This is why CXXXXC is VASTLY superior to CXXC

[Pyrrhos]

Exactly, Tone. Yet I and everyone else that do not swear fealty to "Tight, usually CXXC" get the opposite showed down our throats all the time:

"CXXC will have twice the number of towns and twice the number of citizens" or the variant "The CXXXXC player will only get half the number of towns" or the variant "Play CXXXXC and you will be AI-meat".

All these assumptions are non-arguments because they build on false premises:

False premise #1) No matter which pattern you use; OCP, loose, tight or ICS, AI expansion will ensure that you end up with the same size territory.
(Therefore - Within that territory CXXC gives you twice the number of towns, twice the number of citizens, twice the overall production etc. Conclusion CXXC is superior.)

False premise #2) Because CXXC uses every tile and CXXXXC leaves many unused, CXXC has twice the productivity and produces twice as many settlers and as these settlers have fewer tiles to walk, settle the map much faster, thus negating any advantage CXXXXC might have.
(Therefore - CXXC gives you the same territory as CXXXXC which means twice the number of towns, twice the number of citizens, twice the overall production etc. Conclusion CXXC is superior.)

False premise #3) Every successful player as measured by placement in the HoF or GoTM use CXXC, therefore CXXC is the only way to go and anyone who says different a stuborn, inflexible idiot who refuses to accept fact

Tone, I have proven that under the same conditions CXXXXC is vastly superior to strict CXXC. Another pattern has been presented as CXXC, one that actually is CXXXC according to Ginger Ale's definitions. I have investigated that pattern and found it marginally better than CXXXXC - and said so! Yet there are children such as "TheOverseer" who will never accept that they or their opinions could be wrong and see it as their sacred duty to religiously defend CXXC and persecute other opinions.

On my part, I will accept one fault - I do not suffer fools gladly but there's one thing I find even worse than a fool and that is a pompous, agressive fool such as The Overseer.

 

[Rodent]

CxxC and CxxxxC has its unique advantages so i suggest using 'both'

Settler CxxC early on. capture some land for yourself

when the Power age of CxxxxC i.e Industrial comes the towns in between create workers to reduce themselves forming CxxxxxC.

The reason i suggest keeping the cities is because the pattern formed will be CxxxxxC and there will be some unused tiles and the reduced town can be used to utilize these tiles

Comments please

 

[Pyrrhos]

An AI that used "tight placement, usually CXXC" would soon be "player meat".

Are you able to justify that statement please? My opinion on this statement is the opposite of yours but I don't see how either of us can back it up with evidence.

I'll give it a try:

* The AI uses something like CXXXXXC + culture to maintain it's rate of expansion. City placement is usually not optimal - one tile off the coast or a river is common

* The AI is bad at tile utilisation; improving tiles and managing them

Therefore were it to use CXXC, it would end up with a smaller territory with badly placed towns, badly improved tiles and poor tile management. This would adversely affect AI expansion, research, economy, building of improvements & wonders plus production of military units.

The player on the other hand would have access to a much larger tract of land and would with the superior utilisation of tiles of the human player, have even greater expansion, research, economy, building of improvements & wonders plus production of military units.

Therefore, it is logical to assume that an AI that used "Tight, usually CXXC" would be "player meat" as the player's task would be made much easier whichever level or map played.

 

[Pyrrhos]

CxxC and CxxxxC has its unique advantages so i suggest using 'both'

Settler CxxC early on. capture some land for yourself

when the Power age of CxxxxC i.e Industrial comes the towns in between create workers to reduce themselves forming CxxxxxC.

The reason i suggest keeping the cities is because the pattern formed will be CxxxxxC and there will be some unused tiles and the reduced town can be used to utilize these tiles

Comments please

True.

The problem lies, not only in the gaps created, but also the waste of settlers and improvements built in those towns-for-abandonment such as granaries (worker factories) or barracks (veteran units).

The way I play - at the moment - is capital CXXXXC to get the production (meaning max pop, shields, commerce and beakers) of the full 20 tiles for a "super city". The next "ring" of cities are CXXXXC inwards and "loose" outwards to make full use of the lowest corruption in my empire. From there on it goes down to tight in order to get the military advantages where I come in contact with the AI.

Others have said or suggested that method and it works just fine on normal to huge maps at Monarch/Emperor, where I usually play.

 

[Tone]

I agree that the AI is useless at tile management (but then I guess no one would contend that point!) but I think our difference in opinion hinges on this point.

Therefore were it to use CXXC, it would end up with a smaller territory.

A slightly tighter placement can mean that the towns you have are more powerful earlier on due to lower corruption. Often the difference of having that first ring one tile closer to your capital can make a significant impact on early production and commerce. In particular I am thinking about a new town working a bg or cow where one tile difference can mean that the second shield is lost to corruption. A closer town placement would lead to the first ring being significantly more effective. Does that mean that the AI would still be able to fill up the space? Possibly; possibly not. However I do know that it gives me an edge when playing against a rival that is more widely spaced so why not for the AI, even with its poor management.

Maybe the wider spacing helps more at the lower levels where production is less? I don't doubt for one moment though that the AI could still fill the majority of the space with a more compact arrangement at Deity/Sid and I would certainly find them far more fearsome because of it.

 

[Pyrrhos]

Let's make a test of it then, shall we!

 

[Rodent]

The problem lies, not only in the gaps created, but also the waste of settlers and improvements built in those towns-for-abandonment such as granaries (worker factories) or barracks (veteran units).

Like i said, take care not to build too many improvements in the concerned Cities.

The Towns for abandonment are not going waste. during Ancient and Medieval Eras they are full fledged producers of military in your empire helping you to expand. when Industrial comes these towns are not Abandoned. just reduced to size 6-7 so that the tiles wasted can be used so the culture buildings (if any) stand.

 

[Aabraxan]

"As you might imagine, I have my doubts that CxxxxC is superior on tiny-small maps at DG/Sid. However, as I have never beaten anything even close to those conditions, I can't speak from experience."

Then try it!

. . . . .

When I can beat those levels, I will try it. At this point, though, my losing at DG or Sid would only prove that I could lose at those levels (an already-known fact) and would have little empirical value for a city spacing test.

However, I am currently testing some wider spacing (from my usual) at Monarch or Emperor. I can't remember which right now, but I think it's Emp. I'm using the Celts and shooting for a 20K win, but building enough temples that I'm hoping to take 100K as a fallback. In all honesty, I'm finding the game to be notably harder than usual. I am way behind my usual progress in that game. I'm sure part of that is my unfamiliarity with CxxxxC, but I also feel certain that part of it is due to the city spacing, too.

 

[Aabraxan]

. . . . Another pattern has been presented as CXXC, one that actually is CXXXC according to Ginger Ale's definitions. I have investigated that pattern and found it marginally better than CXXXXC - and said so! . . . .

Pyrrhos, are you referring to the diagram that ZzarkLinux has put up?

If this is the one to which you are referring, may I ask how you come to the conclusion that it's CxxxC, even by GA's definitions? To my eye, you can get from any one city center to another city center in 1 turn on roads. (Not to any other city center, mind you, but to another city center.)

 

[Optional]

If this is the one to which you are referring, may I ask how you come to the conclusion that it's CxxxC, even by GA's definitions?

Ginger Ale's definitions are confusing, and allow for different interpretations. One of the characteristics of tight city placement he uses is, like you say, 'the ability to go from one city to another in one turn, over road, by a 1 move unit.' Zzarklinux in the frame.
Another characteristic Ginger Ale uses for tight city placement is 'each city only gets around 9 tiles.' Zzarklinux out of the frame.

I honestly can't say where Ginger Ale would put Zzarklinux in his definitions. And, although I'm not Pyrrhos, I believe Pyrrhos indeed calls Zzarklinux' city spacing picture CxxxC.

 

[Spoonwood]

Aabraxan,

"However, I am currently testing some wider spacing (from my usual) at Monarch or Emperor. I can't remember which right now, but I think it's Emp. I'm using the Celts and shooting for a 20K win, but building enough temples that I'm hoping to take 100K as a fallback."

Well I think you've decided to play "either I win or I won't try this," sort of games. I can understand this, but then again I don't know how many games I've played a good while in and then quit (meaning I lost) and thought I learned something in the process. By all means, suit yourself... but challenge yourself also.

What's the year for the Celts? What tech do you currently research? Which wonders do you have in your 20k city? Any saves or screenies for us?

Optional,

"If this is the one to which you are referring, may I ask how you come to the conclusion that it's CxxxC, even by GA's definitions? To my eye, you can get from any one city center to another city center in 1 turn on roads"

Ginger Ale lists pros and cons. One of the *pros* of 'tight placement' reads "3 tile separation allows to go from one city to another via roads." In other words, he didn't define loose placement by one turn moverment via roads. Ginger Ale didn't provide verbal definitions, but DID provided screenshots which in effect give us definitions (ICS gets read as CxC, tight as CxxC, and so on). With Zzarklinux's picture we can move between the towns two turns on the diagonal and one turn horizonatally/vertically to reach the next city. For Ginger Ale's picture we can find examples of this for BOTH 'loose' or CxxxC as well as CxxC or 'tight'. So, Ginger Ale's pictures don't help us resolve what sort of spacing Zzarklinux's fall into. Pyrrhos, I'll suggest, thinks that his picture does NOT qualify as CxxC, as the 'xs' in CxxC work both horizonatly/vertically and diagonally. Since, one has to travel an 'x' diagonally, another 'x' diagonally, and then a THIRD x horizontally/vertically, and we have three 'xs', this means Zzarklinux's picture comes out CxxxC. For consistency, Ginger Ale's picture for 'tight' would also come out CxxxC in spots... although it would come out CxxC in other spots. With Zzarklinux's picture, and this method of determining spacing, all spots come out CxxxC.

From the pictures it seems that all styles come out mixed, except for ICS (which seems CxC through and through). OCP seems a mix of CxxxC and CxxxxC, Loose a mix of CxxxC and CxxC, and Tight a mix of CxxC and CxxxC... Loose just has more CxxxC than Tight which has more CxxC.

I'd call Zzarklinux's spacing in his hypothetical 5-city game picture a tight version of loose placement. I don't know why anyone would play a 5-city game with such a spacing though.

 

[Aabraxan]

Aabraxan,

"However, I am currently testing some wider spacing (from my usual) at Monarch or Emperor. I can't remember which right now, but I think it's Emp. I'm using the Celts and shooting for a 20K win, but building enough temples that I'm hoping to take 100K as a fallback."

Well I think you've decided to play "either I win or I won't try this," sort of games. I can understand this, but then again I don't know how many games I've played a good while in and then quit (meaning I lost) and thought I learned something in the process. By all means, suit yourself... but challenge yourself also.

And you decided this because I'm currently testing a game: (1) at one of the two highest levels I've ever beaten (probably the highest); (2) while going for either (a) a 20K VC, which I've never had; or (b) a 100K condition, of which I've only had 1? Help me out with your reasoning here. How does that show that I've decided to play an "either I win or I won't try this" sort of game?

What's the year for the Celts? What tech do you currently research? Which wonders do you have in your 20k city? Any saves or screenies for us?

I'll take a look at it and get back to you.

Optional,

"If this is the one to which you are referring, may I ask how you come to the conclusion that it's CxxxC, even by GA's definitions? To my eye, you can get from any one city center to another city center in 1 turn on roads"

Ginger Ale lists pros and cons. One of the *pros* of 'tight placement' reads "3 tile separation allows to go from one city to another via roads." In other words, he didn't define loose placement by one turn moverment via roads. Ginger Ale didn't provide verbal definitions, but DID provided screenshots which in effect give us definitions (ICS gets read as CxC, tight as CxxC, and so on).

Where I've underlined above, it looks like you've attributed my words to Optional. Not only did GA not define loose placement by 1-turn movement, he specifically listed, as a con: "It takes more than 1 turn to get from city to city with roads and a 1 move unit."

With Zzarklinux's picture . . . . Pyrrhos, I'll suggest, thinks that his picture does NOT qualify as CxxC, as the 'xs' in CxxC work both horizonatly/vertically and diagonally. Since, one has to travel an 'x' diagonally, another 'x' diagonally, and then a THIRD x horizontally/vertically, and we have three 'xs', this means Zzarklinux's picture comes out CxxxC. For consistency, Ginger Ale's picture for 'tight' would also come out CxxxC in spots... although it would come out CxxC in other spots. With Zzarklinux's picture, and this method of determining spacing, all spots come out CxxxC.

We'll have to wait for Pyrrhos to see his answer to my question, but that may be it.

As for Zzark's diagram, I have to respectfully disagree. To move from the central city square to any other city square, the moves go: (1) C-x; (2) x-x; (3) x-C. No, they're not all in the same direction. They might go (for example), E-E-NE (gray to red), but that's CxxC, AFAIK, not CxxxC. You can take a longer route if you wish, but I don't see any need to.

I'd call Zzarklinux's spacing in his hypothetical 5-city game picture a tight version of loose placement. I don't know why anyone would play a 5-city game with such a spacing though.

Surely you realize that Zzark's diagram wasn't really for a 5-city game, right? There are only 5 cities there, but I think the pattern is supposed to be repeated across the landscape.

 

[Spoonwood]

"And you decided this because I'm currently testing a game: (1) at one of the two highest levels I've ever beaten (probably the highest); (2) while going for either (a) a 20K VC, which I've never had; or (b) a 100K condition, of which I've only had 1?"

I said "I think"... and I forgot about that... so I thought wrong. Props to you.

"How does that show that I've decided to play an "either I win or I won't try this" sort of game?"

It doesn't. I admit my error.

"Where I've underlined above, it looks like you've attributed my words to Optional."

Sorry, I missed the quotes there.

"As for Zzark's diagram, I have to respectfully disagree. To move from the central city square to any other city square, the moves go: (1) C-x; (2) x-x; (3) x-C. No, they're not all in the same direction. They might go (for example), E-E-NE (gray to red), but that's CxxC, AFAIK, not CxxxC. You can take a longer route if you wish, but I donj't see any need to."

You're right... I botched that too. What a bad post. Anyways... I know I did get right that there exists a CxxC spacing at one point in Ginger Ale's "loose spacing"... that I saw for sure.

"Surely you realize that Zzark's diagram wasn't really for a 5-city game, right? There are only 5 cities there, but I think the pattern is supposed to be repeated across the landscape."

Yes, but that's part of the problem. Without seeing terrain and luxuries, and especially a full-size or close to full-size empire (say around the end of the ancient age), we simply don't know what *actual* game spacing ends up looking like for a player. Some CxxC players might use a much wider spacing in spacing, or tighter in other spots. So, for a better model of spacing, we need more information than just CxxC spacing. I say this especially because I thought I used CxxxxC spacing... but when I checked some of my own saves it came out more like a CxxxC with some CxxxxC thrown around, and one or two CxxC spacings. We can talk about spacing more like this instead of using oversimplified models, so why not do so?

 

[Pyrrhos]

Ginger Ale's definitions are confusing, and allow for different interpretations. One of the characteristics of tight city placement he uses is, like you say, 'the ability to go from one city to another in one turn, over road, by a 1 move unit.' Zzarklinux in the frame.
Another characteristic Ginger Ale uses for tight city placement is 'each city only gets around 9 tiles.' Zzarklinux out of the frame.

I honestly can't say where Ginger Ale would put Zzarklinux in his definitions. And, although I'm not Pyrrhos, I believe Pyrrhos indeed calls Zzarklinux' city spacing picture CxxxC.

Good spot, but Zzark's pattern is out when it comes to the moves too. I've taken the pix, added one town and drawn lines in purple for three-square moves and black for five-square moves to where the next towns would be.

As can be seen, there are as many towns five squares apart as there are three squares apart. The AVERAGE distance is 4, even if it is possible to move from town to to town but only along certain paths! Therefore, as I have pointed out earlier, Zzark's pattern is neither "loose, usually CXXXC" nor "tight, usually CXXC" - it is a hybrid that has the advantages of both and the disadvantages of neither. IF it is to be placed in one of Ginger Ale's categories, that category is "loose, usually CXXXC" because:

* Each town has, on average, 12 tiles

* The average distance between towns is four even if it is possible to move from town to to town but only along certain paths!

 

[eldar]

It's RCP4. A quick explanation of how to calculate a city's RCP distance:
- "Straight" moves (NE/NW/SW/SE) are 1.
- "Diagonal" moves (N/E/S/W) are 1.5.
- Use the quickest route always.
- Round fractions down.

So from the centre each of the 8 city locations you go 2 "diagonal" and 1 "straight" move for a total "distance" of 4.

It's an idealised form of CxxC that allows every city to get 12 tiles. The same as OCP is an idealised form of CxxxxC that allows every tile to be used with minimal overlap but only once every city is a metro.

Every idealised form will suffer in the face of the terrain available, which will be restricted by coasts, tiles that cannot be settled on, and by AI territory, and pulled out of shape by beneficial factors such as rivers, lakes, choke points, canal points, resources, luxuries.

It is also my opinion that CxxC doesn't require that the average distance between cities is less than 4 (or less than 3.5?), but that you can get from City A to at least two other cities along a road with a 1-move unit in one turn [excepting rivers pre-Engineering]. The above diagram satisfies that condition.

 

[Pyrrhos]

Sorry Eldar, but that is BS-ing of the highest order as true "tight, usually CXXC" is vastly different from this pattern. Furthermore, this pattern does not match the definitions of "tight, usually CXXC" but rather "Loose, usually CXXXC".

You just cannot accept that what you think of is not CXXC, can you! You just cannot accept that you may be wrong, can you! You just cannot accept that, occasionally, in life, we have to admit we're wrong, accept new definitions and move on, can you!

Strewth!

 

[eldar]

Sorry Eldar, but that is BS-ing of the highest order as true "tight, usually CXXC" is vastly different from this pattern. Furthermore, this pattern does not match the definitions of "tight, usually CXXC" but rather "Loose, usually CXXXC".

You just cannot accept that what you think of is not CXXC, can you! You just cannot accept that you may be wrong, can you! You just cannot accept that, occasionally, in life, we have to admit we're wrong, accept new definitions and move on, can you!

Strewth!

1. Please adopt a more reasonable posting style when debunking others theories and opinions. Throwing around lines like "you are BS'ing" is not polite. I for one am getting fed up with your condescending tone towards others when others are far more polite. Reasoned argument should not be responded too with vitriol.

2. If you're going to tell someone that what they have written is "BS", please back up your words with examples or concrete evidence before wading in without yet another post that contains no substance, just another rant at someone who doesn't happen to share the same views as you. You say that it is not "Tight, usually CxxC". Then what in your opinion would "Tight, usually CxxC" be? I've yet to see a single post in any of the three threads on the subject where you have defined what you understand to be CxxC. If I've missed one, so be it - there are hundreds of posts after all, and please provide a link. I'm sorry for trying to help by bringing up GA's article, and highlighting one particular condition of CxxC spacing that is undeniably fulfilled by the spacing in that diagram.

Shall I start to bring RCP into the argument? Because it could be argued that "true" tight/CxxC spacing would be RCP3-style, not RCP4-style. It's true that that diagram is RCP4, and that along the "straight" axes, this would lead to CxxxC spacing. (There - I've just contradicted myself in the space of two posts.) It's just fortunate that it allows CxxC-style travel between cities.

 

[Pyrrhos]

This is true CXXC where every town has no more than a three-tile move to all its neighbours. As can be seen, there are no more than eight tiles per town.

 

[Pyrrhos]

1. Please adopt a more reasonable posting style when debunking others theories and opinions. Throwing around lines like "you are BS'ing" is not polite. I for one am getting fed up with your condescending tone towards others when others are far more polite. Reasoned argument should not be responded too with vitriol.

It would be nice if you understood that you actually are just as offensive and impolite in your posts. You do not - contrary to what you say - back up your opinions with fact, only opinion, and disregard and dismiss what is offered as explanation as it doesn't fit with your pet theories. Since I was not certain that you were familiar with the much more polite "prevaricate and procastrinate", I used the term BS-ing which you most likely would be familiar with.

Furthermore, you edited your post after I had replied to it. (This is possible under certain conditions without the "post edited..." being displayed.)

 

[Aabraxan]

. . . . I said "I think"... and I forgot about that... so I thought wrong. . . . .

Fair enough.

. . . . Yes, but that's part of the problem. Without seeing terrain and luxuries, and especially a full-size or close to full-size empire (say around the end of the ancient age), we simply don't know what *actual* game spacing ends up looking like for a player. . . . .

I don't think there's any question, from anybody, about whether terrain will force a player into a tighter or looser pattern than their "ideal." I think we all adjust spacing to accomodate terrain and resources. Clearly, Zzark's pattern is only possible on flat land and hills, with no marshes, no mountains, no volcanoes, etc.

Good spot, but Zzark's pattern is out when it comes to the moves too. I've taken the pix, added one town and drawn lines in purple for three-square moves and black for five-square moves to where the next towns would be. . . . .

Thanks, Pyrrhos. That clarifies it for me.

Aabraxan, . . . .
What's the year for the Celts? What tech do you currently research? Which wonders do you have in your 20k city? Any saves or screenies for us? . . . .

I don't want to derail Pyrrhos' thread any more than I already have, so I've PM'd you some details.