SGOTM 13 - One Short Straw

I don't think we'll really know which spot is better until we get more map knowledge so it's a bit of an educated guess right now. With that being said, I'd like to try to get our capital settled by the end of this weekend if possible so that we can start figuring out the next 20-30 turns (which are pretty straight forward once we decide on worker vs. WB).
 
I don't think we'll really know which spot is better until we get more map knowledge so it's a bit of an educated guess right now. With that being said, I'd like to try to get our capital settled by the end of this weekend if possible so that we can start figuring out the next 20-30 turns (which are pretty straight forward once we decide on worker vs. WB).

Agreed. I'm actually fine either way and can see the benefits of both locations. Plus, I don't think either option has a major advantage over the other.

Hopefully the those undecided and ? from your list can cast their votes today. As you said, once we decide on where to settle, the first 20 to 30 turns are fairly well laid out.
 
Agreed. I'm actually fine either way and can see the benefits of both locations. Plus, I don't think either option has a major advantage over the other.

THen PH has the advantage of potential additional sea food. By going SIP we close this door. I am more undecided than ever.

minor note: gold info is missing in your previous table (important because many saves will have to mint a turn or two to complete writing, so they can't be compared directly)
 
Looking at the terrain it looks like PH will get 1 more grass hill, which would make it a substantially better production site than SIP.

I think part of the reason SIP is ahead in research is because the SIP tests chose to emphasise production over research, we could discover sailing and settle our second city earlier with PH, but at the cost of significantly delaying our exploration efforts. In the circumstances I don't think this is a very good idea.

With my PH tests, I was never in a situation of waiting for a tech to be discovered before building something, in other words I think that in the early game we are going to be more production than tech limited, so favouring production makes sense.

In addition, one of the advantages SIP has is the early second city, however as that would have to share tiles with the capital it would reduce it's advantage in food, and make the lack of production there worse.
 
minor note: gold info is missing in your previous table (important because many saves will have to mint a turn or two to complete writing, so they can't be compared directly)

Yes, I thought about gold but I had already reviewed several saves and it would take some time to go back and check every save. In general, total beakers (+ gold) gives a relative value with respect to research rate. You can tell which options provide more or less research. IIRC, the timing of Writing isn't important with respect to libraries since no save is in a position to start one right away. It could have an impact on OB though, especially in saves that could have a far-away explorer by T82.
 
Looking at the terrain it looks like PH will get 1 more grass hill, which would make it a substantially better production site than SIP.

Are you basing this on fog gazing a hill to the north? I thought we had already been through that and convinced ourselves that our island ended at the most northern point we can see...
 
Another thought is that we have been settling the island to the west because it has Fish, is close, uses that Clam, etc. But there's also an island to the south (in our test game and in the real game) that could equally use that Clam but would not borrow the GH Mine at all.
There's also another tile to the SW that could claim clam. Clam could claim quatre. :lol:
 
Looking at the terrain it looks like PH will get 1 more grass hill, which would make it a substantially better production site than SIP.
By that line of reasoning, settling on the Corn is an even more superior choice, since it gets access to the same 2 Grassland Hills squares plus an additional Plains Hills square.


In addition, one of the advantages SIP has is the early second city, however as that would have to share tiles with the capital it would reduce it's advantage in food, and make the lack of production there worse.
I don't see why settling in-place's PH can't "count" as a Hills square, but whatever, I'll play along and will pretend that only GH squares are good squares.

SIP gives us 3E, which actually gives us an equal number of Grassland Hills squares as PH. PH can one-up SIP by settling on the western island for the Clam and GH that PH missed, but it might be that there is no hidden Fish for that City, making that western island only good for a Clam plus a GH. Settling in-place "uses up" those two good squares and thus gives us more freedom to settle elsewhere.

Given the excess Food that we have access to though, I don't see why a PH can't "count" as much as a GH can. Or maybe, as Mitchum seemed to guess, you are suggesting that there is another square visible in the fog?


minor note: gold info is missing in your previous table (important because many saves will have to mint a turn or two to complete writing, so they can't be compared directly)
I ran Binary Research and never had to run more than a turn at 0% Science in my tests. So, the amount of Gold left-over was always a negligible amount, valued at less than one turn of 100% Gold slider. I'd hope that the other tests were similarly exectued.


Resource Guessing Game
THen PH has the advantage of potential additional sea food. By going SIP we close this door. I am more undecided than ever.
Okay, let's play that Resource-guessing game.

Let's say that the two hidden squares contain two Fish Resources. Then what will happen? Well, we'll have a capital with 5 Food Resources and 2 Grassland Hills squares... two Hills squares that are great when you're low on Food but not as amazing when you have a huge Food surplus.

What will probably end up happening is that we'll play the opening in a similar way, except that we'll net 2 Fish instead of 2 Clam Resources. We'll probably ignore the Clam Resources for a long time. We're already hitting Happiness + whipping caps with 2 netted Clams (even with SIP netting only 2 Clams), so netting 2 Fish would certainly give us the same problem.

So, what have we gained?

Well, those Fish will be Coastal, which is rare, but okay, it's magical, so just accept this magical gift! ;) Therefore, the Commerce output will be the same.

The difference will be:
PH has +1f.

SIP: 2f1h+1f3h+4h = 3f8h
PH: 2f2h+1f3h+1f3h = 4f8h
So, +1 Food from settling on the PH. Then we'll get +1 Food from each of the Fish.

So, we've gained +3 Food in a City that is already way-over-Fooded.

Yes, the extra Food = more production, blah blah, but really is too much of a good thing really a bad thing?


I am going to claim that if there are 2 Fish Resources there, then DynamicSpirit would have "balanced" them for teams that settled in-place or on the Corn by putting another island to the north. He probably even made it so that this island "forces" you to settle there anyway, like by giving us two Desert Hills Gold Resources there. Even if PH "uses up" those Fish, you'll probably end up giving away those Fish to the Gold City in the long run anyway.


Now, let's play this game again, because it's fun.


What if, instead of 2 Fish, there is one more Clam. Now we've got the same location as SIP except we get "+1 Food" from settling on the PH. Again, it's a net gain, but it gains us more Food, the one input that we have a lot of already anyway. It's nice, but not really game-breaking.


Roll the dice again.

This time, 3E gets 4 Fish and a Grassland Gem Resource.

Aren't you glad that you settled in-place over settling on the PH?


Roll the dice once more.

All hidden squares are barren (empty of Resources) for PH, 3E, and the western island settling.

PH is still a decent location. The western island might not even get settled, though, at least not for a long time, so we will "waste" the GH and Clam over there. Not the end of the world, since we apparently found a better City location than one which gives us 1 Clam plus 1 GH.

SIP is still a decent location. 3E still works as a nice early supporting City and can generate some of our Great People after we get a Library.


Say that we decide where to settle... then what?
Okay, let me take a different track. I think that we need to agree upon a multi-turn set medium-term plan.

I really don't want to play the game as a series of independent votes.

For example "we chose to settle on the PH" as a group, each for our own individual reasons.

Next, "we chose to build WB first," because we voted for this idea as a group.

Then, "we decided to send out WBs 2 and 3 for exploration, because exploration is important."

After that, "we built a Granary, because Granaries are great."


Somewhere down that path of decision-making, we'll find that had we been aware of some of these decisions a bit ahead of time, then we could have planned out a build order and tech path that would have matched it nicely. Instead, for example, we followed a build order that, say, assumed we would use some of early Work Boats for netting, but this assumption broke-down when we sent out our WBs for exploring. Then, we'll find ourselves faced with inefficient whipping of additional WBs or else having our chosen-test-run-path fall behind in its pace.

But, if we'd planned from the start that we'd wanted to make those decisions, we could have picked a test-run or made a new one that got us these extra build items (in this case, extra exploring WBs) as efficiently as possible, say, by using different whip-overflow timing.


That's why I DO NOT think that we'll be significantly moving the game forward by deciding where to settle. Is "Worker first" going to be the best choice for sure? I haven't seen anyone advocating in favour of a particular test run... so, the answer is "no, we'll just have limited our options but we still won't really be proceding beyond the first turn of founding our capital."


What I propose
Let's not just vote on settling location. It won't help. It will "get the game moving forwards," but only for up to 1 turn. Then we'll be stuck, but we will have limited our options.


I think that we should instead vote on a whole series of items at once and then use the results of that voting to decide how to execute the initial 100 Turns or so.


I don't think that I can come up with an exhaustive list of things that need to be decided upon, but let me start:
1. Explorer #1 does What?
We've seen LC come up with a neat plan for scouting around the nearby islands with a Work Boat. We've also heard people talk about how great it is to meet the AIs ASAP. So, which plan do you prefer? Shall we use our first explorer to:
a) Look at a maximum of 2 islands for a potential off-island settling location and then go and find the AIs
b) Follow LC's approach of exploring all of our nearby islands first and we can meet the AIs later
c) Let's meet the AIs ASAP. If our first explorer happens to find a decent off-island settling location, "good for it," but we will not really stop the boat and will try and meet the AIs ASAP


2. Explorer #1 comes When?
It's hard to quantify "when" in terms of turn numbers, but I think that we can subjectively quantify it in terms of Cities settled.

For example, if exploration is the number one priority in your books, then perhaps we should get explorer #1 built before building City 2. Maybe you prefer a more balanced approach with expansion.
a) Explorer #1 should be completed before building Settler 2 (i.e. the Settler for City 2)
b) Explorer #1 should be completed before building Settler 3
c) Explorer #1 can wait to be built until after building Settler 3 (pick this option if an early Explorer #1 is not a priority for you)


3. Explorer #2 comes When?
a) Explorer #2 should be completed before building Settler 2 (i.e. the Settler for City 2)
b) Explorer #2 should be completed before building Settler 3
c) Explorer #2 can wait to be built until after building Settler 3 (pick this option if an early Explorer #2 is not a priority for you)


4. A Library in the Capital comes When?
This point was only briefly raised, but why not quickly vote on it. If it turns out that the team secretly wants an early Library, we should be planning for it in our build orders now, right?
a) The capital should focus on building a Library before Settler 3 is built
b) The capital can build a Library sometime in the future


Feel free to add more items to the list.

If everyone is 100% agreed on a particular test run, then sure, let's forget all of this voting and follow that test run. Otherwise, I am not really convinced that picking a settling location without getting some sort of consensus of the other issues floating around is going to help us more than possibly hurt us.
 
@ Dhoom, LC, Ras, bbp

Indecision = PH. Are you guys OK with that?
Not really.
Indecision = would still like to vote.
Abstain = vote is ignored.

At present, I'm liking SIP the best, thanks to my latest test run. Things went really smoothly:
i. I could get Settler 3 and Galley 1 out faster than any other recorded test game, on Turn 79, if I wanted.

ii. I also had a Granary and a Lighthouse, so infrastructure was good.

iii. Alongside these above two facts, which imply that I had strong production, I also had the top research total out of any of the PH and SIP test runs.

iv. I was able to justify sending out a "permanent" Work Boat explorer, while in all other test runs that I played, I would have advocated sending the first Work Boat explorer back home for later netting.


What are we really looking here for, anyway? Isn't strong production + strong research a winner? If yes, then based on the current test run data, I have to strongly advocate for SIP.


If, however, people would care to vote on some of the other issues that I raised, then we may find that people really care about early exploration above a productive empire and above a fast-researching empire, which may advocate a different test run as being better.
 
@Dhoom - I agree with your logic, but the practical implications are difficult. We are currently doing test runs for 80 turns, which is actually about 1/5, probably 1/4 of the entire game that we will be playing. I don't think it's particularly feasible for us to test all of the possibilities to our satisfaction before proceeding here. Otherwise, we will get bogged down in endless simulations without ever playing a turn. On a related not, it is ill advised to test out and play 100 turns at this point since our game plan can and should change according to the intelligence we gather. For example, comparing demographics at T+1 vs. T0 will allow us to glean information about our opponents and their tile setups, which informs our decisions about wonder construction etc.

The selection of a settling in place is, admittedly, a somewhat arbitrary cut-off much like the T82 cut-off that we've been using for our test runs. We will never have perfect information to make our selection here -- the important thing is that we get an amount of information that is sufficient to allow for a well educated prediction. I merely propose the SIP vs. PH choice so that we can narrow the possible testing options to better refine our course of action. I'm not saying we're going to be wedded to a particular plan just because we go SIP vs. PH. In fact, I think the testing has shown that each spot is flexible and can be molded to meet various needs to a certain extent.

edit: x-post with Dhoom. Also, I can show you my corn save with a T79 settler #3 if that's what you want... It comes with 2 permanent exploring workboats in the T40's and 50's too. I didn't post it because I don't like the long-term viability of the corn spot and it requires us to find a strong city #2 spot relatively quickly (circa T65-70).
 
@ Dhoomstriker

The reason I said that indecision = PH was because there were a lot of PH on the scoreboard and SIP was blank.

The only thing I didn't like about your test run was the rather late exporing workboat. Is ~13 turns enough to find a good location for city #3? Maybe, maybe not. Many of the other saves have many more turns of exploring, which could be used for either exploring the islands nearby or meeting the AI.

To me, it seems that a PH/worker opening provides earlier exploration at the expense of research. Both capitals (SIP and PH) are adequate and can be well leveraged for early REX. So, what is more important, research or early exploration?

EDIT: I think a PH / off-island city #2 will begin to close the research gap a bit due to the increased traderoute income.

Unless someone can come up with a SIP opening that gets more exploration or a PH start that boosts research...
 
Also, my personal preference here is #1 exploration followed by #2 large capital and #3 long-term viability of the capital. My logic is that we want to find the best city sites nearby quickly, and having a large capital will allow us to quickly whip/build settlers to take advantage of those spots. I know I'm shifting my position somewhat from the "SIP lacks hammers" position...

My love of the early exploration is also part of the reason why I'm fascinated with the corn spot.
 
Okay, let's play that Resource-guessing game.

Let's say that the two hidden squares contain two Fish Resources. Then what will happen? Well, we'll have a capital with 5 Food Resources and 2 Grassland Hills squares... two Hills squares that are great when you're low on Food but not as amazing when you have a huge Food surplus.

What will probably end up happening is that we'll play the opening in a similar way, except that we'll net 2 Fish instead of 2 Clam Resources. We'll probably ignore the Clam Resources for a long time. We're already hitting Happiness + whipping caps with 2 netted Clams (even with SIP netting only 2 Clams), so netting 2 Fish would certainly give us the same problem.

So, what have we gained?

Well, those Fish will be Coastal, which is rare, but okay, it's magical, so just accept this magical gift! ;) Therefore, the Commerce output will be the same.

I don't know how to react frankly... It nearly sounds like you are moking it:
magical two fish resources? No. But assuming it can't have one is just wrong and saying that an additional food resource won't matter is too. We have tested whip cycles from size 6 to 4 without monarchy... cool. I hope one can see how an additional food tile can be used in the long run (I doubt we will win the game by t100). Even then it means more growth power.

I was merely stating that since both approachs look close from other point of views, PH has one clear advantage.

Then I don't quite agree with what you propose actually. I don't think deciding our first 100 turns path now right now is keeping options open. Nor is evaluating all possible scenarios for those same turns now and polling on each reaction possible.
 
Out of the test runs, the Corn spot was also the one that gave us the most research overall.


As for early exploration... what stops us from just whipping an extra early Work Boat or two for either SIP or Corn locations? Do we honestly believe that +1 Hammer from settling on a PH (45 Hammers after 45 turns) is the only way that we can get early WB explorers?


Okay, so let's look at the results table. Many of the PH runs have Explorer #1 out around Turn 50, give or take a few turns, with the second one coming out around Turn 57 or 58.


Let's try a SIP run or two and see if we can approximate those WB explorer dates, then compare the Turn 82 results to the PH games.

Isn't that approach going to be the most objective one that we can use at this point?
 
i. I could get Settler 3 and Galley 1 out faster than any other recorded test game, on Turn 79, if I wanted.

ii. I also had a Granary and a Lighthouse, so infrastructure was good.

iii. Alongside these above two facts, which imply that I had strong production, I also had the top research total out of any of the PH and SIP test runs.

iv. I was able to justify sending out a "permanent" Work Boat explorer, while in all other test runs that I played, I would have advocated sending the first Work Boat explorer back home for later netting.

I apparently missed how this save outperformed in most areas, will check it out now.

If, however, people would care to vote on some of the other issues that I raised, then we may find that people really care about early exploration above a productive empire and above a fast-researching empire, which may advocate a different test run as being better.

There you are confusing. Do you really think exploration doesn't matter in setting up a productive (aka big cities) and fast researching empire (aka big cities, trading)?
 
As for early exploration... what stops us from just whipping an extra early Work Boat or two for either SIP or Corn locations? Do we honestly believe that +1 Hammer from settling on a PH (45 Hammers after 45 turns) is the only way that we can get early WB explorers?

I think the difference is that a PH start can afford to send WB#2 exploring because it typically builds worker first and the corn tile is the second food resource. SIP really needs WB#1 and WB#2 to net clams if it wants to grow fast enough...
 
@Dhoom - sounds like a good comparison. My SIP testing didn't give any satisfactory exploration for me because I felt like I needed to rush to push out settler #2 to really take advantage of the spot. I didn't fiddle around with it too much though.
 
Top Bottom