SGOTM 11 - One Short Straw

[Silu]

Since we are at a point where deficit research is imperative and the game has finally really started I guess it's about time to start doing some math on how much and when to beg from the dudes we're at Pleased with, possibly synchronizing some civic and religion switches to get that Pleased status for begging. I'm at work now but could do it tonight. This might be more important than many might think, as the hard-cap on the amount of gold we might get during the course of the game hovers around the same numbers as a GM trade mission or two, depending on our power.

Required info:

- Turns known for each Civ. That I can mostly dig out of turnset reports unless someone has a handy-dandy list made about that.
- If we have made any begs. I suppose any of those would have been mentioned in reports, but still. A failed beg is as important to know as a successful one. I assume none have been made so far?
- Knowing graphs greatly increases the accuracy of begging amount math.

 

[LowtherCastle]

Silu: WOuld you be so kind as to give us the GM Trade Mission formula?

My testing shows that an overseas target gives a bonus. ToA gives a 300g bonus. The population of our capital and the target city gives a bonus, with our the pop of our capital seeming to be weighted more heavily. DIstance seems to not be a factor.

 

[Silu]

It's directly proportional to what the trade route yield would be from our Capital to the target city, so normal trade route bonuses such as being intercontinental or the ToA affect it. Also naturally it's capped. I'll get the exact formula when I get home. Should actually enable accessing the source and XML from work, as things go now I have idle 15-20 minute waits all the flippin' time...

 

[Mutineer]

Great Merchant

For starters, I now agree with the GM next. For one reason: It will allow some cities to build missionaries instead of gold for 100% research.

Note: Building gold in cities without research modifiers is an excellent alternative/supplement to the gold from a GM.


-------------

This problem is very complicated. I had a similar discussion with jesusin in one SG. He thought in terms of BEAKERS. I prefer to think in terms of TURNS.

We have about 50-55K beakers of techs to research. If we had unlimited GPs, we just bulb them. If we had unlimited beaker production in our economy, we just research each tech in 1t. But we don't and that's why this is complicated. We have a limited number of GPs and the max beaker production of our economy is limited. Somehow we have to optimize that and I think the key is to think of turns saved.

Spoiler :

When our economy produces 1250bpt, GS saves 2 turns. When it produces 125bpt, the GS saves 20 turns.

The next GM will give us about 10K beakers of research (I think), but it will take about 12-18 turns (I think) to use all that gold, because our max economy is limited.

That is correct. That why there was no question, GS was better for education (instant beaker). So optimum solution depends, what we can produce more:

If WE can produce more GP then GM needed for continued deficit research, then additional GP come as GS's and used as direct bulb. If not, then maximum effect come from just using GM's for continued deficit research.

Additional bonus come from merchants, producing gold themeless. Problem look complicated, but in reality simple, produce GM's, send on missions and use only when we running out of money. If we starting to collect GM's, switch to GS. We can do more precise calculations as we come near end.

If we keep AI ignorant (we really should not give them tech, the more backward they are, the bigger tech we can take from liberalism and more sure they stay in religion and easier to war them if we forces to), then difference in tech we used liberalism on cower most part of additional tech we could have potentially traded.

Personally I believe we should consider just be-lining to Mass Media at that point, try tech that speed up our research(Improve capital)

Which tech are on our way that improve our capital?
Engineering (Castle, additional trade route if we build in capital), Printing press (more commerce from villages and towns)

I only consider other tech for value of

Literature- National epic. I am not sure we can get any value out of GL, because we absolute it pretty soon.

Calendar - Resources and Mom. Golden age wander does not absolute.

Misic - Free GA

That is it. Other wise I see be lining to Printing press(come try engineering I believe) and then Sci method try astronomy and then directly to Mass media.

 

[Mutineer]

Correction, by some reason I was thinking engineering needed for Printing press, but in game help file sad no, just machinery is enough. In that case we can trade for engineering when AI get it.

Same for guild banking, trade when AI get it and research economy for free GM.

 

[Silu]

Same for guild banking, trade when AI get it and research economy for free GM.

Dangerous. A lone AI having researched Banking won't probably be willing to trade it to us (monopoly whine), possibly losing us the Eco GM.

 

[Mutineer]

Dangerous. A lone AI having researched Banking won't probably be willing to trade it to us (monopoly whine), possibly losing us the Eco GM.

Yes, will need careful observations.
He will need education for that. Just watching is he researching education (How mach he will give for it) will tell as if danger exist.

 

[LowtherCastle]

More thoughts:

1. We will get Radio from Liberalism. No worries there. The AIs are drastically behind in research. I don't see any chance to get Guilds+Banking for free if we don't gift them MC+Mach. I'm not sure how many AIs need to know Calendar/Compass for Justy, etc., to trade them.

2. We are really not far from Friendly with Justy, Izzy, and Zara. The key is sharing Buddhism. Toku and Mehmed will take more work. We might be friendly with all of them by the time one gets banking. We don't that GM for a long time anyway.

3. If we're running lots of merchants in Two Fish, then the market looks better.

 

[Silu]

I'm not sure how many AIs need to know Calendar/Compass for Justy, etc., to trade them.

I can compute this, but as it is quite complex and leader-dependant (no, it's not just the percentage you can read in the XML as some still seem to think) I'd like to wait until a real application scenario for it (i.e. "How many need to know <THIS TECH> for <THIS LEADER(S)> to trade it") before crunching the numbers.

 

[Silu]

Trade Mission gold formula (oh boy) for Epic, Standard:

(500 + 200*TRADE_YIELD) * 1.5

TRADE_YIELD: MAX([MIN([OUR_CAPITAL_POP * 50] , [PLOT_DISTANCE * 50]) * 0.2], [100]) * (SUM_OF_MODIFIERS)

I could list all the modifiers mechanics but it's a ton of stuff... Basically, things that up the amount are high target city population, whether it's connected to capital, whether it's overseas, whether it belongs to another Civ than us, whether target city has the ToA, and the peace duration between us and them. The TRADE_YIELD variable is the same as the actual trade yield in that's displayed in cities. Apparently there is no hard cap, though there's a "soft cap" - the distance doesn't matter beyond our capital population. So if, say, our capital is size 12, distance doesn't up the trade mission amount anymore beyond a plot distance of 12 (keypad directions 2, 4, 6, 8 = 1 distance; 1, 3, 7, 9 = 1.5 distance). For Standard maps.


Legend tells that when you actually have the GM you can check the mission yield for any visible city on the map by queuing a move there with shift and then checking the mission button that appears, never got it to work myself though.

 

[shyuhe]

You can check GM yields for cities that are on the same continent as where the GM currently is. I'm guessing we can expect ~2000 gold for a GM on epic, possibly more.

 

[babybluepants]

Prioritized solution:
1. Do not poprush Two Fish, Rice, and Marble below pop3.
2. Finish the Marble lh asap after the poprushing cycle.
3. Finish the current Rice farm.

That should all be fine.

Any way to get Confucianism into Marble asap?

Dark magic?
I don't see how. We need to be in Pacifism all the way through. All our cities are occupied with important builds.

 

[LowtherCastle]

Trade Mission gold formula (oh boy) for Epic, Standard:

( 500 + 200*TRADE_YIELD ) * 1.5

TRADE_YIELD: MAX([MIN([OUR_CAPITAL_POP * 50] , [PLOT_DISTANCE * 50]) * 0.2], [100]) * (SUM_OF_MODIFIERS)

Thanks. This explains the 300 jumps. I think the formula should have parentheses for Epic, see above. Also note that the +100% for ToA does not show up in the Trade Yield we see on the Governor's Screen, so that adds 300 to the formula. So we have these (realistic) jump points:

1950
Mecca <pop11 ... Delhi <pop15

2250
Mecca <pop11 ... Delhi pop15
Mecca pop11

2550
Mecca pop13 ... Delhi pop17
Mecca pop14 ... Delhi pop16

2850
Mecca pop16 ... Delhi pop18

These hold unless we somehow or another get a few more % from factors not shown.

 

[Silu]

Yeah, it should have parenthesis, sorry about that. Wouldn't make much sense otherwise

 

[Silu]

Estimated safe values to beg, as gold, AI not in financial trouble, as of current turn 157, assuming at least Pleased (further diplo bonuses don't matter):

ZY: 550g
JUSTINIAN: 70g
OTHERS: well under 50g

Okay, seems it isn't quite time to start mass begging yet, except from ZY. Because there's a huge, stochastic cooldown between begs, we should definitely beg from ZY instantly if we can get 150+g unless we want to aim for begging a tech. Should perhaps also beg from Justinian if there's a loose 50g to be had. This isn't an exact science so don't cut these thresholds close. Also all this assuming we have made no successful or unsuccessful begs so far.

And naturally if you beg, also document unsuccessful attempts (yeah I know, I'm kinda repeating myself ) and include the exact amount of a successful beg in a report.

Calculated from this data:

MET TURNS:

ZY: 3
JUSTINIAN: 59*
ISABELLA: 118*
TOKU: 118*
SALADIN: 125
MEHMED: 128

* = some discrepancy, turnset report not clear

CURRENT TURN: 157

Power ratios:

JUSTINIAN: 0.5
ZY: 0.9

 

[LowtherCastle]

Marble

A different solution. The Rice worker cancels action now, rds marble 1t, chops Marble forest. Finish the lh. Work the marble tile at pop5 (Rice best grows at max food anyway, including desert 3f, imo, but then 2f1h before coastal). LH completion is timed perfectly with growth to pop6. Lib and uni pops are no problem.

 

[babybluepants]

Just checking our GP numbers, and I'm pretty sure we could actually pop the GE next (circa 200 AD), assuming Two Fish needs to re-grow asap. I'd guess we don't want that - GM should be next. The GP after that probably should be the GE? It lets us run some scientists in Delhi fairly soon.

@Silu, what's the tech:gold ratio there?

Edit:
@LC,
ok, that sounds good off the top of my head. I'll post a detailed plan tonight, couldn't get around to it yesterday.

 

[Silu]

4. Mecca will expand to the gold hill in about 100 turns.

This comes a bit late, and not like it matters much, but if you were talking about our Gold hill, Mecca needs 100% (Legendary) to reach it, Baghdad needs 80%. Mecca also needs legendary for reaching the other settling possibility tile Gold. For the third Gold near Damascus, we'd have no chance to control that with Damascus popping borders in ~2 turns and having a doubled Madrassa soon. I think the Crab Gold was much better than the Fish Gold.

EDIT: Beakers are in the neighborhood of 1.5 value per beakers (increased or decreased slightly with the amount of Civs that know it, I forget which just now, sorry), and gold is usually 2, so a gold piece is 33% more valuable than a beaker.

 

[LowtherCastle]

Garrisons

1. Unfortunately, Ivory will grab the last Rep +3, so Marble needs a garrison in 7t or less. Fortunately, we can just send the Rice warrior there immediately. Rice won't need it till pop9 at the earliest.

2. In addition to clams to Mehmed for 2gpt, I'd also gift him our copper (separate trade). Gets our resource counter to Mehmed doubled PLUS we have spare turns/production to build warriors in Bombay for now. One garrison may need to go to Oasis.

 

[LowtherCastle]

Further Exploration

Not for this turnset, but before I forget. We still need to explore the barb continent. This could be done by a mace, to gather 10XP for HE, but he might get clobbered. Another possibility is a spy. I like this variant as it could spend the rest of the game reconnoitering the furs and barb units. It would require a free galley for transport.

This map info is vital asap, imo, because we still don't know about using our 2 DoWs. We should strongly consider capturing Aksum. It has +19f, super fast growth and we could us it as another GP farm. Plus, it has a bunch of developing cottages. The DoW would only give us a negmod with Justy.