to complete the NW-NE loop around capital, while heading to Aksum. I think this TS could go to first corn farm completion, yes.

Once again, I'm definitely for BW.

With silver, we should be able to get to wheel--pottery--writing reasonably fast even if we spam out two cities ni quick succession. While we might not get CS from oracle, we should still be able to get CoL or philosophy (my new favorite ).

I don't think CS is gonna happen with the kind of expansion we're talking, unless we take a major gamble and the mapmaker was really generous with AI starts. I think we'll need to seriously think about this aspect of it as soon as the next TS, since a realistic shot has to sacrifice some expansion goals, IMHO.

Why Philosophy? I'm wondering why you like it so much... Esp. with a PHI leader, it's so easy to bulb that, not even significantly slower than the Oracle sling would happen. I personally think CS (massive tech boost early game) > Machinery (major, if we're on an Optics map, esp. if we're semi-iso or close to) > COL/MC (both just all-around good techs to have early, MC for GE gambit, too).

I'm off for two weeks now. Give you guys a chance to catch up with my stupid post count.
Happy spreadsheeting and testing, team (-Silu ).

 

[Mutineer]

That is very good summary, but what it is indicate is that we do not have enough information right now to make calculated decision.
So, Instead I will try to build decision tree. When we have to decide on taking different way and how.

What we need to decide right now?

Only 1 thin, what is next Well or BW.

What is the difference?
personally I do not like to lock down into BW, because it is expensive and close down many alternative path. Yes, we know about all BW goodness it is almost never very bad decision, but it is not always the Best decision.

Lets take a look what happened from here.

When we need to make next decision.

We have 2 track now.

Track 1) Well first.
Track 2) BW first.

First different come on turn we build worker, turn 19, where to move worker.

If we planing to build road after farming corns then first corn to farm should be left one, then top. It could save one move into position to start roading.

If we planing to mine silver then top one first better from conservation of worker turns point of view.

The most reasonable assumption that we need first settler before turn 48 or so.
On turn 35 farmin of corns finished, moving to build road diagonally to top.

Lets consider booth cases for now.

In well case next decision point come when well finished, turn 28.

What to research next? BW is the more reasonable, but there is potential alternative. Masonry. If second city site has available stone build stonehenge in second city with help of stone is a very good move. It will stop zara culturally, easier our expansion.

Problem is, with out current exploration we would not know, our returning warrior, exploring zigzagging going in general east direction(aiming to explore area north of our cultural borders), will not discover enough in time. So, this path is closed for as because of luck of information. So, we will not have alternative to BW.

Next decision come on same turn, as we grow to size 2. What to work next, 1 food 2 hammers tie, finishing warrior in 1 turn, or 3 food 1 commerce tie to continue to grow?

Well it we settle on building settler as fast as possible and not going to mine silver, then in well case we build warrior in one turn and start on settler, switching back to 3 food 1 commerce tie.

So warrior build turn 29, we assume we will send him east, and hoping that our first warrior will explore north.

On turn 35 we finish farm corns and start to road, assuming we roading diagonally NE.

On turn 43 settler is finished, we 5 turn form BW, 2 ties of road build saving up to 2 turns of settling time ( minimum one turn).

Here we can not speculate any longer, any more speculation is lucking information.

Se, Well decision tree is very simple, because of alternative branches are cut of from luck of scouting.

Summary Well first:

Research Well(turn28)-BW. Grow to size 2 turn 28 work one turn of plain forest finishing warrior turn 29, settler finished turn 43. 2 ties of road build. No forest chopped. Expected settling time turn 45-46.


Any other alternatives ways would not give different results form BW first, build settler with out chops at size 2, result would be same, except no roads build worker turns used to build silver mine and farm grassland river side.
----------------------------------------------------------------------------------
Lets look on decision tree from BW first.

Starting point is the same. Turn when worker build.
Farm next chops or farm next silver mine/grass farm.

farm-chops, farming left then top optimum, as cleaning forest north of city could speed up settler movement by one turn, depends on where this settler will be going. In addition farming from let to top start farming 1 turn faster.

Turn 28, we grow to size 2, cross road, if we building settler on size 2 we need to work plain forest 1 turn, if not we better work corn.

Lets look on settler size 2 first.
Turn 36, BW done, next well, as we would want to have ability to connect thinks and I do not see any other alternatives.

Turn 40, chop done, going to settler.
Turn 41 settler finished, With overflow of 7. One chop used. Expected settling time turn 44.

Here we better stop speculations, as we do not have enough information to continue.

Now lets see grow to size 4 then settler. Why 4? It seems we can not get second warrior out before we get to size 4 with out giving away to mach food.
first worker same, left from top, worker turns used for chops.
First warrior turn 30, second turn 38.

Settler could be whipped on turn 41, after chop going in, producing settler on turn 42,
with minimum overflow of 22, depends on what tie we work 2 forest ties or silver hill time for more commerce.

If we do not whip, second chop turn 45 producing settler turn 46.
Here we have to stop speculation.

Now, grow to size 4 and no chops, silver improvements + river side farm.

In that case, worker improve from top to left. First warrior turn 30. grow to size 4 achieved on turn 29 and second warrior produces same turn.

we start to work silver mine on turn 43. and now crossroad, what next ? chop or farm.

Well finished on turn 46, next is fishing-pottery. We will need fishing anyway and it give discount to pottery.

considering whip is no point, as it will stop working our improvement, in that case we might as well build them after chop, making no difference.

Lets look on Mine-chop: settler finished on turn 49 with 22 overflow. 2 turns to fishing. Settling time turn 52

Mine- farm: settler build naturally on turn 50, settling time turn 53


Summary:

Well-BW.
Research Well(turn28)-BW. Grow to size 2 turn 28 work one turn of plain forest finishing warrior turn 29, settler finished turn 43. 2 ties of road build. No forest chopped. we have enough turns to BW to finish 1 more road tie, most likely connecting second city, before returning to chop or improve ties.
Expected settling time turn 45.

BW(36)-well, settler at 2
Grow to size 2 turn 28 work one turn of plain forest finishing warrior turn 29, settler finished turn 41. One forest chopped. Expected settling time turn 44.

BW(36)-well, settler at 4, chop/whip
Grow to size 2 turn 28, warrior turn 30 grow size 4 and second warrior turn 38(need a bit tie micro for that).
chop turn 40, whip turn 41, settler ready turn 42
One forest chopped, estimation settler time turn 45

BW(36)-well, settler at 4, Chop, chop, no whip

Grow to size 2 turn 28, warrior turn 30 grow size 4 and second warrior turn 38(need a bit tie micro for that).
chop turn 40, 45. Settler ready turn 46
two forest chopped, estimation settler time turn 49

BW(36)-well, settler at 4, Silver/chop
Grow to size 2 turn 28, warrior turn 30 grow size 4 and second warrior turn 38 (need a bit tie micro for that).
silver turn 43
chop turn 48. Settler ready turn 49
One forest chopped, estimation settler time turn 52

BW(36)-well(36), settler at 4, Silver/ anything else(farm or road)
Grow to size 2 turn 28, warrior turn 30 grow size 4 and second warrior turn 38 (need a bit tie micro for that).
silver turn 43
Settler ready turn 50
One forest chopped, estimation settler time turn 53

Conclusion:

I believe that we have only 3 valid choices here.

1) well-BW settler at size 2. 2 roads 1 warrior
2) BW-well settler at size 2 1 chop. 1 warrior
3) BW-well Settler at size 4 chop/whip 2 warriors

Only this 3 choices produce first settler fast enough. to be sure we settle before Zara in most juicy spot, if we need to take resource from zara by settling now.

Unfortunately bad scouting decisions leaving as in very tight spot.

If we already know where to settle and be ready to fog bust path option 1 would be best, the most flexible, giving as maximum options what to do with second city, how fast we want third one. How to spend our chops, how fast to improve and start working silver.

Unfortunately I do not believe 2 warrior can do scouting + fog busting for save settling job from current situation.

Any bad luck, like damaged or dead warrior will stop as dead and could greatly slow down our settling.

So, in final decision my vote went to choice N3 and I am very disappointed about that. It is the only safe way form current position, but it is NOT the best possible option we had.

 

[shyuhe]

Thanks for the detailed post mutineer - I do admit that my scouting decision was sub-optimal, as it would be very nice to know what the land looks like between us and Zara right now Hopefully it won't cost us anything in the long-run though!

 

[mdy]

I agree with growing to size 4. If we decided to only go with 1 very early city to reduce issues with barbs and help the economy we could still get the settler out on T44, and start working the silver mine on T47 as we would not need any chops. A second settler could still be built as early as T57 with 1 chop or T58 with just the whip if we needed one.

Note: In my tests I am getting the second warrior and the first settler 1T later than LC's figures for some reason.

A standard CS sling will probably be beyond us given that this is Emperor, however it may still be possible if we bulb Maths. We are philosophical so this would not delay an academy as much as normal. Techs such as MC/Machinery are ones we can get in trade unless we are isolated and need optics quickly.

 

[RRRaskolnikov]

Impressive post Mutineer, looks like option 3 is the best for now.

But unfortunatly, I do fear like some that 2 warriors won't cut it for exploration/fogbusting. Specially if the one warrior we have ventures his ass so far from home (can't predict on which land he will be attacked/eaten). A single lion will crush our plans...

Very short sets (like 3/4 turns) are mandatory so we can adjust function of barbs/land topology/Zara position. We will prolly see other decision trees in the near future (but again, amazing work Mut )

 

[LowtherCastle]

I do admit that my scouting decision was sub-optimal, as it would be very nice to know what the land looks like between us and Zara right now Hopefully it won't cost us anything in the long-run though!

Very short sets (like 3/4 turns) are mandatory so we can adjust function of barbs/land topology/Zara position. We will prolly see other decision trees in the near future (but again, amazing work Mut )

Thought about posting this earlier, but decided to save it for the opportune moment, like now. One of the hardest to learn, but perhaps most important SOPs (standard operating procedures) that MW developed through time was

knowing when to stop, mid-turnset​

This debateably cost us the gold medal at least once, but I'll give you a positive example: In SG10, klarius played first and stopped immediately when he met Gandhi on...T1. We discussed the possibility of beating him to the plains hill and weighed the pros of succeeding and the cons of failing. We decided to go for it. Note that Gandhi's settler WAS ON THE HILL when we settled it...

I am not of the same opinion as Mutineer on the topic of our scouting so far, but if there is a learning point there, imo it is that we could have stopped mid-turnset as soon as we met Zara. At that point, we would have had an opportunity to discuss whether to re-route our scouting.

In this upcoming turnset, all sorts of stuff might happen with our scouting. We might meet someone new, get beaten, get eaten, locate paradise, etc. Any one of those events might be cause for stopping or not. It is up to the player to decide, make the right decision, or learn from overlooking an opportunity.

If there's one thing I've learned from playing these SGs, it's that opportunities have an evil way of disguising themselves. The problem is that each SG scenario is new and fairly unique. It's hard to use experience to predict opportunites. These discussions are most important, imo, because they enable us to foresee opportunities and take advantage of them.

 

[LowtherCastle]

Note: In my tests I am getting the second warrior and the first settler 1T later than LC's figures for some reason.

This is why:

and second warrior turn 38(need a bit tie micro for that).

Basically, when at pop3, you switch from grass to plains forest for 2 turns.

That's the difference between playing your turnset and working your spreadsheet--no rules against "reloading" your spreadsheet. (and do I ever reload my spreadsheet...)

 

[LowtherCastle]

Follow-up to post #135 above.

We all know that klarius is a brilliant player and a code whiz. We all have learned from him. He has a rare talent that might be invisible to most, especially if you've never played an SG with him. I'm not even sure anyone else on MW noticed it. Erkon surely not (just in case you're lurking Erkon...).

klarius always plays "in the moment." With each new turn, he seems to re-evaluate the possiblities and adjust his thinking. This is crucial for an SG. Furthermore, he thinks in terms of the competition. We're not playing against the AI, we're playing against XTeam, etc. So the question is, which team will best adjust to Zara being close by? Which team will best take advantage of any opportunities or difficulties that presents?

That's why I think it's crucial to think in terms of containing Zara. We know DoW is not an option, so what other options do we have? But don't get me wrong, containing Zara does not become our fixed, tunnel-vision goal. It merely becomes one more factor that we keep in mind, along with the others we've been discussing to win this game.

In particular, we don't want to forget our previously stated priority of exploration. We still don't know if there are other AIs we'll be able to trade with. Are there any AIs to the north? Landlocked east of Zara? To the south along some mapmaker-created snaking Antarctica continent? Who knows?

So I don't think of my spreadsheets as "speculating." One of their main purposes is to help us keep an open mind. Prepare for alternative possibilities. One of those is quickly building two more settlers. I want to know, now, how that will affect our research, for example. The only way to know is to carry out the analysis to T60, not stop at T45. Mutineer mentioned another possible containment technique: building cultural buildings.

Anyway, I think you get the idea...

 

[LowtherCastle]

OK lots of posts for what will be a pretty boring set! I've played a few of bbp's test saves and if I play to corn irrigation here's how it will go:

turn 15: still 4turns to worker build - Delhi continues to work corn
turn 19: worker is built, start irrigating corn, build warrior
turn 26: corn irrigated, still 2 turns to pop2, 4 turns to warrior

warrior can scout (safely) while this is going on generally in the direction of Zara. Scouting in that direction seems like the most prudent course of action to me.

Only real debate now I think is whether we tech Wheel or BW. It looks to me from LC's analysis that BW gives us the most flexibility. Comments on this updated plan?

I agree with this. I wouldn't worry about leaving a couple of fogged tiles between our corn and the warrior's current location, safer to go straight to the hill, even if Zara's scout is down there fogbusting. It's always a crapshoot anyway. Playing to pop2 makes sense to me, unless something comes up. I vote for BW. I don't need any more discussion.

Good luck!!!

 

[shyuhe]

Well put LC. I think we're good to go with BW now.

mushroom - just remember to stop if anything unexpected happens - otherwise I would stop at turn 26 since we need to choose where and how we will chop.

 

[mystyfly]

I'm impressed by the effort you guys are putting in However I am more like silu on that issue... So I'm glad we're about to carry on playing soon, I see we're now up to 9.2 posts per turn

One more thing I want to add to our future scouting efforts: When having our WBs journeying the coastline, we must watch out for other landmasses we may reach pre-optics if we have 2 border pops in a coastal city. I remember a few occasions where this was possible in some of r_rolo1's LHC games (games starting in isolation), especially should we find us semi-isolated here... Not that we're new to this kind of situation