Fastest Science Victory

[Blatc]

Pulled off a T33 neighbour elimination: Sweden was nice enough to build Great Library and Watermill. Only 1 other civ saw anything! Now to figure out how to play Liberty-wide SV.

 

[consentient]

I suspect food caravans might be involved

 

[Blatc]

Wasted opportunity on the T33 elimination game; I'm not very competent at Liberty SV. Also, 1 Cultural CS is a tad below average, so I didn't see that ending well. I'm probably better off working on 5 or perhaps 6-city Tradition.

 

[Manpanzee]

I recently played out a full game of Civ for the first time in a while and came out with my second fastest Deity SV -- this one a t206 with Liberty Siam. There were zero faith CS on the map, so I could only buy one GS. Add an extra one there, and sub-200 was probably possible.

Liberty is really interesting -- it feels to me like it offers a lot more variety than Tradition. In this case, I cash-bought a Settler as early as I would have with Tradition, and that felt really strong. I now think that waiting for Collective Rule is a self-imposed hurdle that should be almost entirely disregarded. I also finished Liberty super early and was able to get my GE'd NC on t63, whereas it normally feels like that comes around t80. A start this fast brings the hard-built NC + Academy play back into the picture.

Siam feels like quite a strong Liberty Civ. They're a bit like a pseudo-Poland. Wat culture gives you something like a free Tradition opener, and the midgame Culture boost lets you blaze through Rationalism. Siam isn't as good for policies as Poland, but Poland frequently hits a point of diminishing returns there anyway. The only problem with Siam is that it's extra map-dependent, as a lot of the value lies in the UA.

I'm now wondering how long the list of Civs that theoretically can go sub-200 is. We've got Poland, Babylon, Huns, Korea, Inca, and Shoshone. Then maybe Maya, Siam, Austria, Egypt, Aztec...

Currently fantasizing about an Inca Trad-Liberty build, where you use Aristocracy + natural Inca hammers to help get a wide-empire NC in reasonable time, then maybe time finishing Liberty until after your first GS spawn to get to two Academies unusually early.

 

[Blatc]

Definitely agree that Siam is capable of sub-T200, with a good map and enough of the appropriate type of CS. Austria is definitely capable, as well.

 

[consentient]

Nice game, Manpanzee. Out of interest, did you choose Askia and Suleiman? If so, why?

 

[unrendered]

yay finally got a sub 220 SV. it was close but i managed to get a t219 win on ironfighters babylon GP map (without replays )

t219 screenshot:

Spoiler :

had a few spare gs surprisingly but i had to bulb a little inefficiently to get the t219 win. pretty satisfied overall tho since my previous record was t259.

 

[Manpanzee]

Nice game, Manpanzee. Out of interest, did you choose Askia and Suleiman? If so, why?

I generally like having a couple warmongery civs around -- it seems to make it easier to avoid being public enemy #1. I feel like those guys are some of the less-scary warmongers. I know Vadalaz and others sometimes like having Shaka around, but his hyper-expansiveness kinda scares me. You can sometimes ignore Ottomans/Songhai, but Shaka always needs to be managed.

 

[vadalaz]

I generally like having a couple warmongery civs around -- it seems to make it easier to avoid being public enemy #1. I feel like those guys are some of the less-scary warmongers. I know Vadalaz and others sometimes like having Shaka around, but his hyper-expansiveness kinda scares me. You can sometimes ignore Ottomans/Songhai, but Shaka always needs to be managed.

I agree Shaka is scary, especially on a small map like Great Plains where he can easily reach you with his 3-movement Impis no matter where he spawns. I've stopped using him now, these days I also like to have 2 manageable warmongery civs and keep the world at war almost constantly. Another good thing about the warmonger civs is that they are less likely to propose Arts Funding, so I usually put them in the player 2 slot.

I'm now wondering how long the list of Civs that theoretically can go sub-200 is. We've got Poland, Babylon, Huns, Korea, Inca, and Shoshone. Then maybe Maya, Siam, Austria, Egypt, Aztec...

Spain definitely, maybe Arabia as well? I had only one real attempt with them which didn't go well due to poor timing. I went Chivalry before Education, got greedy and took 4 capitals which led to some serious happiness issues. Camels kill things very fast, so I think it makes sense to postpone the conquest a bit closer to ideology timing, start around T115 perhaps.

 

[unrendered]

what AI civs should i pick when i play GP? up to this point i've just been playing with randoms but i imagine having a few easily befriendable AIs and one warmonger would help out.

 

[vadalaz]

what AI civs should i pick when i play GP? up to this point i've just been playing with randoms but i imagine having a few easily befriendable AIs and one warmonger would help out.

Check this chart for AI diplo values: http://civdata.com/

Some good picks include:
- Sweden - low expansion bias, 10% bonus to Great People if friendly. Usually has good military and can be bribed to attack someone
- Netherlands and Arabia - friendly and good for lux trades. Harun can be a bit annoying with religious spam and William sometimes expands too much
- Venice and Morocco - friendly and good for trade routes. Venice will usually buy at least one CS though
- India, Brazil, Byzantium - friendly and low expansion bias
- Portugal - friendly and likes the top part of the tech tree. You can sometimes steal Education or Astronomy from her. High expansion bias though, so she can be an annoying neighbour
- Babylon - I don't know what it is with Nebuchadnezzar, but he does terribly in almost every game I play, hates everyone and becomes irrelevant by Renaissance. Low expansion bias and can be bribed to attack someone

 

[IronfighterXXX]

I'm now wondering how long the list of Civs that theoretically can go sub-200 is. We've got Poland, Babylon, Huns, Korea, Inca, and Shoshone. Then maybe Maya, Siam, Austria, Egypt, Aztec...

As we know land is more important for fast SV than civilization; so the question is: can land be so good to allow sub 200t SV for all civs (maybe except Venice ?

IMO Yes!

We keep playing great plains maps for fast SV (for ease to get great land), but I think everyone agree that great map with possibility to plant 4-5 coastal, mountainous cities (food cargo ships!) would be better.

Yes, it would require endless re-rolls, but actually I got once this kind of pangea map.

I was playing as Korea for HoF record, but unfortunatelly made stupid mistake (lost settler to barbs) and rage quit

Don't ask me, I don't have starting save from this game

 

[unrendered]

Check this chart for AI diplo values: http://civdata.com/

Some good picks include:
- Sweden - low expansion bias, 10% bonus to Great People if friendly. Usually has good military and can be bribed to attack someone
- Netherlands and Arabia - friendly and good for lux trades. Harun can be a bit annoying with religious spam and William sometimes expands too much
- Venice and Morocco - friendly and good for trade routes. Venice will usually buy at least one CS though
- India, Brazil, Byzantium - friendly and low expansion bias
- Portugal - friendly and likes the top part of the tech tree. You can sometimes steal Education or Astronomy from her. High expansion bias though, so she can be an annoying neighbour
- Babylon - I don't know what it is with Nebuchadnezzar, but he does terribly in almost every game I play, hates everyone and becomes irrelevant by Renaissance. Low expansion bias and can be bribed to attack someone

had netherlands, arabia, morocco, india, brazil, byzantium and portugal in my last game LOL. was wondering why everyone was so friendly. i think at one point i had like 6 DoFs until portugal and morocco denounced me for no reason.

with babylon i have the problem that he sometimes doesn't go irrelevant and goes full science and builds porcelain tower at around t150-160 which can screw over you if you're building it in a low production expansion. then again porcelain's kinda mediocre though unless you're doing RAs i feel, so i doesn't matter too much.

 

[consentient]

As we know land is more important for fast SV than civilization; so the question is: can land be so good to allow sub 200t SV for all civs...We keep playing great plains maps for fast SV (for ease to get great land), but I think everyone agree that great map with possibility to plant 4-5 coastal, mountainous cities (food cargo ships!) would be better...Yes, it would require endless re-rolls, but actually I got once this kind of pangea map.

If you (and by 'you', I mean the fast SV players who are posting theorycraft to this thread) are interested in exploring the theory behind the game, and have so far not shied away from playing each other's maps AFTER the player concerned has submitted good results to HoF, why not make custom maps to practise and find out what is most optimal, before you go through the re-rolling process, so as to help you best identify the ideal conditions?

Or, in other words, if you like I can make you a Pangaea map with 5 coastal locations on which you can practice.

The offer is there if you want it.

 

[Blatc]

I'd be up for playing a modified 'ideal' map, as well as discussing the theory behind it.

Coastal allows for cargo ships for higher internal food and late-game higher gold yield. However, some coastal locations offset this with fewer good tiles (i.e. tiles that yield more than just 2 food, post-Lighthouse). Ideal coastal cities would have minimal 2f tiles, so either lots of water resources and/or lots of decent land tiles.

Benefits of the Great Plains map include:
1) the proximity to other civs, for worker steals and caravan plundering, as well as the possibility of caravans being sent your way after you make peace (limited caravan range early game)
2) proximity of city states for some early gold, at least 1 more worker steal, and a few barb camp missions or worker extractions
3) proximity of civs leads to early boosts to science. I find since playing primarily GP maps that early teching is very easy and the Babylonian GS really isn't necessary (if anything, it tricks me into attempting to build Hanging Gardens and/or Petra, and then getting burned on that!)
4) knowing the general layout of the map from T0 and being able to move and settle with this pseudo previous map knowledge

The very best GP maps offer the above AND also provides enough room for 5+ good cities. Not every city needs to be huge, but you want to approach 20 pop in your smallest city. I think this balance of proximity to AI/CS but enough room to expand would be a tricky part of finding the ideal pangaea map, where it's more likely you'll have room to expand but worker steals and gold income are delayed. A great GP map likely has CS and/or mountains acting as a barrier between your territory and the AI, with the edge of the map acting as a barrier on the other side(s): your first 2 expos can act to block things off (especially if Shoshone) and secure territory so to speak, and then further builds fill in behind.

Sadly, I recently messed up a game where I went 6-city (all mountain) NC with Shoshone and got Scholasticism before Renaissance. The key failing was that I didn't steal nearly enough workers early game. In my best games, I'm getting 6+ workers pretty quickly from the 2 closest AI and a CS, and getting them all back and working promptly.

A downside of GP maps are the lack of jungle, which ideally would be in the 3rd ring (or a bit in the 2nd ring) of your city.

In deciding whether to continue maps or re-roll, I've started putting more emphasis on # of Granary and Stable resources available. Along with luxuries (unique + copies), this should be a key element of putting together an ideal map, though coastal cities would also factor in sea resources. As an example:

Great Plains - resources
*riverside hill next to mountain for capital. Several forest to chop out settlers and early buildings is a nice bonus.
*while riverside hill/mountain is ideal, you can still have an excellent game with just 1 of hill or river for the expos, with most next to mountains (sometimes a non-mountain expo can pick up 2 unique luxuries and a bunch of stable resources inc. horses and is likely worth settling)
*2+ stable resources for most cities, preferably 3+ for capital
*2+ granary resources for most cities, preferably 3+ for capital
*1 unique luxury per city average, with 1 or more extra copies or wonders that give happiness per city average, i.e. 5-city empire with 5 unique luxuries and 4 extra copies and a natural wonder that gives 2-3 happiness. Mining luxuries preferable near capital: these can be connected quickly and sold in peace deals and these tiles are great to work while pumping out settlers.
*circus resources for most cities

Pangaea/Coastal
*would help with Luxury variety
*less emphasis on granary resources, as +food sea resources come into play. Note there is a trade off here with the requirement to tech Sailing/Optics earlier than on GP building the Lighthouse isn't bad but building the work boats takes a lot of hammers.
I think for a pangaea/coastal map to surpass GP, it would need strong food/production land tiles (wheat on plains, salt) to get things rolling and account for the extra hammers required for lighthouse, work boats, and cargo ships (more hammers than caravans). Course, we're talking about the best maps possible, so of course it would have these things

 

[consentient]

With at least one player interested, I will get to work when I can.

Korea seems like the obvious choice, no?

 

[IronfighterXXX]

why not make custom maps to practise and find out what is most optimal, before you go through the re-rolling process, so as to help you best identify the ideal conditions?

I appreciate your good intentions consentient, but for me playing such map is kinda pointless because:

a)the optimal strategy on such map is pretty obvious - tradition for even more growth boost (no need to find out what is optimal)

b)it's very easy to exaggerate and make ridiculously OP custom map (which can't be rolled normally)

c)this map won't be eligible for HoF and while it's not exactly well paid activity (set records for HoF) it's still nice to be on list

 

[IronfighterXXX]

I'd be up for playing a modified 'ideal' map, as well as discussing the theory behind it.

Ideal Pangea map imo:

-space for 4 coastal cities near mountains

-the vast superiority of land resources (esp. mining) over sea resources (but luxes always nice)

-proximity of CSs (esp. religious, culturals) for worker stealings, pantheon

-jungle on 3rd ring

-generally all what you describe as advantages of GP maps (maybe except knowing map shape) + fast cargo ships for cap to get pops fast to fully use all this benefits.

Unfortunately this all inevitably leads to:

b)it's very easy to exaggerate and make ridiculously OP custom map (which can't be rolled normally)

EDIT: example of such almost ideal map (3 coastal, mountainous cities): 127 GOTM Korea on prince - T179 diplo victory: http://forums.civfanatics.com/showpo...6&postcount=55

 

[Blatc]

With at least one player interested, I will get to work when I can.

Korea seems like the obvious choice, no?

It may be, but my biased opinion:

I've started playing HoF style as I understand it (endless re-rolls to get great map, no replaying) and made a couple submissions. In playing as Korea, I struggled to ever get LToP or Porcelain Tower, which is a loss of roughly 1 GW, 1 GA, and 2-3 GS. My opinion may be different if I'd put together a perfect start with Korea, but my current view is they're an A- Science civ, behind Babylon, Poland, and Shoshone whose earlier advantages help you get these crucial wonders.

Ironfighter mentioned earlier how important land was; partly as I'm currently on a Shoshone kick, I will suggest that they make the best use of this land. I'm confident that on a perfectly played game on an excellent but NOT altered map I can get under T190; that may never happen, and I acknowledge some luck is involved, but the potential is there. I expect Poland and Babylon can also go sub-T190.

I find the advantage of knowing the map shape on GP considerable. Starting in the West, DoWing your closest neighbours, you can partly seal off and secure an area that will very likely have ruins to use later for Religion. Being >60 faith towards a prophet before T25 without building a Shrine is great, especially if there's a faith wonder.

 

[consentient]

I wasn't talking about an OP map, just a good one with coastal locations.

The method would be as follows:

1) Load Rotate Start positions Mod and roll a bunch of maps
2) Save all the coastal starts
3) Load each one in turn on IGE and check whether it is possible to settle 4 coastal cities before the AI steals the spots (factoring in 3 Scout starts, blocking, etc)
4) Examine the best of these and add mountains and maybe tweak the capital location for resources slightly.

Sure, it won't be eligible for HoF, but then neither is your replaying of each other's HoF maps. It's just practice, and it will give you a sense of how better to do it when you roll your own maps.

@blatc: The reason I say that Korea is the obvious choice is that in the midgame, you will have:

Cargo 1: Busan to Seoul
Cargo 2: Daegu to Seoul
Cargo 3: Gyeongju to Seoul
Cargo 4: Seoul to Daegu

And maybe more...and then Korea will be able to work those specialists a lot easier.

But hey, it's your map, you can choose.