SGOTM 09 - Xteam

[PaulisKhan]

Here's the current strategy as I see it (these additonal starting techs have definitely changed things up).
Electricity slingshot and run mills under environmentalism for most of the game.

Grow capital to size 20 asap (keeping in mind the need to build settlers and workers)

Find a good grassland/floodplains city with food that can support 16+ cottages, mature cottages and eventually build oxford (I believe this will prove more efficient than doing it in the starting city, although the eventual placement of oxford will depend on the situation in game)
I'm assuming that our empire will soon be large enough for FS to outweigh Bureau.

War/Rex our way to just under the Dom limit at a sustainable rate, while trying to maximise metal resources.
Make sure to pop a GE in time for mining inc.
We may want to consider keeping a GProph in had to shrine the major religion when we capture/found it.

something to consider: large cities mean big trade routes which can make the GLH even more powerful than normal, obviously this will be somewhat map dependent, but we need to keep it in mind (an extra 2 matured cottages per coastal city that dont require citizens to work them is not to be laughed at).


lessons learned form hard experience: Disproportionate amounts of effort tend to go into the last few turns of these types of game, when really it is the decisions made at the start that determine the outcome. If we can keep our economy intact and run at 100% research for most of the first third of the game, we stand a good chance of winning this.
I know that I was recruited for my experience playing space race games and all that experience tells me that it's the teams with the best civ-economists that will be the most competitive.

 

[Frederiksberg]

Here's the current strategy as I see it (these additonal starting techs have definitely changed things up).

Sounds like you have been busy . Do you have any estimate on the total game length? It may be of use when calculating the interest we get on our investments.

something to consider: large cities mean big trade routes which can make the GLH even more powerful than normal, obviously this will be somewhat map dependent, but we need to keep it in mind (an extra 2 matured cottages per coastal city that dont require citizens to work them is not to be laughed at).

You may have a point here. Even the AI cities tend to get big fast, don't they.

lessons learned form hard experience: Disproportionate amounts of effort tend to go into the last few turns of these types of game, when really it is the decisions made at the start that determine the outcome. If we can keep our economy intact and run at 100% research for most of the first third of the game, we stand a good chance of winning this.

You are completely right in pointing out the importance of having a game plan from the beginning. In the late game only a few turns can be won or lost while a wrong strategy early on can be very costly indeed.

If we can keep our economy intact and run at 100% research for most of the first third of the game, we stand a good chance of winning this.

What about the last 2/3 of the game? Having the science slider at 100% doesn't mean that you couldn't make even more beakers with a larger empire running at lower rate. I suppose you are thinking about building commerce to support a higher science rate which is often a good deal as SCT explained in the last SGOTM.

One reason I'm asking this is that there is an interesting dilemma between REX and maintaining a reasonable beaker production in the early game. You might think that maximum REX is best in the long run because growth will be massive with many cities and what you loose early in maintenance will come back with interests later. I'm not sure, however, that things are that simple. One obvious reason is that reducing early beakers will delay the time when beaker and hammer multipliers like Libraries and forges kick in. What does your experience tell you about this trade off between REX and keeping early beaker rate up?

 

[Frederiksberg]

MoM: for extra 3 turns per golden age (which we could have 4-5 of with our Philo trait)

Actually golden ages are more powerful than usual in this game for two reasons:
1) Golden age length is 6 vs 8 on normal speed but 6 turns on quick corresponds to 9 turns on normal.
2) With many windmills, watermills and some forest preserves we will have more tiles with both an extra hammer and an extra commerce during a golden age.

 

[PaulisKhan]

What about the last 2/3 of the game? Having the science slider at 100% doesn't mean that you couldn't make even more beakers with a larger empire running at lower rate. I suppose you are thinking about building commerce to support a higher science rate which is often a good deal as SCT explained in the last SGOTM.

You're right and I should have discussed this in more detail
I mentioned the importance of teching strongly in the first 1/3 of the game because that is when there are the fewest option for generating cash and that is the point where it can be possible to get stuck in a hole. By the end we should be completing 1 tech per turn however early on it's quite possible to get stuck on a tech for 25+ (quick) turns which is a significant fraction of the game
I've lost count of the number of games I've played where overexpansion has crippled me and it's take 100-200 turns (on marathon) to dig my way out. I'm nervous here because of the "no do-over" style of GoTM.
I recall in the deity game where I knocked Ironhead off the board briefly, I was generating -37 gold at 0% research and was forced to whip warriors just to generate enough overflow cash to finish off writing (all of my captured workers had been disbanded). I think in that game I pushed the expansion too far (or kept one too many cities) and it probably cost me 60/650 turns (9% of the game length).
Knowing when to stop will be crucial in this game
Later on we'll have dozens of ways of generating gold to keep our economy afloat so getting "stuck" is much less problematic.

One reason I'm asking this is that there is an interesting dilemma between REX and maintaining a reasonable beaker production in the early game. You might think that maximum REX is best in the long run because growth will be massive with many cities and what you loose early in maintenance will come back with interests later. I'm not sure, however, that things are that simple. One obvious reason is that reducing early beakers will delay the time when beaker and hammer multipliers like Libraries and forges kick in. What does your experience tell you about this trade off between REX and keeping early beaker rate up?

It's a well posed question and I wish I had an answer to it. In fact I'll now probably contradict myself completely
I play by gut feel and if I ever start feeling comfortable I know that I'm not expanding fast enough.
A lot will be determined by the geography of the map and we fortunately have a scout to help bring that info in asap (I think we need to run him as far as possible before he gets eaten by a bear, that will give us a good estimate of the size of our continent, detailed scouting can probably come later).
Right now my gut is telling me that crashing the economy will be tougher to do given our enhanced windmills and watermills compared to a normal game, so we should be as aggressive in our expansion as humanly possible.
I think it's safe to assume that all teams will be using a variation of that same strategy, the first split in the teams performance is going to come from being able to create cash out of nothing early enough that the slider can be kept running high.
The second split will come in the worker micro, and making sure that there are enough workers (there can never be enough!)


Generating cash: partially completed wonders (not as effective as with an industrious civ but still worthwhile if pre-currency, under organised religion or possessing a multiplyer resource)
Whip overflow: Not sure how effective this is on quick speed, I'm hoping that someone in this team is a master of the whip. My usual strategy is to put as many hammers as possible into a 30 hammer marathon warrior and then whip (in emergencies can chop a tree into it as well). The optimal strategy for quick will need some work
Trademission: possible relatively early with GLH and a philo leader
Razing/Capturing cities: self explanatory

If we're rexing fast enough we should need some or all of these strategies in order to stay afloat.
I'll play Shannons new test save and see how far I can push it, though we shouldn't read too much into those results.


additional comments: If we only have access to two civs at the start then it is imperative that we do not destroy either of them too soon. They have to be milked of techs through trade first. Those first few round of tech trading can make or break the game. The only exception to this is if one of the two is Mansa, as he'll trade us his monopoly techs without needing to be friendly.
For this reason we need to know who/what is out there asap as it will determine everything we do later.
I'm not normally a scout builder, but I also normally know who I'll be playing against and what the map will be like in advance.

 

[PaulisKhan]

Actually golden ages are more powerful than usual in this game for two reasons:
1) Golden age length is 6 vs 8 on normal speed but 6 turns on quick corresponds to 9 turns on normal.
2) With many windmills, watermills and some forest preserves we will have more tiles with both an extra hammer and an extra commerce during a golden age.

Definitely, absolutely, 100%, no debate!

Golden ages are king and even more so on quick speed. MoM is a must have, Taj is often a nice way to get the GAge session in play.
Maybe we need to start a list of GP's

1GE for corp
1GM for corp (possibly/probably)
2GS for academy (minimum)
1GS for edu bulb (possibly)
1Proph for shrine (probably)

4 GAges from GP's = 10 additional GP's.

16 GP's total as our base minimum
1 from music (probably worth the detour?)
1 from economics (sometimes it is acceptable to lose this one in favour of beelining railroad and then backfilling the economic branch with trades, but I'm not sure how well the AI will be teching compared to us, very map dependent)
1 from physics
1 from fusion

so we need to generate 12GP minimum from the pool. That should be relatively easy with a philo leader right? (I'm weak on GP generation). The real trick is going to be mixing them up enough to run the golden ages.
A 6th golden age could be on the cards, alternatively that's 5 scientists that could be bulbed early/academised.

 

[PaulisKhan]

I know I'm easily excitable, but this is so much awesome.

Sounds like you have been busy . Do you have any estimate on the total game length? It may be of use when calculating the interest we get on our investments.

I have no real experience with quick speed and the additional starting techs (along with any other surprises) will have a dramatic impact on the game.
If you wanted me to make a best initial guess that could be recursively refined as the game progresses then I'd start with a date of 1400AD (I have no idea what turn that is on quick).
That is a best guess though, not a calculated prediction. I'm guessing that the quick speed and distribution of civs will be more of a hindrance than the starting techs will be gain.

 

[Frederiksberg]

I think in that game I pushed the expansion too far (or kept one too many cities) and it probably cost me 60/650 turns (9% of the game length).
Knowing when to stop will be crucial in this game
Later on we'll have dozens of ways of generating gold to keep our economy afloat so getting "stuck" is much less problematic.

I think I'm starting to get the picture. So generating cash in the early game is important to enable both a fast expansion and a reasonable tech rate. Postponing "multiplier techs" like Writing and Metal Casting can be expensive.

Right now my gut is telling me that crashing the economy will be tougher to do given our enhanced windmills and watermills compared to a normal game, so we should be as aggressive in our expansion as humanly possible.

Yes, the availability of windmills and watermills should tip the scales in favor of faster expansion

I think it's safe to assume that all teams will be using a variation of that same strategy, the first split in the teams performance is going to come from being able to create cash out of nothing early enough that the slider can be kept running high.
The second split will come in the worker micro, and making sure that there are enough workers (there can never be enough!)

I remember CP (Cactus Pete) saying that as a rule of thumb you need at least one more worker than the number of cities you have. Settlers should always be accompanied by a worker so that tile improvement can start immediately in a new city.

Generating cash: partially completed wonders (not as effective as with an industrious civ but still worthwhile if pre-currency, under organised religion or possessing a multiplyer resource)
Whip overflow: Not sure how effective this is on quick speed, I'm hoping that someone in this team is a master of the whip. My usual strategy is to put as many hammers as possible into a 30 hammer marathon warrior and then whip (in emergencies can chop a tree into it as well). The optimal strategy for quick will need some work
Trademission: possible relatively early with GLH and a philo leader
Razing/Capturing cities: self explanatory

We have the option of adopting Police State and get a multiplier on military units. So whipping warriors is a way to generate gold, I think. As soon as we have CoL we can generate GM's for trade missions. Is this more important than generating GS's for Academies? I think city razing is disabled so we have to be careful about which cities we capture - don't want crappy cities that count towards Domination.

EDIT: When Currency is discovered I guess that building gold is the easy answer to keeping the science slider high.

 

[PaulisKhan]

yep to the edit, though partially building wonders can prove to be more efficient than building cash directly when they have additonal build multipliers, you just have to put up with not getting the cash until the AI complete the wonder themselves.
I'm pretty sure you can also partially build the same wonder in multiple cities if gets to 1 turn from finished and the AI still hasn't completed it. This could prove particularly effective if we're building something like a marble/stone wonder under organised religion.
I'll have to double check this in game though if nobody knows for sure.
Also, Murky Waters in the last SGOTM showed that the HE city whipping workboats for cash overflow generated more gpt than building wealth (the Heroic Epic bonus applies to workboats too, as would police state)

That mention of police state bought another point to mind.
Christo Redento is an extremely powerful wonder whos only detriment is that it arrives so late. In this game we unlock the required tech (radio?) with our electricity slingshot.
Now obviously we don't want to go for it immediately after electricity, due to the massive cost, but I think that it would prove beneficial to keep the fact that it is available constantly on the boil.
ie. how quickly can we reasonably bring this online?
Doing so opens up all sorts of efficient whipping/cash buying infrastructure/GP rushing/Drafting massive XP unit rounds etc etc.
I'll try and fit in some test games tomorrow to see how many cities we want in order to tech it in a reasonable time.
It really is a game changing wonder (eliminates the 5 turn delay between civic changes along with the anarchy, although can still only change once per turn). Utilising it properly is going to add another degree of separation between the best teams.
I apologise if it sounds like I'm harping on about so called "game changing opportunities" I just feel strongly that we have a great shot of taking down the two russian teams, despite them having the two strongest space game players and the way to do it is by making sure that we take advantage of every non standard feature of this scenario.

I remember CP (Cactus Pete) saying that as a rule of thumb you need at least one more worker than the number of cities you have. Settlers should always be accompanied by a worker so that tile improvement can start immediately in a new city.

It's also worthwhile roading to any new city sites in advance if we have workers to spare. Having 1-2 traderoutes instantly connected can go a reasonable way to mitigating the initial maintainance cost.
I think with our high potential happy caps, the amount of work our workers will have to do, and the disproportionate movement cost on quick speed, we'll want even more workers than that. There's also a lot of pre-workshopping/mining that will need to be done in preparation for the endgame component build.
We'll want to be irrigating many tiles in all of our new cities, growing them to the happy cap asap while the irrigated tiles all get pre-improved over the irrigation for when we no longer need to grow and want to convert them to hammer/commerce tiles.
These last things are the sorts of micro that I struggle with in solo games, but with short turnsets and frequent discussions I'm hoping it won't be a problem for me here =)

 

[Frederiksberg]

Radio is extremely expensive as far as I remember. I suppose that the benefits of frequent civic changes are not that big in the early game when only a few civics are available. Radio can be partially bulb'ed using GA's:

Great Artist:
Literature
Drama
Music
Polytheism
Monarchy
Mass Media
Radio

 

[ShannonCT]

Cristo would indeed be a powerful wonder for the strategic options it would allow. We could get Radio with Liberalism fairly early. It might also be nice to have an extra GE to partially rush Cristo.

On the expansion discussion, I'm finding that it's not possible to overexpand. The mills are generating too much commerce for maintenance to be more than a nuisance. The bigger question I think is HOW to expand. Usually expanding through conquest is more productive than expanding through settlers, but the quick speed and the early appearance of Xbows and airships, coupled with the quick speed is making conquest very inefficient. An early war is going to have to target an AI with low unit building rate and/or no iron. Alternatively, we can go for unrestrained REX with a rush to Rifling for paratroopers.

 

[Frederiksberg]

Cristo would indeed be a powerful wonder for the strategic options it would allow. We could get Radio with Liberalism fairly early. It might also be nice to have an extra GE to partially rush Cristo.

The Liberalism slingshot to Radio sounds attractive, Radio is more than 4 times as expensive as Liberalism so we get a lot of free beakers. Rifling can probably be self researched in reasonable time so the most obvious alternative would be a slingshot to Assembly Line.

On the expansion discussion, I'm finding that it's not possible to overexpand. The mills are generating too much commerce for maintenance to be more than a nuisance. The bigger question I think is HOW to expand. Usually expanding through conquest is more productive than expanding through settlers, but the quick speed and the early appearance of Xbows and airships, coupled with the quick speed is making conquest very inefficient. An early war is going to have to target an AI with low unit building rate and/or no iron. Alternatively, we can go for unrestrained REX with a rush to Rifling for paratroopers.

The trade off between settling and capturing will to some extent depend on the map. If there is lot's of available land and particularly some juicy spots with rivers and hills settling many cities could be favorable. Do you think we can apply our old tactic of harassing a neighbour civ preventing it from improving the land and resources for a while until we are ready to launch a full scale attack? If the AI is denied the use of improved tiles it should become seriously weakened.

 

[Frederiksberg]

Also, Murky Waters in the last SGOTM showed that the HE city whipping workboats for cash overflow generated more gpt than building wealth (the Heroic Epic bonus applies to workboats too, as would police state)

We did something similar - only we had so high production in the HE city that no whipping was necessary to generate lot's of cash. I believe it was SCT who suggested it.

Christo Redento is an extremely powerful wonder whos only detriment is that it arrives so late. In this game we unlock the required tech (radio?) with our electricity slingshot.

Doing so opens up all sorts of efficient whipping/cash buying infrastructure/GP rushing/Drafting massive XP unit rounds etc etc.

Does a hammer still cost 3 gold in BtS? And if so, aren't the benefits of cash rushing limited compared to pop rushing? GP rushing sound interesting. I suppose it has something to do with running massive numbers of specialists for just one turn in Pacifism/Representation...

 

[PaulisKhan]

Does a hammer still cost 3 gold in BtS? And if so, aren't the benefits of cash rushing limited compared to pop rushing? GP rushing sound interesting. I suppose it has something to do with running massive numbers of specialists for just one turn in Pacifism/Representation...

If we gain christo redento (and Kremlin becomes a very serious option worth going for then) then all sorts of rushing opportunities will become available to us and there will be times when we want a building "now" but don't want to sacrifice the pop to do it.
I'm thinking specifically of cities that are busy maturing cottages but could do with some infrastructure in place to meet minimum building requirements for national wonders etc. Universities and Banks are an example of this. I understand whipping to be very inefficient in high pop cities too.
I can easily be convinced otherwise with a healthy dose of mathematics though.

GP rushing is pretty much exactly what you described, desperate for cash? swap to castes and run merchants in the GP farm at starvation levels for a turn or two under pacifism, a golden age with a philo leader is a great time to do it and the removal of the 5 turn standdown period between civic switches makes Christo powerful even during a golden age.


This raises a question I've never thought to ask before. Is a city limited to 1 pop starvation per turn or can it actually starve more than one in a single turn?
A 20 pop GP farm could generate a lot of GP points for free by running at massive starvation levels for a couple of turns and only lose a couple of pop doing it.
I'm really weak on GP theory though, so it could be a silly question.

-edit- at what point do we start discussing specifics of the first turnset and stop discussing general strategies.
I understand the need to move the scout first before even talking talking about settling. I admit I'm fumbling around in the dark a bit still.


-2nd edit rather than new post- In terms of warring I've always been a fan of spy assisted flanking promoted HA/Cavalry blitzes rather than siege warfare (even on marathon speed). With a good medic unit, immunity to first strikes, strong withdrawal chance and the ability to spread forces between attacks, it's virtually uncounterable. I expect this could be particularly useful on quick speed.
It generally suffers a few more casualties than a methodical siege grind but it's also over far quicker and so war weariness doesn't tend to stick around for long. What is the typical war strategy for this team?

 

[ShannonCT]

The trade off between settling and capturing will to some extent depend on the map. If there is lot's of available land and particularly some juicy spots with rivers and hills settling many cities could be favorable. Do you think we can apply our old tactic of harassing a neighbour civ preventing it from improving the land and resources for a while until we are ready to launch a full scale attack? If the AI is denied the use of improved tiles it should become seriously weakened.

It will be marginally harder since it will take longer to get units in position. But in theory it should still work. If we want to try a phalanx or mace rush, we really want to prevent the target from hooking iron.

 

[ShannonCT]

@PK

Cities can only starve 1 pop per turn, so GP rushing can work.

We can start talking about opening sequences. I've been starting with worker and agriculture, mining, and BW, building warrior to pop2, then second worker to build 2 windmills. Revolt to PS, slavery, and tree-hugging simultaneously. Build settler at pop4 or pop5.

Attack speed will be critical in this game. On quick speed, if your attack is plodding, the AI is going to spam large numbers. Any fast unit with spies and airships sounds like a winning combo. Have you tried paratroopers? We should be able to get them in the early ADs, faster than Cuirs. Horse Archers could come earlier but how many EPs will we have then and can we really expect to keep the AI from hooking any metal?

 

[PaulisKhan]

Of course you are right, I forgot about paratroopers completely. The AI have nothing capable of combating them, there's no way that they will get to rifling in time. Obviously beats the cavalry strategy. Airship the top couple of defenders to minimise required healing time and we can blitz each civ in a couple of turns.
Maybe we should focus on peaceful rex to start then (with possibility of choke on nearest neighbours if we have multiple trading partners around). I don't know what our conquest hammer efficiency will be like vs settler spam on quick speed.

Re start: Do you think it's safe to build the warrior after the revolt? I say this because with anything more than 4 hpt production we get a free hpt on the warrior by building it under police state. Of course then we'd have to dump hammers into something like the barracks while we were trying to grow but I very much suspect we'll want one there anyway.
Alternatively, could we use a second scout more than a warrior?
Anyway, I think I'll devote tonight to bunch of different starting strategies, I guess my HoF is going to be put on the back burner for a while =). In particular I want to compare the "grow to pop 2 while teching to bronzeworking and whip worker under slavery, vs the "build worker first"


I guess I'll experiment on your updated save

 

[PaulisKhan]

component dependent travel time for all game speeds.
http://forums.civfanatics.com/showpost.php?p=7905284&postcount=10

Not entirely useful at the present but I'll never find it again 3 months from now.

 

[Cactus Pete]

Intuitive master of controlled Domination checking in . . . tardy (been completing leif's game -- and not so masterfully -- during limited time available).

Productive discussion. Particularly pleased to note the thoughtful contributions of our new team members. I'm going to learn even more than usual this game.

I'll try my hand a SCT's latest save tomorrow. Couple questions:

1. My immediate inclination is to settle on the northern corn, which would allow either a settler and/or a workboat to be built in (guessing, never tried a Quick game) about 8 turns. I read no mention of this approach in our thread. Has it been tried and rejected, summarily dismissed as obviously sub-optimal due to long-term limitations on the capital's contributions (lots of sea tiles), or is it worth exploring?

2. What effects does the universal knowledge of advanced technology have on the kind of barb units that appear; that is, how advanced, when?

 

[ShannonCT]

1. My immediate inclination is to settle on the northern corn, which would allow either a settler and/or a workboat to be built in (guessing, never tried a Quick game) about 8 turns. I read no mention of this approach in our thread. Has it been tried and rejected, summarily dismissed as obviously sub-optimal due to long-term limitations on the capital's contributions (lots of sea tiles), or is it worth exploring?

Please try this and report back. All ideas should be given a chance at this point.

2. What effects does the universal knowledge of advanced technology have on the kind of barb units that appear; that is, how advanced, when?

I'm not seeing any difference. Land is filling up pretty quickly in the practice game, so the barb threat has been minimal. But then Gyathaar says the barbs will be a bit meaner. Maybe he's given them some other military techs?

 

[Cactus Pete]

Please try this and report back. All ideas should be given a chance at this point. Will do.

But then Gyathaar says the barbs will be a bit meaner. Maybe he's given them some other military techs?

If so, this may make it important to hook up an early military resource.