Are you, in fact, a pregnant lady who lives in the apartment next door to Superdeath's parents? - Commodore

Create an account  

 
Intersite Game - Turn Discussion Thread

Sacrificing a worker seems excessive to me, I'd rather get a village pillaged.
I have to run.
Reply

In Single Player turns are fully sequential so turns are definitely processed before barbarians move. That isn't really a useful test unfortunately.
Reply

(November 11th, 2012, 10:01)mostly_harmless Wrote: Where the heck is T-Hawk? ;-)
Forgetting to check this forum while busy playing Epic 28.

I think but I'm not completely certain that we get the production phase before the barbarians move. In sequential turns (including single player) you definitely do.

Anyway, I think I lean towards a conservative approach here. Let it pillage the village, we don't need to take crazy risks or sacrifices to preserve one turn on the Oracle. If that does delay Mathematics, we could even fall back on Oracle - Metal Casting.
Reply

Getting the village pillaged is WAY better than losing a worker.
Reply

Checked the code and now I'm pretty sure production does happen before barbarian moves in simultaneous turns. Early in CvGame::doTurn, the production phase for each team happens with a call to each CvTeam::doTurn. Then later in CvGame::doTurn, if MPOPTION_SIMULTANEOUS_TURNS is set, it calls CvPlayer::setTurnActive for each player, I think including the barbarians.

I'm a little unclear exactly what happens from there but it feeds through CvSelectionGroup which is responsible for picking and executing missions for each unit. I traced the call stack back from CvUnit::move from CvSelectionGroup::groupMove from CvSelectionGroup::groupPathTo from CvSelectionGroup::continueMission from CvSelectionGroup::startMission from CvSelectionGroup::activateHeadMission from CvSelectionGroup:tongueushMission. But there's a million calls to pushMission and I'm not entirely clear on how exactly that happens, but at any rate pretty sure it's way after the human players' production phase.
Reply

Here's where we stand after moving all units for the turn. First I killed the barb warrior in the east:

[Image: ISDG-236s.jpg]

This mod gives you all kind of extra combat information when you attack. It's totally worth having to load the mod every single time I log into the game. rolleye Anyway, another interesting feature that I hadn't seen before. We won without taking any damage, although it will take two full turns to walk back into Focal Point.

[Image: ISDG-237s.jpg]

Here is our tile micro for the turn. Note that I have made one change here from the way that the micro plan is typed up. It has us working a coastal tile in Focal Point, the lake tile in the capital, and Focal Point working the two grassland cottages on the two tiles east of the capital. The plains cottage southwest of Mansa goes unworked. That made sense before we knew we were going to have to delay the settler from Focal Point, not as much now. I changed it so that Horse Feathers has the lake tile, the capital has the eastern of the two grassland river cottages, and Focal Point has the plains river cottage instead of letting it sit unused. I see no reason to work a 1/0/3 coastal tile over a 1/1/3 plains cottage tile, which is effectively what this amounts to. Focal Point will grow to size 6 at the eot 65 regardless. In fact, here the tile configuration next turn:

[Image: ISDG-238s.jpg]

The capital takes back the corn tile (as Horse Feathers stars a worker after growing to size 4). This lets Focal Point take back the other grassland river cottage, and it grows at end of turn with exactly 28/28 food in the box. We will triple whip the settler there eot 66; there is no way to speed it up any sooner. Mansa is able to pick up the plains cottage tile as scheduled in the plan. Maybe I'm missing something, but I don't see this affecting any of our growth targets. We also need to be working all of the highest commerce-configured tiles, as we are extremely tight on commerce to hit our Oracle target.

There's another mistake in the micro plan as written. We are currently sitting at 177/450 beakers into Mathematics. We are scheduled to get 80 beakers this turn into the tech and be sitting at 257/450 beakers next turn. Unfortunately that's not going to happen; we are at 61 base beakers, which equates out to 61 + 1 = 62 x 1.2 = 74 beakers. That's 6 beakers less than what we have scheduled in the plan. Even without the barb shenanigans, we'd only be getting 78 beakers this turn, so I think we misplaced a number somewhere. I think that we can manage OK, as we can pull some extra beakers out of somewhere on the map, work coastal tiles or not whip or something, but I wanted to point this out because we have so little room for error right now on the commerce front.

If the barb warrior goes after our floodplains cottages, I will be one sad customer. [Image: frown.gif] Cross your fingers that the barb warrior moves any direction other than northeast. But I'm pretty sure that's where it's going to move... I'm kicking myself for not buiding the axe out of Focal Point last turn. We could have moved the thing northwest-west this turn and been in perfect position to nail the thing. Sigh. My bad, everyone.

We have another 14 hours waiting for other teams to play, so feel free to make any last minute suggestions here.
Follow Sullla: Website | YouTube | Livestream | Twitter | Discord
Reply

MM would be unhappy if we move the warrior out of it right? If not we can move the warrior out SW-SW and take a shot at the barb with the C1 warrior from the capital.

In the light of these developments, what are the thoughts of an Oracle -> MC alternative?

Kalin
Reply

Are you following the "best case" or the "worst case" write-up? The worst case seems to be exactly like the one you do and describe, except that we might be unable to work one of the flood plains.
Furthermore, I consider that forum views should be fluid in width
Reply

Great minds think alike, Sullla. Your micro for this turn matches the last sim I posted to a tee. https://dl.dropbox.com/u/15215428/ISDG1/...64-t67.zip

Unfortunately I haven't had time to type it up, and neither the worst case nor the best case micro plan is updated to take account for the warrior moving onto the flood plains.

The salient point in the latest sim is to NOT whip FP on t66, but to keep building a settler at size 6 on t66 and not whip it until t67. This gives us the final few beakers needed to finish mathematics on time.
I have to run.
Reply

Here's my latest sim typed up as a micro plan:
https://docs.google.com/spreadsheet/ccc?...Uh4aXVTQ3c
I have to run.
Reply



Forum Jump: