SGOTM 15 - One Short Straw
- Thread starter AlanH
- Start date
Neither Marble 2E nor Marble 2W looks like a hill to me.
If we settle stone, (which I assume puts our city in a better position), I'm not sure at all which path the warrior should take. I kind of like your idea to explore west - maybe for about four turns so that it can get back to the defensible terrain around marble in time.
LowtherCastle
Deity
- Joined
- May 18, 2005
- Messages
- 23,631
Interesting stuff, ZPV. Where is it specified where animals can spawn?
Mitchum
Deity
@LC
Clean out your PM folder... and clean your room while you're at it.
Clean out your PM folder... and clean your room while you're at it.
In CvGame.cpp, in the function CreateAnimals, we see the function call
Code:
pPlot = GC.getMapINLINE().syncRandPlot((RANDPLOT_NOT_VISIBLE_TO_CIV | RANDPLOT_PASSIBLE), pLoopArea->getID(), GC.getDefineINT("MIN_ANIMAL_STARTING_DISTANCE"))That function is in CvMap.cpp, and those parameters do what you expect - it picks a random plot on the designated landmass which isn't a mountain, isn't fogbusted, and is not in the 5x5 square around any unit.
edit: the way it chooses the random plot, means that when there are very few (say 1 or 2) valid tiles, the barb spawn rate is lower than you'd expect.
edit2: if you're talking about the terrain they can spawn on, it's in UnitInfos.xml. In particular:
Lions spawn on (featureless) Grass, Plains, Desert.
Bears spawn in Forests.
Panthers spawn in Jungles.
Wolves spawn in Forests and (featureless) Tundra and Ice.
LowtherCastle
Deity
- Joined
- May 18, 2005
- Messages
- 23,631
I cleaned out my PM a bit, my undertaker will have to clean out my room...@LC
Clean out your PM folder... and clean your room while you're at it.
I'm happy enough to settle on stone, due to overall map position.
I think we'd go 1SE - 1S with the warrior then, before pausing on T1.
I think we'd go 1SE - 1S with the warrior then, before pausing on T1.
Dhoomstriker
Girlie Builder
- Joined
- Aug 12, 2006
- Messages
- 13,452
Most of the rules look fine. I think that we agreed last game that it was more of a hindrance than it was helpful to pause the game before uploading it.babybluepants said:
However, it is extremely helpful to pause the game when you are just looking around at the map... if you misclick, the click isn't registered until after you unpause the game... so, theoretically, you could just close the game without unpausing and avoid the misclick... that said, it would be nice if someone (bbp?) with access to reviewing uploaded saved games could test out such a scenario and see if it gets logged as having performed an action or not... techincally, it shouldn't count, but we don't want to rely on it not counting if it does count as an action.
Generally, this advice is good, but sometimes you want to choose to use the City Governor, since you know that when the City grows in Size during your turnset, the City Governor will do the right thing (since you tested that fact in a test game). So, I might instead say "Do not use city governors except when you have explicitly determined that doing so is the best thing to do for a particular city." Typical Dhoom wordiness for ya there.babybluepants said:
The other rules look fine to me.
It may feel silly, but trust me, it will feel far sillier to have our Warrior die and then either be unable to risk settling our second City unescorted or else risk settling it and losing the Settler to a Barb Animal.babybluepants said:
In other words, I'm okay with our Warrior fortififying for as many turns as it takes (even for 15+ turns) just so that he has the greatest chance of staying alive to protect the settling of City #2.
Ensuring that we can settle City #2 ASAP is our initial Warriors primary job... anything else comes secondary, in my mind, even if that fact makes for a boring first turnset (the first player could always do 2 turnsets, in that case).
Those two Barb Animal types are the scariest. Perhaps we should even aim to maximize the number of Forest squares that we are spawn-busting near the Marble or Stone Resource that we don't aim to settle on, at least maximizing the spawn-busting on the turn that Barb Animals first appear.ZPV said:
Do Barb Animals appear everywhere at first or does it take them a few turns to spawn? I.e. Should we consider keeping our Warrior in place for multiple turns if we go with the idea of covering nearby Forests/Jungles with our spawn-busting during the time of Barb Animal spawning, or would we only need to have the Warrior in place for that first turn of spawning?
I don't do it myself, but it seems like a good rule in principle.
No, I won't abuse my powers that way. Not least 'cause I'm too lazy to.
Ok.Generally, this advice is good, but sometimes you want to choose to use the City Governor, since you know that when the City grows in Size during your turnset, the City Governor will do the right thing (since you tested that fact in a test game). So, I might instead say "Do not use city governors except when you have explicitly determined that doing so is the best thing to do for a particular city." Typical Dhoom wordiness for ya there.
I think it could be a matter for team consensus, since we typically outline every city action in the PPP anyway.I agree. It's safety first if we go for the immediate settler. I'm just wondering how many turns of exploring we get, and I think we should pre-establish it. It may seem like more than it really is, but a bad lion injury can really cut it down.
Dhoomstriker
Girlie Builder
- Joined
- Aug 12, 2006
- Messages
- 13,452
Eh? Well, maybe we can ask Neilmeister to do it... it's more about finding out if pausing actually works or not. If, however, you pause the game, misclick, and the game still "records" the misclick, then pausing is dangerous, as you won't know about a misclick that got recorded as happening but never actually happens since you will close the game down.
In other words, I just want to make sure that by pausing the game, we don't accidentally end up being counted as reloading when a misclick occurs, rather than what I expect would happen, which is that no move is recorded until the game is unpaused (at which point clicks made when the game was paused seem to "resolve themselves" and start occurring).
Well, have we already decided on building a Settler first? I think that the testing done showed that Settler-first was the way to go; are we happy to go ahead with that or are we going to want to run some more tests once we see the fully-revealed big fat cross?
For the record, I'm leaning toward settling on the Stone both because we suspect that the southern river leads to the Coast and because of the revealed terrain in the minimap of the starting screen seems to indicate that there is land to the north, which implies that a Palace settled more to the north would at least allow for a "ring" of Cities to the north of the Stone.
Then again, the revealed area could be on another continent entirely, so it's hard to say with certainty.
I don't really know which way is best to move the Warrior for the first few turns--I would be more interested in giving input after Turn 2 or Turn 3 or so... when we still have a couple of moves remaining before Barb Animals appear and we need to decide if we're going to maximize early exploration versus trying to spawn-bust the Marble Resource area at the time of Barb Animals spawning.
We shouldn't be revealing any information that the mod gives us on the back end, unless AlanH does it. If you wanna ask anyone - ask him.
I'm actually not sure what you mean. I don't typically pause games myself. You're saying that buttons hit while paused are recorded as real actions, even though they wouldn't actually happen? Not sure why. What you're describing seems pretty minor anyway. It's common to have entire turns replayed in SGOTM (as LC did at the end of last game), so there should be some leeway for a single click - though I don't understand why that would even record.
I didn't think so, at least not definitively. As far as I remember, ZPV and I are the only ones who've stated tested opinions on it so far.
beestar
subdeity
Hey guys, checking in.
I believe looking at the animation of the river will show us the direction to the coast (once we get the savegame of course).
Are we definitely settling on both mineral resources? As I understand it, the argument is that these give us 3 hammers in the city tile(?), and we don't have anything better for a worker to do, so we might as well build settler first.
It seems to me that we are ruling out the possibility that a better site could be nearby -?
I believe looking at the animation of the river will show us the direction to the coast (once we get the savegame of course).
Are we definitely settling on both mineral resources? As I understand it, the argument is that these give us 3 hammers in the city tile(?), and we don't have anything better for a worker to do, so we might as well build settler first.
It seems to me that we are ruling out the possibility that a better site could be nearby -?
LowtherCastle
Deity
- Joined
- May 18, 2005
- Messages
- 23,631
Interesting. I use low resolution so I've never paid attention to this, so naturally I had to test it. What you said appears to be true when Sid makes the map. However, if the mapmaker worldbuilders the rivers, the flow direction is the direction he builds it. Regardless of coast, lake or anything else.
I wbed both:
Code:
^
!
<--o-->
!
v
!
v
-->o<--
^
!LowtherCastle
Deity
- Joined
- May 18, 2005
- Messages
- 23,631
Back to Square One for fog-gazing. We'll just have to open the save before we decide on our warrior moves. This might affect our choice of rock too, I guess, though I'm not sure how.
Yes, the rivers go by mapmaker chosen direction.
I don't see what we could possibly fog-gaze. Too much of the BFC's is invisible.
I've attached my previous tests for reference. Described here:
I don't see what we could possibly fog-gaze. Too much of the BFC's is invisible.
I've attached my previous tests for reference. Described here:
Spoiler :
Dhoomstriker
Girlie Builder
- Joined
- Aug 12, 2006
- Messages
- 13,452
Fair enough. I shall set about doing just that.
I had only observed it before in a test game, but now I checked out the behaviour a bit more in detail.
For example, if you pause the game, move a Worker, move a Warrior, and move a Settler, then unpause the game, all 3 units will simultaneously make their moves.
So, the game records the clicks but doesn't execute them until the game is unpaused.
Technically, the actions don't resolve themselves until you unpause the game, so the HOF Mod really SHOULDN'T be recording actions until you unpause.
It's more likely to come up by accident, if at all. For example, when the game is paused, units won't move, population won't be whipped, and even reasigning a citizen from a Grassland Farm to a Grassland Hills Mine won't actually take effect... so one could easily accidentally misclick and not even notice that one misclicked. If the HOF Mod's recording was set up correctly, then it shouldn't matter that you misclicked, as you should just be able to close the game and then whatever happened during the pausing won't ever happen, so you'll never gain extra info from it.
Rather than being a way to cheat, which it is not, it is more worrying about the HOF Mod being overly strict and potentially recording things that didn't actually end up happening.
The game keeps track of these things in memory but doesn't actually execute them until you unpause--if you close the game, those actions will never have occurred. As long as the HOF Mod treats actions in a similar manner, then there is no real issue with pausing the game.
While we wait for the saved game, there is only so much to talk about in terms of the actual game.
As for LC's bit about building Rivers in the World Builder... it is hard to build Rivers effectively in the World Builder. I have done it but it takes a lot of work to figure out the interface. My gut instinct is that Neilmeister didn't change the Rivers but instead moved us to a place on the map where he liked the geography of the existing Rivers. So, I'd only eat a small grain of salt and would be willing to mostly trust* the direction of flow of the water.
* As much as you can trust any Map Maker that is.
The important part to keep in mind is that building one fast Settler (or Worker) only requires us to settle on top of ONE of the two 3-Hammer squares.beestar said:
We will have already built a faster-than-normal Settler by the time that we go to settle our second City (regardless of whether that Settler was built as our first build item or not).
So, technically, no, we don't need to settle on top of the second Stone or Marble Resource. By then, we'll have revealed more of the surrounding area to be able to determine whether we'll want to settle on top of or simply near to the other such Resource.
Being riskier with our Warrior COULD allow us to find a potentially different location, but given that both areas are green (having a lot of Grassland squares), both areas would allow us to share the Corn and Flood Plains square between our Cities, and both areas are close to each other (low risk for defending a Settler with only a single Warrior) I'm in favour of simply setting on the goal of settling on or nearby the other Stone/Marble Resource for our second City.
If there is a 4-Flood-Plains-Hills-Gold-River location nearby, we can still rush to pump out yet another Settler and by then, have a couple of extra Warriors to ensure that we can settle such an awesome location quickly anyway.
My thinking:
With settler-first we can't really look for a different location, since our warrior will be locked into fog-bust duty very quickly. This is why I was initially saying that the quality of land available at the two rocks matters. What my T50 tests show is that settler-first isn't massively better than worker first, if at all. I was interested in that opening if it proves to be significantly faster in early production. Being able to settle an awesome city, other than the opposite rock can make worker-first considerably better. Unless if we're willing to risk building a settler and looking for the best spot.
With settler-first we can't really look for a different location, since our warrior will be locked into fog-bust duty very quickly. This is why I was initially saying that the quality of land available at the two rocks matters. What my T50 tests show is that settler-first isn't massively better than worker first, if at all. I was interested in that opening if it proves to be significantly faster in early production. Being able to settle an awesome city, other than the opposite rock can make worker-first considerably better. Unless if we're willing to risk building a settler and looking for the best spot.
Dhoomstriker
Girlie Builder
- Joined
- Aug 12, 2006
- Messages
- 13,452
I'm not really comfortable with taking that level of risk. If our Warrior dies, then Settler-first is going to turn out terribly.
Exploring when you only have a single Warrior is often dangerous anyway, while waiting to explore when you have multiple Warriors around is a whole lot safer, due to the mutual spawn-busting that they offer each other, greatly reducing the chances of you having to fight a battle with a wounded unit. So, I honestly don't mind losing some early exploration turns.
That said, delaying our first Settler (by going Worker first) might allow for a safer settling of the second Stone/Marble, due to Hinduism having (potentially--I haven't checked) expanded our Cultural Borders so as to make the other rock be covered by our Cultural Borders and thus making settling there safe from Barb Animals without requiring a Warrior escort.
In that situation, we COULD take more liberties with our Warrior and go off exploring with him until he dies... that said, I still like to have an extra spawn-buster around since having an extra Warrior makes all other Warriors that much safer, meaning that we'll probably have a much easier time against the Barbs if we do keep our first Warrior alive.
As we saw with last game, delaying the meeting of the AIs actually gave us a relatively easy warring situation with very few Longbowmen, so again, I'm not in any hurry to go out and meet the neighbours.
If you're researching Agriculture first, it can be nice to meet a couple of neighbours in order to save a Flask or two, but if you're beelining a Religion, then you won't gain any bonus Flasks for meeting the AIs early while they conversely could gain some from you.
The Devil's Advocate argument there is that I have observed that when you meet the AIs early on, they tend to prefer piggybacking off of your techs... which often means that if they were going to compete for a Religion had you not met them, they will almost certainly NOT compete for a Religion if you do go and meet them.
If this game were a random game (not one created by a Map Maker) and if we wanted to found an early Religion, then I would advocate going out to meet the neighbours, just to increase our chances of being the ones to found the Religion. However, as long as we plan to beeline Polytheism for Hinduism, then I'll safely trust that Neilmeister set up the map for us to be successful in this endeavour, and as such, I see no reason to rush out to meet the neighbours.
If there is an AI that is close enough for us to perform an Apostolic Palace gambit, then said AI will come to meet us soon enough anyway, without us having to risk losing our Warrior.
Well, here's an argument in favour of sending our Warrior toward the rock that we don't settle, with the goal of maximizing the visibility of the terrain around that other rock with two Warrior moves.
If, for example, we reveal no additional Food Resources, we might not choose to settle on the second rock, in which case we'd probably lean toward Worker-first. On the other hand, if we see another Agriculture-improvable Resource within the other rock's fat cross, then racing toward Settler-first and then pumping out two Workers (possibly with some Warriors in between) could work out in our favour.
