SGOTM 13 - Gypsy Kings

Thanks bc - good game and very good analysis.

I have three other small analytical observations to make:
  • At 0% slider, R1 and bc had respectively 10 and 6 :science: on their next tech, with respectively 90 and 55:gold: in the bank increasing at respectively 32 and 25 :gold:/turn. So if we give bc the extra turn to T151, he's behind by 3 :science: and 10 :gold:.
  • Of the cities where whip-:mad: was going to impact on how they were going to do things in future, at T151 R1 had 35/18/0 turns left of whip-:mad: in CC/FH/PC and at T150 bc had 36/24/2. So I estimate that bc has done his relevant whips 6 turns net earlier, which you might say is 40% of 45:hammers:, or 18:hammers:. So I don't think either game has whipped in such a way as to unbalance the comparison.
  • If R1 did whip down 4 population per bc's comparison approach, if those 4 population were each working a lighthouse-coast, that would also reduce R1's T151 science advantage of ~8:science:/turn.

So with T150 production my latest test game is up 17 food in bins, 28 hammers, 59 food/hammers (extra settler vs. extra worker)

... and down by 3:science: and 10:gold:.

That extra worker does have a significant value, and I don't expect he'd go fully to waste, but with only one galley available, each additional thing to get moved around tends to subtract a bit from the value of all the other things that now aren't as effective either. This is very hard to estimate and depends how AI trading goes. Certainly a second galley should be on the planning table soon after that trireme. R1's next priority would be another settler. bc's is might be a third worker, but if we had found another nearby 2-seafood city site, then I'd want a settler for that, first.

I have more infrastructure that I've whipped a bit earlier (so whip anger might decay a few turns earlier)

I think that's a wash, but even if you accepted my 18:hammers: estimate and deducted it from your total output, your 8-cities approach is still ahead. (You could argue that the effectively-earlier whip is worth more than the raw :hammers:. Also that it should be added, rather than subtracted!)

biggest difference is producing 1 more settler vs 1 more worker
(more cities busts more fog so less likely to get as many enemy galleys)
(more cities increases maintenance that isn't quite offset by trade route and tiles, but situation will improve quickly with currency (and somewhat soon whipped courthouses)

In this scenario, Vicky has quite few cities. In the real game, she may have more (so more trade route opportunities), or we may meet more AIs (so the 8th city actually gets foreign trade routes).

Ron has a research advantage at T151 of 8 sustainable / turn.

I think that stems from his higher total population, rather than some systemic difference. Consider the effect of 4 people working lighthouse-coast.

I hope this helps people decide what is the better option.

I think it's fairly close, but your fast-expansion approach has demonstrated an edge.

If there's still doubt, we should also look at the rates of :food: and :hammers: production for the two empires, working equivalent tiles, and adjusted somehow for population. My expectation is that the extra trade route, extra free central tile and extra worked seafood will offset the extra maintenance to put bc's game slightly ahead there too.
 
I am OK with the faster settler approach....I still think we should try to work in extra WB's ANYWHERE we can if given the opportunity. In the short run, if we need to put hammers on something in CC while waiting to grow, I would like to see more WB's. I know they will pay back very quickly once we get them to their destinations.
 
I am OK with the faster settler approach....I still think we should try to work in extra WB's ANYWHERE we can if given the opportunity. In the short run, if we need to put hammers on something in CC while waiting to grow, I would like to see more WB's. I know they will pay back very quickly once we get them to their destinations.

OK, great. I do agree that WBs are a priority. I think SM will not have time to build a granary until it has built 3-4 WBs for this reason. Forge->Library->GLib in CC is also a priority. Fast settling is a priority. Who builds WBs when can be revisited after Thorn's turn set, when some more data will be available.

@Thorn: I think most people have made some detailed commentary on your PPP by now. I think the main play testers now agree that the contents of my PDF of bcool's plan in post #713 should supersede things that have been written and said about tile management and production for CC and FH. Can you please synthesize the comments into an updated PPP and hopefully we can sign off on them fairly soon?

Making sure we wake up the worker near SM to get on the galley on the correct turn will be important, as is taking the granary off the SM build queue, switching to 0%:science:, and switching tech to Aesthetics.

Plan for Paired Clams
I think the main point we have not explicitly resolved is the opening build and tile plan for PC, which will happen in Thorn's turn set. My conclusions of post #739 were predicated on it being supplied with both its workboats. The latest test games from R1 and bc were (I think) building both WBs for itself. I think an intermediate course is best, such that PC gets the T126 WB from SM, and either builds one itself or gets a second one later from SM or BF (we can decide this after Thorn's turn set when we know more). I think post #723 demonstrated that this intermediate plan seems best for the empire. I've done some other testing today that agrees with that general point. We can get started on a wonder after LH-granary-warrior by about T151, which seems like useful times for Maoi or Parthenon.

Given that plan, because we know we want a fast LH and that there's no rush on a second WB, it is best to start working the unimproved Phill onto a lighthouse when PC founds on T117. We switch to the Pmine when it builds on the middle of T122 to finish by T128. That turn we work a lighthouse-enhanced un-improved clam, and move on to the lighthouse-enhanced net when the WB arrives from SM T129.
 
...
Plan for Paired Clams
I think the main point we have not explicitly resolved is the opening build and tile plan for PC, which will happen in Thorn's turn set. My conclusions of post #739 were predicated on it being supplied with both its workboats. The latest test games from R1 and bc were (I think) building both WBs for itself. I think an intermediate course is best, such that PC gets the T126 WB from SM, and either builds one itself or gets a second one later from SM or BF (we can decide this after Thorn's turn set when we know more). I think post #723 demonstrated that this intermediate plan seems best for the empire. I've done some other testing today that agrees with that general point. We can get started on a wonder after LH-granary-warrior by about T151, which seems like useful times for Maoi or Parthenon.

Given that plan, because we know we want a fast LH and that there's no rush on a second WB, it is best to start working the unimproved Phill onto a lighthouse when PC founds on T117. We switch to the Pmine when it builds on the middle of T122 to finish by T128. That turn we work a lighthouse-enhanced un-improved clam, and move on to the lighthouse-enhanced net when the WB arrives from SM T129.
I concur with this plan for PC. I played R1's T108 Test Game and got the same results by sending the first WB from SM up to PC. SM is able to complete its Monument and build a WB before its borders pop. SM can then build a second WB for its own Fish.

Change of topic: In playing through the Test Game, most of the other AI Civs do not have Metal Casting yet, which we took from The Oracle. MC gives us an early military advantage in being able to build Triremes, but most of the latest test games are focusing on peaceful expansion, and I don't think we've been building that many Triremes in the test games. We most likely will not be able to get fair value tech trades, but it may be possible to get techs like Alphabet and Iron Working by trading MC. What does everyone think about trading away MC? I suppose the best course of action is to stop and ask for team input for any tech trades.
 
Plan for Paired Clams
I think the main point we have not explicitly resolved is the opening build and tile plan for PC, which will happen in Thorn's turn set. My conclusions of post #739 were predicated on it being supplied with both its workboats. The latest test games from R1 and bc were (I think) building both WBs for itself. I think an intermediate course is best, such that PC gets the T126 WB from SM, and either builds one itself or gets a second one later from SM or BF (we can decide this after Thorn's turn set when we know more). I think post #723 demonstrated that this intermediate plan seems best for the empire. I've done some other testing today that agrees with that general point. We can get started on a wonder after LH-granary-warrior by about T151, which seems like useful times for Maoi or Parthenon.

Given that plan, because we know we want a fast LH and that there's no rush on a second WB, it is best to start working the unimproved Phill onto a lighthouse when PC founds on T117. We switch to the Pmine when it builds on the middle of T122 to finish by T128. That turn we work a lighthouse-enhanced un-improved clam, and move on to the lighthouse-enhanced net when the WB arrives from SM T129.

I did supply a workboat from bronzed fish to Paired Clams, but I built a workboat 1st using the unimproved plains hill and then mine, then whipped its lighthouse at pop 2. I didn't try building the LH first with the plains hill / mine. That sounds like a good plan.
 
Change of topic: In playing through the Test Game, most of the other AI Civs do not have Metal Casting yet, which we took from The Oracle. MC gives us an early military advantage in being able to build Triremes, but most of the latest test games are focusing on peaceful expansion, and I don't think we've been building that many Triremes in the test games. We most likely will not be able to get fair value tech trades, but it may be possible to get techs like Alphabet and Iron Working by trading MC. What does everyone think about trading away MC? I suppose the best course of action is to stop and ask for team input for any tech trades.

I'm still struggling to imagine any scenario under which we would see anything that would lead to short-term military action. We have no metal until IW (one way or another) unless we're desperate enough to settle Bronzed Fish on the copper. Then we'd have to build at least about four galleys and 8 axemen to have a fair chance at rushing Vicky's capital and keeping it (say). (IIRC that's 4*75 + 8*52 = 716:hammers:. Pyramids are 750:hammers:) Then they'd have to travel at least about 10 turns to get there. They'd need at least one guard trireme - either barbs or Vicky at 40% will be delighted to risk a galley for our galley and two axemen, especially when their galleys are cheaper than ours. Then we'd have to pay for the city without Vicky's trade routes, and follow up with a lot more military to clean Vicky out. Slowing down the initial assault to get IW+Maths+Construction for catapults seems more plausible, but has the same intrinsic problems. All this has to be better than peaceful expansion.

So I don't see that granting the AIs access to forges, workshops and triremes is a big deal in the time frame up to which they might tech it for themselves.

It does give them access to the Colossus. IIRC, we can inhibit them from teching MC by putting some hammers on the Colossus, but that doesn't stop them from having a go if they've already got the tech. Personally, I think we have time for only one of GLib or Colossus unless we pop an early GEngineer. One of my test games bulbed Astro in T211. PC could be in a state to start the Forge by about T150 (180/7 = 25 turns without whipping). We'd get a second workshop going at some point. Assuming we get copper somehow, then there's another 375/(2.25*9) = 18 turns. Maybe those turn counts would go down a bit with some worker-Duckweeding. T185 would be a lower bound for producing the Colossus, so the window of utility is 20-50 turns, much of which will clash with times when it will be desirable to be running merchants in cities like Marbled Clam or Piggery.

So I'm not a fan of the late Colossus, too.

Until we get Aesthetics about T135-138 we won't have any other tech that could possibly trade for things like Alpha or IW. So if we want to trade before that, MC is our raw material. I did notice in test games that when we didn't get into the trading game, the AIs stayed more backwards. Perhaps it is in our interest to wait until we have met all/most of the AIs and then trade. That will clarify the religious blocks, and maybe wars we should be bribing with MC. Slower AI tech will mean that the window for which they've had knights has been shorter when we arrive with muskets and trebs/cannons, and so we'll need fewer pikemen.

It's only at about T150 that we are significantly interested in someone who'll trade us Maths or Animal Husbandry. We might get lucky and have some shrapnel to give an AI for AH. I think our workers will have enough to do until then without IW, but if we were to get it earlier, we'd prioritise different work down at BF.

Having traded away MC, there was a game when I could have traded later to get Engineering, and that's not to be sneezed at. Perhaps trading early MC to someone likely to tech Engineering is a good thought.
 
I did supply a workboat from bronzed fish to Paired Clams, but I built a workboat 1st using the unimproved plains hill and then mine, then whipped its lighthouse at pop 2. I didn't try building the LH first with the plains hill / mine. That sounds like a good plan.

With external WB supply, LH-first is good because it provides earlier access to LH-nets and (pre-granary) working nets+Pmine at size two is fine. (My tests showed nets+Pmine at size two was doing slightly better than two nets at size two.) That combo builds the second WB somehow or other, and moves onto the granary. Details of the optimization will depend on what we decide to do later in PC (e.g. Duckweed workers onto Parthenon). If we're building a wonder without Duckweeding, then the hammers on the wonder are maximized if we delay the granary 2-whip as long as possible while our :)-cap is still 4.
 
It does give them access to the Colossus. IIRC, we can inhibit them from teching MC by putting some hammers on the Colossus, but that doesn't stop them from having a go if they've already got the tech. Personally, I think we have time for only one of GLib or Colossus unless we pop an early GEngineer. One of my test games bulbed Astro in T211. PC could be in a state to start the Forge by about T150 (180/7 = 25 turns without whipping). We'd get a second workshop going at some point. Assuming we get copper somehow, then there's another 375/(2.25*9) = 18 turns. Maybe those turn counts would go down a bit with some worker-Duckweeding. T185 would be a lower bound for producing the Colossus, so the window of utility is 20-50 turns, much of which will clash with times when it will be desirable to be running merchants in cities like Marbled Clam or Piggery.

So I'm not a fan of the late Colossus, too.

If we decide to bulb down the philosophy - paper - education - (tech gunpowder) - (tech/trade engineering) - printing press - chemistry - liberalism for steel path then we would delay astronomy for some time and get significantly more benefit from the Colossus. Right now I think this path is better than quickly bulbing astronomy. Especially if we have the GS to bulb this path (and even bulb optics and some astronomy before finishing chemistry).

We could have steel and astronomy at about the same time with this path and could use more GS to accelerate our teching (which could maximize the advantage of the GLib and the food in our capital) and make the colossus a decent wonder. However with this path I would want both the Parthenon and the colossus which might be asking too much. I think it still might be better even without the colossus...

Of course I should put my time where my mouth is, so next week I will play out a long term test game to compare with mabraham's long term test games...
 
@Thorn: I think most people have made some detailed commentary on your PPP by now. I think the main play testers now agree that the contents of my PDF of bcool's plan in post #713 should supersede things that have been written and said about tile management and production for CC and FH. Can you please synthesize the comments into an updated PPP and hopefully we can sign off on them fairly soon?
Okay, I'll have an updated PPP soon!
 
PPP v2.0. I think I captured all of the new updates. Changes are in green. Please let me know if I missed anything. This is new to me and I will be posting here during my turn set looking for help and providing information. I don't want to throw anything out of balance! :eek: I will not start the turn set until everyone agrees it's okay.

PPP Turn 108 to Turn 118

Diplomacy
Sign Open Borders treaty with Vicky
Trade or Gift Clam

Research
Start Aesthetics
Literature
Set science to 0% then to 100% if needed by end of turns

Civics Changes
None

City Builds

Clam Chowder

1. Forge till T111-then switch to settler. Whip settler on T114 then switch back to Forge
2. Library

Fish Hills

1. Workboat whip T109
2. Pyramids

Stone Mountain


1. Monument (remove granary from que)
2. Workboat


Unit Moves

Worker 1

1. build Mine
2. start workshop on SM grassland but move onto galley with settler

Cook

1. fog bust going due east
2. continue east exploration looking or AIs

Galley 1

1. Go to 1SE of FH and wait for settler

Warrior 1 > sentry in Fish Hills
Warrior 2 > sentry in Clam Chowder

Magellan after whip
1. Explore crab island then go west


Unplanned Pauses (if any)

* when anything not talked about/new happens!!!
 
PPP v2.0.
Spoiler :
I think I captured all of the new updates. Changes are in green. Please let me know if I missed anything. This is new to me and I will be posting here during my turn set looking for help and providing information. I don't want to throw anything out of balance! :eek: I will not start the turn set until everyone agrees it's okay.

PPP Turn 108 to Turn 118

Diplomacy
Sign Open Borders treaty with Vicky
Trade or Gift Clam

Research
Start Aesthetics
Literature
Set science to 0% then to 100% if needed by end of turns

Civics Changes
None

City Builds

Clam Chowder

1. Forge till T111-then switch to settler. Whip settler on T114 then switch back to Forge
2. Library

Fish Hills

1. Workboat whip T109
2. Pyramids

Stone Mountain


1. Monument (remove granary from que)
2. Workboat


Unit Moves

Worker 1

1. build Mine
2. start workshop on SM grassland but move onto galley with settler

Cook

1. fog bust going due east
2. continue east exploration looking or AIs

Galley 1

1. Go to 1SE of FH and wait for settler

Warrior 1 > sentry in Fish Hills
Warrior 2 > sentry in Clam Chowder

Magellan after whip
1. Explore crab island then go west


Unplanned Pauses (if any)

* when anything not talked about/new happens!!!
Looks about right.

I'm not certain if the Fish Hills Work Boat whip should be on T108 when you start, or T109 after it grows to size 5. I've played it both ways in testing.

If/when you get the new city founded on the two clams site, I believe we want it to start on a Lighthouse while working the plains hill (city founded on the plains tile 1SE of plains hill, so that both clams are in the first ring of the BFC).
 
Thorn, we have been testing very specific usage of tiles for each city to get our very specific outcomes for settling times and the long term finishing time for the pyramids. I hope you will be willing to look at the pdf linked to this post http://forums.civfanatics.com/showpost.php?p=10322839&postcount=713 and make a checklist for all the build/tile changes each turn as well as when to whip.

I hope you will play out a practice test game and post that for our review before you play.

This very detailed consensus of what to do with our turn sets is the way I believe most of the team wants the game to proceed. In later turns I doubt we will dictate the tile usage of all the cities, but especially early on these details have large consequences.

So can you look at the pdf and play out a test game? In addition to these details there are other factors that your PPP covers, like opening borders with Vicky and gifting her clams. But there are details about even how to move the workboats and galley that we should probably discuss as well.
 
Thorn, we have been testing very specific usage of tiles for each city to get our very specific outcomes for settling times and the long term finishing time for the pyramids. I hope you will be willing to look at the pdf linked to this post http://forums.civfanatics.com/showpost.php?p=10322839&postcount=713 and make a checklist for all the build/tile changes each turn as well as when to whip.

I hope you will play out a practice test game and post that for our review before you play.

This very detailed consensus of what to do with our turn sets is the way I believe most of the team wants to the game to proceed. In later turns I doubt we will dictate the tile usage of all the cities, but especially early on these details have large consequences.

So can you look at the pdf and play out a test game? In addition to these details there are other factors that your PPP covers, like opening borders with Vicky and gifting her clams. But there are details about even how to move the workboats and galleys that we should probably discuss as well.

I agree! ^ ^ ^ ^ ^ ^

I think the WB in FH gets whipped THIS turn, T108 as Griff says also.
 
Thorn, we have been testing very specific usage of tiles for each city to get our very specific outcomes for settling times and the long term finishing time for the pyramids. I hope you will be willing to look at the pdf linked to this post http://forums.civfanatics.com/showpost.php?p=10322839&postcount=713 and make a checklist for all the build/tile changes each turn as well as when to whip.

I hope you will play out a practice test game and post that for our review before you play.

This very detailed consensus of what to do with our turn sets is the way I believe most of the team wants the game to proceed. In later turns I doubt we will dictate the tile usage of all the cities, but especially early on these details have large consequences.

So can you look at the pdf and play out a test game? In addition to these details there are other factors that your PPP covers, like opening borders with Vicky and gifting her clams. But there are details about even how to move the workboats and galley that we should probably discuss as well.

I downloaded the pdf and looked through it yesterday. Rest assured it will be with me at all times! As far as the test game I don't mind playing through one and posting the results but which one do I use? :confused:
 
I downloaded the pdf and looked through it yesterday. Rest assured it will be with me at all times! As far as the test game I don't mind playing through one and posting the results but which one do I use?

this post (link below) has the most accurate test game with the game up to T108 simulated accurately. You can use this to practice.

http://forums.civfanatics.com/showpost.php?p=10324112&postcount=718
 
But there are details about even how to move the workboats and galley that we should probably discuss as well.

For sure. So we can discuss and/or agree, here's some detailed plans for boats.

One idea I tossed in a while back is that we might have time to double back with the galley to get some more of the tile visibility we missed last turn. Now that we have our plan for when the first settler is built (whipped T113 to build T114), we know that we have time to get visibility on one such tile:
Spoiler :

The numbers show where the galley will be at the end of that turn. So on T109 we have the visibility we want, T113 we arrive on pickup station (whipping settler and waking the worker after he does his thing on the workshop, so he's ready to go next turn), T116 the galley arrives SW of the Phill (gives maximum visibility in case there's a barbarian), unload worker first onto Phill (he's more expendable if there is a barbarian we can't see), then settler wherever. T117 we settle 1SE of Phill (build LH and work Phill tile) as we start the Pmine. Where the galley ends up T118 will depend what Magellan finds...

Magellan is whipped T108 in FH and his first two moves are easy. After that it's Thorn's baby :) I think Magellan's first priority is to help decide the location of the city that will use the crab. If we find silver and deer up there, we might settle that city even ahead of Marbled Clam! So, we want to get visibility on as many tiles as possible if they could be in city sites that could use the crab:
Spoiler :

If the coastline heads east, then by T112 Magellan won't learn anything new and should double back to the west to check out out the western city-site tiles. If it's a small island, keep going around it anti-clockwise. Otherwise he probably turns back on T113. Whenever we're done checking out possible city tiles, head W or SW (e.g. down the western edge of that little island west of the crabs).

@Thorn, please consider these points for your PPP (e.g. copy and paste)
 

Attachments

  • T108 galley plan.JPG
    T108 galley plan.JPG
    147.9 KB · Views: 138
  • T108 Magellan plan.JPG
    T108 Magellan plan.JPG
    178 KB · Views: 143
I had a look at the territory near Cook in the real save. There are some fresh water lakes just in from the coast, because various of the tiles have access to fresh water.
Spoiler :

I dunno whether that might suggest anything to anybody.
 

Attachments

  • T108 fog-gazing lakes.JPG
    T108 fog-gazing lakes.JPG
    139 KB · Views: 142
We've got company on the score graph... From the culture graph Sporks look like they put out a monument, and then an 8-:culture: wonder at about the time we did. However they're behind on the power graph - maybe they didn't take MC?
 
There are actually 2 "corner" tiles on copper island that need busted. The one pointed out by mab and the one directly south of where the galley is now. We have time to get them both cleared just barely. The galley will be 1/2 move behind mab's plan, but it would not affect settling turn of PC.

This discussion brings up another point.

IMO, NEVER use automove for any units. I hate the way the computer picks a path, and it never factors in upcoming roads for land units.
 
There are actually 2 "corner" tiles on copper island that need busted. The one pointed out by mab and the one directly south of where the galley is now. We have time to get them both cleared just barely. The galley will be 1/2 move behind mab's plan, but it would not affect settling turn of PC.

Seems great. The next settlers are due T123 and T125, so the galley has more than enough time to get to the SW departure point to settle MC.

This discussion brings up another point.

IMO, NEVER use automove for any units. I hate the way the computer picks a path, and it never factors in upcoming roads for land units.

Definitely. When scouting with boats, move one tile at a time. That way you choose which tiles you see, and sometimes that information changes which tile you wish to end up on. For example, if Cook was following a north-going coastline, and the first move found an island to the east, it'd be silly to have booked in a second move north already.
 
Top Bottom