SGOTM 12 - One Short Straw

Looks like my suggestion for an extra month paid off for you guys. I take full responsibility for your win. :D:D

Well done guys!
 
Thanks, everyone! :)

First impressions of the Ducks' final save:
  • They built GLH in the Fish-city equivalent in 625 BC (this is 8t after Roosevelt in our game).
  • This seems to have driven them to plant a whole bunch of island cities early - one on the horse island north of Moscow, two in place of Bahamas, 1-tile island west of Stalin, all of Bermuda, Jamaica, Cuba and Yaroslavl'. They also planted Silver before Siberia.
  • They took forever to settle the GP-Bananas-Pigs area and didn't use any of GP, Pigs or Hammer (which were 3 of our most important cities).
  • Moscow got OU+NE+TGL.
  • Fish is exactly the same, except it also has HE.
  • Siberia is IW, but not other wonders and watermilled instead of workshopped.
  • Coal plants everywhere, no 3GD. They apparently Libbed AL before corps, so this makes sense.
  • No SoL, HS or MoM.
 
I did as you said when I left Fifth Element. I tried to make them lose but they just wouldn't listen to me... Maybe FE shouldn't have listened to me either! :lol:
Nothing bitter on my side, Mitch.
It's just that you're goin' to win the Gold (let's wait the official results) after you left FE.

But we have had a very good time this game. Definitely Dhoom's style was too much for our taste.

BTW we started thinking to the GLH, probably the most powerful wonder on this map.
But not even the time to discuss it. Roos built it on turn 61. And in NY, not even in Washington.

Then i've tried many starts, repeating roughly the same opening and this never happened.

Also, not a single wonder in France. I expected to capture some... nothing.
 
Congratulations, guys! I've been following for a long time. Among other things, your tech stealing was amazing and I learned a lot. And that photo-finish with the Ducks was real drama! :crazyeye: Actually, with so many teams pushing it so close to the end, it's been non-stop entertainment for us in the peanut gallery!
 
If the French built you a shrine that is a lovely Paris to take! woot.
 
From Plastic Ducks:
This game again is showcase how cottage is overrated. We laid down 1 cottage in capital and OSS has 0!
:yup: We didn't put a single cottage down anywhere all game. To be fair, there was a bit of a heated debate about cottages somewhere in the middle of the thread, but my tireless cottage-hating rants were too much for the others to bear. :D
My impression from watching both games is that OSS has done an amazing job of MM to get the absolute optimal immediate result out of every individual action (even when such a decision caused them to deviate from a longterm aim), whereas PD was much more focussed on keeping all actions consistant with the main plan, and not nearly as worried about optimizing the direct payoff of every action.

In many cases, the OSS discussions left me with the feeling that they could not see the forest for the trees.
:rolleyes: Please... You can't win a 1750 AD Space Race on Emp, with a fallout-gimped start to boot, without a sound strategic approach. Saving a few worker turns won't do much. What we did is consciously keep our strategic and tactical options open throughout and adjust the approach as each TS gained us further knowledge of the game. I think this is the primary reason for our great success in this game and last. It's LowtherCastle's influence entirely and he, in turn, learned it from klarius by his own admission. He now has gold medals with two entirely different teams for a reason.

This takes tremendous team effort at constant analysis and the discipline to completely reconsider painstakingly developed plans when they begin to seem sub-optimal compared to the alternative.

Just look at the way the endgame developed - we kept minimizing the tech path and infra builds every mini-TS, as it became more and more apparent that we can finish faster than originally anticipated. Had we stuck with a pre-imagined plan for the last 20-25t, we'd probably finish 5-10t behind PD instead, as kossin had predicted.
 
To be fair sometimes I looked at the trees rather than the forest, but I knew the others would be able to decide if the tree was in the right forest or not :)
 
I would be surprised if someone could get a better result than us without the GLH. Probably with advanced knowledge it could be done (but it would be tough).

Some of the criticism I've read that might have some merit...
1) We got Sushi before Mining (although I think at the time we decided to get Sushi first because there were espionage timing issues)
2) We didn't get Assembly Line early (might have set up a better Liberalism tech?) But again I think we were optimizing espionage and the steal of Communism and hoping someone would tech it for us and thus get it on the cheap with espionage.
3) We didn't make GP Farm/Pigs Production powerhouses. Well those cities could have become big time production cities but not with the aggressive GP plan we had. Near the end we probably could have diverted some workers to convert them, but by then it didn't matter too much
4) We didn't build enough infrastructure (grocers, markets, harbors?) with the whip
5) The gift of Galley City was a very good "mistake" but giving Gandhi those extra techs with the trade probably did accelerate Roosy. Which may have been the difference in killing Roosy early and not.
--But an accelerated Roosy with the GLH teched what for us (Replaceable Parts?, Economics, Corporations, Riflery, Democracy, (Constitution), what else am I forgetting?, banking, guilds (we lost the GM but got those techs at less than 0.25 cost with steals)


Other criticism/speculation
6) Ducks went for Roosy instead of churchill. I think Churchill even though he was a harder target was the better choice for us because Roosy teched a bunch for us whereas I doubt Churchill would have come close to helping as much as Roosy did (since he had the GLH).

7) We didn't switch to Free Religion as early as ducks, but again those GAs really did pay off more than the Free Religion bonus might have. We didn't sacrifice that much running specialists compared to workshops. Scientists at 100% science bonus are better at the end then grassland workshops during a GA with 100% hammer bonus. And those spies we were running were even better. Engineers aren't too far behind. The priests weren't that great but we only ran those in 2 cities (London and Paris).

8) The biggest missed opportunity was an earlier switch to communism. I think we missed out on ~7 turns of State Property since we could have anticipated Roosy taking one of Mao's cities and we could have taken Xian in LC's turn. And thus could have stolen Communism and switched to State Property in the GA as he predicted.

9) Ducks were critical of Versailles build. I guess we have to look at the numbers again but even if Versailles only netted 15 pre-inflation gold per turn it pays off in ~20 turns.

I wonder if we were dedicated to the GLH build that the AI might have delayed the GLH build (like we tested with the Oracle--the AI know when someone has started a wonder and this seems to influence their tech choices and willingness to go for the wonder themselves)
 
I was going to ask you guys if you thought LC's Galley City "mistake" helped or hurt the finish date. It certainly led to more useful AI's, but probably forced you to lib Bio instead of AL.

I thought that the way you leveraged that "mistake" was one of the best examples of the principle that bbp is talking about: a good game is not about laying down the best road map from beginning to end, but rather weighing strategic options, and then iteratively optimizing gameplay as the changing situation narrows those options.
 
How did you play out the Espionage? It seems to me you needed:

- a decent research partner AI who can be steered into techs that you can afford to wait for (FDR not Gandhi?)
- a whole bunch of Espionage multipliers to make it worthwhile, which pushes Democracy earlier
- and possibly a Great Spy (how to generate when GPs are also needed for Academy, shrine, bulbing, Sushi, Mining Inc., etc?)

That seems like a lot to add to the existing space race priorities. I personally felt it was better to suppress the AI so that they'd be backwards and wars would be quicker/cheaper, but obviously spying worked out well!
 
Galley City was a great mistake, IMO. :D TBH, had LC not done that, we would have worked out a variation on it anyway (either another settler to gift or a big tech soon after). This way it worked out great, because the city's primary function was as port towards France and we could still use it that way.
 
How did you play out the Espionage? It seems to me you needed:

- a decent research partner AI who can be steered into techs that you can afford to wait for (FDR not Gandhi?)
- a whole bunch of Espionage multipliers to make it worthwhile, which pushes Democracy earlier
- and possibly a Great Spy (how to generate when GPs are also needed for Academy, shrine, bulbing, Sushi, Mining Inc., etc?)

That seems like a lot to add to the existing space race priorities. I personally felt it was better to suppress the AI so that they'd be backwards and wars would be quicker/cheaper, but obviously spying worked out well!

We knew early on that the other AI had advanced technology that we wanted to steal.
--so a decent research partner AI really wasn't required to benefit from espionage this game. It turned out Roosy teched quite a bit for us and with relations as difficult as they were and the reluctance of AI to trade monopoly techs, the espionage steals from him worked out even better than we thought they might.

We knew that we wanted corporations this game and espionage would be important so we built courthouses faster than perhaps a normal game. And we ran spies as much as we could even without espionage multiplier buildings the spies were better than scientists since we ran representation (so +4 :espionage: and +4 :science: is better than +6 :science: usually)

We captured a lot of espionage buildings (Beijing for example came with Intel Agency and Scotland Yard) So we really didn't build too many espionage buildings except in cities with Scotland Yard or where we needed a Mongrel GP (spy,priest or engineer)

The passive espionage from numerous courthouses, and captured espionage buildings and spy specialists in many cities ended up being enough to steal almost all of the advanced techs the AI's started with plus some more that Gandhi bulbed and Roosy teched for us.

We never did use a Great Spy for infiltration mission but it was discussed. We ended up with enough great people for ( Academy, shrine, 3? bulbs, Sushi, Mining Inc., 5 GAs, and 3 late GS bulbs basically wasted for extra points and Laser Tech) We worked hard to generate the great people and ran pacifism for a large part of the game (slavery for quite a while then switched pacifism and caste system, quick dip into organized religion when we needed to whip a bunch of buildings, then pacifism, then only near the end Free religion for the last few space ship technologies)

Nationalism was also a very useful tech for us this game (for the +25% espionage bonus and drafting mostly although the low cost of the civic made it competitive with Bureaucracy especially with such a large empire) The +2 :) from barracks wasn't bad either.
 
How did you play out the Espionage? It seems to me you needed:

- a decent research partner AI who can be steered into techs that you can afford to wait for (FDR not Gandhi?)
- a whole bunch of Espionage multipliers to make it worthwhile, which pushes Democracy earlier
- and possibly a Great Spy (how to generate when GPs are also needed for Academy, shrine, bulbing, Sushi, Mining Inc., etc?)

That seems like a lot to add to the existing space race priorities. I personally felt it was better to suppress the AI so that they'd be backwards and wars would be quicker/cheaper, but obviously spying worked out well!

We knew early on that the other AI had advanced technology that we wanted to steal.
--so a decent research partner AI really wasn't required to benefit from espionage this game. It turned out Roosy teched quite a bit for us and with relations as difficult as they were and the reluctance of AI to trade monopoly techs, the espionage steals from him worked out even better than we thought they might.

We knew that we wanted corporations this game and espionage would be important so we built courthouses faster than perhaps a normal game. And we ran spies as much as we could even without espionage multiplier buildings the spies were better than scientists since we ran representation (so +4 :espionage: and +4 :science: is better than +6 :science: usually)

We captured a lot of espionage buildings (Beijing for example came with Intel Agency and Scotland Yard) So we really didn't build too many espionage buildings except in cities with Scotland Yard or where we needed a Mongrel GP (spy,priest or engineer). And Security Bureaus from Democracy were never built. Democracy was stolen but that was only for the Statue of Liberty.

The passive espionage from numerous courthouses, and captured espionage buildings and spy specialists in many cities ended up being enough to steal almost all of the advanced techs the AI's started with plus some more that Gandhi bulbed and Roosy teched for us.

We never did use a Great Spy for infiltration mission but it was discussed. We ended up with enough great people for ( Academy, shrine, 4? bulbs, Sushi, Mining Inc., 5 GAs, and 3 late GS bulbs basically wasted for extra points and Laser Tech) We worked hard to generate the great people and ran pacifism for a large part of the game (slavery for quite a while then switched pacifism and caste system, quick dip into organized religion when we needed to whip a bunch of buildings, then pacifism, then only near the end Free religion for the last few space ship technologies)

Nationalism was also a very useful tech for us this game (for the +25% espionage bonus and drafting mostly although the low cost of the civic made it competitive with Bureaucracy especially with such a large empire) The +2 :) from barracks wasn't bad either.
 
Galley City was a great mistake, IMO. TBH, had LC not done that, we would have worked out a variation on it anyway (either another settler to gift or a big tech soon after). This way it worked out great, because the city's primary function was as port towards France and we could still use it that way.

It was a great gift and since it was surrounded by fallout and our cultural, Gandhi ran specialists in this city and gave birth to 2 early Great people here and bulbed Philosophy and Theology? I believe for us. This and its close proximity to our capital allowed us to steal those techs for much less than it would have cost us to research them ourselves. (We would never have teched Theology ourselves, but stealing it was so cheap and that opened up the Hagia Sophia, a wonder that really helped in this game).
 
That seems like a lot to add to the existing space race priorities. I personally felt it was better to suppress the AI so that they'd be backwards and wars would be quicker/cheaper, but obviously spying worked out well!
I think our wars were quick and cheap enough. ;)

What bcool said about espionage. Gandhi had bulbed Philo early, giving us Theo/Philo/Nat/Comm/Const/Steam + typical trade fodder like Guilds/Banking/RP as sure things by the time we made the decision to invest in spying.
 
I would be surprised if someone could get a better result than us without the GLH.
Me, too. The timing involved in the Communism-to-Physics era was quite complex to predict. I think we could have had less warfare stalling in the Orleans-Gaungzhou era with a few more units, which is something that could be further optimized.

1) We got Sushi before Mining (although I think at the time we decided to get Sushi first because there were espionage timing issues)
2) We didn't get Assembly Line early (might have set up a better Liberalism tech?) But again I think we were optimizing espionage and the steal of Communism and hoping someone would tech it for us and thus get it on the cheap with espionage.
I think the espionage timing (and the fact we were so gung-ho about stealing Steam Power from Churchill) led to this. As did the early investment in CH.

Ducks didn't research Medicine until 1550 AD, at which point we had Sushi in all significant cities and are growing them well into 20's. By 1590 AD, their 100% research is 3500bpt and our 50% research was 3200bpt (we break 4200bpt @ 100% with -100gpt more than them, but no significant cash savings). We are also ahead by the cost of Industrialism in overall research at that point. I think that the sheer quantity of Sushi resources and the fact that the game is actually quite a bit shorter than it appears as late as 1500 AD make Sushi-first, large cities and lack of late game infra whipping quite viable. Needs further examination, though.
 
Back
Top Bottom