Master builder poll

[vondrack]

Originally posted by delmar
Scouting is also very important. For this reason I usually build a few warriors while the worker finishes the tile improvements that are necessery for the aforementioned optimized setup. If I feel that the warriors produced during this period are not sufficient for scouting, then I build further warriors in the second city. This is often not necessary because I find sea in one or more directions, limiting the scope of scouting by warrior. In any case, I wouldn't delay the start of the escort/settler production much after the optimized setup is ready for the sake of building scouts, as having more than 4 units per city for a prolonged period indirectly means that we can't keep our research set to 100%, which I consider an extremely important thing this early in the game.

Good point about the upkeep cost. Two scouting warriors are the minimum, three a convenience, four an overkill.

Originally posted by delmar
Some voiced their opinion that a strategy this early is useless because it will change anyway. I don't agree with this statement, as I can hardly imagine any situation where I wouldn't want to build settlers for a long time and obviously the settler can't go alone so there must be an escort as well.

This is a misunderstanding. What I and perhaps others were trying to say was that elaborating and adopting a detailed strategy prior to even climbing that hill east of Legopolis is premature. Discussing scenarios that will turn to be void as soon as we move Howard (our 1st Warrior) two tiles away from Legopolis is a waste of time.

Originally posted by delmar
I realize that one of the major issues at hand is what that certain "escort" should be, and that it seems to depend on many yet unknown things. For example whether we are alone or not. My answer to this question would be that we should shoot for a strategy that will work in most cases. Specifically I wouldn't dare to count on being alone, therefore I would not even consider a strategy that needs changing if we meet another civ.

Going for a strategy that works in most cases may easily end up using a strategy suboptimal for a given situation. That's why I feel that gaining at least minimal knowledge of our surroundings is needed before adopting a long-term strategy.

Originally posted by delmar
I realize that one of the major issues at hand is I don't know whether the 2-warrior escort is good enough for the case where we are not alone. I never build many warriors, only 3-4 for exploring and suppressing early unrest in the capital, for the simple reason that their upgradeability sucks. An exception to this rule would be if I wanted to have a bunch of swordsmen, which I usually don't. Therefore I am tempted to say that the 2-warrior version is inferior.

Agreed, I also lean toward using 1 Merc instead of two warriors. Upkeep and production costs are a factor.

Originally posted by delmar
At the same time I can see the adventage of having 2 warriors in each city permanently, for supressing unrest and for basic protection. They might be also useful, after upgraded to swordsmen, to take out enemy units before they can attack.

We will not need that many units for military police, as the game is on Regent (only the fourth citizen will cause an ungarrisoned city go into unrest). As for swordsmen, I would strongly suggest using mobile forces (chariots and later horses) instead of slow moving swordsmen. Swordsmen are unable to withdraw after a victorious battle even if on a road. Horsemen are able to both retreat and withdraw. Of course, this requires horses...

Originally posted by delmar
Yes, I noticed that city spacing thing, too. I am a bit surprised that some folks are on one hand afraid that someone will steamroll us before we can say "Numidian Mercenary" and on the other hand are worried about not being able to use all the 20 tiles around a city. Well, all I can say is that we will have what the majority wants...

City spacing is a matter of personal preference and mostly a heart issue. I guess nobody will disagree with that two tile spacing is perhaps the most effective way of using land early in the game. OTOH, most people would agree that being aggressive is better that being passive... and yet, we are 99% sure we are not going to wage agressive wars, unless absolutely necessary. We are builders and those tend to focus on the future greatness, if possible. Cluttering our empire with cities that are able to use half their tiles only just does not feel well (even if it is a sound strategy).

Besides, I do not consider the ability to reinforce in one turn that important. Two turns is most often enough. You just have to plan ahead and as soon as there is a threat, move reserve troops into a 1-turn reach of the possibly threatened site. Completely surprising attacks are rather rare, as cities can "see" their full two tile radius all the time.

Originally posted by Tiberius
PS You should find somebody to update the “Guidelines for the President/VP” thread after dec. 20.

I will do that, as I will be available and actually having more time than now.

[Tiberius]

Without running scenarios that would calculate exact shield production and such, so blindly I'll do it this way:
- 3 warriors
- settler
- 2 warriors
- settler
- Mercenary (or maybe barracks-mercenary)

The reason for this is that until we build our second settler, it is very unlikely that anybody would attack us. By the time our second settler will have been built, the explorations should give us the answer: are we or are we not alone and if not, we should start building mercenaries. One warrior for each settler ought to be enough, while the other 2 would do the exploration (1 defending in Legopolis)

[redstar1]

Sounds good. Need those warriors for scouting badly.

[quantum_mechani]

In answer to why I think plains are equal to grassland:

-After despotism is gone the extra food allows us to to mine juicy mountain and hill resources, as well as producing settlers faster if there is still room to expand.

-If there is no fresh water for irrigation then plains are much worse than grasslands.

-Grassland often has bonus squares mixed in.

[delmar]

For some reason (perhaps because of Vondrack's comments) I used to think that I am one of those careless people who usually get slaughtered in the 10th turn of the MP games. Now, however, I am very surprised about these plans, and I want to tell you that I am typically much more careful. I don't know if it is right or wrong, I just want to let you know that it never occured to me that perhaps I should be more opportunisitic and less careful and yet it appears now that the majority of the team is.

In particular, I don't like this plan:

Originally posted by Tiberius
Without running scenarios that would calculate exact shield production and such, so blindly I'll do it this way:
- 3 warriors
- settler
- 2 warriors
- settler
- Mercenary (or maybe barracks-mercenary)

because I think that the first 2 warriors should go out scouting (as everyone is advocating the importance of scouting, I also thought that this is a rather conservative approach and I was preparing some arguments why we shouldn't send 3-4 warriors scouting), and if we send the first 2 warriors scouting, then by the time we have the first settler we will have only 1 warrior to defend it and the capital.

I mentioned what I though would be a brave approach on another thread already, but I think this is a better thread to discuss this so here it is again:

- 2 warriors (scouting)
- 1 unit (warrior or merc) defending the capital
- 1 merc or 2 warrior for escort
- 1 settler
- repeat the last 2

Unfortunately I don't have time now for a detailed analysis, so just take it for what it's worth. I am not saying this is an absolute must, only that it is safer than what we seem to be heading towards and yet I wasn't sure if it was safe enough.

[delmar]

We will not need that many units for military police, as the game is on Regent (only the fourth citizen will cause an ungarrisoned city go into unrest).

I think this statement is either inaccurate or incomplete. I believe that on Regent difficulty, there are 2 citizens born content. The 3rd will revolt, unless we have a temple, a luxury resource, a garrison, or some gold spent on happiness. I would like to request that someone else double check this because this might be very important in the near future.

As for swordsmen, I would strongly suggest using mobile forces (chariots and later horses) instead of slow moving swordsmen.

My personal preference is for horses as well, not only because they can withdraw but also because they have a great future (in terms of upgrading). Note however that it takes a whole lot more research to get horses than to get swordsmen, and the swordsman's attack value is 3. Consequently, the swordsmen option seems safer in the short term. Btw, did we already start a discussion what to research next? I think it should be either The Wheel or Iron Working to enable us to grab the resource (or realize that we don't have any and plan accordingly).

City spacing is a matter of personal preference and mostly a heart issue.

This sounds like "it doesn't really matter". I am guessing that you mean "it is just not worth to have a great strategy if it looks ugly", and I might even agree on this. Nonetheless there are options that might be significantly different and that we can consider. Once I read an article on CivFanatics that showed that the so called Infinite City Sprawl technique (not sure about the exact name, the essence is that you build as close as you can) can be extremly effective. I never bothered to read the details because the idea of building cities on top of each other just simply doesn't match my taste. But since then, I do opt for closer spacing if I have a shed of doubt where to put the city.

One very imortant aspect of this is corruption. You might think it doesn't bite you while you have only 2-3 cities, but the truth is that you might need to improve an extra tile and keep an extra base population (with increased potential for disorder, etc, etc) to get to an optimal net shield output if you place the city one tile further away, because corruption kicks in one shield earlier than it would have one tile closer to the capital. (Tsk, tsk: good topic for modelling in a scenario! ).

Besides, I do not consider the ability to reinforce in one turn that important.

One turn is not important. Being close is, because you might need to send troops from the opposite side of the country, not only from the direct neighborhood.

[vondrack]

Originally posted by delmar
I think this statement is either inaccurate or incomplete. I believe that on Regent difficulty, there are 2 citizens born content. The 3rd will revolt, unless we have a temple, a luxury resource, a garrison, or some gold spent on happiness.

My statement was completely inaccurate.
What you say is correct. The third citizen throws the city into disorder. Fortunately, there are dyes nearby Legopolis...

Originally posted by delmar
This sounds like "it doesn't really matter". I am guessing that you mean "it is just not worth to have a great strategy if it looks ugly", and I might even agree on this.

delmar, seriously, you can be real good in understanding what people mean, just keep trying hard... Yes, what I meant was that even though I (and perhaps many other builders) know that ICS is a very effective strategy, it produces a VERY ugly empire.

Originally posted by delmar
Nonetheless there are options that might be significantly different and that we can consider. Once I read an article on CivFanatics that showed that the so called Infinite City Sprawl technique (not sure about the exact name, the essence is that you build as close as you can) can be extremly effective. I never bothered to read the details because the idea of building cities on top of each other just simply doesn't match my taste. But since then, I do opt for closer spacing if I have a shed of doubt where to put the city.

Should other teams find out we utilize ICS, they would probably gang up on us. AFAIK (never been into MP before, but read quite a lot about it), ICS is something MP players try to prevent others from using, even if it takes to cooperate (exactly because it is so effective).

Originally posted by delmar
One turn is not important. Being close is, because you might need to send troops from the opposite side of the country, not only from the direct neighborhood.

True, but you reinforce the city under attack from the closest neighbouring city and in turn reinforce that closest neghbouring city from another city that is... you get the idea. Units seldom cross the whole empire, usually it works like the domino effect: A is under attack, B reinforces A, C reinforces B etc.

[delmar]

Some details about a possible warrior-warrior-warrior-(merc-settler)*repeat strategy:

3,750 BC    bw 10      warrior 5        

3,700 BC    bw 9       warrior 3    worker moves to bonus grassland    

3,650 BC    bw 7       warrior 1    worker starts road               Legopolis grows to 2

3,600 BC    bw 6       warrior 3    worker finishes road    

3,550 BC    bw 4       warrior 2    worker starts mine    

3,500 BC    bw 3       warrior 1        

3,450 BC    bw 2 80%   barracks 10  worker finishes mine    

3,400 BC    bw 1       barracks 7   worker moves to S grassland    

3,350 BC    wh 14 100% switch to Merc 4    worker starts road    

3,300 BC    wh 12      Merc 3       worker finishes road            Legopolis grows to 3

3,250 BC    wh 9       Merc 2       worker moves to SE grassland    

3,200 BC    wh 8       Merc 1       worker starts road        

3,150 BC    wh 7       Settler 6    worker finishes road        

3,100 BC    wh 6       Settler 5    

--- from this point on, worker doesn't deal with capital ---

--- builds road towards next city site                   ---

--- at this point, we have 5 units so we are losing 1 gpt   ---

3,050 BC    wh 5       Settler 4            

3,000 BC    wh 4       Settler 3            

2,950 BC    wh 3       Settler 2            

2,900 BC    wh 2 90%   Settler 1                                     Legopolis grows to 4

--- at this point, we are breaking even ---

2,850 BC    wheel 1    Merc 6                                      Legopolis back to 2

At this point, we are losing 1 gpt. We have about 3-4 turns to build a new city if we want to keep science at maximum.

From this point on, we can repeat the Merc/Settler combo with small adjustment: population grows every 7 turns, Merc/Settler ready every 6, so we will have to insert an extra Merc or something from time to time.

Advantages: worker spends minimal time on improvements around capital, good research rate as unit upkeep is minimal, moderate security (1 unit per city guaranteed all the time), moderate scouting (two warrior full time).

I leave the disadvantages as an exercise to the reader. One issue I am sure will raise some eyebrows is that this plan doesn't include mining the grasslands (except the bonus one). I think this is an advantage for the time being but maybe not so.

[ZargonX]

Tiberius: I am sorry to hear that you are withdrawing from the race. I am sure you would have made a great Master Builder! There are always future elections, right? Also, I appreicate you throwing your support behind me; I hope I can live up to it!

[delmar]

ZargonX, do you plan to use this thread for further discussion or (since you are on the roll anyway ) will you open a new thread for Master Builder discussions as well? I think it would be good.

[Tiberius]

The red X is my idea about where should we put our 2nd city. It would be a coastal city, with 3 bonus grasslands, 2 hills and luxuries. There are chances that we can build the Colossus. Also it'd have only 2 overlapping tiles with Legopolis (outlined with blue in the picture).

Of course by the time the settler will be ready, we might decide for a better position (maybe one tile to the south or south-east from the red X), but right now it seems the best position for me.

[Tiberius]

About the build order issue:
- delmar, you seem to assume that if we are in the vicinity of one of the other teams, they have built already 3 warriors and are heading in our direction with all 3 of them! Yes, it'll be only one warrior in the capitol for a few turns, but the other warriors won't be at the other end of the world! Howard will scout the easter sorroundings and then protect our settler and then the new city and scout southward only after that. The 2nd warrior will scout probably west and we will know is somebody is approaching from there in time.

If we want to win this game, we will have to take some risks from time to time. This one seems to me a not so big, yet profitable risk. Thus I'm against waiting 6 more turns before we build the settler.

What should we build after the first settler depends very much on what could the scouters discover: should we find out that we have neighbors, we'll build a merc; are we alone: warrior. I've run a simulation myself and the difference between building 2 warriors + 1 settler and one merc + 1 settler is 2 turns. (after the warrior-warrior-warrior-settler sequence).

PS Please stop calling our capitol Legopolice. It is Legopolis.

[Spiffor]

During the informal chat that took place last light, we thought of replacing warrior #3 by a merc. That would make our settler ready in 3000 BC.
For the record, if we produce warrior #3 instead, our settler will be ready by 3150 BC.
If we produce a settler immediately after warrior #2, our settler will be ready by 3250 BC.

It is a matter of personal preference, but I think we should found city#2 ASAP : we found a suitable city site, we should use it almost immediately (ie in 400 years )
Having our settler ready by 3000 BC will mean we're ready in the same time than anybody else. But we, unlike the others, have a pop. boost because of our cattle, and I think we should abuse from it shamelessly.
Imagine, in 3000 BC, we could have 2 cities in great locations, and incense in our borders ; while the others will have their capitol back to 1, and a settler barely ready.

[Tiberius]

What do you mean by ASAP? After the 2nd warrior or after the 3rd warrior?

Why should we build a merc. when we don't even know if we have neightbors!

[redstar1]

I say we build a settler after warrior #2. That patch of jungle right at the end with a border on the water and access to the grassland and dyes looks a perfect spot....