SGOTM 13 - One Short Straw

[LowtherCastle]

I do think we need the 2 explorers - that was the assumption. You don't think this is important?

Explorers, of course, but when? Before we improve the clams? Thing is, corn+clam first is still not a no-brainer for me. If the warrior reveals another resource on T0, then this discussion restarts, but there are plenty of possibilities, including fishing-sailing and wbs+lh.

An early 19bpt is nothing to sneeze at compared to 14 or 16bpt.

 

[shyuhe]

How many exploring workboats do people want to send out? Feel free to couch your response to a triggering event, like 1 workboat before first settler, 2 workboats before first off shore city, etc.

IN FACT, I will make the bold claim that if we plan to research Priesthood, delaying meeting these AIs can often give us less competition for building The Oracle, since AIs will be less inclined to research this tech simply because it is cheaper for them to do so, due to another player (us) already having "bitten the bullet" on researching this tech as a "monopoly" tech.

This is actually reverse. If you get priesthood and put a hammer into the Oracle before the AI research priesthood, they will avoid it. The AI kind of cheats and knows what you're building and your one hammer discourages the AI from beelining priesthood for the Oracle. You can push some pretty late dates with this strategy. Besides, I'm pretty sure the AI doesn't take tech cost into consideration when it chooses a tech.

 

[babybluepants]

Yes, we're talking about timing. Perhaps I misunderstood, but I read your post as suggesting we skip exploring altogether until after the 3E settler. IIRC, my WBs have been: clam - explore - explore - clam, all before building a settler. Worker first and corn farm is implied.

I realize the rush to explore is partly mitigated by going SIP-3E, since we don't need to find the second city in that case.

Edit: xpost w shy

 

[LowtherCastle]

If you get priesthood and put a hammer into the Oracle before the AI research priesthood, they will avoid it. The AI kind of cheats and knows what you're building and your one hammer discourages the AI from beelining priesthood for the Oracle. You can push some pretty late dates with this strategy.

We talked about this last game, but no one was sure. Are you sure about this? I thought someone got a counterexample to succeed.

Yes, we're talking about timing. Perhaps I misunderstood, but I read your post as suggesting we skip exploring altogether until after the 3E settler.

1. I'm not "suggesting" anything in any of my posts (other than that we do testing), just to be clear. I'm throwing out ideas that I think have potential.

What we're searching for is the optimal opening, right? That oftens comes from figuring out what we can eliminate from the initial queues. For example, eliminating the wkr-first gives us a settler sooner, if we want to poprush it, because we can grow faster. So then the question is, how important are early land improvements? Needs to be tested and compared with other tests.

Or, what about fishing-sailing? Is wbs+lh a power move? If so, what sequence of wb1..gran...?
Or, what about fishing-pottery? Is wbs+gran a power move?

This map is kind of unusual, at least for me. So my years of playing novel SG scenarios tells me to find out in advance hwo to approach the scenario to avoid later regrets. This was my first major SG Lesson Learned (from SG4 -- Gandhi to space).

2. I was talking about no explorers till after 3E (about T49 latest).

 

[Dhoomstriker]

This is actually reverse. If you get priesthood and put a hammer into the Oracle before the AI research priesthood, they will avoid it. The AI kind of cheats and knows what you're building and your one hammer discourages the AI from beelining priesthood for the Oracle. You can push some pretty late dates with this strategy. Besides, I'm pretty sure the AI doesn't take tech cost into consideration when it chooses a tech.

Your point and my point still hold side-by-side (although I have yet to see the evidence about the whole "put 1 Hammer in" trick, the possibility of this trick actually working does not make my comments invalid)... you can be the first to tech Priesthood and put a Hammer in The Oracle independently of what the AIs are doing in terms of research.

However, by watching what the AIs research by starting games where I World Buildered myself a Great Spy for each AI in order to infiltrate them, they would have a tendency to try and research my techs... if I set my research slider to 0%, they'd tend more to go off on their own.

This effect can easily be tested by playing a game where you are near to a religious zealot... if you go and meet them, they are more likely to try and research your starting techs if those said techs differ from theirs, than they are to prioritize an early Religion. As such, you can actually "convince" AIs to delay founding early Religions by going out an meeting them.

AIs DO consider costs and, as a somewhat unrelated point, they also look at techs up to three levels deep when deciding which tech to research next. There is a large random number factor that affects their decisions, but "piggybacking" off of another player's research does have an impact on their decisions.


I'm not necessarily saying that we have to delay exploration, but I am simply providing an additional reason for why doing so could be helpful... thus, if we find that there is a reasonably-sized trade-off with early exploration versus early Clam-netting, we can "feel a bit better" if we decide to go with the early netting.

->"My guess on AIs:

Catherine
Bismarch
Victoria
Charlemagne (to model hasbourg - Austria)
Joao or Isabella
Ragnar or Washington"

No Romans? Elba Island seems to be off of the west coast of Italy...

-> "Looking more closely at the starting screeny, there is prolly a mining ressource on the green UNFORESTED hill west of the starting position.
It means we prolly want to settle that island fast if it's iron or copper. "

Should Copper prove to be on the island to the west, would the consensus be to settle a City there or would we possibly consider waiting to get our Copper several turns after learning Mathematics (i.e. by building a Fort there, which is an improvement that takes a large number of Worker turns to complete)?


Also, if we do not find Copper nearby, would we plan to slot in Animal Husbandry? Iron Working even?

I thought... when I saw the start:
-Moai

While Moai is good, would we put it in our capital? Consider that we are unlikely to found a Religion that gets placed in our capital, so we are thus unlikely to build Wallstreet there if we get that far in the game. Will we plan to build Oxford if we get that far? In a non-Wonder-spam-in-the-capital game, would you actually build the National Epic in the capital? I'm asking about these other National Wonders because it could be quite feasible for us to have an "open National Wonder slot" for Moai in the capital... at least a capital that in encouraged to work a lot of Coastal Moai squares will get Bureaucracy benefits from working a lot of Hammers (for and island City) and a reasonable amount of Commerce... that is, if and when we make it to Bureaucracy.

 

[RRRaskolnikov]

All good points dhoom.
Though I don't think anybody talked about delaying clams for exploring wbs. Though maybe the third and fourth should be set to this task.

Romans: more an inspiration for Nappy than a reality, so I am not sure. Honnestly that list was a wild guess But maybe we are indeed close to the continent!

About resources: let see the map! My comment was based on the abscence of forest on that tile, but when u think it's an hand made map, it makes no sense.

 

[Dhoomstriker]

Okay, I played around a little bit with the test saved game. Here are some initial observations:
1. Settling in-place kind of sucks in terms of being able to improve Mines... we can only improve one such Mine, and only after we get Bronze Working, which takes a while to research, since we don't know Mining to begin with.

2. Somewhat mitigating the above point, our Cultural borders will have expanded over the Plains Hills Forest square for +3 Hammers per turn by the time that Fishing comes in (if we were to research Fishing first). The amount of Hammers is no different from working a Grassland Hills square.

3. Settling on the Plains Hills square kick of sucks in terms of Health... since we are Charismatic, our Happiness cap is 6, and as was discussed earlier, we get 5 Health to work with... but if we settle on the Plains Hills square, we won't get +1 Health from having 2 Forests in the big fat cross, so we actually only have 4 Health to work with from the Plains Hills location.

4. Settling on the Plains Hills square almost "forces" our 1st or 2nd tech to be Mining, so that the Worker will have the ability to get us a decent production square going for Hammers pre-Bronze-Working.

5. Since a Work Boat and a Lighthouse have identical costs, building either one seems to be a wash when it comes to working 2 Coastal squares. However, if we are going to work three Coastal squares (say, by settling in-place), then building Work Boat + Lighthouse is better than Work Boat + Work Boat, since we will get the following Food yields from our Clams:

Scenario i. Working two coastal squares:
i. a) Work Boat + Lighthouse = 5 + 3 = 8
i. b) Work Boat + Work Boat = 4 + 4 = 8
8 = 8

Scenario ii. Working three coastal squares:
ii. a) Work Boat + Lighthouse = 5 + 3 + 3 = 11
ii. b) Work Boat + Work Boat = 4 + 4 + 2 = 10
11 > 10

6. If we don't settle in place, we'll probably work a Mine instead of a 3rd Coastal square, so it doesn't really matter which one of a Work Boat or a Lighthouse we build second out of those two types of build items.

7. We grow CONSIDERABLY faster than whipping anger can dissipate, so the more 2-pop and 3-pop whips that we can make, probably the better off we will be, at least in the early game when Happiness is limited.

8. Teching Mysticism and "wasting a population point" can, for the short-term, keep us Happiness neutral: -1 for Whipping, +1 for a Monument. Once the Whipping unhappiness wears off, we'll have +1 Happiness to play with.

9. Getting Pottery and 2-pop whipping a Granary can be quite useful, but from my limited testing, I think that we'll at least want to build 1 Settler before we build a Granary (the same probably can said about a Monument).

10. A Settler can be 3-pop-whipped at City Size 6. We can feasibly grow to Size 6 and 3-pop-whip the Settler 1 turn later (or immediately if we previously spent a turn putting Hammers into the Settler, but then you'd be doing so at a smaller City Size, so it's probably more efficient to grow to Size 6 first and then put 1 turn of Hammers into the Settler). All other early build items appear to be 2-pop-whips, so aiming to 3-pop-whip Settlers might be a good way of maximizing the value of our excess Food.

 

[Dhoomstriker]

11. Settling in-place leaves us with the decision of sending our Worker with our Settler or sending our Worker to the western island first, in order to Mine the Grassland Hills square.

12. Settling on the Corn gives us the most Hills squares to work, but doesn't synergise well with the idea of running Scientists.

13. Settling in-place could give us two Cities that, after we expand to another couple of Cities, can run Scientist Specialists for 2 early Great Scientists... Academy + Philosophy? Later, we can feasibly switch to Caste System + Pacifism and use both of these Cities to run 4 Scientists each.

14. Pacifism is only helpful if we run a State Religion (potentially angering some AIs).

15. Pacifism is only helpful if we can first spread our State Religion to the Cities which will run Scientist Specialists... which likely (unless we get lucky with auto-spreading of our Religion) means that we will have to build either a Monastery plus a Missionary or else switch into Organized Religion (thus requiring that we tech or trade for Monotheism) and build a Missionary. At worst case, we'll need to build 2 Missionaries, but I think that it is safe to assume that we will found at least one of Confucianism or Taoism and that at least one of them will appear in our 2 Great Scientist Cities. If you want more than 2 Great Scientist Cities, we may need more Missionaries.

16. I kind of like the idea of building an early Academy AND Lightbulbing Philosophy, if we think that we can afford to give up 4 squares' worth of Hammers or Food-used-for-whipping relatively early in the game, in order to give us a head-start on running a ton of Scientists (8 Scientists) and totally killing our production and Food input in two Cities, just so that we can get Astronomy really early. It's definitely an early-game trade-off for a great mid-game result.

17. Astronomy unlocks Observatories, which, if we still care about Science at that point, can actually help to justify possibly skipping the Academy that I just finished pushing-for above, which could mean that we'd only need to sacrifice Food and/or Hammers in one City early on (for Lightbulbing Philosophy), allowing us to REX much better.

18. Astronomy will obsolete Monuments, so we probably don't want to over-build Monuments... one in the capital will be useful since our capital will have a lot of useful squares to work; other Cities will have to be decided upon on a case-by-case basis.

19. The above point implies that we really should be skipping Stonehenge.

 

[babybluepants]

That makes me think - Astronomy beeline doesn't work with TC, btw.

 

[Dhoomstriker]

That makes me think - Astronomy beeline doesn't work with TC, btw.

A fair point. However, I think that if we beeline Metal Casting, say via The Oracle, particularly if we have access to Copper, then we'll give The Colossus a long enough lifetime.

I seem to recall Gosha advocating building The Colossus even in a game where he got Astronomy early on.

While you can look this issue as Astronomy disabling The Colossus, I think that a better way to look at it is that if we can build The Colossus, it can pay for itself VERY quickly. It costs 375 base Hammers. With the required Forge and with Copper (we can probably skip building it if we lack Copper), then the actual cost is 375 / 2.25 = 167 Hammers.

Let's say that we are working 3 Clam squares and even, say, only another 3 Coastal squares total. That's +6 Commerce per turn, which comes out net-neutral after 28 turns... although "Hammers now" or "Commerce now" is of more valuable, so let's say that it takes, I don't know, 35 turns to fully pay for itself, were we to otherwise spend those Hammers on Wealth.

If we ignore the Great People Points (which we will probably want to ignore if we are going to try and generate Great Scientists... meaning building it in an auxilliary City), then we just need to ensure that we'll use it for more than 35 turns on Epic... not too hard to believe.

So, assuming that we take longer than the number of turns to build The Colossus (say, in a base 8 Hammer City, that's 167 / 8 = 21 turns--and a lot less if we 2-pop-whip other build items and overflow some Hammers into the Wonder) plus 35 turns = 56 turns or less on Epic to research or trade for Code of Laws, Calendar, Compass, Optics, Machinery, and other techs that we might want like Currency and Alphabet, then we should still target building The Colossus.


However, the location of WHERE to build The Colossus (as well as The Oracle) becomes an important issue.

Should we agree to aggressively a Lightbulbing strategy, accidentally getting a non-Great-Scientist for one of our 3 (or possibly 4 if we build an Academy) Great People is going to be incredibly costly.


So, can we afford to sacrifice one of our two on-continent Cities as a non-Great-Scientist-making City if we settle in place, since we'd put The Oracle in one of these Cities?

Is it even possible to compete on building The Oracle in a 3rd settled City?

Would a good compromise be building our second City off-continent, if we do settle in-place, so that we could still have two on-continent Cities that are dedicated to generating our last 2 Great Scientists without having any Great People pollution? If we do so, is it possible to get an off-continent City (settled as our second City) built-up quickly enough in order to compete on completing The Oracle?

 

[shyuhe]

Fair criticism of my points. I guess it's a bit of a wash without code digging...

I wouldn't be concerned about astronomy obsoleting the Colossus. Inter-continental trade routes will more than make up for the loss, assuming we actually need astronomy (why else would we tech it?). We may have a 10 turn window where our overall bpt drops from obsoleting the Colossus but as soon as we make contact, it'll bump back up.

 

[Mitchum]

Hopefully we'll almost be done with research by the time we learn Astronomy, right?

Edit: By almost done, I mean that we'd have to finish CS (assuming we bulb Astronomy) and maybe slow research Engineering at 0% science.

 

[LowtherCastle]

Lexad, gonna join us?

 

[LowtherCastle]

Are you talking Normal or Epic turns? There's no way you're getting granary + settler by T50 on Epic. You couldn't even tech Pottery/BW by then. The fastest we were getting for Fish/Mine/Sail/BW is T58 in EP's test, so maybe 1-2t faster for Pottery.

I haven't tested anything in game because I don't have access yet (hopefully tomorrow), but my spreadsheet tests have SIP T49(epic): rax(part)-wb1-wb2-wb3(part)-settler plus fish-mng-bw-pottery***. What I haven't tried yet is fish-pott-mng-bw, with the granary coming earlier.

Dhoom's lh analysis done similarly for granaries might show when the granary optimally comes in teh sequence.

***EDIT: Okay, I just noticed that the settler was T49, pottery T51 for that test.

 

[babybluepants]

Ok. I tried testing this quickly last night. You can definitely do T49 by 1pop whipping the settler from size 3. Don't really know what to do with the first build in that scenario. Your hammers will decay before you wanna actually complete anything but WBs. I guess a warrior first is better, at least it could be completed without investing too many hammers?

Essentially, your suggestion seems to sacrifice early production and exploration for research.

 

[LowtherCastle]

Who cares about the first 10h? Let them decay. We gain them back on the first wb anyway by being at pop2. But thi is not about sacrificing production at all. It's about leveraging slavery. You put food-hammers into a worker, I put them into a pop point. The earlier wb starts growing our pop sooner AND generates 2cpt. Those poppoints are slave-hammers. That's why I keep saying the question is whether improving the land tiles later hurts us or not.

EDIT: Put another way, bbp, you are sacrificing growth (slave-hammers) and research by building a worker. So the question is, do we get a sutiable return on the worker?
EDIT2: If we settle on teh plain hill, the worker has
11t farm+road corn
10t move, mine and road grass hill #1
10t move, mine and road grass hill #2
---
31t

So the worker becomes idle (or performs non-essential busywork) on T50.

 

[Dhoomstriker]

Hopefully we'll almost be done with research by the time we learn Astronomy, right?

Edit: By almost done, I mean that we'd have to finish CS (assuming we bulb Astronomy) and maybe slow research Engineering at 0% science.

I would tend to agree that we should be fine facing even Longbowmen, Crossbowmen, Macemen, and Muskets with that level of tech, and even Grenadiers or the odd Rifleman, should it come to that and should it be one of the last AIs that we have to attack.

One implication of stopping research at that point, though, is that we will not be teching to Nationalism, which would mean that we would not be able to use Drafting. I'd be fine with doing so, particularly because drafting Macemen isn't the be-all-end-all approach anyway, but it's worth keeping in mind in case someone had their heart set on Drafting.

Who cares about the first 10h? Let them decay. We gain them back on the first wb anyway by being at pop2. But thi is not about sacrificing production at all. It's about leveraging slavery. You put food-hammers into a worker, I put them into a pop point. The earlier wb starts growing our pop sooner AND generates 2cpt. Those poppoints are slave-hammers. That's why I keep saying the question is whether improving the land tiles later hurts us or not.

A thing to keep in mind is that BUFFY misreports Hammer decay by displaying Normal Speed decay rates even for Epic Speed, so it is possible that when the interface tells you that you will lose Hammers, you will not lose them at all.

Hammers on a building (such as a Barracks) will take 60 turns * 1.5 turns for Epic Speed to decay, so if you are adamant about not losing Hammers, then this method allows you not to lose Hammers on your build item.

I was always building a Worker first in the testing that I did (sometimes building a Work Boat partway through completing the Worker), but I could see how skipping the early Worker if we settled in-place could possibly be a good approach. I'll have to play around a bit, but options for the first build item in place of a Worker include not only a partially-built Warrior, but also a partially-built Settler.

While I showed that a Lighthouse can be marginally better than 2 Work Boats if we plan to work 3 Coastal squares, the Lighthouse does come at a cost of requiring us to research Sailing, which may not be worth delaying a different tech that we want to get, such as Bronze Working or Pottery.

When you think of all of the possible variables, even just picking one single settling location and working from it has countless ways for you to play the first 60 or so turns.

Many approaches are "obviously poor," such as building 10 Warriors in a row, but what I meant was that there are still a large number of combinations that have the potential to lead to a good result... so, if anyone is thinking that "oh, two or three people have run a test or two, and thus I shouldn't bother trying to do any testing myself," then I say "think again, because your testing will be of value."

 

[LowtherCastle]

I would add that any comments one person makes don't rule out possibilities necessarily. For example, I've been testing SIP without wkr first and comparing that to PH with a wkr. Well, maybe the best solutino is PH without wkr first. I don't think anyone's tried that yet.

 

[Mitchum]

Am I looking at the wrong test save? It appears to be at Normal speed rather than Epic... Also, Random Events have not been disabled.

 

[babybluepants]

But thi is not about sacrificing production at all. It's about leveraging slavery. You put food-hammers into a worker, I put them into a pop point. The earlier wb starts growing our pop sooner AND generates 2cpt. Those poppoints are slave-hammers. That's why I keep saying the question is whether improving the land tiles later hurts us or not.

Unless my calculations/tests are off somehow, at T60 PH-Worker (compared to SIP-WB):

• is ahead by approx. 30 overall hammers (counting your initial 11h and my 22h warrior, both of which are wasteful at this stage)
• is ahead by probably at least 15 WB turns exploring
• pop 5 compared to 2-3 in capital
• better tiles to work during whip regrowth, including an extra 5fpt tile
• no whip anger compared to 2
• worker available to chop that forest
• per-turn production of 7f8h13c compared to 9f2h16c.

SIP-WB is ahead in research by quite a bit, ofc. Hence, I say you're sacrificing production for research. Not saying that's wrong, just trying to evaluate. I'm happy enough to test any approach, as tiny island games are hardly straightforward on the MM level. Intuitively, though, avoiding the best food tile seems a bit nutty to me.

My approach potentially runs into trouble after this, as you point out. Namely, the worker could be idle and needs to be shipped quickly, but you really wanna send a warrior with the first galley trip. Also, if it turns out there's no great site on nearby islands, or that a worker isn't needed there at all, or that a WB explorer can't even get very far to begin with... I'm potentially late getting to some important techs (namely Pottery-Writing), as well.

Looks like we'll get the save tonight, btw. Someone could just move the warrior and we can continue this discussion with better info...

You wanna sign up? Don't give Alan another potential headache.