SGOTM 12 - Fifth Element

[pnp_dredd]

multiple crossposts!

I worry that we're aiming at earliest Oracle at the detriment of an Academy (from a cheap library. with 2x gold in the capital's BFC!).

Stockpiling gold DOES have a real world effect: it will delay techs. Sure, we'll tech them more efficiently, but we'll get them a number of turns later. Depending on what techs they are, this could be very important.

 

[havr]

PPP looks good to me.
That will be a record time to come up with a PPP for FE

 

[Fluroscent]

I worry that we're aiming at earliest Oracle at the detriment of an Academy (from a cheap library. with 2x gold in the capital's BFC!).

That is not actually what is happening. We are not doing this strictly to get a better Oracle date, but rather to avoid losing out on working the gold tiles for a bunch of turns. We are simply looking at an opening that will leave us in the best possible position. Consider this:

-We finish academy before Oracle, and gain however many beakers it produces in the turns it takes to finish Oracle/CS. However, we lose out on the beakers we could have had by working the gold rather than employing scientists.

-We finish Oracle first, and work the golds asap. Then pop the GS while stocking gold (should be another 12-15 turns, depending on the details of micromanagement)

It is a simple comparison really, which approach yields the most total growth/production/commerce?

My tests (and BLubmuz' too it seems) indicate that we actually gain more total beakers AND an earlier Oracle AND more "stuff" (defensive units/SH failure gold/workers+settlers) if we delay the academy in favor of working the golds.

That said, I will test the no 2nd worker, academy ASAP, delay improving golds opening to see if it has merit.

Stockpiling gold DOES have a real world effect: it will delay techs. Sure, we'll tech them more efficiently, but we'll get them a number of turns later. Depending on what techs they are, this could be very important.

We can always consider if techs are worth delaying, but employing binary research is not going to slow us down on anything critically important here as I see it.

 

[unclethrill]

If you want to continue until fishing too, that would be ok by me (more pressing enter and waiting).

I don't have a problem with that, if everyone else agrees (can't imagine anyone else wanting to just press enter 7 more times) but for now I'll plan to stop and post at the end of AH. If everyone agrees then I can continue Tuesday (unless CIV V gets in the way ) or Wednesday.

 

[beestar]

Sounds fine to me!

For demographics it would be great to get a screenshot of the zero-turn situation, and then turn 1 and then maybe a halfway point. Not sure how much we'll get out of it, but worth a look.

Test gamers: Is it possible to get an extra workboat and Chariot (or warrior) out early, to scout? Knowing the geography is going to be important, especially how close the nearest AI is, and how many share our continent, and how good/bad the AI land is, and whether they have a Modern Armor I think we have a bit of leeway in our Oracle sling, as the AI will be teching slower than usual with their fallout squares and their reduced number of starting units. I realize this will be slightly less efficient, but it would really increase our flexibility and ability to plan the long term.

 

[Fluroscent]

I tested a couple of builds going AH-Fish-Writing-Wheel, and building wkr-wb-warrior-library (asap academy). I was able to do the CS-sling on T76 without even settling StP. An approach that leaves us with roughly the same tech status at T79, but a huge loss in production (no extra warriors/wbs or anything).
I also tried with chopping Oracle in StP. Got Oracle on T79, and roughly the same status as my earlier attempt, but I had to chop three forests instead of one, and production was somewhat lower in Moscow.

 

[BLubmuz]

I've tried a variant:
Settle StPete ASAP, with the settler right after worker-WB.
Built the library there before than in Moscow.
Agri after Writing (but writing after the Wheel).
Once improved and roaded the SE (not the W) Cows, roaded to StPete, then roaded/scrub the rice, scrub/mine the PH, farm the rice.
As soon as StPete grows to 3, the library is completed and it can run 2 sci with almost no production and stagnating.
Moscow can work 2 gold and run 1-2 sci. Academy and Oracle (in Moscow) will arrive on turn 80-81, with Agri in and no chops (no BW).
There's no way to have a very early Academy without renouncing to build workers and some other stuff. And Moscow must take in charge those tasks.

Running Binary Research is mandatory!

 

[unclethrill]

I know that we want the Oracle sooner than later so that we can get CS but my best guess for an AI getting it if they have a similar start to ours would be around T90-100 so anything in the 80's should be fine. just my 2cents!

 

[unclethrill]

As soon as StPete grows to 3, the library is completed and it can run 2 sci with almost no production and stagnating.
Moscow can work 2 gold and run 1-2 sci. Academy and Oracle (in Moscow) will arrive on turn 80-81, with Agri in and no chops (no BW).
]

I thought the whole idea of getting StP out early was to build the Oracle there and not pollute the gene pool. If we are going to build the Oracle in Moscow and stagnate StP, then why even rush to settle it. The last think we need is another silverado dragging us down.

 

[Fluroscent]

I also tried popping a GS from StP to see how it played out, but it did not feel competitive. Then again, I didn't skip the warrior . I don't think we can delay the warrior longer than after worker-wb.

 

[BLubmuz]

I know that we want the Oracle sooner than later so that we can get CS but my best guess for an AI getting it if they have a similar start to ours would be around T90-100 so anything in the 80's should be fine. just my 2cents!

Sure like hell! The AI is busy building archers, workers and settlers.

I thought the whole idea of getting StP out early was to build the Oracle there and not pollute the gene pool. If we are going to build the Oracle in Moscow and stagnate StP, then why even rush to settle it. The last think we need is another silverado dragging us down.

StP stagnating at 3 it's far better than Silverado, since once popped the GS it can restart growing and producing without any delay.
There're different theories here about the polluted gene.
I assumed we don't care to have random GPeople, so i tested accordingly.
In any case 1-2 GSs will help a lot more than a Shrine, so i think it's worth avoid to pollute. And StPete can build a temple and run a priest, to increase the chances to pop a GPro.

I also tried popping a GS from StP to see how it played out, but it did not feel competitive. Then again, I didn't skip the warrior . I don't think we can delay the warrior longer than after worker-wb.

As you probably noted, after my tests, i changed my opinion about the need for an early warrior.
I discarded safety, since, as i say above, the AIs have other things to do than threat us.
But that warrior can scout the land right W of the desert, to locate a spot for the 3rd city. We can wait, since we have to complete the oracle before to send settler out, but it's better have knowledge.
So, i can try to squeeze it in my tests.

Now, seen your input, i'll try to optimize the settling of StPete with an early Academy.
Let me see what i can do.

Since i know UT have problems in taking screenies, here are 2.
It would be nice if he can post the same screenie at the end of his TS.

Spoiler :

Here is the F9/Demogr. screen

And the F9 where you can see the MP. You can note it does not belong to a city.

 

[Fluroscent]

I still can't see an opening that requires building a library in StP being as good as a delayed academy. Like I said, it makes little difference if we complete academy around the same time as CS or 10-15 turns later, since we can just stockpile gold. Academy has to be either asap in Moscow, or delayed to work the golds more as I see it.

And I can understand not feeling like the warrior is needed, but I think we do need it to scout around.

Another reason I like the gold/growth approach is that we are IMP, and thus when we work more hammers, we will build the settler very fast (I build it at size4).

We also (likely) want to have StP build chariots/axes (with barracks). I really don't think we can be sure we are safe for as long as in the testgame. Nearby AI may have better land.

 

[BLubmuz]

I think we need to find an halfway.
I've just completed a test completing the warrior, 2 workers and the settler.
Despite working the gold early, i finished the Oracle on turn 84, the worst attempt 'til now.

Fact is that Moscow can't grow and work the gold, not speaking of working 2 gold.
And it cannot work both gold and keep 2 Sci: you need to switch to refill the food.

This very early Academy risks to be worst than no Academy.
I remember SG9, where FE was the first to complete the Oracle (i mean in the SG competition), we built an early Academy, but we were in a horrible lack of workers and settlers. And it was another middle-of-the-pack game, despite a great late game and an apparently great start.

Please remember that settle StPete early pays one TR for each city, which even the costs of maintenance but raises the research.

Now i try with a mix of the best from my attempts, to see what i can have.

EDIT:
I updated some stuff on page 1 and mainly the roster please tell me if it can work.
To try a good scheduling, we can consider 4-5 days/TS. So please, try to foresee if you can play around the resulting time or if we need some adjustement.

EDIT2:
I've tried a schedule considering 5 days/TS, with 10 turns/TS with the exception of the first and second.
It doesn't work.
I think that we can set a limit between turn 240 and 280 for now, see the HoF date of a great Space Player.
Thus:
a) we use 4 days/TS and we'll be very determined to post the PPP and play on time
b) we'll set longer TSs, around 13 turns each.

 

[Fluroscent]

Fact is that Moscow can't grow and work the gold, not speaking of working 2 gold.
And it cannot work both gold and keep 2 Sci: you need to switch to refill the food.

You should not state something as a fact when I have done the very thing I got Oracle T79 that way.
If we don't worry about delaying the academy until ~T95. Which I am saying is almost as good as an academy at roughly the same time as Oracle, due to binary research. I have not seen any convincing argument that it needs to be earlier. The science gained by an earlier academy is just plain less than we gain by working the golds (both) early and continuously.

I think we should make some more precise comparisons now, and I'll attach my best results at T79:

Builds:
Moscow -> Worker- wb- worker (working pasture+fish)- warrior- library- settler (size4 with 2x gold, pasture, fish)- wb-> stuff; Stuff for me was wb (scout)- warrior(MP for StP)- SH (incomplete - 114 = 114)- warrior (9)
StP -> Oracle (working mined PH- except one turn of silk to delay it)

Worker tasks:
Worker1 -> scrub/pasture/road E cows-> scrub/mine riverside gold-> scrub/mine/road 2nd gold-> road to PH-mine PH. Then clear grass fallout for health, then chop forest for StP, then whatever.
Worker2 -> scrub/mine/road 2nd gold (joint with 1)-> road to PH (can save a turn by moving ahead of the other)-> mine PH-> Return to improve other cow.

Tech:
AH->Fishing->Wheel->Writing->Myst->Medi->PH->BW->Math->CoL

Notes:

-This build works really well on timing. Growth happens to match improvement of tiles, and StP gets settled the same turn the PH-Mine completes, with the road in place.

-On T79, we are 1T from growing to size 7 in Moscow. We have accumulated 24/100 GPP, and can work 2 scientists when growing if we switch to confu (without stopping working both gold). With two turns lost to revolutions (confu+bur/sla), we will have the academy on T95.

-Moscow will almost stagnate at size 7 while running specs, so this is a good time to begin spamming settlers+workers. Expansion should be easy with bureau in place to allow a large cash flow.

-Of course, this builds main strength is the great production, allowing failure gold from SH, an early wb scout, and warriors/chariots/axes if needed.

 

[BLubmuz]

You should not state something as a fact when I have done the very thing I got Oracle T79 that way.(...)

Yeah, after you sto... took my idea (should have read something like "not good" somewhere) to build the Oracle in StPete. And why not start SH in StPete, break for the Oracle and COMPLETE it?
GPro points, not bad.

But there's an hole in your plan, IMO: to have full benefit from Bureau, Moscow has to work the gold. So, while i mostly agree with you, it's better to run specialists *before* CS, not after. But trying to work the gold at the same time.
As i said, the problem will solve once we have a LH. And i think we must go for the GLH (9 turns well spent in Moscow).
No Pyramids, unless we have stone or we have nothing better to do.
I was already forced to build them without stone and they really are a pain.

All the above while i'm trying to find a good way to do it. Yours is, almost.
My attempts 'til now too, since by turn 81 i has the CS and the Academy. But we can do better

 

[beestar]

I'm just going to stay out of this and let you micro planners sharpen the opening

BLubmuz, thanks for planning the roster and making sure we can finish before the deadline - this SGOTM is a lot shorter, in calendar time, than I thought. I'd suggest that the first few turnsets after Unclethrill go to the detailed planners (BLubmuz, Fluroscent, pnp_dredd, cas?)

 

[BLubmuz]

I'm just going to stay out of this and let you micro planners sharpen the opening

BLubmuz, thanks for planning the roster and making sure we can finish before the deadline - this SGOTM is a lot shorter, in calendar time, than I thought. I'd suggest that the first few turnsets after Unclethrill go to the detailed planners (BLubmuz, Fluroscent, pnp_dredd, cas?)

I think that there's more need of MM as the game advances. At this stage, with only one worker to manage and almost nothing to do in moving the citizens, it's better some member less inclined to MM will take in charge.

You know your cookies , feel free to adjust the roster accordingly, then i will edit it.
What about the other issues i raised on the timing?

BTW, that capital "L" in my nick was to ensure it wasn't changed with an "i".
Sometime i like to write it in this way: BLuBMuZ™.
But you can avoid to use the capital L. Also, just Blub is OK.

 

[beestar]

Hmm, 5 days/TS seems really short, especially if we have contentious times. Maybe we should make it 15 turns standard. The player can always pause halfway through if needed.

I see you've noticed I am very careful about capitalizing BLubmuz ... to be honest the reason is that I thought maybe your name was B. Lubmuz I also have to remember to deliberately misspell Fluroscent every time I write it - the drawback is that I recently wrote a short paper on fluorescent light bulbs, and I kept writing "fluroscent" [sic]!

 

[BLubmuz]

Hmm, 5 days/TS seems really short, especially if we have contentious times. Maybe we should make it 15 turns standard. The player can always pause halfway through if needed.

I see you've noticed I am very careful about capitalizing BLubmuz ... to be honest the reason is that I thought maybe your name was B. Lubmuz I also have to remember to deliberately misspell Fluroscent every time I write it - the drawback is that I recently wrote a short paper on fluorescent light bulbs, and I kept writing "fluroscent" [sic]!

ha! Blub is a sort of joke from my family name a friend of mine invented loooooooooooong time ago. Since then it was my battlename.

I'm goin' to try this 15 turns thing, for now keeping the 5 days. Let's see how it works!

Take this, Fluro:
- Oracle in StPete turn 81 (completed, not IBT) with BW in
- Academy in Moscow turn 79
- 3 workers
- barracks in Moscow
- 1 warrior, 1 chariot, 1 axe, 1 spear (it should have been a warrior)
- 23 g. from SH failure. BTW, it was the first time i see it completed before the Oracle and it was the first time i try to build it... strange. It was also the first time i see a foreign unit near our borders, a Chinese WB.
- all the tiles in Moscow BFC (but one) scrubbed, roaded and improved, both nets operational
- 3 tiles in StPete BFC same
- 1 road N and 1 S to connect possible cities
- 1 possible city site (not in this map, just to simulate for the real one) scrubbed
- no chops

I tell it myself: great work of MM


EDIT:
I tried the 15 turns/TS and it barely fits with 5 days/TS. Please verify it (look only to the last turn, i wait to update the first to have decided).
It's a good way for everyone of us to play 2 TSs. I don't see more than 15 turns viable on Normal speed.
If we need to discuss in deep before some TS, we'll try to shorten the time between that one ant following one.

If you have any idea/comment, welcome

 

[unclethrill]

I plan to play in the next hour or less.