G-Minor XCVII

[Morcar_olmig]

Cro,

Thanks for sharing this! You seem to tend to capture city states even if they are cultural and founded on hills. Is it really worth it or you sell them later?

To be honest - i think this is option to consider - instead of building settler you just capture it with your military (which you build anyway and you don't halt your city growth). But if we compare the cost of the settler and the courthouse - it's 500 vs 600 in gold or 106 vs 100 in hammers. Looks like almost the same, so i think that's probably worth it if CS occupies nice spot and not cultural

 

[Bleidraner]

Well just tried 4 cities Tradition to t150, PP135, gonna miss Atomic by 2 measly turns, and with 2 votes this will take me over 200 so will retire and try again. But this is a much stronger attempt than any of my previous early conquest ones, so. I am convinced (and more so after Vadalaz's game) that this is the way to go. I would like to try for GL next time, but on a 4 cities Trad I cannot find space in the Build queue for it. Needs to be done before t55.

 

[Cromagnus]

Cro,

Thanks for sharing this! You seem to tend to capture city states even if they are cultural and founded on hills. Is it really worth it or you sell them later?

To be honest - i think this is option to consider - instead of building settler you just capture it with your military (which you build anyway and you don't halt your city growth). But if we compare the cost of the settler and the courthouse - it's 500 vs 600 in gold or 106 vs 100 in hammers. Looks like almost the same, so i think that's probably worth it if CS occupies nice spot and not cultural

It's true, the rush-buy cost is comparable, but 4gpt in maintenance is non-trivial. Honestly, I don't really value cultural CS all that much for warlike diplo. You only need Tradition, Rationalism + 1pt in Patronage. That's really not so many policies. I tend to prefer to capture faith CS first, then military, then cultural, then maritime, then mercantile. As to why I went after that CS on the hill, well, I had to take what the map gave... it was the only CS that bridged the gap to an AI. When 3 of your 4 cities are planted by the AI, you have to go after the one in the good spot even if it's a tougher capture. In general, I go after CS because on Emperor the AI starts with units, so until siege towers you have to wade through units. If your first capture is a CS, it's quick, and that gets the ball rolling. Plus you can usually get a worker + a tribute (often another worker) before capture. Once you have siege towers you can just ignore the AI's units, and they'll disappear when you take the capital anyway.

I'm probably not going to have time to finish out that game, or try any more attempts, which is a shame. I rather enjoy the fun of playing Assyria the way they were intended!

 

[Bleidraner]

Yes same here. i have a good game (looking 185ish) going but will not finish on time. These games where multiple strategies are viable almost deserve a Major.

 

[Cromagnus]

Yes same here. i have a good game (looking 185ish) going but will not finish on time. These games where multiple strategies are viable almost deserve a Major.

Agreed. I don't think anyone predicted that a no-brainer strategy for the major would be discovered so quickly, or that the games would go so quickly. Normally a Marathon game would take much longer. This Major actually took less time than a Standard game, and once there was a single, undeniably best approach, that was that. I didn't even bother to submit once I realized how simple the winning strategy was. I prefer maps where luck is less relevant than effective play, or at least where luck isn't SUCH a factor. Like the Minor, where multiple strategies were viable, took time to explore and refine, and it wasn't a 150-turn difference between the obvious best strategy and the second-best. Oh well, can't blame the gauntlet choices though, I wouldn't have predicted it would end up this way either. But yeah, I could really have used a whole month for this minor, and I was bored of the major after a week.

 

[vadalaz]

I thought that outlining the optimal strategy in the thread for the G-Major was pretty fun. But I agree that once it was discovered, there has been little incentive to keep playing. Going for a sub-100 is the last bit of motivation there I guess, but yeah it's just about getting a lucky map roll at this point.

Still trying to figure out how to play this one though. I think in the perfect scenario you build Great Library for Theology and steal Civil Service from someone who beelines it, then go for a ~t90 Edu... Or even steal both Theo&CS and GL Education. Now that would be awesome, ~T80 Education on Emperor as Assyria. Doubt that's possible though, AIs don't tech that fast and that's a really late GL anyway.

edit: a more realistic plan would probably be to steal Drama, research Philo and finish GL all at the same time. Probably around t55-60. I think Emperor AIs should have Drama by that time. Then tech up to CS and hopefully steal it, though it's a long shot. Stealing Currency or something like that along the way could work though.

 

[Bleidraner]

I think what Cro means is that outlining the strategy on the GMajor was fun... while it lasted. Once the optimal strategy was outlined and executed, best time was a combination of absolutely optimal play (every movement counted) and map luck, so once you get a result you think demonstrates your own limits, that's it for that gauntlet. As the optimal strategy was a 100 turns long, it was over pretty quickly. But who could have guessed a Marathon culture game was gonna be done in 100 and nothing turns?

On this one, the minor, we definitely have not explored the limits of the gauntlet. There is little doubt that the optimal strategy is a late war playing a tradition game, although there is some doubt. Also, the Tradition play is also unclear, 4 cities, 3 cities GL, does late war mean t80 or t60 and so on.

IMHO, both were a lot of fun, but this one needed more time and the other, no so much.

 

[Cromagnus]

Yeah, Bleidraner was right in his interpretation. The Minor at least left room for debate about which technique would win, although I think in the end, it became clear to me that the "1 founded, 3 captured" approach isn't viable. Or at least it's too reliant on map luck. Simply put, the odds that you will capture 3 cities with near-optimal placement is very low. Even on a bad map you can usually find 3 decent expo spots. And in the end, the amount of re-rolls necessary to get a perfect set of captures (and ideal tech steals) makes the strategy less viable, even if it could compete. I think it can in theory compete if luck runs your way, but it's too reliant on luck.

However, "2 founded, 2 captured" and "3 founded, 1 captured" are all actually quite viable IMHO. You need the army anyway, and you can usually find one CS + one capital, or one AI expo + one capital, that are worth capturing. The ability to plant one city of your own to bridge the gap is key.

In my experimentation with "1 founded, 3 captured", I discovered the most reliable way to play the map was to move your settler 4-5 times in one direction and plant on the best spot you find. This *usually* gets you close enough to an AI capital and a good CS. And it ensures that you're the near-optimal distance/placement for long-term growth. I haven't entirely given up on that approach. After all, I consistently was able to achieve ~t106 Edu via 3-captured and 2-captured approaches, which leads me to believe that with more refinement I could have posted sub-t100 Edu eventually.

However, the truth is, I'm sure I could consistently do better than that with 4-founded.

But, there's something really gratifying about stealing 10+ techs and achieving fast Edu despite teching Mathematics before Writing. :-D

Anyway, fun Gauntlet guys and grats to the winners!

 

[Denniz]

Gauntlet Results:
G-Minor XCVII - Diplomacy, Emperor, Standard, Lakes, Standard, Assyria (Ashurbanipal)
1st vadalaz - 1230 AD T-183
2nd Manpanzee - 1430 AD T-203
3rd Bleidraner - 1540 AD T-218

Congratulations!