SGOTM 13 - Gypsy Kings

Checking in also but busy for a while today.

Given some testing I did earlier, settling on the PH seemed far and away better than any other option, and it wasn't even close.

Some things I definitely agree with: Slavery is a must. Exploration is a very high priority.

Click to expand...

YES AND YES!

Try testing settling on PH, I think you may find better results!!!

2. Given that we have such an abundance of , might our alternative be to aggressively settle our neighbouring islands?

Click to expand...

I totally agree here from my early testing. We need to make getting OFF this island our highest priority IMO. Napoleon won't conquer the world if he stays on Elba island.

T104 Oracle is indeed very fast!

I don't have the ability to play test games at work (alas). We might consider 3x WB then Settler before the first worker even. Plant that second city 3E of the starting location, and it already has visible access to some good shared food resources + 2 mines - I would suggest building the Oracle here.

Wow, I can see I'm going to be a total newb here. I play Civ with a lot less thought about decisions than this! I'm going off to try Grifftavian's test game now. Maybe I'll learn something. Right now the only I advice I can give is we should leave our first city named 'Paris'.

Thorn said:

Right now the only I advice I can give is we should leave our first city named 'Paris'.

Click to expand...

AlanH said:

In this game you are playing Napoleon of the French. Following your defeat by the Sixth Coalition you have been forced to abdicate and exiled to the remote island of Elba. It is time for you to take your revenge on the European powers that have defeated you. Are you up for the challenge, or are you going to end up like the real Napoleon, defeated and condemned to finish your life in exile for the second time?

Thanks for this scenario go to DynamicSpirit.

Click to expand...

In keeping with the spirit of the game description, we could go with Elba as the name of our first city!

I'll give settling on the Plains Hill a try next!

Thorn said:

Wow, I can see I'm going to be a total newb here. I play Civ with a lot less thought about decisions than this!

Click to expand...

We were all noobs once. For me, it was back in SGOTM7. Please don't be afraid to speak out about any ideas you have - often players can get stuck into old habits and it takes a fresh perspective to come up with something amazing. It's great to have you with us

Also, welcome back old GK team members! And a big welcome to Mabraham who is also with us for the first time

I managed to improve on my PB (which is now 1650BC, for those who think in years). I have attached a series of saved games for anyone who wants to try to improve it. It started from my first test save (next post).

Key features:

• One off-island side city founded, without warrior defence
• Finished Oracle 2 turns after Fishing-Mining-BW-Sailing-Myst-Medi-PH-Writing (i.e. no Pottery) having not sacrificed much/any along the way, i.e. there's only about 2 turns of scope for improvement for Oracle->CoL. Oracle->MC from Pottery could be a bit faster
• In this test, Izzy+Marble build the Oracle T91, so I hacked her in World Builder to get the final data. I have observed an AI on this map in another test build the Oracle T94.
• There is a bit of fluff in the how-to below, so scope for improvement
• Only one chop contributed, and the PFH mine got up only just before completion. Rather than the settler, we could build a second worker and a monument, which will get the Oracle several turns faster, but sacrifice empire expansion and future capital

Somebody else can contribute some useful data by setting up a game with 6 Industrious AIs, just pressing Enter and finding out when Oracle, Pyramids and Colossus are built over several (10?) iterations. Then same with 6 non-Industrious AIs. Then we'll have a feel for reasonable target dates and thus fuel for the discussion.

The how-to
T0 SIP, work corn, fishing, rax (just in case the last long enough for us to come back to it), warrior wanders around before returning to capital
T10 Mining, switch to WB
T11 Growth to 2, work PFH and GF
T20 WB->clams, WB, work PFH and clams
T21 BW
T31 WB->clams, WB, work both clams
T34 Growth to 3, work both clams and PFH
T40 Slavery, Sailing
T41 Whip WB
T42 WB->clams, WB, work both clams
T43 Growth to 3, work trip clams
T48 Growth to 4, work trip clams and PFH (1 > 1 but 5th population is useless from , so get WB out ASAP)
T51 Worker, start WB exploring
T54 Mysticism, start lighthouse (think trip clams+PFH still right, to get worker faster)
T55 Whip lighthouse
T56 Back to worker, work trip clams
T61 Meditation, galley, improve corn
T62 Growth to 4, work trip clams and corn (make sure of 2-pop whip on galley)
T65 Growth to 5 yuck, work trip clams, corn, GF
T69 Growth to 6 yuck:+), road corn
T70 Whip galley (2 pop)
T71 Priesthood, settler, finish corn road, work trip clams+corn
T72 worker->galley
T73 worker->GH
T74 mine GH
T78 Writing, whip settler (1 pop)
T79 settler->galley, work trip clams
T80 Oracle, worker->galley->capital, galley heads off with settler (risky, no protection - maybe settle island), work trip clams and GHmine
T81 worker->PFH
T82 mine PFH
T85 Growth to 5mad work two clams, Gmine, PFH
T86 un-, work two clams, GF, GHmine, PFH
T92 Alphabet, finish PFH, work trip clams, GHmine, PHmine
T93 Growth to 6mad, whip oracle, work trip clams
T94 Oracle, claim victory, etc.

Final screenshot - NB has spoiler info for my first saved game (next post)

OK, here are my three test saved games. They're each on quite different map types, and quite different again from Grifftavian's test game. The first test game corresponds to the game reported in the previous post, not that it matters much.

Nice job with the testing Mabraham.

Oracle is sometimes taken by the AI T91 or T94? Yes, it would be good to investigate a number of trials to see if this is very common behaviour.

I understand it is desirable to play test games because right now there really isn't anything else to do. But I'd recommend restraint when it comes to inferring how the actual game will behave. One thing that may be an important factor, is that I think there is a high chance we are starting this game completely isolated.

What I do like is the good planning for the first 20 or so turns - where we basically DO know all the little details. In this respect I am in favour of:
1. Either warrior or barracks first to grow to size 2,
2. Growth should coincide with learning fishing, at which point we begin a workboat.
3. How to get that first workboat out ASAP - looks like the optimum is to work gf+phf.

I think both Mabraham and Griff are testing the same things here, assuming settle-in-place.

What the warrior reveals on turn 0 might make us inclined to move the settler, ie, settle on the ph. I can think of some pros and cons for this move:
Pros to settling on ph:
extra 1/turn more than makes up for lost output from moving for the turn.
That extra also accelerates the rest of our opening.
The island to the west is available for settling, which will claim the missed clams.

Cons to settling on ph:
Capital loses one source of clams.
Cannot settle additional cities on starting island (fog gazing masters - do we know for sure that the island doesn't continue to the north?)
An additional 2 forests

adrianj said:

(fog gazing masters - do we know for sure that the island doesn't continue to the north?)

Click to expand...

If I had to guess I would say no based on the description of the game. And if it is that small of an island I would think building a warrior first might not be necessary? Would there be a threat from barbarians at this level if it is just a small island?

Thorn said:

If I had to guess I would say no based on the description of the game. And if it is that small of an island I would think building a warrior first might not be necessary? Would there be a threat from barbarians at this level if it is just a small island?

Click to expand...

If I hold my laptop screen up close and squint through my bifocals it doesn't appear to me that the home island has any land to the North. IF that is the case, we shouldn't have to worry about land based Barbs on the home island. I'd be more worried about Barb Galleys pillaging the Clam Boats. That might make an Oracle>Metal Casting slingshot more favorable, in order to get out a Trireme or two to defend the Clams.

Grifftavian said:

A long term solution to the lack of hammers for a SIP Paris would be The Moai Statues. IIRC, SIP will have 12 water tiles in the BFC (3 of them Clams). The one time I did get The Oracle built in my test game, I build Moai Statues first, chopping all the forests into Moai, and then build Oracle by working hammer enhanced Clams and two Mines. The downside to that approach was that when I met the first AI Civ just after I had completed The Oracle, they had 5 or 6 cities to my two. So while a SIP Paris may be a good site for Moai, I wouldn't recommend building them until we have several more cities up and running, hopefully cities with better production that have secured Copper, Stone, Marble, Horses and/or Iron.

Click to expand...

Maoi-capital is a good suggestion. At size 6 it will take at least 38 turns to build with the max 10-turn to build available (375 total, some chops and whips possible). Then at size 6 we'd work corn + 3 clams + 2 coast tiles to get 5 per turn back, so payoff in 75 turns. If we get HR (somehow) or Pyr->Rep then we can get higher rates of return, of course. Obviously, it would be enhanced by Colossus and Bureaucracy (but the latter clashes with Nationalism later). The longer the game goes with us working the coast tiles, the better the total return. However in an SGOTM, the question is more about applying sufficient force early enough, rather than maximum force eventually. My general feeling from forum-reading and experience is that the Maoi is useful in long games, or to make a marginal city site playable.

At the magic size 11 for Caste System (below), assuming infinite health and happiness, with Moai+forge+Colossus+granary our working population returns (ratios are // post-modifiers, net of city costs) 14/12/27 on corn+clams+coasts, or 11/19/21 on corn+mines+clams+coasts. We're still whipping that food to hammers at about 3 to 5, so call it 0/37/27 or 0/39/21 per turn on average, after a lengthy building program. Note Bureaucracy hasn't been factored in. If our capital is spending 30+ turns on a Moai, and the second city at least that long on forge+Colossus, who's building our settlers?

Alternatively, the capital can sit at size 6 building non-whip settlers (149) at 19+ per turn. That's 4.8 settlers over 38 turns. Then run Caste System (whenever we get HR or Rep or something) so that at size 11 our working population returns (ratios are //// net of city costs) 0/0/6/21/21 net. Those will rapidly become piles of , of course. (Can anyone estimate a rate?) Note that Pyramids/Parthenon/NE/Pacificism haven't been factored in. Adding any one of Pyramids/NE/Pacificism doubles one of those 21 values. A library in the capital in this scenario is slightly better value than the former scenario. Ditto academy.

I think the latter generally offers faster returns, and we're after a fast game. In particular, it offers faster returns, and we need to get a serious tech lead for conq/dom. In the latter, we spam out a number of cities suited to our terrain (preferably 6+ for the theatres we'll want later), and then knuckle down to scientists. Maybe our second city is trying for forge->GE Pyramids (or straight ), maybe not. Our side cities will either run some merchants to float the economy (if they have food and no hammers), or build the necessary stuff (if they have hammers) since we can't be whipping.

bcool said:

I think Settling the 2nd city on our initial island gives us a nice solution to this happiness problem for the capital. The 2nd city can use the corn and 1 clam that the capital can't use when we whip it down to 2 pop. The can both grow reasonably well sharing and we can use the good tiles all the time without running into as many happiness issues. They both can be heavily whipped multi-pop whipped with max OF and remain at relatively low populations...

Click to expand...

Interesting idea, and contrary to my preconceptions. As you say, the crux of it is splitting the population over two cities to leverage the extra into more working of the good tiles. I will try it out.

Depending on early scouting, I think there will be two scenarios to consider here:
1. We are isolated, and require Astronomy to get anywhere. In this case we definitely need lots and lots of . Possibly a smaller, more efficient empire to begin with?

2. We have a coastal link to the mainland. In this case we are much more limited by production.
2a) We can peacefully expand onto the mainland. In this case we prioritise settlers/workers/barb defenders and get over there.
2b) We can reach an AI by galley. In this case we should start planning to take over their lands ASAP.

Reaching the mainland and setting up a new colony (eg, with Forbidden Palace) should be a priority.

mabraham said:

Interesting idea, and contrary to my preconceptions. As you say, the crux of it is splitting the population over two cities to leverage the extra into more working of the good tiles. I will try it out.

Click to expand...

I suspect you could get a settler out by approx T50 if you did it before worker. Given the lack of useful things for the worker to do (3 clams isn't so bad without even farming the corn) it might work out well.

adrianj said:

Checking in.

Had a quick read:

Some devil's advocate stuff...

Click to expand...

adrianj said:

1. Are we *sure* that Oracle is even something we want? Oracle is 225 , compared to CoL cost of 819 , or MC cost of 1053 . So we get something like 4:1. What are our alternatives?

Click to expand...

Pyramids (750) is a leading alternative. We'd have to be using something like the poly-overflow strategy of Duckweed that bcool is suggesting. If we prioritise them, then IMO we're running a Caste-fueled SE next, and once we're magically at 11 population, those become an extra 212/turn and a pile of , so payoff in something like 25-35 turns.

adrianj said:

2. Given that we have such an abundance of , might our alternative be to aggressively settle our neighbouring islands?

Click to expand...

It's either the top priority or an adjunct to it. Absent GLH, we need a way to pay for it. Currency for the extra 2-3 trade route would be essential, with CoL courthouses high on the list. That means slavery, and our would have to come largely from the capital, perhaps enhanced by Colossus and Bureaucracy.

Thorn said:

Wow, I can see I'm going to be a total newb here. I play Civ with a lot less thought about decisions than this! I'm going off to try Grifftavian's test game now. Maybe I'll learn something. Right now the only I advice I can give is we should leave our first city named 'Paris'.

Click to expand...

Welcome Thorn and mabraham! I hope your experience here will prove to be FUN first and foremost. As someone else said, playing is fun, winning is MORE fun! I am looking forward to hearing some new voices, no offense intended to any returning member of course. I also want it to be perfectly clear, This is NOT, MY team! I sort of fell into the leader role somewhere back around SG 3-4 when most every other member quit for one reason or another. If ANYONE doesn't like how ANYTHING is going, please speak up. I like to operate inside Democratic principles, this is not a Monarchy.

My initial thoughts after reading through what has been posted already and merging that with what I believe to be true about winning quickly will follow.

Try not to whip anything other than Settlers/Workers UNLESS it is an emergency. Never whip for less than 2 pop UNLESS it is an emergency. Try to maximize overflow if at all possible. Do not be afraid of red faces, they will be whipped away sooner or later. Always work your power tiles no matter what form they come in!

In my initial tests, settling on the hill seemed far superior to settling in place. The workboats get built sooner, that seemed to increase research rate and the whole process sped up from there.

There are a few reasons I like settling on the hill. That extra hammer really adds up over the course of the game if you don't sacrifice much to get it. I know in this circumstance we actually lose a forest chop, but I still like it. The second reason is total assumption, but I am betting that there is not really enough land to make 2 cities viable on this island. Thirdly, all trade routes will be overseas trade routes if we settle off the island ASAP.

I am not sure about the need for Astronomy, but I am sure that we will need a Navy at some level, if nothing more than barb defense. An Oracle for MC seems like a distinct possibility to solve the latter problem. IF we need Astro, that opens up a completely different set of issues. Exploring early will be a must, I would even go so far as to say send WB's in 2,3, or possibly 4 directions depending on what we encounter with the first 1-2. The seemingly obvious answer to the Astro question would be 4 GS's to bulb Machinery>Optics>Astro(2). The problem, where to generate them? Capitol city will have lots of food and be able run many Scientists under Caste System, but then we need to generate hammers for all other purposes in cities 2-3-4...etc. The other option is to run Scientists in multiple cities at the same time, and rotate the the building chores as the GS's are born. But I really think this is planning too far ahead as Adrian said. At this point, we can barely plan 20-30 turns if that, and where to settle city #2 is really out of the question!

I am going to give a go at the new BOTM, and see if I can survive the barb onslaught!

I played another Test Game, using mabraham's 2nd test map, and settled on the FPH on turn 1 this time. I managed to get The Oracle on Turn 85/1875 BC this time. Unfortunately, I was still researching Pottery when The oracle was whipped, so wound up taking Monarchy in lieu of Metal Casting. The morale of this story: After Midnight is a cool song, but not an optimal time for playing Civilization.

Downside is that I was still at just one city on turn 85 so aj's correct about concentrating on The Oracle at the expense of REXing out 2 or 3 more cities. I also hadn't met any of the neighbors by turn 85 in this game. Probably need to try a test game and concentrate on REX, see how many cities we can get out before crashing the Economy. But first some much needed Beauty Sleep!

Editted Autolog...

OK I forked from my earlier how-to after the T48 save. I consider the fourth WB vital for exploration.

Oracle in second city
T51 Settler
T54 Mysticism
T56 expires
T57 Whip settler (2 pop)
T58 Settler->PFH, lighthouse, work 2 clams
T59 Settler->GH 3 E of Paris
T60 Orleans (builds lighthouse working clam), 0% science
T61 100% science, Paris works 2 clams+corn
T62 Medi, Orleans switches to monument
T66 Paris grows to 4, works 2 clams+corn+GF
T67 Paris switches to worker
T69 Orleans grows and whips monument
T70 Orleans builds lighthouse
T71 PH. Paris whips worker (2 pop)
T72 Paris builds second worker, worker farms corn
T73 Paris whips worker (1 pop)
T74 Paris builds lighthouse, second worker farms corn
T76 Corn farmed early
T77 Done PH. Orleans pops border, grows to 2, works corn+clams. Workers to PFH. 0% science.
T78 Workers mine PFH. 100% science
T81 Paris grows to 2, works 2 clams. Orleans whips lighthouse, works clam.
T82 Orleans grows to 2, works clam+corn, builds Oracle (with 23 flowing in)
T83 Finished PFH, one worker to GH.
T85 Orleans border pop
T86 Paris whips lighthouse (1 pop)
T87 Paris grows to 2, builds galley. orleans grows, works clams+corn+PHmine
T89 Paris whips galley. GHmine done, Orleans not ready for it yet
T90 Paris builds monument
T91 Done writing. 0% science
T92 Paris grows to 2, works 2 clams. Orleans grows to 4, works clam+corn+PHmine+GHmine. Still 12 turns before I can 2-pop whip an Oracle.

This is too slow. I can't see a way to make it considerably faster in Orleans, even if I omit the unnecessary monument. Life is limited by worker turns and tech rate even though Paris has been burned to the ground.

Oracle in capital
T54 Mysticism, lighthouse
T55 Whip lighthouse
T56 Paris grows to 4, work 3 clams+corn
T58 Paris whips settler.
T59 Paris builds worker
T61 Meditation. Found Orleans 3E of Paris, works clam, builds galley. 0% science.
T62 100% science
T68 Paris builds worker. Worker to PFH.
T69 Worker chops PFH
T71-72 Start PH somewhere in here
T70 Orleans grows to 2, works clam+corn
T73 Finish PFH chop
T74 Orleans whips galley. Worker mines ex-PFH
T75 Orleans builds lighthouse
T77 Writing. Could whip Paris down to 1 for second worker, but think unwise. Orleans grows to 2, works clam+corn. 0% science.
T78 Paris builds Oracle. Worker->galley 100% science.
T80 Worker farms corn. Worker mines GH
T83 Paris grows to 3, works 3 clams. Orleans works corn+PHmine
T86 Orleans whips lighthouse (1 pop)
T87 Paris grows to 4, works 2 clams+PHmine+GHmine. Orleans grows to 2, works clam+cornfarm. Oracle is 2-pop-whippable in 21 turns. Two chops make that about 14 turns.
T90 Pottery
T91 Orleans grows to 3. First chop done.
T92 Worker->galley
T93 Worker back to mainland, roads corn. Orleans builds settler.
T95 CoL
<time passes>
T101 Paris whips Oracle (2 pop)
T102 Oracle. Orleans grows to 4, whips granary

This is also too slow for the Oracle, but it does provide a nice springboard for the future. We have a second worker, and an extra mine. The second city is a lot more useful than the second city in my T94 Oracle, even considering the 8 turns difference. The two cities really do allow for / caps that are effectively higher.

Ronnie1 said:

Try not to whip anything other than Settlers/Workers UNLESS it is an emergency. Never whip for less than 2 pop UNLESS it is an emergency. Try to maximize overflow if at all possible.

Click to expand...

I generally agree, but I'd suggest two exceptions. An Organized lighthouse is a 1-pop whip and if it adds several then that may well be best. My T94 Oracle did one of these immediately Sailing was available, for +3/turn. A granary is also an abnormally high priority, because it roughly doubles the rate of return of future whips, as well as keeping more tiles working more.

Ronnie1 said:

Do not be afraid of red faces, they will be whipped away sooner or later.

Click to expand...

Can you explain further? IMO, growing unhappy can be OK (all it costs is 2 per , and that can be cheaper than not working your food tiles) but generating with over-whipping needs to serve a fairly important short-term priority if you could otherwise usefully work at the -cap.

Ronnie1 said:

Always work your power tiles no matter what form they come in!

Click to expand...

Oh yes.

Ronnie1 said:

In my initial tests, settling on the hill seemed far superior to settling in place. The workboats get built sooner, that seemed to increase research rate and the whole process sped up from there.

Click to expand...

OK, but do note that in my T94 Oracle SIP, I could work the PFH. My workboats arrived T20, T31, T42, T51. In Grifftavian's T85 Oracle settling on PFH, his workboats arrived T27, T42, T43 (and he needed no fourth WB). The critical difference here was that at size two, I could work PFH+GF for 250/turn. He could work unimproved clams+GF for 412/turn.

• My production was higher before the workboats were online (which is when are high-priority).
• I was initially behind on , but made that up fast when I got clams improved faster, and was probably ahead at T42.
• His output was higher early (he got BW a turn earlier than me) but lower later (my third clams meant I got Writing without Pottery on T92; he was due to get Pottery on T94).
• During a bit over half of his Oracle, he was returning 8/turn from his two mines+central square. During most of my Oracle, I was returning 7/turn from island GHmine+PFH+central square, with the PHmine coming online for the last couple of turns, so the settling decision doesn't have a strong effect on the application of hammers to the Oracle.

IMO, where the settle-on-PFH decision shines is in the long term, once there's a second city using the neglected clam and we've got 3 of the available hills mined and settled for +1 on the fourth. The unknown is how good that second city might be... if it's clams plus one or two land squares, that's pretty poor.

A possible SIP scenario is settling a city 2S 4E of the settler, which can use the otherwise-wasted GH 3E of settler. IIRC Grifftavian's test save had this available.

Ronnie1 said:

There are a few reasons I like settling on the hill. That extra hammer really adds up over the course of the game if you don't sacrifice much to get it. I know in this circumstance we actually lose a forest chop, but I still like it.

Click to expand...

But aren't we going to conquer fast enough that the higher returns don't come in

Ronnie1 said:

The second reason is total assumption, but I am betting that there is not really enough land to make 2 cities viable on this island.

Click to expand...

I'm tending to agree. The SIP and city-3E option might be a better springboard for Pyramids + expansion without the Oracle, however.

Ronnie1 said:

Thirdly, all trade routes will be overseas trade routes if we settle off the island ASAP.

Click to expand...

True, but AI contacts (if any) will do better than that, and we do show a profit for an on-island second city before making AI trade routes.

Grifftavian said:

I played another Test Game, using mabraham's 2nd test map, and settled on the FPH on turn 1 this time. I managed to get The Oracle on Turn 85/1875 BC this time.

Click to expand...

Nice work!

Grifftavian said:

Unfortunately, I was still researching Pottery when The oracle was whipped, so wound up taking Monarchy in lieu of Metal Casting. The morale of this story: After Midnight is a cool song, but not an optimal time for playing Civilization.

Click to expand...

Given that you didn't use the Sailing you researched, you certainly could have Oracled MC or CoL if you had been playing with your whole brain

Another improvement is to build barracks for the first 10 turns, and then leave it on ice until you came back to it later. The hammers on it take at least 50 turns to decay, so you can transfer some of those onto the Oracle, instead of a warrior.

Grifftavian said:

Downside is that I was still at just one city on turn 85 so aj's correct about concentrating on The Oracle at the expense of REXing out 2 or 3 more cities. I also hadn't met any of the neighbors by turn 85 in this game. Probably need to try a test game and concentrate on REX, see how many cities we can get out before crashing the Economy. But first some much needed Beauty Sleep!

Click to expand...

Yeah, even if you worked up to T94, you were not going to get galley+settler+travel done. So there's a serious trade-off to consider, once we have some AI wonder-finishing dates for data.