Testing out CxxC vs Metros for Space Race ability

[Othniel]

Timerover's second map is excellent.

Which map is this, exactly? Timerover's original or my edited one?

I have some reservation about rigorous editing of the initial map by Othniel: You might be editing towards making it more suitable for your style of play. Since you've marked yourself as a tight spacer, you might unconciously have made it more suitable for that.

I was consciously subduing my subconscious as I made my edits .

Seriously, as I changed things around a bit, I made sure to never count tiles or do mental city dotmapping. I just made the map more generally "good" by swapping harsher tiles for grass and plains and by adding a handful (three or so total) food bonuses to AI start locales. I also added some fresh water to AI start locations that were previously dry.

I would hesitate to call these edits "rigorous", but I also know how self-deceived people can be. If you or others feel that I edited the map too much, I can accept that.

Perhaps you would like to mod your proposed map slightly, using the changes I mentioned before?

 

[Optional]

My main thought was that adding food to your start location to facilitate a 4 turner and thus having higher settler production would make tight city spacing a lot easier.
But I'm starting to speak beyond my own competencies as a player here. When I look at a start position, I don't see whether there's a 4, 5 or 6 turner in there. Some players spot this immediately.
I think you should have some decent food around, but not quite supertop. How that looks like I'm not quite sure of. I was hoping for some insight from more experienced players than me. It seems like I'm the only person that has reacted to the maps so far. I don't feel quite comfortable with that monopoly.

 

[TheOverseer714]

IMO, bad terrain affects wider spacing more than tight, but if you aren't doing strict spacing, it can be worked around. I'd like to see a screenshot of the starting terrain, before I make any further judgements.

 

[Spoonwood]

"http://www.civfanatics.com/civ3/stra..._placement.php"

I think this works as a way of defining different city location styles... however I would feel wary about all the advantages and disadvantages of each style coming as necessarily correct. For instance, a commercial civ won't have *as much* of the corruption "con" apply to them for ICS.

 

[Othniel]

I'd like to see a screenshot of the starting terrain, before I make any further judgements.

Can someone post a screenie or two of the proposed starting continent? Preferably a zoomed out one as well. If no one does this in the next few hours, I'll do it this evening.

 

[timerover51]

Othniel, the reason that I liked some of the other maps that I generated is that they had multiple continents, and a fair number of islands. I like to play those kinds of maps, but they would not have been suitable for the test.

I am not sure about bending over backward to avoid giving the AI equal starting positions. The more you get away from a straight generated map, the less your results will apply to the average game. Remember, the purpose of this test is guidance for the average player on a unmodified generated map.

As soon as your start modifying maps, the spacing argument becomes much more open to dispute, especially if you start modifying resource and terrain yields like I do. With that I can have 10 to 12 one-unit per turn cities with the minimum research time running by the Industrial Age. That is using town/city population of 8/16, and not even going to Metros. I set pollution clean up to 4, so one worker needs one turn to clean up, so no more massive worker stacks dealing with pollution. I do that so that I can enjoy playing the game and work on other things besides micromanaging a hundred or so cities. Besides, I do not like tight city spacing, having grown up and lived in the Chicago metro area about all of my life. Alaska is much more to my liking.

June 12: I looked over your changes, Othniel, and do not see any problems with them. I am interested in seeing what the results are.

 

[Fugitive Sisyphus]

I always use pack my cities in tight(CxxC). If I am playing a small map, an island map, or just map with a lot of water I don't have enough space to go CxxxxC. If I am playing a big map I will only be able to claim one or two luxuries so my non-core cities can't grow big enough to take advantage of even 12 tiles for much of the game due to happiness problems.

For fun I just started a random emperor Standard/70% Water/Continents map with a random Civ(got the Maya coincidently) and even though I tried to spread my cities out I could do no better than CxxxC. I simply didn't have enough space( or river tiles) to go CxxxxC.

I attached an image of my 'empire' to show you what I mean.

Edit: Though I admit I do love how I have to only keep track of a few big cities as opposed to many small ones.

 

[Othniel]

Forgot about posting screenshots last night. Here they are now:

timerover51's Original Map and My Edits Contrasted:

Spoiler :

Overview:

Spoiler :

Edited Overview:

.....

Spoiler :

Original Start Area:

Spoiler :

Edited Start Area:

.....


Then, we have Optional's, umm, map option :

Spoiler :

Overview:

Spoiler :

Start Area:

Spoiler :

My Proposed Edits to Optional's Map:

So, these are the map options. If people want to chime in on what map they prefer, awesome. If you have proposed edits, please let me know.

Vote Options:
1) timerover's original map
2) timerover's map with my edits
3) Optional's map
4) Optional's map with my edits
5) One of these maps with different edits

I think I will make a choice on the map in a few hours from now. Any feedback will be taken under consideration. If no one responds, I get to make a choice all by me lonesome and as arbitrarily as I want.

EDIT:
I'm re-attaching timerover and Optional's original maps pre-edits. They have posted them elsewhere in this thread but I wanted to solidify all map info in one place. If anyone wants to check out the maps in the editor for more info, feel free to do so.

 

[Optional]

I suggest if someone responds - and this is me responding - you leave a little bit more time for discussion.
I haven't looked at any of the maps in all that much detail yet. Btw, the missing aluminium on the map that I generated is something I had missed.
The main things I have considered are:
- Plenty food but less luxuries would be better for tight placement, while plenty of luxuries but less food would be better for spacious placement. So; striking a balance between food and lux resources I would say is essential.
- Settlement options; a bit of space around you (there is concensus about this, so I don't need to explain it).
- Two continents, 4 lux each, a decent start spot (also consensus here, if I'm not mistaken).
- An average landscape, also a landscape that the random generator would make.

The first two criteria I would consider the crucial ones.

Othniel, you've done some landscape surgery to make the map in general 'better'. Why is this? It doesn't look good, you manipulating the map before you start playing, and you're not really explaining it either (unless I've missed something, which can easily be the case).
I'm all with you on moving the starting spot slightly more central in Timerover's map. Also, to have at least one food resource nearby seems fine to me. I don't see the need for the other stuff with regards to the credibility of this test.

I don't want to be drawn into a vote yet. Not that I don't have any line of thought: I would go for Timerover's map, and I agree with a slightly more central position for the start, maybe pop in a sweet water lake here and there to make for a more flexible settlement pattern, and with making sure you've got at least one food resource nearby. More editing I wouldn't do. Average maps have some desert, they have some jungle, they have the odd island, etc. I don't see how this skews the test. When I was doing some random generating, I tried so avoid irregular patterns, islands, swaths of jungle, etc, but they appear anyway. It turned out to be hard enough to generate a map that at least ticked the main boxes.
I rather went with a map that ticked the main boxes than to do a lot of editing. Hey, I did edit as well, you know; I replaced a volcano in your starting box!
But the bottomline I would have is; you must like the map you start playing. You should be doing this for your enjoyment. You're not a scientist looking for a cure for a lethal desease.

 

[Othniel]

Thanks for all the feedback, Optional.

I suggest if someone responds - and this is me responding - you leave a little bit more time for discussion.

Here's the deal: I'm really pinched for time the rest of June. I have conferences, a backpacking trip, weddings to attend, yada yada yada. I leave tomorrow for the conference. Bottomline, I not sure when I'll be able to devote much time to this until July, when I should be really free.

I want to get these test games to a certain point before I leave and get super-busy. In my mind, that certain point was,

• Map picked out, posted on this thread, and opened up to public discussion / dotmapping.

If that can happen, then the thread will hopefully take care of itself somewhat until I get more time. My fear is that if this map choosing business is left hanging while I'm gone, then a lot of interest in this test will fade away. Already, there is a large decrease in volume of response. Is the fear unfounded? Maybe, but I'd really like to get a map chosen tonight so I can post it before I leave.

If the map choosing decision is too rushed, on the other hand, I realize it can make the test itself pointless. So I am trying to strike a balance here. Hopefully others respond soon so I get a clearer picture of people's opinions.

The main things I have considered are: [...]
The first two criteria I would consider the crucial ones.

I'm still not convinced that having lots of food favors one strategy. In general, getting more settlers out there would favor both, I think. I realize Metros need fewer settler in sum, but the quicker I settle even Metro style, the faster those metros start working on improvements.

Othniel, you've done some landscape surgery to make the map in general 'better'. Why is this? It doesn't look good, you manipulating the map before you start playing, and you're not really explaining it either (unless I've missed something, which can easily be the case).
I'm all with you on moving the starting spot slightly more central in Timerover's map. Also, to have at least one food resource nearby seems fine to me. I don't see the need for the other stuff with regards to the credibility of this test.

Fair criticism.

I edited the map in two primary ways:

• "Enhanced" AI start locations to make them more challenging
• "Improved" the terrain around my core

I think improving the area around my core might be the controversial one. My thought is this: if I have tons of desert--and I'm not AGRI so desert is poor for me--that is part of my core area, it is going to skew the test. It will skew the test because I'll have crappy cities in low-corruption areas. It will be hard, I feel, to say one style is better than another when I'll have half a core of good cities and half that won't be productive no matte what. Maybe metros are better for this because they have more tiles to choose from, but I don't know. The hypotheticals are honestly starting to hurt my brain.

Basically, I think I want to go with your map (see below for more info). I haven't edited basically at all. If I use it, I can avoid some of these skewing questions, I think.

I don't want to be drawn into a vote yet. Not that I don't have any line of thought: I would go for Timerover's map, and I agree with a slightly more central position for the start...

I'm actually favoring your map at the moment because:
1) the starting continent is smaller
2) the start is more centrally placed without edits

For timerover's map, I think the starting continent would support a very large number of science farms and quite simply kill the metro style. If I wanted huge number of farms, I wouldn't have disallowed overseas conquest and a large starting continent circumvents that disallowment. Your map seems to give metros a little more of a chance, I think.

I still think we should remove a bit of that marsh from the northern AI start locations on your map. That much marsh is not average, in my opinion, and will make those AIs into absolute weaklings. Make the marsh into jungle if we still want the terrain to be harsh, but marsh really only hurts the AI, not the human for the most part.

I also edited that game to have grass underneath rather than plains. We can keep the plains if people want. The reason I made this change was to give my capital +4fpt potential, rather than only +3fpt. The extra food makes a 6-turn settler factory possible. This is a far cry from a 4-turn settler factory, but it's a lot easier to grow with 4fpt than it is from 3fpt. 4fpt is also pretty average, IMO.

Hey, I did edit as well, you know; I replaced a volcano in your starting box!

But the bottomline I would have is; you must like the map you start playing. You should be doing this for your enjoyment. You're not a scientist looking for a cure for a lethal desease.

I'll best enjoy whatever map is the consensus map. Having tons of people dismiss the test because the map was too this or that would be frustrating.

 

[TheOverseer714]

I'm okay with Timerover's map with the edits, it seems good-average, not the massive swaths of rough terrain on a poor-average map. I would rate Optional's map as fair-average, it has more mountains and other stretches of bad terrain. I agree that time is of the essence, if this waits 4 weeks, there may be less interest in the results.

 

[Optional]

I more or less agree with your whole post, Othniel.

The interest in this thread is not really high at the moment. Trying to reach a consensus on the map at the moment would mean trying to reach a consensus between you and me. Actually, there's not even much wrong with that, as we seem to differ enough in our views.

I feel strongly against rushing a map decision now. When Timerover's and my map were put up, I was hoping that a discussion about map criteria would start. There are still too many objections about both maps to be made at the moment. The main objection I have against my own map is that the AI does have a bad deal there, with all that jungle and marsh in the north of the continent.

I've done some more random generating, this time with the climate setting changed from 'average' to 'wet', to avoid big stretches of desert. It gave me a map with one continent almost free of jungle, one you could easily put 4 strong start positions in. The continent seems a little bit bigger than the other one, but there's not much between them. The map is probably better than my previous one for providing a more challenging AI - provided they don't commit early suicide-DoW's, maybe we should keep the Zulu and the Germans out of this test, they're pathetic sometimes.
If you'll allow me, I'll put it up here. Will that cause more delay? It looks like we're heading for some delay anyway, I wouldn't worry that much about that. I can try and keep the discussion in this thread going as best as I can while you're away.

Edit: I'll just throw this map in now, to not lose more time. I had some technical trouble designating the start position, but I was thinking about the most northwestern start in the map. I haven't had time yet to look at the details, maybe Aluminium is missing yet again, who knows?

 

[Optional]

I agree that time is of the essence, if this waits 4 weeks, there may be less interest in the results.

Would we have results already that quick?

 

[TheOverseer714]

Would we have results already that quick?

Probably not, even though almost everyone plays faster than I do I guess if it is interesting, it will stay that way. Besides, I think it might make an good article in the strategy forum, or even the updated war academy.

 

[Othniel]

Thanks guys. You are encouraging to me. Thanks to timerover also for his work on the maps.

Optional, you are right that there is plenty more to discuss about map settings. Tell you what, I'll wait until Wednesday of next week to make a choice. I should have some time on that day to post.

The weekend will allow time for discussion. Interest in these sorts of things tends to ebb and fall and who knows, maybe some people will have time to give thoughts over the weekend.

I highly doubt that I'll get the test done within 4 weeks. I'm sure I'm even slower of a player than Overseer. If Overseer feels he's slower than me, I'll challenge him to an arm-wrestling contest for the title. My right arm is powerful from moving my mouse around for hours on end.

Regardless of how much time it takes or if interest is so low that I make 8 pages of posts in a row with no one else commenting, I will get this thing done. It's a promise. Hopefully the journey will keep the interest high even as I drive for final results.

 

[TheOverseer714]

I highly doubt that I'll get the test done within 4 weeks. I'm sure I'm even slower of a player than Overseer. If Overseer feels he's slower than me, I'll challenge him to an arm-wrestling contest for the title. My right arm is powerful from moving my mouse around for hours on end.

Oh yeah!, said the Snail to the Tortoise

 

[AutomatedTeller]

I guess you can go with the CxxC in that thread, but I think that the discussion becomes much, much less interesting then, because that's NOT what most of the guys who are proponents of CxxC mean. Generally, they mean off-set, just like most people who propose CxxxxC actually mean an offset scheme that has a tile or two overlap, not one that leaves 4 tiles unused.

Except in games where they expect no city to get TO size 12, of course.

 

[Othniel]

I guess you can go with the CxxC in that thread, but I think that the discussion becomes much, much less interesting then, because that's NOT what most of the guys who are proponents of CxxC mean. Generally, they mean off-set, just like most people who propose CxxxxC actually mean an offset scheme that has a tile or two overlap, not one that leaves 4 tiles unused.

I definitely plan on looser Cxxc that nets 12 citizens for most cities.

 

[Spoonwood]

"I definitely plan on looser Cxxc that nets 12 citizens for most cities."

I've started to think this test won't prove all that much... especially in light of this comment, looking at my own *supposedly* "CxxxxC" maps, and looking at *supposedly* "CxxC" maps. How many supposedly CxxxxC players end up going with something more like a loose version of CxxxC because of terrain considerations? How many supposedly CxxC players go with a tight version of CxxxC, again because of terrain considerations and to maximize tile usage at size 12? If the suggestion there that most of the better players usually use some form of CxxxC (other than for the 20k city and for 100k games), doesn't the only real question become whether a tight version of CxxxC works better or a loose version of CxxxC works better? And since both "styles" end up hovering around the same basic pattern... how much does the question really end up mattering?

 

[Aigburth]

"I definitely plan on looser Cxxc that nets 12 citizens for most cities."

I've started to think this test won't prove all that much... especially in light of this comment, looking at my own *supposedly* "CxxxxC" maps, and looking at *supposedly* "CxxC" maps. How many supposedly CxxxxC players end up going with something more like a loose version of CxxxC because of terrain considerations? How many supposedly CxxC players go with a tight version of CxxxC, again because of terrain considerations and to maximize tile usage at size 12? If the suggestion there that most of the better players usually use some form of CxxxC (other than for the 20k city and for 100k games), doesn't the only real question become whether a tight version of CxxxC works better or a loose version of CxxxC works better? And since both "styles" end up hovering around the same basic pattern... how much does the question really end up mattering?

I think the two styles to compare are planning for 12c/C and planning for full metros, I doubt many people would always place their cities according to a strict pattern as there are terrain considerations to take into account such as making best use of fresh water sites. I play the 12c/C style but may have a mixture of CxC, CxxC and CxxxC in my core as in the exapmle below: -

Spoiler :