City placement redux (for the third time) - Page 3

badams52 · February 19, 2003, 01:40

Demographic comparisons

Capt Dizle · February 19, 2003, 01:45

Ralphing looks more powerful than the 3 tile but is still far less efficient than my Holy 5 by 5. Do I have to explain this in more detail?

badams52 · February 19, 2003, 02:03

Quote:

Originally posted by Theseus

What makes this so powerful, IMO, beyond early production and later metros, is the *flexibility*.

And yes, I'd use this in MP all day long. If anything, early power is that much more of an issue.

I agree with Theseus, This is certainly something you would use in MP. Arguably, I could have had a larger military force in the 3-Tile game, but Ralphing supports a larger military and makes it much easier to have a larger miliatary. It also allows the "keeper" cities to focus on temples and libraries. I didn't notice strong corruption at the first ring of keeper cities, but they did produce improvements at a slightly slower rate. This least me to believe that Ralphing should only go to the second ring of keeper cities, and maybe not even that. I'm still trying to see what happens in Monarchy to see if you can still Ralph between the first and second keeper city rings.

One other interesing aspect was how I was able to get 14 cities when Ralphing in the same time I got 9 from 3-Tiling. I think for the most part it's because the camp cities only build barracks, military, workers and settlers whereas in the 3-Tiling, I was more likely to switch to a temple or library as needed.

BRC · February 19, 2003, 02:08

I'd say that they are pretty close. Still, the fact that Ralphing is more flexible later on makes me believe that I will be using this more often. You guys are amazing.

badams52 · February 19, 2003, 02:09

Quote:

Originally posted by jimmytrick
Ralphing looks more powerful than the 3 tile but is still far less efficient than my Holy 5 by 5. Do I have to explain this in more detail?

jt, an image would be helpful.

Capt Dizle · February 19, 2003, 02:18

Quote:

Originally posted by badams52

jt, an image would be helpful.

When I have time I will do more than provide an image, I will provide a full "proof" supporting my claim that the Holy 5 by 5 is paradigm.

Harovan · February 19, 2003, 04:13

Ralphing?

Dominae: I don't claim the credit to have invented something new. Indeed, the idea of camps has been around forever, especially in games on higher difficulties, where the chances to outexpand the AIs gets less. The credit for the camps goes to Aeson and Vel, what concerns Civ3, and I'm sure in the earlier Civ games were similar strategies possible (haven't been on Apolyton yet at this time).

The pattern I published was merely a try to create a combination of both worlds, i.e. to get the benefits out of an ICS-like build in the early game with the benefits of "OCS" (bad word) with gigantic cities and a reduced city number corruption in the late game after military expansion. Thus, it was just my proposal to answer Theseus' question in the first post of this thread.

Theseus · February 19, 2003, 11:16

"If I see further, it is because I stand on the shoulders of giants," wrote Isaac Newton to Robert Hooke.

There is no question that temporary camps, as a concept, have been around for a long time.

But I posed a question, and Sir Ralph answered it... 2/3/5 city spacing, or Ralphing, is f-cking fabulous.

Almost (but not quite) as good as mixed-unit Armies.

nbarclay · February 19, 2003, 12:02

I just had an interesting thought: why not make some experimentation with city spacing a part of Apolyton University? We could either have different people volunteer to try specific city patterns in AU 206 (and encourage replays with alternative city spacings to see how they change things) or have a series of two or three AU games each focusing on a specific city build pattern. A maximum-land pangea map with the player starting a bit away from the coast would probably be ideal for this type of experimentation.

Nathan

nbarclay · February 19, 2003, 12:39

By the way, there are two other advantages to not planning on having most cities get much if any past size 12. (1) Sanitation and hospitals can be lower priorities. (2) Happiness is less of an issue, making happiness buildings and wonders less important and reducing the impact if warfare temporarily causes a loss of access to luxuries.

Nathan

Dominae · February 19, 2003, 14:00

Quote:

Originally posted by nbarclay
A maximum-land pangea map with the player starting a bit away from the coast would probably be ideal for this type of experimentation.

There's a very valid point put forward by a few posters on this thread (alexman and myself come to mind) that city spacing is primarily influnced by the available terrain (the "lay of the land" as I put it). Clearly on a big Pangea map anything that resembles ICS will be more powerful than any other strategy. Well all know ICS is the best. This is why badams' test, although a commendable effort, is heavily biased: the underlying terrain in AU105/205 is just "too good".

Again, I think the temporary camp idea is good, but it's use and exact placement of cities depends a lot of the terrain available. That's why I'm not sure this revolution is all that revolutionary (sorry for being the spoilsport).

What would be required to show me that a particular pattern is better than others is many successful games across multiple map types.

Dominae

Dominae · February 19, 2003, 14:04

Quote:

Originally posted by nbarclay
By the way, there are two other advantages to not planning on having most cities get much if any past size 12. (1) Sanitation and hospitals can be lower priorities. (2) Happiness is less of an issue, making happiness buildings and wonders less important and reducing the impact if warfare temporarily causes a loss of access to luxuries.

I completely agree. I took the opposite approach and asked what advantages are there to having cities well beyond size 12. The answers that I got were: Tanks in 2 turns, less micromanagement, and faster SS components. The first two I consider a matter of personal preference. The third is valid in the 2% of games where a space race is actually a race. Anyone care to offer any other advantages?

Dominae

Capt Dizle · February 19, 2003, 14:20

Who has proved that ICS is the best in standard multiplayer?

BRC · February 19, 2003, 14:20

Quote:

Originally posted by Dominae

I completely agree. I took the opposite approach and asked what advantages are there to having cities well beyond size 12. The answers that I got were: Tanks in 2 turns, less micromanagement, and faster SS components. The first two I consider a matter of personal preference. The third is valid in the 2% of games where a space race is actually a race. Anyone care to offer any other advantages?

Dominae

Wonders also (I know, similiar to SS parts). I haven't done tests, but it seems logical to me that you would rather have the least number of cities working the most tiles, for the purpose of less corruption. True, there is the opportunity to waste more shields when completing a project, but commerce is never wasted. You'll also be paying less upkeep on buildings if you have less cities, more than making up for the upkeep on hospitals in these metropolises. The defensive bonus is minor, but who knows when you might have to fend off an attack from the AI.

I, personally, also gain benefit from having less to worry about in my empire. Checking cities is much easier when I have less. This may actually encourage me to do it more, thereby increasing the payoff. I know that it's a stupid reason, but it is a realistic one, and one I need to consider. You know that I played with the wide spacing earlier. In my current game, I just don't feel like my empire is as powerful as it used to be. I know that the tight spacing is what let me get to my current position (own the whole continent, 50% of world), but, at the middle of the Industrial Era, I think its starting to hold me back now.

nbarclay · February 19, 2003, 14:27

Quote:

Originally posted by Dominae

I completely agree. I took the opposite approach and asked what advantages are there to having cities well beyond size 12. The answers that I got were: Tanks in 2 turns, less micromanagement, and faster SS components. The first two I consider a matter of personal preference. The third is valid in the 2% of games where a space race is actually a race. Anyone care to offer any other advantages?

(1) A size 19 city and a size 18 city (39 tiles) provide essentially the same gross shields and gold as three size 12 cities using the same tiles, but with two thirds the building maintenance costs. (Of course the flip side to this is that you only have two thirds the cultural buildings.)

(2) Dividing the same land among fewer, larger cities results in less corruption due to number of cities.

In strictly economic terms, the ideal city placement would probably be designed around building fairly well-developed size 12 cities through most of the medieval era and then being able to disband them and use all the tiles elsewhere for bigger cities once hospitals become available. But I'm not willing to do it because the idea of disbanding a large, well-developed city strikes me as completely ludicrous. I've heard of ghost towns where everyone left when the mines or whatever it was that gave the town a purpose disappeared, but I've never heard of a ghost city where 800,000 plus people suddenly decided to leave. (I checked one of my saves, and 800,000 seems to be the pop size that goes with size 12, plus some extra for extra food stored up.) And besides, the extra culture from the extra cities makes up somewhat for the economic disadvantages.

Nathan

Konquest02 · February 19, 2003, 14:35

Quote:

Originally posted by nbarclay
I just had an interesting thought: why not make some experimentation with city spacing a part of Apolyton University? We could either have different people volunteer to try specific city patterns in AU 206 (and encourage replays with alternative city spacings to see how they change things) or have a series of two or three AU games each focusing on a specific city build pattern. A maximum-land pangea map with the player starting a bit away from the coast would probably be ideal for this type of experimentation.

Nathan

Quote:

Originally posted by Dominae
What would be required to show me that a particular pattern is better than others is many successful games across multiple map types.

This could be a nice topic for AU 206. We would then have multiple games (although all on the same map...) and we could experiment on city placement. I'm pretty sure all people will not "ralph" the same way so we would get to compare...

I may set this one up, so give me a call when you are ready! (if you want to!

)

--Kon--

Konquest02 · February 19, 2003, 14:38

I think ralphing disadvantages religious civs as your cities do not get as big and require less happiness buildings. Since cathedrals are cheaper for religious civs, they would be able to maintain huge cities happy more easily.

Dominae · February 19, 2003, 15:22

Re: big cities (size 14+).

Production-wise, I'm not convinced, as you can always create bigger cities my giving them more than their share of tiles (I actually do this). This is easily done by irrigating some other tiles in case there is a lack of Food. Rarely do you need to produce more than 3 SS parts at a time (less for Wonders), in which case it doe not really matter how productive the other cities are (it's their Commerce that matters at that point in the game).

Regarding upkeep costs, that is definitely a good point. Not enough to get me to disband some of my 3-spacing cities, but important nonetheless.

Dominae

Theseus · February 19, 2003, 15:28

Kon, isn't that backwards? The whole point IS to have larger cities and metros as you progress through the game.

The traits that get the biggest bang for the long-term buck will be REL and COM in terms of city performance, and IND in terms of empire performance.

Short term, MIL is rewarded, as this starts as an ICS strategy.

badams52 · February 19, 2003, 15:52

Quote:

Originally posted by Dominae

Again, I think the temporary camp idea is good, but it's use and exact placement of cities depends a lot of the terrain available. That's why I'm not sure this revolution is all that revolutionary (sorry for being the spoilsport).

What would be required to show me that a particular pattern is better than others is many successful games across multiple map types.

It's true that the land available makes a difference in what spacing you'll use. But I think most of the time I find myself on a continent with lots of space and little time to REX. Using OCN, if I'm lucky I'll get to start the 2nd Ring, but usually I get one ring and then have to decide who to declare war on if I want to expand. I used OCN my first time on AU105/201 and it's true I got a little more land then using the 3-Tile or camp systems, but I was also playing Regent at the time.

Maybe I'm wrong, but it seems the amount of land you settle before war factors more around how you keep the AI away from settleable land rather than the spacing of your settling.

I don't know about the rest of you, but once I'm done REXing, I just take the AIs cities, I don't bother to raze or disband them and waste my time resettling cities to fit my paradigm.

So when I play, the real question becomes how do I gain the most of my surrounding terrain before I run out of space. Ralphing seems to be a good option, but I was wondering if you could get away with having fewer camps and keep the early 3-Tile effect. I might more prefer a system more like this:

. . . . O . . . .
. . . . . . . . .
O . . . . . . . .
. . . C . . . . O
. . . . . . . . .
. . . . O . . . .
. . . . . . . C .
O . C . . . . . .
. . . . . . . . O
. . . . O . . . .

Each Core cite (O) is connected to the 6 in it's ring by 3 camp sites which get disbanded later. It's a little looser than 3-Tile, since only 3 cities instead of 4 are connected for quick defense but it's still flexible.

And one thing else I like about Ralphing when I tried it was the ability to fudge on camp sites. Though Ralph gave a concrete tessellation plan, we know from experience that given mountains and coast, you can't follow it exactly, and knowing that camp sites are temporary gave me the ability to move them one tile NW,SW,NE, or SE without hurting the grand idea behind the scheme of the placement.

edit: made layout easier to read

Jawa Jocky · February 19, 2003, 15:54

Quote:

Originally posted by BRC

Wonders also (I know, similiar to SS parts).

In my current game, I just don't feel like my empire is as powerful as it used to be. I know that the tight spacing is what let me get to my current position (own the whole continent, 50% of world), but, at the middle of the Industrial Era, I think its starting to hold me back now.

These are two problems I encounter in almost every game. This is why I want to drop a few cities in the industrial era.

Konquest02 · February 19, 2003, 16:00

Maybe I wasn't clear...

Goes back and reads himself
I definetly wan't clear... In fact, I got it backwards...

What I was trying to say is that the advantage a civ gets from ralphing in terms of easiness to manage cities is less important for religious civs than for everybody else...

Vs. 3-tile spacing, ralphing tends to yield bigger cities at the end of the line. A bigger city is harder to keep happy and in WLTKD. It is easier for a religious civ to maintain big cities. Therefore, it is easier to ralph with a religious civ than a non-religious one.

I see that I am contradicting myself, but I think I am right this time around.

--Kon--

Jawa Jocky · February 19, 2003, 16:00

Quote:

Originally posted by Dominae
Re: big cities (size 14+).

Rarely do you need to produce more than 3 SS parts at a time (less for Wonders)

Dominae

But here is why I don't need to drop cities.

Do my usual tight Palace core and use the AI's looser spacing for the FP core.

badams52 · February 19, 2003, 18:07

Quote:

Originally posted by BRC

Wonders also (I know, similiar to SS parts). I haven't done tests, but it seems logical to me that you would rather have the least number of cities working the most tiles, for the purpose of less corruption.

That reminds me, while Ralphing, I found it harder to build the early wonders. Less shields per city and the need to build more settlers used up more resources than 3-Tile.

Dominae · February 19, 2003, 18:24

Quote:

Originally posted by badams52
That reminds me, while Ralphing...Less shields per city and the need to build more settlers used up more resources than 3-Tile.

I believe this is critical. Some testing will decide the issue (in my mind, at least).

Dominae

vmxa1 · February 19, 2003, 19:54

Quote:

Originally posted by badams52

That reminds me, while Ralphing, I found it harder to build the early wonders. Less shields per city and the need to build more settlers used up more resources than 3-Tile.

This is what struck me. Burning up settlers early is not free. In a smaller map where the civs are going to be on me soon, I can see the benefits of being able to crank out troops, but if I am not going to be pressed early, will I be better off?
That chart that was put and every one said yes that shows 3 tile as weaker, did not show that to me.
I thought I saw the numbers as being very close and the 3 tile was going to out do the other from that point forward, as it will not be disbanding anything and many of the cities have structural improvemnts in place.

badams52 · February 19, 2003, 20:44

After playing both, ralphing seemed to me to be a great warmonger's strat, notice having more units while 3-tile would be better for the builder.

I was surprised that each ended up with the exact same income from cities. I was able to lower the science rate in 3-tile and keep the same science since I had one library in Thebes at the time, but the extra amount got eaten up with all the temple improvements (I love early temples with a religious civ).

BRC · February 19, 2003, 21:33

Quote:

Originally posted by vmxa1

This is what struck me. Burning up settlers early is not free. In a smaller map where the civs are going to be on me soon, I can see the benefits of being able to crank out troops, but if I am not going to be pressed early, will I be better off?
That chart that was put and every one said yes that shows 3 tile as weaker, did not show that to me.
I thought I saw the numbers as being very close and the 3 tile was going to out do the other from that point forward, as it will not be disbanding anything and many of the cities have structural improvemnts in place.

I got the impression that 3 tile would slightly pull away from Ralphing, but that when you hit the Industrial Era, Ralphing would come on stronger. The question becomes, Is it worth it??

However, with Ralphing, you are storing up population at the cost of shields and commerce. However, this population (assuming decent terrain) will grow faster which may translate into higher production and commerce later.

Theseus · February 19, 2003, 23:09

As experimentation takes place, we will start to see what does and does not work in practical application, including terrain effects and city build order.

As much as I've been cheerleading this concept, I worry about the opportunity cost of using Settlers this way.

But the power is there... going way back, this changes "the care and feeding of neighbors" and extends to "fun and games with modern armor."

Ohhhh yeah.

BRC · February 20, 2003, 00:19

Quote:

As much as I've been cheerleading this concept, I worry about the opportunity cost of using Settlers this way.

That is the only thing that bothers me too. I'm hoping that the inclusion of 2 tile spacing (which is new to me) with a couple of cities helps make up for the population transfer. However, as each town grows, shouldn't the payoff be greater later(in total # of population points)? Also, does this push back an early offensive?