Sandman's Diplo Fastest Finish

[Sandman2003]

Quick question, what are you trading mid-turn when going to IA?

I haven't deliberately researched an optional tech, yet. Sometimes an AI will research monarchy for you which you can pick up while trading MA techs. I usually can still use republic and literature since I only need to trade that to one or at most two of five civs at beginning of MA. If enough AI learn the same tech, and don't know each other, then you will get a good discount on the price. Then there are a bunch of luxes and resources that you have and the AI don't, so it is worth connecting to the AI so you can trade these if necessary. I, of course, sell the AI the initial techs until I have cleaned them out of anything they can pay me. Finally, I am finishing the MA in positive gpt, 4 turn research mode, so if necessary a lump sum payment is possible. I don't think this would be a breech of superslug's double negative gpt ruling, if in the same trade session (same turn) you trade that AI another tech so that you get all your money back. Don't quote me on the last bit though....

 

[Smirk]

Very sneaky SirPleb, I think I'll be using that trick myself.

I'm fully convinced now giving up on small, with only 5 other civs it leaves too much room for error on getting those age techs, not to mention the fewer huts available.

I'm experimenting with stuff other than wet/warm, with the hopes of speeding up both finding the other civs and getting those huts. Also I really don't need lots of land like a milk run so can do without all the jungle covered grassland. At least I theorize...

 

[Sandman2003]

I've played further from my good start today. I didn't use negative gpt trades since they've been ruled out. (As I think they should be.) I did use running at a deficit without penalty to keep my research rate as high as I could. I've reached the Industrial Age at 610BC. I've been lucky to get a free required second level tech at entry to both the Middle Ages and the Industrial Age. With seven rivals it has been easy to get and trade for the first level techs in each era, luck was with me in getting something useful after that. I'm not sure how soon I can get to four turn research in the IA. My Golden Age has ended and I'm showing six turns for Industrialization. I hope to get that down to five.

610BC entry to IA, that is 18 turns in front of my latest effort at the same point, and probably up another 4 based on picking up a necessary 2nd tier tech at IA beginning. Very impressive, as usual SirPleb! I look forward to your full writeup when finished.

 

[LulThyme]

Very sneaky SirPleb, I think I'll be using that trick myself.

I'm fully convinced now giving up on small, with only 5 other civs it leaves too much room for error on getting those age techs, not to mention the fewer huts available.

I'm experimenting with stuff other than wet/warm, with the hopes of speeding up both finding the other civs and getting those huts. Also I really don't need lots of land like a milk run so can do without all the jungle covered grassland. At least I theorize...

Yeah I ve been thinking about something similar (as im trying std russian on 20K to see how faster tech rate helps) making it normal or even arid greatly speeds your scouts...

 

[SirPleb]

I've finished and submitted my current game. Diplomatic win in 210AD I'll do a writeup in a day or two.

 

[Sandman2003]

Diplomatic win in 210AD

That is awesome! I must admit that privately I had estimated the best date at around 300AD, and Ronald's game appeared to support my theory. Yet you have come in and smashed my theory, and my date.

Needless to say I don't think I can beat that, so I am going to be true to my word and try some other diplo categories at the other end of the difficulty scale. I doubt any other difficulty level can match what is achieveable on chieftain though!

 

[bed_head7]

Possibly Sid. You won't get the techs from huts, but you'll get out of the AA fast enough and then from then on you might be able to manage better than 4 turn research as the AI would actually do something for you.

Or maybe Deity to strike a balance. Techs from huts, still help from the AI even though you won't be able to manage super fast researchn alone.

Of course, Chieftan may very well be the best place, but you might be able to get close on those higher levels.

 

[superslug]

I've finished and submitted my current game. Diplomatic win in 210AD I'll do a writeup in a day or two.

I recommend a dedicated thread for your write-up. You have quite a following as a player and the pending (monster) update is likely to spike forum traffic pretty high.

 

[Ronald]

I've finished and submitted my current game. Diplomatic win in 210AD I'll do a writeup in a day or two.

Well, I had the bragging rights for one day

Congratulations, very well done.

I am very interested to compare our two games.

Ronald

 

[Sandman2003]

@bedhead7 - Maybe it is possible on sid to beat chieftain finish dates, but the problem for me, at any rate, is staying alive and with the tech leaders through to the Modern Age.

I am thinking about having a go at emperor and maybe deity, although I am not expecting to beat SirPleb's cheiftain time on those difficulty levels. However, I should be in with a shot for a fastest finish for the specific difficulty level if everything goes according to my plans!

 

[bed_head7]

Oh, perhaps not. I took your statement about other levels to mean no one could match a Chieftan date, and I wasn't sure that was true. I read your stories, and those were all months ago so I imagine you have improved, so I am not even sure that Sid is much out of your range if you get a great starting spot. And as you won't be playing maps that don't have great starting spots, maybe you should give Sid a shot?

Of course, you would be going Archipelago with Byzantines. Or maybe not. At least, that is what I would go for. You have the ocean scout and a chance to have control over trading through AA. Toss the Greeks and Koreans from the game, and you still have six scientific civs in the game and an alphabet monopoly. I haven't tested it out on Sid, so it might not be doable without a lot of luck, but Writing SGL seems like a possibiliity.

 

[Sandman2003]

I am going to give SID a go soon, and I will use mapfinder to find an excellent start. I don't know if I have improved much since my last story though - summer time here brings competing demands on my time. I haven't finished any of the last 4 GOTMs for example, hence looking at HOF in the first place. When I do try SID, I will run with any victory condition, although maybe for nostalgic reasons diplo could be a preferred option. I will probably post a story on it as well! As far as I can tell, pretty much any completed game on SID will find a place in the HOF tables right now to boot...

 

[Sandman2003]

I have been accumulating some emperor level starts with mapfinder. In the meantime, I started wondering just how strong the best diplo dates posted on the in-between difficulty levels were. Intuitively you would expect the best dates achieveable on a difficulty level to form some sort of bell curve as you move from the advantages of chieftain in very chep techs to higher difficulty levels eventually the AI start being useful so as to help reduce the date on later levels.

Looking at Warlord and Regent, after the most recent update, Chamnix stills holds both these positions with a 660AD and a 610AD finish respectively. If my bell curve theory is correct, then it stands to reason that the regent date is somewhat stronger than the warlord date. I decided to test out the warlord date in a trial game.

I reasoned that my best chance would lie in adapting the fast tech strategy from cheiftain to warlord. Of course, you are immediately faced with two issues from the difficulty transition - slightly less useful goodie huts, and a massive 66% increase in self research tech costs.

To me the tech cost rate was the big killer. Especially combined with the fact that negative gpt research will cost valuable buildings, so this isn't really an option on warlord! I also totally agree with SirPleb's comment that an ultra early MA entry is self-correcting. That is due to the time your undeveloped empire takes to research MA techs, you give most of the advantage back.

The answer: Tiny maps. Yes, you will lose a lot on the AA, but a powerful start should ensure that much of this time is caught up in the MA/IA. As this was just a trial, and my first Tiny map game ever, I didn't even run mapfinder for it, but instead manually checked through about half a dozen maps, until a suitable food+river start emerged:

Wheat plus flood plain and enough BGs for a four turn settler factory. No cows, so weak by HOF standards, but sufficient for my trial game.
The AI I played against were Greece, Persia and Babylon to maximise trade opportunities, and I also turned down AI aggressiveness to least aggressive (ok, I was scared of the nasty warlord AI!). Huts did not yield a free settler or a free city, but the first tech (the wheel) did cough up an SGL in 3350BC. At that point I decided to play the game out.

Owing to the lack of GHs in total on a Tiny world, I considered that every tech trade I could do was vital. I therefore held off building the Pyramids while my scouts searched for contact with Persia. Naturally they were the last tribe met, and so the Pyramids were not built until 2710BC. Here's the shortened timeline of key events:
2710BC – Pyramids from SGL, all three tribes met
2470BC – Pop writing
1870BC – Complete the jump to pick up republic, revolt – 5 turn anarchy
1750BC – Pop poly
1700BC – Become republic
1275BC – Learn construction to enter MA
Get full first tier in trade
Theology was our free tech
370BC – SGL from chemistry
230BC – Colossus built, GA kicked off
210BC – Copernicus built
170BC – Newton’s Uni built with SGL
30BC – learn metallurgy to enter IA. Pick up Nationalism and medicine from AI
Free tech = fascism
150AD – Forbidden Palace built
290AD – Declared war on Bab

300AD – Conquered Babylon
320AD – Take Elippi and get gems
370AD – Peace to the Babs
420AD – Get Steel and SGL
470AD – Complete ToE
630AD – Lose Babylon Library
640AD – Lose Moscow marketplace
AI get rocketry and computers, we get fission
660AD – UN victory

A few additional points of note:
1) I learnt code of laws, and deliberately held off popping huts until starting philosophy research.
2) In the AA, I only had to pop one first tier tech, and I was also able to trade for IW.
3) Fantastic luck on entry to MA, getting all the way to Theology for free!
4) Major Doh! moment number one - I forgot to construct the forbidden palace until almost the end of the MA!!
5) War with Babylon was fought because they had the closest coal, and two additional luxes to boot. Also they hadn't even built the infrastructure to connect these all up!! 5 longbows were sufficient to conquer Babylon (coal), and the two lux cities, then peace secured two more cities. The massive defences consisted of two spears in Babylon, and only one elsewhere! I also got them back to gracious by the time the game ended.
6) I wasn't sure whether you would get second tier optional techs as your free scientific tech, until this game!
7) The last two city losses were the last two turns of researching flight when for the first and only time in the game we researched into negative gold - tough penalty. The rest of the time I avoided this by trading with the AI when they had resources (Usually just techs), and converting scientist specialists to tax collectors on the last turn of a tech, until the tech rate climbs to more than one turn, etc.

In addition to the FP, I actually made some iother bungles in this game, such as not building tight enough around the capital (I blame newb on Tiny for that), originally choosing a poor colossus site etc. Overall, while I doubt these selctions will challenge SirPleb's 210AD game, I am convinced that 400AD is achieveable on a properly played map with reasonable luck.

 

[LulThyme]

IIRC for an exp civ, the huts on warlord are almost exactly the same as on Chieftain. I think just half chances for settlers or something like that.
Not sure about non-exp.
Im pretty sure that going to small world would give a better time with 5 vs 3 scientific civs, more starting techs, and more huts, even though the tech rate might be slightly slower...

 

[Sandman2003]

@LulThyme - I just rechecked boogaboo's post on page one of this thread. You are right, for expansionist the odds are exactly the same on GH popping on both cheiftain and warlord.
As to tiny vs small, I am unsure about the relative merit. One supercity can account for a lot of research on Tiny, although the 5 sci civs do give much better odds on age changeover.

At the start of the MA for example, the odds of getting a useful MA second tier tech are:
On Tiny: 2/9 for AI clearing out the firts tier for you x 2/3 for your freebie to not be chivalry for 4/27 or about 14.8%

On small: This is a bit harder, and it has been a while since I have had to do any serious probability calculations, so I could be wrong, but I make it- 50/81 for the AI clearing the first tier x 2/3 again for your freebis to not be chivalry for 100/243 or about 41.2% (almost three times better than tiny!)

Though, if it is only odds you are worried about, then the answer is play enough games and you will eventually get the combination you want (not that I really want to spend the rest of my life playing out warlord games though...). Small also gives more GHs to pop of course, as well.

 

[LulThyme]

Thats a lot of advantage..
Even if you lose say 10-15 turns to slower research or even more, you save on average probably around one tech per age from the more scientific ais and a lot from huts too...
Dont forget that as you increase the diff, AIs will actually research.
If youre lucky they might get invention while you do upper path or something like that.
Now if you get chivalry, gifting to them increases the chance they get invention instead of taht so its not TOTALLY wasted... (though still a bummer for sure )

 

[Sandman2003]

You are making some good points LulThyme. This whole strategy in regard to tiny vs small to me is only a debate for warlord level. My emperor, deity and maybe even Sid attempts will be on standard size with max sci civs.

At warlord level, I have noted that the AIs do do some research, in the Ancient Age . IW can typically be traded for rather than researched or popped. By the MA, they are so far in your dust, even if you gift them everything you have, and wasting time researching monarchy, that they don't help any further.

As far as the value of each free tech goes, I see them as four turns research each, because to have any chance on these scientific quests (the same theories apply to the space race also) you must be at 4 turn research by the end of the age. So in this way you could actually calculate the expected cost in free techs from each reduction in map size, and weigh that against the expected number of turns lost before reaching 4-turn research.

With warlord it is actually slightly more complicated than that, because you need to stay pretty much in positive gold while researching due to building penalties.

I haven't tried calculating these things out yet, so my comments are based on intuitive feel. Maybe I should try a smal to "improve the feel".

 

[LulThyme]

Btw you calculation for getting a useful 2nd tier tech were right.
I also wanted to point out that there is more to this.
For example, even if they dont get the whole 1st tier, you cna still get a 2nd tier tech... So the important thing is that the overall average of useful techs you get is greatly increased...

 

[Sandman2003]

I have done some more calculations (and hope they are right as well). At the entry to the MA you will get either two, three or four free techs. Ie the AI will either all get the same tech, and you are guranteed a useful second tech, or the AI get mono and feud and you pick up chivalry, for two useful techs; or the AI get three and you get chivalry, or the AI get two different techs and you either get the third 1st tier, or you get a useful second tier for three techs; or best of all you get four useful techs.

Odds are:
Tiny: Two 2/27 Three 21/27 Four 4/27
Expected gain 3.074

Small: Two 10/243 Three 133/243 Four 100/243
Expected gain 3.370

Gain in terms of research turns is: 1.184

Which isn't very much really. Of course if you really want all 4, you have to play three times as many games on Tiny to get the lucky streak...

Now I just need to work out the odds at the IA and ModA, plus make some sort of estimation on the advantage in GH popping on small.

 

[Sandman2003]

I am not actually sure what I am yrying to prove with my above analysis. I am also having some difficulty genrating very many two cow starts on Tiny, so I am almost ready to try warlord on small. I might try and see if with enough enticement, the AI might research an MA tech for me, and maybe an early MA war or two to add to the city expansion, and availability of specialists. hmmm