GOTM133 Zululand - Final Spoiler: Game submitted or abandoned

[mad-bax]

GOTM133 Zululand - Final Spoiler: Game submitted or abandoned



This is the final spoiler for GOTM133 Zululand. In order to qualify for reading further down this thread than this post or posting yourself you need to have completed and submitted your game, or abandoned it.

I hope that after a fairly "delicate" opening position that the human player found plenty to challenge them on their way to a fast UN victory. Timing of the GA, getting across the water, preserving your diplomatic reputation and all this with a tribe not ideally suited to this kind of victory condition.

How did you cope?

 

[tR1cKy]

frist

I enter the Middle Ages in 330 BC, with China down to 2 cities and the Egyptian OCC next in my list. In 230 BC, Beijing falls, the Pyramids are captured and China is no more. Moreover, the 2nd leader is generated and it is kept spare for now. Two turns later, Egypt is gone as well and i become the exclusive owner of the whole homeland - because the Greek OCC is on an island SW of the continent.

The next task is to assemble an adequate fleet and ship units oversea to invade India. Meanwhile, i'm researching at full speed up to Astronomy and Navigation. Thanks to the Great Lighthouse, i won't need Astronomy to safely ship troops oversea, but i'll need Navigation to enjoy the full set of luxuries both home and abroad, a goal that i want to reach as soon as possible.

Production is a careful balance between military and improvements. I'm giving equal priorities to libraries and markets, because my #1 goal is to maximize the number of happy people through the empire. Of course research and production potentials are also important, both for obvious reasons. In this regard, i'm moving a fair amount of cities to either fit into the RCP 4 (Shangai, Sparta) or into the second ring at RCP 8 (Canton, Nanking, Corinth, Alexandria, Elephantine).

By 30 BC, my fleet is ready to land forces on the Indian coast. At this point, all the planned relocations have been carried out, all the core cities except for one have a library and half of them have a marketplace while the others are, at various stages, building them. I'm also building universities in Zimbabwe, Athens and Ulundi, the three most popolous core cities. On the research front, i'm 5 turns away from Astronomy.

The Indians aren't a big deal, but there's a complication with them: eons ago, India had sneak-attacked Russia and captured Smolensk, just 3 tiles away from their capital. It would be a waste of time to go there by myself, so i decide to ally with Russia and hope they take care of it by themselves. Meanwhile, in 10 AD France finally researches Feudalism, so i trade for the tech and decide to ship the spare leader oversea to rush Sun Tzu in a former Indian town, in order to be able to upgrade swords directly on the front line.

By 130 AD India is reduced to 2 cities, Smolensk (come on, Catherine!) and another small settlement detached from their former homeland. With an alliance still in place and Russia yet to carry on their task i decide to move against France. They are quite a different beast, with pukemen defending into size 7+ cities and a large cultural potential (read: flips) but my numbers are enough to subdue their cities one by one.

In 190 AD i complete my research on Navigation and, with a harbor already built oversea, i can finally collect 6 luxuries and trade for the remaining 2. Research continues at a slower pace toward Banking. Next turn Carthage completes Invention and i happily trade for that tech. One turn after the Russians finally manage to take over that crappy settlement near Moscow but i have to wait for the alliance to expire (or for them to sign peace) to take over what's left of India.

The near totality of attacks against France is carried over with elite units, and during the battle of Paris my quest for the 3rd leader is finally over. Orleans is moved one tile NE, to fit into the center of the new RCP 4+8, and the Forbidden Palace is rushed there. Too bad for Sistine Chapel, which is lost forever (read: who cares?).

In 280 AD i take over the Indian OCC and in 320 AD France is gone as well. Besides the capture of the Greek OCC, which has to be performed in the early Industrials, my military campaigns are over. At this point i've just researched Banking and, while my original core is pretty well developed, my second core is obviously lacking on almost everything, so i decide to keep research at a minimum rate (to Gunpowder) in order to spend the necessary cash into developing properly that area. I'm still 160 tiles below the Domination limit, but, with a mix of cultural expansions and settling, that gap will progressively shrink to just 2 tiles in the early Industrials, and with 99% happiness.

In 420 AD Russia completes Gunpowder. I buy that tech immediately and resume research at full speed up to the Industrials, which are reached in 620 AD. Meanwhile, Zimbabwe has built Copernicus (by bare hands) and is building Newton's (also by bare hands) while Athens is prebuilding for Theory of Evolution. Russia and Greece are gifted to parity and they pick Medicine and Steam. Alas, the price required for Steam is outrageous, to the point that it would hamper my rush to Electricity, so i decide to research Nationalism, in an (ultimately futile) attempt to have the AS research one more tech besides Replaceable Parts.

As soon as Nationalism is known i trade for techs, attack the Greek OCC, get a disastrous SOUR (you need at least one to have a spoiler worth its name ) and then finally destroy them next turn with the help of further reinforcements. Research goes to Electricity, followed by Scientific Method, followed by the upper branch up to Combustion, followed by the lower branch up to Radio.

Alas, after completing their optional tech (Economics, and i was expecting it), Carthage decided to go for Sanitation, and Russia was simply too slow to research Replaceable Parts, so for the time that tech was known it was clearly too late to hope for the AS to complete either Mass Production or Flight in a useful time, and i had to research everything else by myself. Even so, i was able to keep 4-turns research during the whole Industrials, with the sole exception of Atomic Theory completed in 5, and also to not lose any turn by timing Evolution with the completition of a research. Moreover, the added benefit of hospitals allowed me to build a few selected metroes and squeeze in a few more happy people.

Progressively, slaves were joined into the new core cities and, later on, the majority of the national workers followed suit. One by one, all my non-metro cities, including the few hopelessly corrupt ones, were brought to size 12 with full happiness. At the end of the game, the base score provided by people was slightly higher than the one provided by territory, and with a tile count that was kept less than 10 tiles below the Domination limit for a whole era.

With only one scientific AS left, i set up two prebuilds for the UN, one to be completed at the projected date for reaching the Modern Ages and one to be completed 5 turns later. I choosed two cities with mountain tiles in their radius, to be able to tune up the production per turn without losing population and/or the WLTKD status.

In 1100-1110 AD interturn i'm into the Modern Ages. Russia is gifted to parity but fishes Computers, so i switch the Palace prebuild to Smith's and the Hoover Dam prebuild to Palace. Five turns later, Fission is researched, the Palace prebuild turns to the UN, just in time to complete it, and i score a Diplomatic victory in 1160 AD. The end date isn't great, but the score... well, let's just say that to win this competition you have to set a new all-time high for Diplo victories, and not by a small margin

Very enjoyable map, from the beginning to the end. Thank you very much mad-bax for having set it up.


Here's the research log from the Middle Ages onward:

_330bc: Engineering (trade, Russia); Monotheism (trade, Greece);
_190bc: Theology (research);
__90bc: Education (research);
__70bc: Monarchy (trade, somebody);
__10ad: Feudalism (trade, France);
__70ad: Astronomy (research);
_190ad: Navigation (research);
_210ad: Invention (trade, Carthage);
_320ad: Banking (research);
_420ad: Gunpowder (trade, Russia)
_460ad: Chemistry (research);
_500ad: Physics (research);
_540ad: Theory of Gravity (research); Chivalry (trade, Carthage);
_580ad: Magnetism (research);
_620ad: Metallurgy (research);

_660ad: Nationalism (research); Steam Power, Medicine (trade, Russia);
_700ad: Electricity (research);
_740ad: Scientific Method (research); Economics (trade, Russia);
_780ad: Industrialization (research);
_820ad: Corporation (research); Steel, Refining (ToE);
_860ad: Combustion (research);
_910ad: Atomic Theory (research);
_920ad: Sanitation (trade, Russia);
_950ad: Electronics (research);
_970ad: Replaceable Parts (trade, Russia);
_990ad: Radio (research);
1030ad: Mass Production (research);
1070ad: Motorized Transportation (research); Communism (trade, Carthage);
1110ad: Flight (research);

1160ad: Fission (research); Computers (trade, Russia); Military Tradition (trade, Carthage);


And here's the final kill log:

1100bc: Greece is OCC
_710bc: Egypt is OCC
_230bc: China
_190bc: Egypt
_280ad: India
_320ad: France
_680ad: Greece

 

[greatbeyond]

The end game was really anticlimactic until the UN vote came up. After clearing out all the AI in my half the world I settled in to keeping the AIs happy and building research and happiness. I gave no thought to invading the other continent assuming that something would go wrong with the UN vote as a result. I made contingency plans to gift an island city in case any of the AIs were in danger of being eliminated.

At the end The UN vote was 4 for me with Carthage voting for their self circa 1776AD.
This was much later than I thought I might come up with and in a replay I did manage a victory roughly 300 years earlier. However It pales next to tRicKy's results.

What was the secret? I didn't trade away any recently discovered techs to the AI and completed the UN while they were still several techs back. In fact I held off on the ToE and ended up taking Radio and Flight with it. If I had been going for space, I probably would have held out for Rocketry and Computers. I also finished most techs in 4 to 6 turns near the end. Perhaps trading with the AIs would have been better.

The early sneak attack by China certainly slowed things down. I wonder if anyone coexisted peacefully with the AIs on the home continent?

 

[Lanzelot]

and during the battle of Paris my quest for the 3rd leader is finally over.
...
Russia and Greece are gifted to parity and they pick Medicine and Steam.

Lucky bastard...
I didn't get an MGL in the entire game, and both Russia and Greece drew Nationalism, a tech which I didn't even bother to buy. That may account for the 10 turns difference between our games: I finished in 1260 AD.

Otherwise our games went pretty similar, as I already noted in the first spoiler. Only difference towards the end: I used ToE to pick Radio and Fission as my free techs, instead of Steel and Refining. You could have saved another turn, if you would have done the same, right? (But you were hoping for Russia or Greece to pick Fission, in which case you would have saved 5 turns, right? I had been thinking quite hard, whether I should try that or rather take Fission with ToE, but after the disappointment with Nationalism, I got so disgusted, that I just eliminated Greece and decided to "play it safe"...)

 

[Lanzelot]

To answer your question, greatbeyond: with neighbors that close, I made "proper defenses" an early priority. That delays any sneak-attack ambitions they may have. An attack came anyway at some point, but I was prepared, just took the war happiness and their cities...

 

[tR1cKy]

Lucky bastard...
I didn't get an MGL in the entire game, and both Russia and Greece drew Nationalism, a tech which I didn't even bother to buy. That may account for the 10 turns difference between our games: I finished in 1260 AD.

Well, in a certain sense But you know, besides what i've written in the spoilers, i've been sneak-attacked three times, two by Carthage and one by Russia. In all cases, i've limited myself to kill off the intruders and then i signed peace as soon as possible. Plus there's the final attack against the Greek OCC. All these attacks have been carried over by elite units, and if you add up all the elite victories of my previous campaigns you get a number around 60. Statistically, 3 leaders are less than you can expect from such a big amount of elite wins.

About the picks, i've done a small research to see how much the random factor really impacted on my score. I've used the base score of the remaining autosaves, added the victory bonus resulting from that particular turn, and put the result into the jason score calculator. Well, the results are surprising: turn after turn, i was slowly gaining jason points rather than losing them.

 

[tR1cKy]

Is there a problem with the results? Normally, they would be out by now.

 

[mad-bax]

No - No problem. Real Life is intervening somewhat at the moment.

I have published the results now, and will put together the results thread as soon as I can get to it.

I apologise for any inconvenience.

 

[tR1cKy]

No need to apologize at all.

However, the results seems just too high to be credible. Not only the top score (13.6k, come on!) but even my own score, by a comparison to previous games, seems unwarranted.

My opinion is that, for this particular game, the MaxScore factor of the Jason formula has been largely underestimated, and that all the scores need to be scaled down at least by a 10%.

 

[Più Freddo]

I had similar thoughts. It seems, to the layman, as if the factors for the Jason score haven't been calculated or selected correctly for this game.

Which sent our friend Tricky, like Goethe said, "Himmelhoch jauchzend, zum Tode betrübt." (Heavenly joy, deadly sorrow.) Dostoyevsky countered in The Brothers Karamazov with something like this: "like a weak youngling he had fallen to the ground, like a warrior armed for life he had again risen".

 

[Lanzelot]

And I was so happy that finally some people had been able to achieve a score of over 10K with a non-Conquest/non-Domination victory condition...
(Remember our discussion a couple of years ago...) Usually this happens only very seldomly with single extraordinary games.

 

[mad-bax]

There is only one human input, which is the ocean cossing pre requisite. Even this is checked when the file is processed and you are notified of any difference of opinion.

The issue, if that is what it is, is that the Jason formula contains assumptions about the scenario which do not necessarily hold for customised games. For gotm scores should only be used as a measure of comparative efficiency in reaching a particular vc. One certainly should be careful in comparing different games.

I will double check the crossing parameter anyway.

 

[Più Freddo]

Yes, there is no doubt about the ordering of results. I see it more as an awe-full expression of the sublimity of the game.

 

[tR1cKy]

I'm not questioning the order of results either. Drazek's gold is well deserved, as anyone could see by itself by comparing the two games in the replay utility.

on a side note, this guy is like a ghost, he never bothers to post any spoiler, and this is kind of disappointing because such games, IMO, do really deserve some more detailed insight, but hey, you can't force people to post spoilers if they don't want to

Back to the point, from what i understand of the Jason scoring system, there is a "MaxScore" factor unique to every game, and although it is ultimately generated by a formula, nothing would prevent the "game master" from adjusting it by hand when such factor clearly falls short of giving healthy results because of the limitations of the formula itself.

I wouldn't bother at all about this "doping effect" on the Jason score, if it wasn't for the all-time higher and faster table. It's basically impossible to beat a 13.6k score, unless, some day in the future, another map with the same doping effect will materialize out of the RNG chaos. This kind of defeats the very purpose of the all-time highs.

For the same reason, i would support a re-evaluation of the scores set in the ancient GotM 15 (Han Dinasty) where the player started with 2 settlers instead of one. Obviously, that is another case in which the formula, by its very same structure, fails to take into account some peculiarities of the game that have the effect of an added bonus. The 13k top conquest is clearly doped, as it is the 920 AD Spaceship victory, that is also furtherly doped by the use of the faraway palace exploit.

Of course, i can happily live and keep playing even if no action is taken at all

 

[mad-bax]

The Jason scoring system has been used for a long time now, and right or wrong it would be difficult to change it now. There is no appetite to change it amongst the staff either.

A discussion was started about it some time ago and the thread was universally ignored, so I assume that the playing community don't care much either.

This scoring system is just a model, which seeks to avoid games milked to turn 540 getting the highest score. Life is too short after all. This it does successfully, but not perfectly.

Using the firaxis scoring method, the highest scores are map dependent. Some maps will score more highly than others. The same is true of the Jason system.

It is not really possible to inspect a map and decide what the highest possible score would be, and then normalise it to some arbitrary number. What is possible is to adopt a new assumption. For instance, we could assume that the highest scoring game was perfectly played, and generate a factor which would convert that score to a consistent arbitrary number - say 10,000. That factor would be applied to every score. If we had enough players, you could apply that logic to each VC.

Reality is that a 12K game on one map is not necessarily as well played as a 9K game on another map.
Also, A 12K domination is not necessarily as well played as a 9K 20K culture game on the same map. By inspection, it is clearly easier to get a higher score with some VC's than others.

So, whichever way you cut it, and whatever model you choose, because it is not possible to calculate the true best score for any map and any victory condition, you can only compare your score with other scores on the same map, played to the same victory condition.

For me, both the Jason and Firaxis methods are meaningless, arbitrary models. Fastest finish dates are at least a measure of efficiency, but even then, the best played games of chess are not necessarily the shortest.

If I had my way, there would be no scoring whatever, and everyone would play to the same victory condition. Players would be ranked by finish date only.

End of diatribe.

 

[Più Freddo]

I think playing for score has a merit. Having lots of territory and happy citizens is reasonable goal, as is reaching some limit as soon as possible. That's why we have awards as well as medals.

 

[mad-bax]

I think playing for score has a merit. Having lots of territory and happy citizens is reasonable goal, as is reaching some limit as soon as possible. That's why we have awards as well as medals.

I understand PF. As I have said many times, I just build the maps. You lot decide everything else. My opinion FWIW has less value than anyone elses.

I still think that Jason is just a race to the domination limit plus the theoretical number of turns required to get to the chosen VC. So we may as well play domination every game.