SGOTM 15 - Kakumeika

mabraham · Mar 25, 2012

WastinTime said:
MythBusters!

I think I've read at least 4 times now from almost every team member how "highly inefficient" whipping is compared to working a GH. And were using a quote taken out of context from DaveMcW:
"It is always inefficient to kill a citizen working a mined grassland hill.
At size 6, it becomes inefficient to kill off a mined plains hill.
..."

I have great respect for his posts too, but I've had strong doubts about this one. Our blind faith in this alleged myth is poisoning our decision making.

Our empire is working only about two Gmines right now (we only have 3 complete!) and we've been whipping pretty freely, so if there's poisoning I'm not sure we've seen it yet

WastinTime said:
Let me define the myth : "You are always better off working GH instead of killing a citizen no matter what size your city is! Even if you have a granary!"

Not only is this a myth, but in most cases, it's actually the opposite. It's more efficient to whip even if you DON'T have a granary.

First off, his use of the word inefficient refers to the break-even point for total hammers. So 30 hammers received instantly from slavery is considered equal to 30 hammers gathered over 10 turns. This can be far from true. It is often HIGHLY efficient to have those hammers 10 turns sooner. However, I'll give him the benefit of the doubt and put NO value on early hammers. For example, let's say you have 1 turn left on a Library and
whip it. You don't get the library any sooner.

Sure, the analysis has to include the utility of having the whip-hammers now. The analysis also needs to take place well after the whip, so that the un-whipped population has time to show its value. The size of the happiness cap is also relevant. Neglecting any of these factors leads to an unbalanced analysis.

WastinTime said:
But that's not the biggest problem with this myth. It only applies to the case where you are working a GH with your very last food. (i.e. your city is now stagnant). The myth holds true for this one specific case where your city would need to grow back at +1 food. An only in the rare case where there is no value to getting your hammers early.

So the key to efficiency is related to the excess food/how fast you can re-grow.
If you have a granary, and if after you kill a citizen, you have only +2 food, it is still more efficient to whip!

Common Example: with granary, 36/72 food: You're working Corn + minedGH. Building a settler on marathon speed with 90 hammers to go (10 turns).

Compare 5 turns later....

No whip:
you would have no settler and still have to wait 5 more turns.

With the whip:
You have already re-grown to size 2, you've had your settler all this time, AND you've been building something else (at 1 hammer/turn)

Myth...Busted!

This example has a fully-improved city whipping at size 2, which is the most efficient size for whipping (fewest food to grow new population to convert to hammers). That's pretty artificial. You can get successive settlers out faster with 1-pop whipping such a city. You can also do that with successive settlers 2-pop whipping a similar 4-pop city *if* you have both Gmines and Gfarms available. By the time you get to size 6, at normal speed with a settler at 70/100 and a half-full food box and enough Gfarms and Gmines, whipping can get out settlers turn 1, 5, 11 and 17 compared to 2, 7 13, 18 without whipping... but there's 27 turns of whip unhappiness stacked up and that becomes limiting soon.

All the above pre-supposes settler spam, which have the benefit of being able to whip multiple population. For spam of 35 :hammers:

axes, whipping is poor because you can only whip one 30 :hammers:

population at a time, unless you can contrive to get at most 4 hammers of overflow. Spamming something like maces is easier, but the happiness cap is a critical factor in the length of time that constant whipping can stay ahead of non-whipping.

There are secondary effects with whipping or not - you have to pay more maintenance for larger cities, but having more population is useful in a some AI interactions, or bulbing techs.

Whipping does require more improvement infrastructure, because you need to re-grow on food tiles and want to work mines at the target size. This is where tile-sharing shows its strength. In theory, Delhi and PigGems could run whip cycles half out-of-phase with each other. While one is re-growing the other is using the mines.

bcool · Mar 25, 2012

PPP micro feedback

I think CowTown should whip its granary T80 (I know it isn't an ideal time but I think it is worth it in this case)
CowTown can put the whip OF into the pyramids T81
T82 CowTown starts an axe
T83 Pyramids (if it seems safe enough while Delhi builds the settler, axe is started so can whip or chop a forest if it gets ugly)

Delhi (forget the library in Delhi, getting the pyramids and representation benefit is better than the library)
T80 Pyramids.
T81 Settler (to avoid growing into unhappiness and allow CowTown to build the pyramids)
T82 Pyramids.
T83 Settler

I might have more to say, but I've been pretty sick and don't have much energy to do detailed testing.

Also nothing in the PPP on military unit movement. I think the axe on the move should be sent to the forest near DeerGold rather than SheepTown (where it seems to be heading)

mabraham · Mar 26, 2012

Walter_Wolf said:
There is some level of uncertainity of how the things will develop considering we are at war with 2 AI's, and some things need discussion, so I am posting a rough draft of the micro plan for the next turnset for the team. I would add the missing rest of the PPP tomorrow and add/change the things the team feels need to be improved. I might even play a few opening turns in 24 hours from now (if we get a consensus about them) to check the AI moves, and eventually make some changes to the PPP according to what happens.

Walter_Wolf said:
The goals for this turnset are:
Build an exploring trireme and try to sneak through hammuragawa's borders.
Build a settler in Delhi at size 6 till whip anger decreases.
Settle the crabs/banana/marble site with it.
Setup a "threat event" possibility for Tokugawa with our military presence in a 5x5 city radius around Osaka and Tokyo before peace negotiations for better peace treaty odds/demand options. A road 1S1E of Sheeptown is needed for this (I am not sure about this anymore).

The suggestion was for a road 1S1W of Sheeptown, to faciltate sneaking an axe to 2S of Sheeptown to get the chance of a better peace deal. However, I'm expecting to have to pay for peace if we want it... is Sailing possibly large enough? Do we have any tech to trade short of MC? Are we afraid enough to care?

In short, I wouldn't rush for this road yet.

Walter_Wolf said:
Tech

Finish research writing, and put slider to 0% untill we have a library or two.

Yeah. 0% writing, then 100% writing, then whatever will finish it. Then 0% maths.

Walter_Wolf said:
Cities

Delhi
T80 - build settler working 2GH mines, FP farm, corn, copper mine and GL forest (or a priest specialist)
T82 - switch build to library untill growth to size 7.
T84 - growth, add GL mine (from pigsgems). Switch back to settler.
T86 - continue library build
The settler might also be whipped (with OF into the mids or something) if we are OK with it.

I like bc's suggestion for whipping CowTown. As a side effect, it means Delhi can have the Gfarm T80 onwards. If PigsGems cooperates by giving up the Gmine T83 (in favour of river plains, which ekes out enough hammers for its whipping plan) we can do the following in Delhi (corn FPfarm copper throughout) to eke out the settler T86:
T80 pyramids (2mine 1farm)
T81 settler (2mine 1farm)
T82 pyramids (2mine 1farm)
T83 settler (3mine 1farm)
T84 settler (3mine 1farm)
T85 settler (3mine 1farm)
T86 new build (either settler or pyramids, not library)

Then we either rush Pyramids for the Rep bonus, or get out another settler working priests (available T93 either for fish-marble site or peace-gift 3E of CowTown) to finish Pyramids and pop 100% GPro around T97. Since settling the marble-crab site is likely to slow the tech rate slightly (working marble quarry is OK for us though) I'm hoping that we can line up Maths chops around T95 onto Pyramids

Walter_Wolf said:
Deergold city
T80 - continue building axeman working deer, gold and farm.
T81 - growth, add GL forest.
T82 - switch GL forest to GL mine.
adding mines/farms after growth.

Yep

Walter_Wolf said:
Sheeptown city
T80 - continue the granary build, working sheep and gold
T82 - (switch build queue to library or finish granary?)
T83 - growth, add GL mine
T87 - build library
The corn flip timing is unknown, but if it happens in this turnset, I'll switch to it immediately.

Yep. Add a note to consider whether Cowtown will want the extra Gmine as Sheeptown switches to corn.

Walter_Wolf said:
Cowtown city
T80 - continue granary working deer, cow and delhi's GL farm
T82 - growth, add GH mine
T83 - build axeman till the war is over. Some whipping is possible, or a 2 pop whip of an axe with max OF into the mids for failgold maybe.

Prefer bc's plan.

Walter_Wolf said:
Pigsgems
T80 - build granary working pigs, gems and GL mine
T81 - build trireme
T84 - whip trireme. OF into a workboat working pigs and gems.
T85 - build workboat.
T86 - growth to size 3, whip workboat, work pigs and gems.
T87 - build 2nd workboat.

Yep

Walter_Wolf said:
Units

Settler from delhi settles the green dot from the test game screenshot (1N of crabs that is).

Eiffel
T80 - finish the mine 1S of cowstown.
T81 - move 1S, build mine.
T83 - move to 1S of deer, build mine (for cowtown).
T87- move to 1W of gold, build mine.
I've tested if a road on the hill 1S1E of sheeptown helps with threatening toku, and the axes still need 2 turns of movement to get there ('cause of crossing the river), so mining is Eiffel's job for now.

Yep, caveat the comment pertains to a road 1S1W of SheepTown and Toku is likely to squat on 1S1E anyway.

Walter_Wolf said:
Fritz
T80 - move 1E, mine
T82 - move 1N, mine
T86 - (move to 1N1E of pigsgems, build mine)

This T86 mine is completed by Hoover and Karl on T82 below.

Walter_Wolf said:
Hoover and Karl
T80 - move to 1N1E of pigsgems, build 1T mine each
T81 - move 1N, build (finish) mine.
T82 - move back 1S, finish GL mine.
T83 - move to 1S of Deergold city, build the road.
T84 - move 1W, mine
T86 - move 1W, build farm

T83 do not put a road 1S of Deergold city (where there is no advantage until we build a lot more roads), put it on the hill 1S1W of DeerGold. Roads on diagonals are normally preferable.

I suspect we want another Gfarm ASAP (and to switch DeerGold to work it, maybe setting up an OF whip for Pyramids), so H&K can
T83 road on the hill 1S1W of DeerGold
T84 start farm 1S2W of DeerGold.

We can squeeze out that 5-turn farm one turn earlier if Fritz helps with
T82 move 1N, mine and wake
T83 move on roads to 1S1W of DeerGold, mine and wake
T84 farm 1S2W of DeerGold.

Walter_Wolf said:
If we whip pigsgems (and I think we should), the T82 move is not necessary for hoover and karl, and all the subsequent actions happen 1 turn sooner. The mine 1N1E of pigsgems can be improved with Fritz after T86.

I'd like to finish the mine 1N1E of PigsGems so that it has a Gmine when it wants one and we are comfortable committing all three of these workers around DeerGold to improve and prepare for post-Maths Pyramids chops. That micro will be easier if there's no question of running back to PigGems.

Axeman 5 is lurking in the middle. I think he is more useful in SheepTown than near DeerGold. 1-2 woodsman axes on a forest will likely throw back anything Hammu will bring before DeerGold can build more axes.

Walter_Wolf · Mar 26, 2012

I hope mabraham will chime in while the rest of the team is (probably) asleep

.

Spoiler :

DanF5771 · Mar 26, 2012

It's a little bit confusing, both parties determine EndWarValues and they represent their benefit of not continuing the war.

I.e. if a weak player is under a lot of pressure and his cities are threatened plus his powerful opponent was able to accumulate a lot of war success, then this player will calculate a very large EndWarValue. On the other hand the strong unthreatened player will determine a very low EndWarValue.

During peace negotiations both teams put their determined EndWarValues in the balance -- but on the opposite side of the balance! Thus the weak player needs to add other stuff (techs, gold, cities, ...) on his side to achieve a 'fair' balanced trade.

Therefore our goal is to lower our EndWarValue and increase Toku's EndWarValue as much as possible so that he has space to add techs on his side for compensation.

If we endanger more cities than Toku, our EndWarValue gets reduced via the presented division by 3. We should further avoid any war success for him. Even though Toku clearly has the higher power this division and our better war success so far will lead to a difference of the EndWarValues that should allow us to get Archery from him.

In addition to this reduction of our EndWarValue, there is also the dubious doubling of Toku's EndWarValue I mentioned for the case that Toku started a total war. AFAIK we can't tell whether that is true or whether he chose a limited war, but if it is true we might even get about 3/4 of Math from him (rough estimate!) -> we need ~100 :science:

invested in Math, less if Hammu gets Math too (e.g. in a trade from Toku for IW).

Walter_Wolf · Mar 26, 2012

I had some lunch break time today, so I made the V0.1 PPP. I have to test it tonoght, and might play tomorrow in about 34 hours from now.

The goals for this turnset are:
Build an exploring trireme and try to sneak through hammuragawa's borders.
Build a settler in Delhi for the the crabs/banana/marble site.
Defend against the AI's.
Setup a "threat event" possibility for Tokugawa with our military presence in a 5x5 city radius around Osaka and Tokyo before peace negotiations for better peace treaty odds/demand options. A road 1S1W of Sheeptown is needed for this. (?)

Tech
Finish research writing in 2 turns (100% T80, and as much % needed T81).
Put slider to 0% untill we have a library or two (?).

Cities

Spoiler :

Units

Spoiler :

Stopping conditions:
We get a game breaking military threat on Sheeptown or Deergold.
New AI met.
AI builds the Pyramids.

Every turn:
Save game.
Taking screenshots of new tiles deffoged, resources found, demographics screen, barbarians and AI units appearance etc.
Note new AI's cities founding, wonder completition.
Note forrest/jungle growth and other stuff of major interest using the Alt-S option.
Check for AI culture in sattelite view.
Note hinduism spreads if any.

mabraham · Mar 26, 2012

DanF5771 said:
It's a little bit confusing, both parties determine EndWarValues and they represent their benefit of not continuing the war.

Arrgh you're right. I assumed from the name of AI_dealVal that it was only called for AIs... but I can now see where it isn't. Sigh.

DanF5771 said:
I.e. if a weak player is under a lot of pressure and his cities are threatened plus his powerful opponent was able to accumulate a lot of war success, then this player will calculate a very large EndWarValue. On the other hand the strong unthreatened player will determine a very low EndWarValue.

During peace negotiations both teams put their determined EndWarValues in the balance -- but on the opposite side of the balance! Thus the weak player needs to add other stuff (techs, gold, cities, ...) on his side to achieve a 'fair' balanced trade.

Therefore our goal is to lower our EndWarValue and increase Toku's EndWarValue as much as possible so that he has space to add techs on his side for compensation.

If we endanger more cities than Toku, our EndWarValue gets reduced via the presented division by 3. We should further avoid any war success for him. Even though Toku clearly has the higher power this division and our better war success so far will lead to a difference of the EndWarValues that should allow us to get Archery from him.

In addition to this reduction of our EndWarValue, there is also the dubious doubling of Toku's EndWarValue I mentioned for the case that Toku started a total war. AFAIK we can't tell whether that is true or whether he chose a limited war, but if it is true we might even get about 3/4 of Math from him (rough estimate!) -> we need ~100 invested in Math, less if Hammu gets Math too (e.g. in a trade from Toku for IW).

OK, so is the practical advice for WW to run 0% science for a turn, and then 100% science to run down the bank?

mabraham · Mar 26, 2012

Walter_Wolf said:
I hope mabraham will chime in while the rest of the team is (probably) asleep .

Spoiler :

These are the worker's orders in the next few turns with your suggestions added (at least I hope they're OK).

Eiffel
T80 - finish the mine 1S of cowstown.
T81 - move 1S, build mine.
T83 - move to 1S of deer, build mine (for cowtown).
T87- move to 1W of gold, build mine.

Hoover and Karl
T80 move to 1N1E of Pigsgems, mine, wake.
T81 move 1N, mine.
T82 move 1S (1N1E of Pigsgems), finish mine.
T83 move to 1S1E of Deergold, build road (idk if they will have enough movement points to build the road at this point, so if they don't, something different is in order).
T84 move to 1S2W of Deergold.
T85 build farm.
T87 move to 1S1W of Deergold, mine.
T88 - move to 1S1E of Deergold, mine.
T89 - move to 2W of Deergold, farm.

Fritz
T80 move 1E, mine.
T82 move to 1S1E of Deergold, mine and wake.
T83 - move to 1S1W of Deergold, mine and wake.
T84 - move to 1S2W of Deergold, farm.
T85 - move to 1S1W of Deergold, mine and wake.
T86 - move to 1S1E of Deergold, mine and wake.
T87 - move to 1S of Deergold, prechop for pyramids fail gold after math.
T89 - move towards pigsgems for another farm 2S of the city.

I'm not sure about the T83 Hoover and Karl ability to build the road. They start from one, follow it on the tile 1N1W, and end on the hill which is accross the river, Don't have the game to test atm.

One of the reasons to build the T82 mine with H&K is that they are on the road to benefit from it T83. Fast Workers have 3 movement points. Hills and forests consume two. Forest-hills consume 3. Moving from road to road consumes 0.5. So on T83 they have movement remaining with which to build the new road. Then Fritz can use the road to reach the same tile.

The details from about T86 onwards depend on the timing of Maths. We'd prefer to start pre-chops 2 turns before Maths is due, so that the forest graphic goes away after we have acquired Maths. I expect that with a new city settling, completing Maths will be outside your turn set (i.e. T93-5), and so it is probably not right to start any pre-chops. I'd suggest getting up the mine 1SE of DeerGold, and the farm 2W of DeerGold, and probably road 2W of DeerGold and road on the chokepoint forest 2W1N of DeerGold.

mabraham · Mar 26, 2012

Bedtime here now.

I really think we need some team feedback on when we want to get the Pyramids. bcool and I each seem to have come over to the other's point of view in a somewhat half-hearted fashion. There's not much point discussing tile micro until we have this issue decided.

We can get Pyramids T90-2 before Maths, get a 95% GPro around T95 and get AP maybe T102-4. That limits our fail gold options considerably, but gets us Rep on our priest-push T90-5. We won't have any libraries then. A big downside is the chance of a GEng. We'd then have to whip a temple and mass priests in Delhi and get a ~97% GPro about T107 and GEng-bulb the AP.

Or we can delay Pyramids until several turns after Maths completes (so around T96-8) which gives a window for chopping ~4 forests onto the Pyramids for 67 fail gold each. We time a 100% GPro for the turn before the Pyramids finishes (since you first get wonder :gp:

on the turn you apply the last hammers to it; thanks STW). That will push the AP date back a few turns, so maybe T107-9. This creates a window for two settlers before the Pyramids. The main downside here is that we might lose either wonder.

DanF5771 · Mar 26, 2012

mabraham said:
OK, so is the practical advice for WW to run 0% science for a turn, and then 100% science to run down the bank?

Yes, the suggested 100% T80 in WW's PPP is impossible anyway (17 :gold:

in bank but total expenses = 18).

bcool · Mar 26, 2012

mabraham said:
Bedtime here now.

I really think we need some team feedback on when we want to get the Pyramids. bcool and I each seem to have come over to the other's point of view in a somewhat half-hearted fashion. There's not much point discussing tile micro until we have this issue decided.

We can get Pyramids T90-2 before Maths, get a 95% GPro around T95 and get AP maybe T102-4. That limits our fail gold options considerably, but gets us Rep on our priest-push T90-5. We won't have any libraries then. A big downside is the chance of a GEng. We'd then have to whip a temple and mass priests in Delhi and get a ~97% GPro about T107 and GEng-bulb the AP.

Or we can delay Pyramids until several turns after Maths completes (so around T96-8) which gives a window for chopping ~4 forests onto the Pyramids for 67 fail gold each. We time a 100% GPro for the turn before the Pyramids finishes (since you first get wonder on the turn you apply the last hammers to it; thanks STW). That will push the AP date back a few turns, so maybe T107-9. This creates a window for two settlers before the Pyramids. The main downside here is that we might lose either wonder.

I'm happy to come back to my initial point of view--delaying the pyramids for math enhanced fail gold chops. (Edit: especially if there is a chance we get math in the peace deal)

Tentative plan...
--Get the pyramids to a few hammers from completion in Delhi.
--Get the GPro (asap? by T90) Then we can start the AP in Delhi while other cities build the pyramids for fail gold. This means we research meditation before we finish math.
--Start the AP in the capital until we finish math and we fail-gold chop forests into the pyramids in other cities.
--Then finish the pyramids in the capital and continue the AP.

I've attached a test game.

Rather than build 2 settlers in Delhi as mabraham suggested. I have almost finished the pyramids and on T90 I can start the AP. I don't think this delays the AP significantly at all and we have a 100% GPro. And we can maximize our fail gold on the pyramids. It does delay the pyramids, but the benefit is limited I think for an early pyramids (the few priest specialists we would run and perhaps the engineer specialist after the GPro is born won't amount to that much research (~50-75?) Equal to one post-math fail-gold chop. There is some chance we are beaten to the pyramids in this plan, but that wouldn't be the end of the world. This plan doesn't delay the AP (I believe).

I think we build one settler in Delhi, almost finish the pyramids there as I detailed above. And then depending on how the war goes build a 2nd settler in CowTown.

edit plan used in Delhi in attached test game

T80 pyramids (corn, copper, FP, 2 Gmines, 1 Gfarm
T81 settler (same tiles)
T82 pyramids (same tiles)
T83 settler (corn, copper, FP, 3 Gmines, 1Gfarm)
T84 same
T85 same
T86 pyramids (corn, copper, FP, Gfarm, 3 priests)
T87 same
T88 pyramids (corn, copper, FP, Gfarm, Gmine, 2 priests)
T89 same

Tech path (I think this what I did)
Writing 0%, 100%
math 0%, 100%, 0%, 100%
Meditation 0%, 100%, 0%, 100%

(this sets us up for a lucky math trade with Toku in peace deal and gets us meditation in time to bulb theology with with arrival of GPro when we finish meditation)

bcool · Mar 26, 2012

DanF5771 said:
Yes, the suggested 100% T80 in WW's PPP is impossible anyway (17 in bank but total expenses = 18).

If there is a chance that we can get math in a peace deal then it makes sense for WW to switch from 0% to 100% researching writing and math. So start math before we finish any libraries.

Walter_Wolf · Mar 26, 2012

I just finished a test with 2 axe whips in deergold and cowtown each. Accumulated 138 and 145 gold I think. Whipped the trireme and workboat according to plan and another scouting WB T88/89 not sure.
Started the priests earlier than bcool suggested, and bulbed theo T88 (too early?). Mids 1 turn from completition in Delhi, and can start AP T89. Delayed the settler by a few turns. Settled T89, but the WB is late anyway. It gets there T90 I think. Nets built T91.
built the road 1S of sheeptown for Toku threat as after whipping in cowtown Eiffel had some spare time (not in a hurry for the mines).
Whipped the granary in sheepgold, and started a library. Worker plan according to PPP.

I'll write the details down tomorrow, as I have to go to work in 5 hours (low sleep mode engaged).

Tell me if it sounds good, and If it's worth bothering to write down.

I believe delaying the mids is worth the failgold opportunity, and besides all, much depends on the war developement.

cheers

bcool · Mar 26, 2012

Walter_Wolf said:
I just finished a test with 2 axe whips in deergold and cowtown each. Accumulated 138 and 145 gold I think. Whipped the trireme and workboat according to plan and another scouting WB T88/89 not sure.
Started the priests earlier than bcool suggested, and bulbed theo T88 (too early?). Mids 1 turn from completition in Delhi, and can start AP T89. Delayed the settler by a few turns. Settled T89, but the WB is late anyway. It gets there T90 I think. Nets built T91.
built the road 1S of sheeptown for Toku threat as after whipping in cowtown Eiffel had some spare time (not in a hurry for the mines).
Whipped the granary in sheepgold, and started a library. Worker plan according to PPP.

I'll write the details down tomorrow, as I have to go to work in 5 hours (low sleep mode engaged).

Tell me if it sounds good, and If it's worth bothering to write down.

I believe delaying the mids is worth the failgold opportunity, and besides all, much depends on the war developement.

cheers

I didn't start the priests earlier because I wanted the settler out of Delhi as fast as possible without growing into unhappiness. (so I worked the mines in Delhi during the settler building phase)

The 6th city increases our maintenance from 15/12 city maintenance/civic maintenance to 19/15 (an increase of 7) (at least in the test game)

Before the crabs are netted the 6th city can work the marble so we are producing 5 commerce and 5 hammers for an increase of 7 in maintenance.

I think 5 hammers per turn is worth the 2 extra maintenance. (no inflation yet and no building multipliers for research yet so maintenance is basically equivalent to commerce for now)

I also liked have a significant amount of research (~187) into mathematics so that we might get it in a peace deal (researching meditation earlier to allow the Theology bulb earlier might make that less likely).

shulec · Mar 26, 2012

Here are my thoughts on some of the issues you guys discussed:

Peace with Toku
I don't think Tokugawa is going to be giving us anything for peace. We will have to give something most likely. I think he will only give us math if he is down to one city and the power ration heavily favors us. If we had currency, a gift of 40 gold to him would probably do it. We need to have a settler on standby if we do get peace to settle and gift to him. We are in no position to eliminate him from the game. Do we want him to be our friend?

Pyramids
I don't want to delay them for more than a few turns. I am fine with a 5% risk of a GE. If this happens, we need to consider self researching theology. Theology is only about 75% of the beakers of Machinery. A machinery bulb is more bang for our buck. We don't need to be the first to Theology, just the first to build the AP. If we are worried about the AP we can always use the GE for the AP.

I also think we will really want to run representation for the additional research from scientist and spy specialists. We may need Police State too.

When to build the AP
With regards to the diplomatic victory vote, it doesn't matter when we build the AP. There seemed to be concern about this at one point. We will be nowhere near the point where we will have given Hinduism to all the AI. We just need to be the first to build it. There is only benefit for early AP if we want to use it for something ASAP.

bcool · Mar 26, 2012

shulec said:
Here are my thoughts on some of the issues you guys discussed:

Peace with Toku
I don't think Tokugawa is going to be giving us anything for peace. We will have to give something most likely. I think he will only give us math if he is down to one city and the power ration heavily favors us. If we had currency, a gift of 40 gold to him would probably do it. We need to have a settler on standby if we do get peace to settle and gift to him. We are in no position to eliminate him from the game. Do we want him to be our friend?

We'll see if we can threaten him, perhaps we will get lucky. No sense not trying for some small tech or part of math if the code suggests it might be possible.

shulec said:
Pyramids
I don't want to delay them for more than a few turns. I am fine with a 5% risk of a GE. If this happens, we need to consider self researching theology. Theology is only about 75% of the beakers of Machinery. A machinery bulb is more bang for our buck. We don't need to be the first to Theology, just the first to build the AP. If we are worried about the AP we can always use the GE for the AP.

I also think we will really want to run representation for the additional research from scientist and spy specialists. We may need Police State too.

The additional fail gold from the pyramids will accelerate our research far more than running representation a 5-7 turns earlier with only a couple of specialists.

shulec said:
When to build the AP
With regards to the diplomatic victory vote, it doesn't matter when we build the AP. There seemed to be concern about this at one point. We will be nowhere near the point where we will have given Hinduism to all the AI. We just need to be the first to build it. There is only benefit for early AP if we want to use it for something ASAP.

We do want to build the AP as fast as possible for enforced peace if nothing else. I would much rather use the AP then give Toku a city for peace if it comes to that--which I highly doubt. The additional hammers for the Kashi and other holy buildings aren't too bad either (especially if we manage to build a monastery or 2). Since we have to build it first and we don't know how good the AI team start is, I don't have any problem rushing to build the AP as quickly as possible. I would much rather not risk a 5% engineer and delay Theology and the AP significantly. It is crippling to our game strategy to lose the AP. It is not crippling to our game strategy to lose the pyramids.

shulec · Mar 26, 2012

I am not interested in the pyramids earlier for representation sooner. My concern is about losing the Pyramids.

Walter_Wolf · Mar 27, 2012

This is the (non optimal) city build order I used in my test game last night. I'll try to check where is it possible to improve it.
Please tell me if you approve this kind of whipping. We don't need math to start accumulating failgold from the mids.
As bcool pointed out, a faster settler and later G priest is better, so I'll look into getting it done that way.

cheers

Spoiler :

WastinTime · Mar 27, 2012

I've been reading, but not posting because I approve of the way things are progressing. Good work. 100% prophet is best w/ delayed pyramids -- fine.

bcool · Mar 27, 2012

I looked up Walter_Wolf's old wonder testing from the last sgotm to see what kind of risk the pyramids might be if we finish it ~T98 instead of ~T90.

In his testing with 4 industrious AI on Emperor the earliest pyramids went was T95(T96 by new reporting I believe) and the average was T114 (T115 by new reporting).
http://forums.civfanatics.com/showpost.php?p=10832510&postcount=571

Now we don't know what the start was like for the AI in this game, and our wonder building in sgotm15 might speed up the other wonder dates, but from Walter_Wolf's testing finishing the pyramids T98 is quite a low risk.

In our game the Great wall went T54 I believe (1800 BC) which is a bit slow based on Walter_Wolf's testing. So I expect the risk of delaying the pyramids is even lower. The reward is quite high ( at least 200 gold). We need to take some risks. And we are not playing a long game (hopefully) So the loss of the pyramids to us (which would give at least 300 fail gold above what we would hope to get if we finish it) will give us what we need to power our teching to the mid game where we hope to win the game.

The loss of the pyramids in a space game is much worse.

Plus if we lose the pyramids and tech currency then we can work river grass hill mines instead of scientists while building wealth.

A river grass mine is almost as good as a representation empowered scientist (if you ignore the gp point that is). A river grass mine 3 hammers building wealth is worth about 3 raw :science:

. The 1 commerce is 1 raw :science:

. And the 1 food is about 2 hammers assuming we are whipping off excess population.

So we get about 4 raw :science:

and 1 food/ 2 hammers vs. a rep scientist that gives 6 raw :science:

With the pyramids we can accelerate our research sure but it isn't a dramatic effect unless you look long term. The fail gold will help us during our critical teching phase whether or not we actually build the pyramids or not.
I think building the pyramids ourselves will be slightly better than losing it. It is slightly better because then we have more chance of a great engineer to help with machinery and we get the rep happiness bonus (which will save us on MP and let us grow the capital a bit more than we otherwise could). But in terms of teching to the necessary tech we need to win a religious game, I think losing the pyramids might be better for us.

SGOTM 15 - Kakumeika

Deity

Deity

Deity

King

Emperor

King

Deity

Deity

Deity

Emperor

Deity

Attachments

Deity

King

Deity

Grrrrr... I AM the force!

Deity

Grrrrr... I AM the force!

King

Deity

Deity

Similar threads