SGOTM 13 - Gypsy Kings

[bcool]

Comments in blue.

Okay, here's my rough draft PPP. It's not complete and need mega feedback! I will update this post as changes are made.

PPP Turn 108 to Turn 122

Diplomacy
Sign Open Borders treaty with Vicky

Research
Start Aesthetics
Literature
We are going to want to go 0% research for almost your entire turn set I think. Set it on Aesthetics, and go 0% research until T120 or T121. Then 100% on Aesthetics in your last 2 turns.

Civics Changes
None

City Builds

Clam Chowder

1. Forge
2. Library
This is probably going to be quite complex in your turn set. I don't think we have settled on the exact best path here, but we are going to interrupt the forge build and whip a settler soon. Then some combination of switching/whipping the forge/ another settler and even possibly a turn on a library. We definitely won't finish the library in your turn set, and maybe not the forge either.


Fish Hills

1. Workboat
2. Lighthouse
I think we are whipping the workboat T109 and then starting on the Pyramids. Possibly interrupting the pyramids and starting a worker near the end of your turn set.

Stone Mountain

1. Monument
2. Workboat
I believe everyone agrees on this. Continue working the Grass Hill Mine and build the monument and then start on the workboat.

Unit Moves

Worker 1

1. build Mine
2. move to CC if needed or build workshop on SM grassland
will finish the Grass Hill Mine, then move North to start a workshop on SM grassland. Will board the galley before it is finished most likely and travel with a settler to the 2 Clams site to build the Plains Hill Mine.

Cook

1. fog bust going due east
2. continue to fog bust
the term fog bust usually refers to having a unit remain stationary in the fog near to your cultural borders to prevent the appearance of barbarians. Cook really is scouting for AIs and trying to make as much progress East as he can. As I understand the recommendations, he cuts Western corners on islands as they would only reveal possible routes that go West, but doesn't cut corners on Eastern corners to reveal diagonal land connections that might connect us to more AI.

The new workboat that will be produced in Fish Hills is missing. I think we want this FH workboat to scout the Crab Island then go West like Cook is going East.


Galley 1

1. ferry Worker 1 to CC if needed
Galley 1 is coming back cutting through Stone Mountain I believe to await the boarding of a settler from CC and the worker on Stone Mountain's island. Galley 1 would wait 1SE of Fish Hills.

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


Unplanned Pauses (if any)

* when anything not talked about/new happens!!!

Spreadsheet details <<< not complete and need lots of help!

Also, do we want to have any diplomatic interactions with Vicky? Open Boarders treaty comes to mind.
I will wait for the change requests to pour in.

 

[Ronnie1]

by bc.....mabraham, I would be interested in comparing the worked tiles plans in your test game with mine. I think there are some interesting choices in how we whip. I was running quite a few turns with 1 unhappy in the capital, but I was getting the settlers out pretty fast and the other infrastructure up. I could have easily built a trireme instead of my warriors (obvious more immediately useful). while a 3rd worker in FH might speed up the pyramids 1 or 2 turns it might not be as useful as getting a settler out of FH earlier than we could otherwise.

I agree in principal with bc here...I was running 1 or even 2 in CC, all my settler whips were started with 2....The really cool thing about CC is that with a forge and 2 , if we work the best available tiles get 18/t, so in 2 turns we are able to 2 pop whip with a max overflow of 55 hammers which works out really great!

In my test games, I got settlers out T116,126,135,146...forge on T127, library T140.

I'm going to have another run through later to see if I can squeeze another WB out since we need another sooner since we decided to settle 2 Clams with both resources in the skinny square.

by mab.....My settlers came out T117, T127, T136 and T146. This last one is probably worth delaying a bit and managing to get out some kind of boat, since the earliest we can get Literature to start GLib is around T149.

I actually started Glibrary on T147 in my last test.

@ Thorn....for this next set, it will be important to check every city and tile every turn. One of the things I noticed in my testing, when a citizen turns from to , it happens IBT and he actually gets to work the best available tile. So sometimes I had to make that tile available in anticipation of that change. The main thing is to play with the test game....play the same 10-20 turns over 4-5-6-etc time until you are comfortable with what is taking place. It doesn't take as long as you might think, and will give you a much better feel for what is taking place.

 

[mabraham]

@ Thorn....for this next set, it will be important to check every city and tile every turn. One of the things I noticed in my testing, when a citizen turns from to , it happens IBT and he actually gets to work the best available tile. So sometimes I had to make that tile available in anticipation of that change.

I disagree. The tiles that are worked are those selected when you end the turn. If there is growth, or transition to happiness, you do not get work from the new guy. He is allocated something to do by the time you get see the city next turn, but you can count back from the contents of the boxes this turn and last turn and see what work was done, and he did none.

Certainly we need to be checking that the new guy is following the plan - i.e. turning off the city governor, but my observations in building this spreadsheet have been that he doesn't need to be planned for.

For example, on T95 CC was unhappy and grew happy while building a settler. We got two free on the settler from not feeding the unhappy guy on T95. On T96 he worked the GForest, so fed himself and produced a .

Note that I recently observed that you don't pay for either when building settlers and workers - which complements our strategy of growing into and switching to these builds!

 

[bcool]

Juggling build queues is somewhat painful, and I have not thought of a good way to do that (yet). Please check out the recommended mechanics on the CC spreadsheet around T50.

The most important area that wasn't on the previous PDF is the tile matrix. Hopefully it is fairly self explanatory. It is set up so that your current set of tiles will continue forever until you actively change them. However, if you go back to earlier turns after making changes on later turns, you can end up with a mess. When you're moving guys from tile to tile, just type in the actual number in the appropriate light grey cell. When you want to undo that effect, don't just type in the old number. Instead, copy a nearby blank light grey cell onto the cell you want to undo. This will mean the propagation of tiles worked over time will keep operating properly, which just typing numbers in won't do (in general).

There are numbers of rows that are worth "hiding" when you're engaged in different tasks, if you find yourself short of screen real estate.

Hope it's useful!

Re: spreadsheet. It is an impressive start at simulating the game. However, as it is now, I don't think I'll be able to use it productively. I think I figured out how to juggle build queues, but reverse engineering your spreadsheet was not as easy as I thought it would be. It quite impressively complex, but of course this makes it a bit more difficult for us to just jump in and use.

I was trying to come up with a way to better manage juggling build queues. Maybe you have 4 possible builds per city, and you have a cell that you toggle to "1" if that is what you want to build that turn. So in formula you have a if "cell x = 1" then process hammers (and food) for that build. If no "1" then it keeps the old values stored in that build. Of course you would have to have some way to reset the values when the build finishes. It is tough to do elegantly.

 

[mabraham]

Re: spreadsheet. It is an impressive start at simulating the game. However, as it is now, I don't think I'll be able to use it productively. I think I figured out how to juggle build queues, but reverse engineering your spreadsheet was not as easy as I thought it would be. It quite impressively complex, but of course this makes it a bit more difficult for us to just jump in and use.

I was trying to come up with a way to better manage juggling build queues. Maybe you have 4 possible builds per city, and you have a cell that you toggle to "1" if that is what you want to build that turn. So in formula you have a if "cell x = 1" then process hammers (and food) for that build. If no "1" then it keeps the old values stored in that build. Of course you would have to have some way to reset the values when the build finishes. It is tough to do elegantly.

Agreed. I have just implemented something along these lines, so that you can juggle a build back up the queue after it gets the work done on it, and the new build back down. Builds that actually finish will just fall off the queue like you'd expect. So once the queue is populated with the next bunch of builds (e.g. we're building a wonder and have a few settlers queued), Duckweeding a settler on is simply a matter of putting a "2" in a box to move the wonder to slot "2", and another "2" in an adjacent box to take the old contents of slot "2" and starting building it. Then we whip it at the right time, it falls off the queue, and the contents of slot "2" move into place to get the overflow. Two boxes are needed for instructions so that all possible queue changes can be accommodated. I'm testing this at the moment, so do watch this space.

Back at T50 or so, when we moved a workboat off the queue to do a settler whip then workboat whip, you just move the workboat into slot 3 and move the settler (in slot 2) down. Then there's nothing to do but whip twice and then we're back to the original workboat build.

 

[mabraham]

Please find attached an updated version of the spreadsheet and a corresponding PDF manual. They're both zipped to avoid running foul of the forum's rules on attachment sizes.

I moved the section that manages empire-wide resources to its own worksheet, and implemented a much more user-friendly way to manage the build queues. Hopefully that will make the spreadsheet much more useful as a planning tool for anybody keen enough to try it.

There's also an 11-page manual with screenshots and descriptions of what is going on. Constructive feedback on that and the spreadsheet are still warmly welcomed!

Hopefully we can get on and agree on a plan for this turnset now!

edit Fixed minor glitch with CC build queues. Updated version to 0.2.1.

 

[mabraham]

Comments in blue.

I agree with everything bc said about the PPP.

The FH workboat should initially scout to the east of the crab island. There are three potential city sites for Crabby that pick up the crab. I think this scouting workboat should go along the eastern coastline of the Crabby landmass until we have seen all the BFC tiles that we are likely to see. For example, if the coast goes straight north, then we want to go to 3N of the crab, so we've seen as many of the BFC tiles as possible of a hypothetical city site N-NW of crab. Once we are at that point, we'll have to decide whether to keep following that coastline, or double-back along the coast to see the western side of that island. I suggest we put that in as a discussion stopping point during Thorn's PPP.

 

[mabraham]

We need some information on the constraints we want to satisfy over the next 20 turns with our settler builds so that we get decent value for our galley and worker. We have more than enough time to get Paired Clams settled before worrying about marble connected. We just need marble up by T145 or so, so that means Marbled Clams by about T134, which is 26 turns after this turnset starts. Heaps of time.

Our worker heading to Paired Clams and our galley bringing it back will make the best use of their time if the next settler arrives at about the time they are arriving back. So here's a screenshot of the timing. Numbers are located on the position of the worker or first settler at the end of the indicated number turns since the whip of the first settler.

Spoiler :

Here we see the galley take settler and worker over to the PC site, unload both on the PH. On turn 4 the settler walks SE and plants PC, while the worker builds the mine. The mine is up on turn 9, and the worker boards on 10 for the galley to start to move back. So if the second settler is whipped on any turn up to 13, then it'll be boarding by turn 14, to land on the marble on 15 to move to the Gforest on 16 to settle on 17. If the second settler is whipped on turn 14, then it can board at the start of turn 15 and has to stay on the galley. The galley can move further on turn 16, to unload the settler on the Gforest on 16, to settle on 17. Either approach is workable, so long as we can afford the extra galley travel time.

What we do next will vary a bit.

By now, we have a second worker out of FH one way or another. If we've been able to trade for IW, then we want to get Bronzed Fish settled ASAP. This second screenshot describes the further minimal bounds on the galley. Numbers are now the location of the galley or the BF settler at the end of the turn. I've assumed that the second settler was whipped earlier than T14. Otherwise, we have to add two turns for the extra galley movement.

Spoiler :

So we need to whip the third settler on turn 16 at the latest to get it (plus worker) out to BF to get going as soon as we can, given MC as fast as possible after the PC settler.

Let the turns those three settlers are built be a, b and c. For best results, we have

a = 108+x
b <= a+13
c <= b+3

thus

a = 108 + x
b <= 121 + x
c <= 124 + x

So far, the three published plans have produced settlers from CC on

• T113, T125, T142 (bc)
• T116, T126, T135 (R1)
• T117, T127, T136 (mab)

So, bc's second settler is a turn early, and the third is rather late. R1's second settler is three turns early, but his third is three turns late. mab's second settler is three turns early, and his third is three turns late.

So I'm going to look at tile optimizations wrt my plan on the spreadsheet to see if I can push my second settler a bit later and the third settler a bit earlier.

If we haven't be able to trade for IW, then there's no rush to settle BF. Instead I guess we settle Piggery, but the constraints are much the same.

A secondary constraint would be whether we can have the second worker out in time to build the mainland Gworkshop before turn 15, so as to board with the settler to head off to BF. (Can board a turn or two later if heading to Piggery.)

 

[mabraham]

So far, the three published plans have produced settlers from CC on

• T113, T125, T142 (bc)
• T116, T126, T135 (R1)
• T117, T127, T136 (mab)

So I'm going to look at tile optimizations wrt my plan on the spreadsheet to see if I can push my second settler a bit later and the third settler a bit earlier.

A secondary constraint would be whether we can have the second worker out in time to build the mainland Gworkshop before turn 15, so as to board with the settler to head off to BF. (Can board a turn or two later if heading to Piggery.)

OK, I've come up with a tile-and-build plan that gets settlers T117, T130, T136 and the worker from FH T124. The second settler is being whipped on "13" from above (=T129), and then another on about turn "19" (=T135).

I can move the second settler to produce on T131, which means the galley travels further to plant Marbled Clam at the same time, instead of spending the same time waiting for the third settler, but the tile usage in CC is not as good. It is very hard to regrow to 6 after 3-whipping the second settler in time to spend two turns on the settler to 3-whip it again. Putting more food in before the first or second settler in order to regrow faster just means you spend more time while growing to 8 before the second settler, which is counter-productive. I do not think there is an optimal solution unless FH can build the third settler in a reasonable way.

Meanwhile on T124, FH got out a second worker, which has time to put out a 9-turn mainland workshop and get on the galley with time to spare before the T136 settler is built. So that is all good. Also, I am still slow-building the Pyramids on T150 (and could whip them on T145 to finish T146). My count of "good tiles worked" is the same, and is almost identical.

The gain with this plan is that I spend more time working tiles in the T127-129 time, rather than whipping early and having the settler wait around for the galley+worker to be ready.

The leading alternative is to bring the galley home ASAP from PC, get out the second settler ASAP (e.g. T127) and take the T124 worker from FH across to organize the marble. (That worker will have time to put a turn or two on a mainland workshop for something to do.) The galley scampers back to get the PC worker (who's had time to build the mine and most/all of a Pworkshop) and races back to pick up a T136 settler (say) to go and plant at BF at exactly the same time as the first plan. (The earlier 19 number on the BF site didn't include the 3 turns' delay that I've now decided is inevitable).

Again, numbers are turns after the T116 whip on the first settler, mostly galley locations:

Spoiler :

Now MC is up three turns earlier, we've (perhaps mostly) built a PC workshop instead of a mainland one, the MC worker will be free sooner, and BF is not delayed. Pyramids won't be delayed earlier.

If we were building Piggery instead of BF, then there's less rush on the galley, and we can probably finish the PC workshop.

Actually, I'm liking this plan better than the delayed-second-settler idea.

 

[mabraham]

Galley 1 is coming back cutting through Stone Mountain I believe to await the boarding of a settler from CC and the worker on Stone Mountain's island. Galley 1 would wait 1SE of Fish Hills.

Actually, when we commit to a build plan for the next settler, we should count the galley moves and see if the galley has time to double back and de-fog those two corners we missed off the copper island. Even a T115 settler is late enough to allow this "safety-play".

 

[Ronnie1]

I disagree. The tiles that are worked are those selected when you end the turn. If there is growth, or transition to happiness, you do not get work from the new guy. He is allocated something to do by the time you get see the city next turn, but you can count back from the contents of the boxes this turn and last turn and see what work was done, and he did none.

I must have made a calculation error somewhere...it sure seemed like I picked up extra hammers.

 

[bcool]

Okay...

I think this is a competitive option. Mabraham's galley optimization got me to thinking...

This is an alternative plan up to T150
Whips pyramids T150, but! 8 cities (18 pop) with all workboats in place. Great Library with 100+ hammers into it. 2 workers.

issues
-Needs... another worker for Bronzed Fish (since we probably would have traded for IW by now)
-and needs another worker for Pigs (since we probably would have traded for Animal Handling by now)
-Needs a trireme (but one is ready to whip with max OF in CC)
Will need another galley soon.
-I do slow down the pyramids for faster expansion, but I can still whip it T150. Fish Hills can recover from the whip by slowly building infrastructure while CC takes Pmine to build the Great Library and Pigs takes its fish to quickly get up into productivity.
-whip anger in CC is 36, whip anger in FH is 24

I believe it compares favorably with other test games, and made better use of the 8 good tiles shared between FH and CC.

I will report tile usage maybe tonight if people agree with me that it is a competitive option.
edit: here is the link
https://spreadsheets.google.com/ccc?hl=en&hl=en&key=tCSc96mbUfYhOoMR5F0sn0A&authkey=CKPhkaMM#gid=0

didn't record the other cities but made notes on what they built for whom.

Other important milestones
Forge finishes T126
Library finishes T139
Settlers finishes T113 settled T117 (Paired Clams)
finishes T124 settled T128 (Marble Clams)
finishes T127 (in FH) settled T130 (Bronzed Fish)
finishes T144 (in FH) settled T147 (Crabby)
finishes T147 (2 pop in CC) settled T150 (Pigs)
2nd worker finishes T133 (in CC)
Literature finished T148
T143 Marble hooked up
The Great Library has 126 hammers invested into the build

------
I played with mabraham's spreadsheet and found an error in my first reported use of tiles in my old test game. So it definitely seems to be accurate so far. I wasn't completely clear on how to add a new build (a build that wasn't in the build queue list already).
I do find test games to be easier to explore new plans. I'm not comfortable enough yet to "see" the ideal whips and tile usage from just the spreadsheet yet.
Build queue worked very well however, and the guide was really nice. I think I should have printed it out, I probably could have found how to add the new build to the queue. I remembered reading it but couldn't find it again (I hate reading from the computer --at least anything longer than a page)

 

[Ronnie1]

I played another test run through to T151 using some of the changes that have been presented....

I think it is pretty solid for the following reasons.
1) The WBs get out quickly to speed growth
2) The Settlers are out soon enough yet still don't stress our single galley too much.
3) The galley has time to shuttle our worker(s), however they do have to wait a little.
4) 2 Clams gets a great start and early Moai start
5) Marble is connected in time for GL and use by city at Size 2.

Again...I made no trades (OB) with anyone except Vicky and traded for no techs.

Things I think we still need to be clear about going into the next set.
1) Diplomatic options with new contacts, or stop orders on new contact?
2) Magellan path...around "Crab Key" from the east for sure, then which direction? NW > SW always going WEST IMO.
3)Religions...If we get an autospread to us...do we adopt?

Spreadsheet is here...99% confident on tile usage...https://spreadsheets.google.com/ccc?hl=en&key=t4HX422byDNVy2buM1OhXQw&authkey=CPf5jCU&hl=en#gid=0

Here is the closing save file, details, 6 cities, 17 total pop, CC started GL T151, Pyramids finish T150(no whip), anarchy T151, 2Clams completed LH, Granary starting Moai T151, SM has LH and working Granary, Marble has LH and working Granary, Copper working LH, Piggery to found T152.

EDIT: X-posted with bc...I like his BETTER

 

[Ronnie1]

I like MOST of bc's last test better than mine...with a few exceptions. 2 Clams is less advanced for some reason, we should be able to fix that. Also, Marble did not start with a Lighthouse, which would have advanced it also.

I would like to see a spreadsheet of the tile use to evaluate if I could please.

 

[mabraham]

I must have made a calculation error somewhere...it sure seemed like I picked up extra hammers.

Possibly free food (converted to hammers) from building a settler/worker while or . These things are very easy to miss - but I've had to make a lot of close observations while making and fixing this spreadsheet.

 

[mabraham]

Things I think we still need to be clear about going into the next set.
1) Diplomatic options with new contacts, or stop orders on new contact?
2) Magellan path...around "Crab Key" from the east for sure, then which direction? NW > SW always going WEST IMO.
3)Religions...If we get an autospread to us...do we adopt?

1) I think we should stop on new AI contacts. Unfortunately, we can't save the game while still in the initial negotiation screen, but if feasible, the player might be able to keep the game open for a short while to post and get any feedback. At least one time I met Joao, said "sure, peace is great" and then immediately had him demand a tech. Having had the group discuss the religion/liking/trading/resource matrix seems likely to be fruitful. We can't pre-plan for all eventualities.

2) Yep, start with the east coast, do as much as we need to de-fog the potential Crabby city-site BFCs, then discuss heading back around to the west coast.

3) I'd generally say we don't accept a religion, even if they ask. It's not going to auto-spread to more than a city or two, and we derive a negligible benefit from switching (one , and maybe some AIs liking us more). We could get a further negligible benefit from OR (25% to buildings) but that costs us 100% of our empire for a turn. Meanwhile, any AI not of that religion will hate us. I don't much care about keeping them happy, but it would be silly to deny ourselves an early potential tech-trading partner for a negligible gain. Once we have more of a feel for the religious matrix, we might consider switching, but it would have to be with a view to bee-lining Philo to run Pacificism. That means we need the religion in at least CC, and preferably FH and some other city that might pop a late-game GPerson (e.g. Piggery).

 

[bcool]

editted this post http://forums.civfanatics.com/showpost.php?p=10321701&postcount=692 to include tile and build plan

 

[mabraham]

Okay...

I think this is a competitive option. Mabraham's galley optimization got me to thinking...

This is an alternative plan up to T150
Whips pyramids T150, but! 8 cities (18 pop) with all workboats in place. Great Library with 100+ hammers into it. 2 workers.

issues
-Needs... another worker for Bronzed Fish (since we probably would have traded for IW by now)
-and needs another worker for Pigs (since we probably would have traded for Animal Handling by now)
-Needs a trireme (but one is ready to whip with max OF in CC)
Will need another galley soon.
-I do slow down the pyramids for faster expansion, but I can still whip it T150. Fish Hills can recover from the whip by slowly building infrastructure while CC takes Pmine to build the Great Library and Pigs takes its fish to quickly get up into productivity.
-whip anger in CC is 36, whip anger in FH is 24

I believe it compares favorably with other test games, and made better use of the 8 good tiles shared between FH and CC.

[/S]I will report tile usage maybe tonight if people agree with me that it is a competitive option.
edit: here is the link
https://spreadsheets.google.com/ccc?hl=en&hl=en&key=tCSc96mbUfYhOoMR5F0sn0A&authkey=CKPhkaMM#gid=0

didn't record the other cities but made notes on what they built for whom.

Other important milestones
Forge finishes T126
Library finishes T139
Settlers finishes T113 settled T117 (Paired Clams)
finishes T124 settled T128 (Marble Clams)
finishes T127 (in FH) settled T130 (Bronzed Fish)
finishes T144 (in FH) settled T147 (Crabby)
finishes T147 (2 pop in CC) settled T150 (Pigs)
2nd worker finishes T133 (in CC)
Literature finished T148
T143 Marble hooked up
The Great Library has 126 hammers invested into the build

Looks quite good. Will be keen to see details.

I am concerned about the lack of a trireme. I have seen barb galleys by T137, and often by mid-140s. So I think we may need to have a contingency trireme or worker plan instead of the Crabby or Pigs settlers.

It is worth remembering that a workboat costs 45 and a trireme 75 . Each workboat saved pays for over half of the trireme, and we're about 90% to keep the trireme. While we've got only one, we'll have to be pretty cunning about placing it somewhere from which it can be a "rapid response" unit. The galley also costs 75 and is only about 60% to survive, so if our risk is only one workboat, we should the galley as a lure only. As far as I can see from my test games, unless the barb starts its turn on an improved seafood, it will prefer to chase something rather than pillage.

 

[bcool]

I have 2 turns on a trireme T140 and T141 so could whip that in the capital if necessary. would cut into the Great library however. Perhaps a double whip in Bronzed fish might be able to get a trireme out in an emergency as well.

The galley path was effective but I didn't keep careful notes... drops off worker and settler at Paired Clams, comes back in time for settler for Marble (might waste a turn waiting), picks up settler from FH for bronzed fish, loops around to Paired Clams to pick up worker (after worker finishes mine and workshop there), then picks up other worker from CC, drops both off to quarry marble, then waits for them to finish, takes workers drops one off for forest chop into Great library, and picks up settler and takes it with worker to crabby, then picks up settler and drop it off for Pigs,

then can go back get worker chopping forest for crabby and drop it off for pigs or Bronzed Fish.

 

[Ronnie1]

snip...nope