SGOTM 12 - One Short Straw

While we're on holiday...
Duckweed said:
While you have been looking at the total culture produced a turn, I usually pay more attention to how much more culture increased during a time, which could tell you the most important factor of the game. I'd like to see you guys play to the best results.
Here it is:
Code:
DATE		SCORE	CULTURE	CPT
180 - 1200	2,127	10,215
192 - 1320	2,527	12,730	210
194 - 1340	2,719	13,435	353	
202 - 1420	3,248	16,025	324
206 - 1460	3,712	17,350	331
208 - 1480	3,875	17,350	0	
212 - 1510	4,358	18,620	318
218 - 1540	4,403	21,155	423
223 - 1565	4,666	24,410	651
228 - 1590	5,096	32,290	1,576
233 - 1615	5,969	49,710	3,484
239 - 1645	6,813	79,385	4,946
244 - 1670	7,931	116,125	7,348
250 - 1700	9,031	166,110	8,331
260 - 1750	10,724	258,420	9,231
T206-208: 2t revolt into FM+
T194-212: drop off in cpt is odd. Most likely it's new and/or newly-captured cities hiring and firing artists?
Not sure what else he means, other than probable usage of culture slider in the end game. Sushi should give somewhere in the order of 4-5k per 50 cities, and they must have 500 in base culture, if not more. Our total commerce in GA is currently 1200c, so theirs must be approaching 2000c in the end. That would be about 20-30% slider, assuming all cities with Sushi, but could be more without 50-city Sushi.
 
They might have captured the sistine chapel which would add a lot of culture too since they would be running 10+ specialists in a lot of cities.
 
What I find interesting is the HUGE spike in power at the same time. Power is a function of military units, some techs, some buildings and population (every even pop in each city gives just 1K, so this can't account for too much of it). Do you think they built that many units near the end to make their power spike like that? It seems like overkill (i.e. taking a bazooka to a fight when a pea shooter would have done the trick).

EDIT: I guess modern warefare does add a lot of power:

18000 soldiers – SAM Infantry, Artillary
20000 soldiers – Infantry, Stealth Bomber
22000 soldiers – Mobile SAM
24000 soldiers – Jet Fighter
26000 soldiers – Gunship, Mobile Artillary
28000 soldiers – Marine, Navy SEAL, Submarine
30000 soldiers – Paratrooper, Tank, Panzer, Destroyer, Stealth Destroyer, Attack Submarine, Tactical Nuke
32000 soldiers – Mechanized Infantry
40000 soldiers – Modern Armour, Battleship, ICBM​
 
Culture: I'm just really curious about his "most important factor in the game".
Good point, bcool. That might be part of the T244 spike.
Similar - Broadway/R'N'R for the happy.

Power: :dunno:
3 mil soldiers is really a lot for this kind of game. The more interesting thing is just how consistent that graph is - they seemed to build up at a steady rate throughout the late game.
One thing comes to mind - they may have boosted Gandhi and Roosevelt much more heavily then we did, for trading/stealing (resulting in a need for a big modern army). I think we've actually slightly out-expanded them to this point, and beaten them to Sushi. Yet, they're slightly ahead in score - which presumably means tech lead.
Another thing: Nationhood and barracks everywhere - we've talked about this. 50 cities = 150,000 soldiers.
 
CURRENT FALLOUT COUNT = 178
Spoiler :

RUSSIA:
55 in borders + 4 soon-to-be + probably some to flip in America and Coventry

AMERICA:
33 in borders + 2 on the island NE of Seattle

INDIA:
36 in borders

OLD RUSSIA:
4 (1 in Moscow + 3 in neutral territory on StPete-Nov island)

ENGLAND:
11 (all in Coventry)

FRANCE:
0

CHINA:
0

NEUTRAL TERRITORY:
33 (not including previously listed in America/Russia)
Spoiler :

2 (crab island N of Old China)
6 (island N of Moscow)
3 (island N of Yaroslavl')
2 (2-tile island E of Iron Spice)
1 (coal island)
6 (island SE of Novgorod)
4 (crab island SE of coal island)
4 (SW of Coventry)
3 (island east of London)
2 (island W of Yekaterinburg)


UNEXPLORED TERRITORIES:
  • Island b/w London and Yekaterinburg
  • Island N-NW of Old Russia
  • Island-ocean zone W of India
  • Island NW of Seattle
  • Island N of Seattle
Most of these will require land units or satellite to view, btw.


In counting them, I did this:
Spoiler :

Civ4ScreenShot2609.JPG

Civ4ScreenShot2610.JPG


Worthwhile, or later? Kinda messy - I'd find it hard to play. Didn't save, but didn't take very long either.
 
Some general thoughts (feel free to trash):
  • (1) After AL-Astro-Physics we have no immediate need for any techs (I think). Industrialism is obviously nice to have, so we see just how screwed we might be for Aluminum (one under Stalingrad?), but not necessary just yet. Should we just invest max in corp spread after this?
  • (2) We need to re-figure our gpp now.
  • (3) We need some kind of analysis of cities, with a plan for production and research multipliers, as well as happy buildings that are worth doing prior to wealth building. This is the really difficult bit, I find.
  • (4) Some kind of useful analysis of research rate and gold loss over the next 30-40t. Is that possible to do accurately?
  • (5) Longer term Which cities are building SS parts, and what sort of production capacity do they have?
  • (6) Longer term War plans for Stalin and Roosevelt / fallout-clearing plan.
  • (7) What else?
(2) If we were doing division of labour, bcool could take this over - since he had it most figured out previously?
(3) I started a summary of cities today, but had no idea how to usefully organize the data. Then I remembered this:
Something like this for each city would be useful (at least for me):

Moscow
Target Population: 25 (currently 21 w/+7:food: and -1:yuck:)
Primary function(s): Production, Military, Science, mutt GPs
Corporations: Already has Sushi. Needs Mining ASAP.
Great People this city will generate a great person on T211 and again for the 5 person GA.
Infrastructure: all science multipliers, all hammer multipliers (including levee), all health buildings, we should never reach the happy cap unless WW gets really bad
General: work all improved tiles but make sure not to get too much OF, hire spies/engineers then fill in with scientists as long as the city can still produce a future great person, after that, just run scientists

I don't know if the above is true, but having something like this would make all of our planning much easier. Maybe this is too much to ask...

I'll work on it, but it won't be until Monday in all likelyhood. Would this help you guys too or should I just shelf it?
I've spent some time today looking at cities. How about this:

MOSCOW
Current population = 24
Current health = 25
Current unhealth = 25 (-1 Forge)
  • Hospital = +3 (200h)
  • Recycling Centre = +1 (Forge) +4 (Fact/Coal/RI) (300h)
  • Factory = -1 (250h)
  • Coal Plant = -2 (150h)
  • Research Institute = -1 (250h)
Final health = 29
Current happy = 30 (including +2 from Nationhood)
Current unhappy = 24
  • Theatre = +1 +1/10%
  • We don't really wanna spend hammers on stuff like temples and colosseum
Final happy = 31+slider
Food surplus working all good tiles = 19fpt @ Pop 19 (will increase with more Sushi resources)
Target population = 29 (if we get more Sushi food, and other cities don't justify slider, just nix stuff like grass farm, etc.)
I think we wanna build Fact/CP/Hosp asap. That's a total of 600h, which Moscow can do in exactly 5t if it starts now - (62h OF + 92h GA turn + 320h 4 non-GA turns)*1.25. Or, slower if we really need the wealth-building to finish Physics.

Someone else can improve the graphics on the above. :D

(4) I'm gonna try to look up formulas for city maintenance, civic upkeep and inflation, and wire some spreadsheets.
Spoiler :

Civic upkeep:
Current cost = 190 (63 Rep + 0 Nat + 64 Caste + 63 FM + 0 Pac)
With the same civics, we're looking at an increase of approximately 1.42gpt per every new city and 0.36gpt per every new population point. This doesn't include inflation, though. So, for something like 50 cities and 600 pop (compared to current 37 cities and 394 pop), we'd have 282gpt (compared to current 190gpt) * InflationRate.

Distance maintenance:
To follow

Number of cities maintenance - couldn't find formula. :dunno:

Corporate maintenance: need cities' target populations for this.
It's a function of population and number of resources.
With current resource number +1 pop = 2.71gpt pre-CH (i.e. 1.35gpt with CH)
With current population +1 reso = 18gpt pre-CH (9gpt with CH)

Unit maintenance - should remain fairly stable, aside from inflation, ofc.

Don't quite know how to estimate research rate.
 
Yes. I'll take it. :)

On a related note, I have a competition project due on Jan 7th. Most likely will take all my free time over ten days leading up to it. Ideally I can play a solid TS asap this week, and then I'll have to disappear a bit.

An initial PPP for discussion shall be posted later today... I think getting to Physics is almost a mini-TS all its own. It'll be easier to plan longer term once we're back in GA, hopefully for good.

Edit: LC, can you play the next?
 
Here is a more detailed plan for the ggp production. Let me know if you guys have any questions. I tried to put in comments to explain. Also put in comments about what specialists to run and estimated when we would have buildings built that allow more specialists.

I made a plan that doesn't depend on the GE from Fusion (which might mean we get too many engineers, but we need to run more specialists in our mongrel cities so this is a calculated risk)

I made educated guesses on the odds of what great person might be born, but didn't actually calculate them. We are going to have to live with 5-7% chance of getting screwed again, but I have noted where we could spike a city for backups if we do. Beijing would be focused on a Great spy for example, but we can spike the number of specialists with GS or whatever starting T227 to give us another shot at a great person of the type we need.

Also we are going to have to be careful not to run too many specialists in cities like Rheims so they don't pass our mongrel cities.

It is getting to the point we might have to check specialist numbers every turn and pause every time we don't get a great person that we expected.
 

Attachments

So, you're proposing:

3GP GA:
Current GM
Current GE
Physics GS
Start circa T220-1

4GP GA:
Current GM
1800 Mongrel Paris (GP/GE/Spy)
2000 GS Moscow
2200 GA GP Farm
Start circa T232-3

5GP GA:
2400 GS Pigs
2600 Mongrel London (GP/GE/Spy)
2800 GM Fish
3000 Mongrel Bahamas (GP/GE/Spy)
3300 GA GP Farm
Start circa T244-5

Maybe Moscow is better focusing on hammers in the next little while (infrastructure and then either wealth of more execs). Could we pop a GS in Pigs at 2200, and then one in Moscow later? It would presumably give us better GS odds in Moscow and allow Pigs to work on a 3600 faster.
 
ECONOMY
Spoiler :

Research cost to complete: 1033b (AL) + 6240b (Phys) = 7273b total = 6061b base

Current treasury: 616g

Current stats: 695bpt @ 0% w +822gpt; +166bpt w -153gpt per 10% slider

Current costs: 1902gpt
Current tax revenue: 1444gpt @ 100% gold
Total revenue (pre-multipliers): 1701gpt @ 100% gold
Approximate multiplier effect = 1.25x, not including Wall Street

Approximate loss of commerce without GA: 200+ base > 1207 base down to 1000c
very roughly, this should equal 1/6 difference in above numbers, so:
Drop in tax revenue = -300gpt @ 0% slider, or 30gpt per 10%
Drop in research rate = 28bpt per 10%

Approximate stats non-GA: 695bpt @ 0% w +522gpt; +138bpt w -183gpt per 10%

Wealth builds:
Moscow = 116gpt
Fish = 26gpt
Siberia = 191gpt
Cuba = 12gpt
Bahamas = 42gpt
Shanghai = 23gpt
The Crabbes = 7gpt
Hammer City = 100gpt

City captures: 2 immediate
Cash trades: unlikely?

Finishing Physics on T+5

T216 50% slider > 1525bpt w +57gpt > Research left: 4536b Treasury: 673g
T217 40% slider > 1247bpt w -210gpt > Research left: 3289b Treasury: 453g + capture cash (300g?) = 753g
T218 50% slider > 1385bpt w -393gpt > Research left: 1904b Treasury: 360g
T219 30% slider > 1109bpt w -27gpt > Research left: 795b Treasury: 333g + capture cash (hopefully)
T220 0% slider > 795bpt (hire more specialists) w +522gpt > Research left: 0b Treasury: 855g+
T221 Start golden age

This requires keeping up the wealth-building and not spreading corps, though. We could probably either stop wealth-building in Moscow/Siberia/Hammer one at a time, or spread a few corps, but not max out.

General question: we need to rush to Physics. Do we give up some turns getting Factories and Power if needed? Which cities out of Moscow, Siberia, Hammer are more important to set up?
 
Maybe Moscow is better focusing on hammers in the next little while (infrastructure and then either wealth of more execs). Could we pop a GS in Pigs at 2200, and then one in Moscow later? It would presumably give us better GS odds in Moscow and allow Pigs to work on a 3600 faster.

I actually like Moscow going early because it has decent odds for a GE, Prophet. If we get one there we can save it and use the GS in Pigs.
I gave Moscow 3 turns at only 5 specialist during the gap in the GA to build infrastructure / mining execs.

By the way Pigs can accelerate its GP production a lot I think it has 22 sustainable specialists, I had to slow it down a bit, so it can produce a 3600 great person if necessary no problem.

By the way I think specialists are the key to getting physics on schedule. All ocean tiles (except in Bermuda) should be converted to specialist now (we lost the Colossus and many cities don't need the ocean tiles to grow anyways with Sushi). Especially all cities that likely will not produce a great person can convert them, the ocean tiles, to merchants/scientists based on the buildings they have.
 
Gandhi War

Very loosely:

T216
Bombard Vijayanagara and Agra (how many hits needed?)
Take Vijayanagara
Take Agra if possible (probably not?)
Galleys from Agra drop back to Newcastle for pick up

T217
Take Agra if not already
Drop off on Madras island
Move towards Delhi

T218
Take Madras
If there are spare units from Madras, load galleys for Varanasi
Move towards Pataliputra
Move towards Delhi
Drop units aboard galleys on hill Delhi-NW

T219
Drop some off at Varanasi, pick up others from Madras for drop off next turn
Arrive at Delhi wheat

T220
Bombard Delhi
Take Varanasi

T221
Take Delhi
Bombard Pataliputra
Move from Varanasi towards Pataliputra
Probably CF

T223
Redeclare
Take Pataliputra
Take Bombay with Delhi stack

Question: any clever way to cut off reinforcements to Delhi from the south?

Edit: maybe the above isn't gonna happen, though. He does have quite a few units milling around, and I don't really know what to expect for our unit loss. I think we ideally wanna take 5 cities asap, including Delhi (so we can work a hill town :lol:). Might need a short break to regroup before that's done.
 
By the way I think specialists are the key to getting physics on schedule. All ocean tiles (except in Bermuda) should be converted to specialist now (we lost the Colossus and many cities don't need the ocean tiles to grow anyways with Sushi). Especially all cities that likely will not produce a great person can convert them, the ocean tiles, to merchants/scientists based on the buildings they have.
Yeah, I agree. A bit hard to estimate, though. When you say "on schedule", what do you mean? T220? T221?

I actually like Moscow going early because it has decent odds for a GE, Prophet. If we get one there we can save it and use the GS in Pigs.
Clever. Ok, we'll do that.
 
Back
Top Bottom