SGOTM 11 - Fifth Element

Mitchum · May 7, 2010

OK. I have run a few conclusive tests (I think).

Test 1: I made a huge square grassland island. I put a city in the middle with zero culture, so that it was a 3x3 square (having the city expand would complicate the picture a bit since the first expansion would cause the city to be a non-square). I stationed warriors around the city such that there was a 1 tile area around the city where barbs could spawn. The warriors were assumed to be fog-busting a 5x5 square even though they could only see in a 3x3 square. I removed all barbs from the map so that they would spawn and be evenly distributed around the map.

I had made the island a bit too big and a barb spawed on T1 along the edge of the continent, 3 tiles away from a warrior, as expected.

Test 2: I made the island a smaller so that warriors would be 2 tiles away from the edge in all cases, eliminating the possiblity of a barb spawning on the edge of the continent again. I removed all barbs from the map.

A barb spawned 2 tiles away from the cities cultural borders, just outside the cities visible range. It was 3 tiles away from the nearest warrior

Test 3: I reduced the size of the island so that the warriors were now 1 tile closer to the city. Now, the visible range of the city (1 tile out) and the 5x5 coverage of each warrior covered the entire island. I removed all barbs from the map.

No barbs spawned after repeatedly ending my turn and removing the barbs from the map, as expected.

Test 4: Now I wanted to test the visility of the city. I placed lakes around the city such that the city could now "see" two tiles in all directions. I removed all barbs from the map.

No barbs spawned after many turns.

Test 5: One final test. I removed the lakes along the northern part of the city but left them along the south. This left some "dead" space where the city could not see and a warrior was not within 2 squares. I removed all barbs from the map.

A barb spawned in the dead space that was created on T2.

Conclusion: The tests proved that Dhoomstriker knows more about fog-busting than the average Joe. :goodjob:

1. Units fog bust a 5x5 area regardless of whether or not they can see these tiles.
2. Cities only fogbust the tiles that they can see.

Mitchum · May 7, 2010

BLubmuz said:
OK guys, tell what you want, but my moving fogbusters worked well in my test.
Thus, i can't see why not in the real game.

Test 6: Since I had a perfect map for a test, I tried this out. I took Test 3, the test in which I had 100% coverage of my island. I then added 5 extra tiles due north of the northern warrior. I then moved that warrior back and forth, north and south, trying to fog-bust those extra squares AND the ones between the city and the warrior at the same time.

After about 6 turns, a barb DID spawn to the north, even though my warrior had been there the turn before. That barb could have just as easily spawned between the warrior and the city, which could be much worse in our game as that barb could march straight into our city. I don't think this method of "active" fog busting will work based on the results of this test.

The reason that you may not have seen a barb in your game is that your movements were only affecting 2 or 3 tiles. My test was affecting 5 tiles. So, over time, there is a chance (a good one actually), that a barb would spawn in one of the tiles that you are not fogbusting on that actual turn but the turn previous.

BLubmuz · May 7, 2010

I got the answer from AlanH about Tata:

I'll leave him in your roster for now in case he shows up.

Since he knows where you are, I don't see much point in contacting him. If he wants to play he'll reappear. If not, he won't be much use, and it will be difficult to find a replacement at this stage.

I've replied that i'll wait another week, then i'll ask him to cancel him from FE.

BLubmuz · May 7, 2010

Mitchum said:
Test 6:
The reason that you may not have seen a barb in your game is that your movements were only affecting 2 or 3 tiles. My test was affecting 5 tiles. So, over time, there is a chance (a good one actually), that a barb would spawn in one of the tiles that you are not fogbusting on that actual turn but the turn previous.

It seems the only reason. Since you got the environment ready, is too much ask you to reproduce our situation?
If i have to follow my lazy nature, it would be better for me keep the fogbusters in place, but if you can test my effort worths something, i'll be glad to attain that hard job

Mitchum · May 7, 2010

BLubmuz said:
It seems the only reason. Since you got the environment ready, is too much ask you to reproduce our situation?
If i have to follow my lazy nature, it would be better for me keep the fogbusters in place, but if you can test my effort worths something, i'll be glad to attain that hard job

What better way to test "our environment" than testing with... wait for it... our environment (i.e. our test save). I suggesting removing all barbs from the game using WB. Then, move back and forth. If no barbs, remove them again. Repeat. If you do this long enough, a barb should spawn.

OK, hold on. Let me try it... :mischief:

OK. I played with our test game. I moved both of our warriors back and forth between locations a and b labeled on the map. I had to go into the WB several times to remove several barbs, which would cause them to spawn again in the available space. Note that this could happen any time in the game. If Zara kills a barb, a new barb will spawn in any open area on the map. Here is what happened:

You can see when the warriors are at location a, I've labeled their dead spots as "dead a" and when they are at location b, I've labeled their dead spots as "dead b".

While at location a, a barb archer spawned at "dead a" since the southern warrior was too far west. In addition, the barb appeared between our warriors and our city, the worst possible scenario.

I think we need to stay at the "b" locations. That way, if a barb does spawn, it will be in the far west and it will likely attack our fortified units and die. Even if it wins, we'll have more time to react.

Now, we have two problem areas. One is temporary and the other is semi-permanent. The tile 3S of the pigs is a dead spot until Silverado's borders pop, which should be a mere 3 turns after learning Monotheism. No big deal. Also note that the cow tile is no longer a dead spot once the borders pop.

A bigger issue is the single tile NE-NE of the pigs. This tile is not covered by cultural boundary sight or our fog busters. Of course, the south and east are wide open unless Zara is fog-busting those areas in our game. We'll just have to be aware of the potential for a barb to spawn here and head for our city unless we want to actively fog-bust that tile. Or, our stone city may very well fog bust that square for us, depending on where we settle it.

Mitchum · May 7, 2010

Before you ask, I ran ANOTHER test. This time I put roads between fogbusting locations a and b (using WB, not workers). Then, on every turn, I moved the warrior from a to b and then back to a, so that it could fog-bust all tiles with this motion.

NOPE! When I ended my turn at location b, a barb spawned in a "dead b" square. So, barbs spawn based on which tiles are fog-busted (or more appropriately spawn busted as Havr says) at the end of your turn, not based on where you've been during your turn.

BLubmuz will be happy to know that "active" spawn busting does not work! So, he can be his normal lazy self and leave the warriors fortified in the forest...

Dhoomstriker · May 7, 2010

BLubmuz said:
Fogbustind moving seems to work.

Please create a new game, more along the lines of the test that I suggested. Using our test game (the one that mimics the real game) as your testing bed will not provide solid test results.

"Why not?" you might ask. Well, the Fog Busting that I laid out is already pretty close to optimal. The whole point is that Barb units are less likely to spawn in those tiny uncovered areas. Exposing a square here and then exposing another one back over there, by moving back and forth, really won't give Barbs a lot of space to spawn. So, you won't easily get conclusive results. Plus, you won't know if a Barb unit spawned just out of reach and then moved when you didn't see him... go with the test that I suggested, as it'll tell you with 100% certainty that if you see a Barb, the movement trick does not work.

While doing this i could count no less than 6 mixed barb warriors and archers in the E.

Exactly. That's why movement with our new Warriors won't be "scouting" so much as it will be "creeping slowly." Your goal will be to get into a Forest, stay there for a couple of turns, then move into the next Forest and stay THERE for a couple of turns. You'll slowly push out the "unspawnable area" this way. If you just start scouting, you'll lose us our Warriors left, right, and centre, faster than you say what I just wrote.

t61: ... research to 0

Good.

t62: 21 in the bank, BW will arrive in 10 (-2g/t) note: this won't change 'til the end of my TS

Am I right in thinking that a test game gives you this fact?

t63: both warriors completed, both heading E Silverado builds another, Dehli settler in 12
the exp. borders of Silverado help to fogbust S of the pig's lake

t65: the mine is completed, moved citizen there

Yes, it will be up to you to manually move the citizen to the Mined Plains Hills square. You'll be the first player who will need to move a citizen to an improved square that hadn't previously been worked, so make sure that you're watching for that improvement to be completed and aren't just focused on the exploring Warriors, or else you'll miss a turn of working it. You might even be able to get the game to automatically do it if you turn on the city governor (which is worth checking out in a test game).

the warrior from Dehli goes NE, the one from S. straight E walkin' S of Dehli. Still inside our borders.

I didn't look to see the exact squares that you are talking about, but where possible, say, in order borders and where there is no risk of encountering an enemy, and if doing so doesn't take you out of your way, try not to walk on an unforested square that is horizontally or vertically next to a Forested square. It's a very minor point, but doing so can slightly increase the chance of Forest growth occurring in the unforested square. Don't go out of your way to do it, but when you have the choice, keep this point in mind. As an example, it is always safe to walk in a Forest, as a Forest can't grow where one already exists, so you wouldn't be messing up the regrowth possibility for such a square.

t71: farm completed: undecided if move there to gain 1C or stay on a 1f+2H to gain more overflow edit: probably i got the answer myself: move on the farm, then move on the 1f+2H the last turn i build the settler (t.73)

What are you talking about? I already proved that what you are saying is untrue. If a Settler is in your build queue, then all Food produced that turn will be converted into Hammers. Period. No "if"s, "and"s, or "but"s. Period. If the Settler is completed at the end of that turn, all remaining Hammers (which includes Food that was already converted into Hammers) will overflow. So even on the last turn of building a Settler, you'd want to work the Irrigated Grassland River square, as you'd get 1 free Commerce while getting the same amount of Hammer overflow. Try a test game if you don't believe me--it's a very simple test, and I already ran that test to prove it.

Biz is behind with 2 cities.

Bismarck played REALLY stupidly this game--he improved 3 squares in his capitol and then built a Settler while at Size 1. I don't think that we can count on the AIs in the real game being this stupid. They are stupid in many ways, but on average, won't be THAT stupid.

t74: this has to be decided: in my test i started a settler in Dehli and another warrior in Silverado.
It's what i think it's better, but we need more scouting and to discuss this.

Which Warrior # would this one be? Knowing that would help us to decide. Until I know, I can't say either way which Warrior you are talking about building or not building. Unless it was the Settler that you were talking about not building now. I'm a bit unsure which unit you were talking about.

That brings about another question... should we rename Warrior 3 to Warrior 2 or should we keep with the numbering system, while knowing that we lost 1 Warrior to Barbs?

Dhoomstriker · May 7, 2010

BLubmuz said:
t73: our worker start mine a Forested GL Hill (11 turns) PH in 5.

Mitchum said:

In general, I think this is a bad idea. It's better to chop first and to build the mine second. The reason for this is two-fold. Let's say we have 2 workers (which should happen soon). If we want the second worker to help, he'll lose a turn moving onto a forested hill. However, if the forest is already chopped, he can move onto the hill and start helping with the mine on the same turn. Also, if you change your mind a few turns into building the mine straight away (11 turns) and decide we need the chop sooner, you have to start over (Chop = 5 turns and mine = 6 turns) and lose those initial worker turns.

Click to expand...

Mitchum is right. If we had a very limited amount of 1F + 2H squares to work, then you might consider mining the square. But we don't have a very limited amount--we have other such 1F + 2H squares that we can work.

The ONLY reason why you MIGHT want to build a Mine on the Forest is if you do not want to chop the Forest now and only want to pre-chop it. HOWEVER, if that's what you are doing, then I would argue that doing pre-chopping on a Hills square makes for VERY INEFFICIENT pre-chopping with a Fast Worker: the whole point of using a Fast Worker for pre-chopping is to take advantage of the fact that Forests on flatland still allow you to chop after moving onto them. A Forest on a Hills square takes up all 3 of the Fast Worker's movement points, making the square the kind that you only want to move onto once for chopping purposes, i.e. the kind of square that you shouldn't pre-chop from but should only "chop" from.

You can optionally put a Mine after chopping or optionally come back later. If you come back later, with the Forest gone, the square will only use up 2 movement points, so you can still have the Fast Worker build a Mine on the same turn if you pre-planned your movement to that Hills square properly.

We need to be ready to defend our cities to the east around T100 when those archers start marching toward our cities.

The more Warriors that we have "outside of our cities" and "past our cities" as a "vanguard," the less Archers in our borders we will have to face. That is the reason why I am not a big fan of having a Warrior sit inside of our capitol, when said Warrior would be better off forming a stronger vanguard line, pushing out the Barb line further and standing in a Forest far enough from our borders that if he does die, we'll have time to replace him before our city gets captured. Sitting in a city is a "last-ditch" effort and often loses you the city.

Dhoomstriker · May 7, 2010

BLubmuz said:
OK guys, tell what you want, but my moving fogbusters worked well in my test.
Thus, i can't see why not in the real game.

I want conclusive proof.

It is easy to prove that the theory fails. It's a lot harder to prove that it works. A LOT of testing will have to be done before you can conclusively say that it works.

To REDUCE this amount of testing, I gave you a Chariot scenario. Building 3-4 such islands and moving 3-4 such Chariots would let you do the testing a lot FASTER and MORE EFFICIENTLY, but you'll still need a lot of testing to actually prove that it works.

I haven't read down to Mitchum's testing results yet, so he may have already proven things one way or the other for you. But if not, you'll have a lot of work to do to in order to "conclusively prove" that this "moving units around" theory is more than just hot air.

Dhoomstriker · May 7, 2010

BLubmuz said:
Our "scouts" will reveal nothing, or almost by turn 74.

Is that because they won't have reached many of the unexplored (hidden) squares by the time that Settler 3 is built? Or was there some other reason that I didn't understand?

OK for the farm. I'll switch only the turn before the settler is finished, then back.

I hope that I have been able to convince you otherwise (i.e. don't switch even on the turn just before the Settler is complete, as doing so won't help in the way that you think it will). I also hope that you keep your PPP up-to-date when you realize that points like this one need to change.

BLubmuz · May 7, 2010

Dhoomstriker said:
Please create a new game, more along the lines of the test that I suggested. Using our test game (the one that mimics the real game) as your testing bed will not provide solid test results.

No need to do so.
Even if i think that Mitch stressed a bit the test (well done, BTW) and even if in my test game (see below) no barbs appeared, i'll stick to the original plan to keep our fogbuster fortified in the forests.

Am I right in thinking that a test game gives you this fact?

I wrote my raw PPP on Notepad, while plying Mitch's latest test.

Yes, it will be up to you to manually move the citizen to the Mined Plains Hills square. (...)

I got a rule while playing TSs, a rule i hope everyone of us will follow:
- Open the F10 and make a round of all the cities (no need to do this with just one city and one blocked, but to give the idea)
- Open F4 and look for any trade opportunity (too soon for this, but...)
BEFORE hit enter every damn turn!

Which Warrior # would this one be? Knowing that would help us to decide. Until I know, I can't say either way which Warrior you are talking about building or not building. Unless it was the Settler that you were talking about not building now. I'm a bit unsure which unit you were talking about.

That brings about another question... should we rename Warrior 3 to Warrior 2 or should we keep with the numbering system, while knowing that we lost 1 Warrior to Barbs?

I need to open the save, i'll let you know.

Dhoomstriker · May 7, 2010

Mitchum said:
Before you ask, I ran ANOTHER test...
BLubmuz will be happy to know that "active" spawn busting does not work! So, he can be his normal lazy self and leave the warriors fortified in the forest...

Thanks for doing the testing! :goodjob:

Wouldn't he have been happier if his Chariot Theory had proven to work?

Unfortunately, the Lazy Programmer Theory won the day. :mischief:

Mitchum · May 7, 2010

Dhoomstriker said:
The more Warriors that we have "outside of our cities" and "past our cities" as a "vanguard," the less Archers in our borders we will have to face. That is the reason why I am not a big fan of having a Warrior sit inside of our capitol, when said Warrior would be better off forming a stronger vanguard line, pushing out the Barb line further and standing in a Forest far enough from our borders that if he does die, we'll have time to replace him before our city gets captured. Sitting in a city is a "last-ditch" effort and often loses you the city.

The obvious exception to this is if we need a unit in a city to keep the "We fear for our safety" cry babies out and working tiles rather than sitting on their butts in the city, eating 2F and not contributing to the cause.

Irgy · May 7, 2010

BLubmuz said:
t71: farm completed: undecided if move there to gain 1C or stay on a 1f+2H to gain more overflow edit: probably i got the answer myself: move on the farm, then move on the 1f+2H the last turn i build the settler (t.73)

I thought someone had confirmed that you don't need to do this switch to get the overflow? I can test it myself, but you're more likely to believe it if you test it. Easy enough to check, just save the test game the turn before the settler is built, try both, and see if one gives you more overflow hammers or not.

Irgy · May 7, 2010

Dhoomstriker said:
... for anything but whipping a Wonder.

Whipping Wonders is not very efficient Food-to-Hammer conversion.

What you'd be better off doing is spending 1 turn working non-Hammer squares while building an Axeman. The next turn, as long as you didn't mess up and have more than 45 Hammers to go on the Axeman, then you can whip the Axeman for 2 population points. Axeman = 52H - 45H for a 1 population-point whip + 1H for the second population point = 6H is the maximum number of Hammers that you can invest into the Axeman without goofing up the trick.

Essentially, you'll get as many as 44 Hammers of overflow per population point whipped. Whipping directly into the Wonder will only give you 45H / 2 = 22Hammers (the floor operation truncates the half-of-a-Hammer). My way gets you an Axeman for free for whipping 2 population points. However, my way also requires you to stop working the Mines for a few turns and will still garner your unhappiness like your way will, so you'll only want to perform the whipping near the completion time of the Wonder, as we would with your way. We'll just start the process a lot sooner with my way.

I must be missing something here. Why is whipping wonders a less efficient food to hammers conversion than whipping something else? Is it that you don't get the marble bonus on the whip? If so, are you sure about that? I'm pretty sure I've confirmed getting the forge and OR bonuses on whips before, but I must say I've never checked the resource bonuses explictly.

I understand the axeman trick though. I used a similar trick once to build the globe theatre in a city with 3 food resources and no production whatsoever (literally 1 hammer). I couldn't get enough population to whip the theatre on its own, but I built it with whip overflow from a wide variety of buildings and units until I could get enough hammers invested to whip the theatre itself.

Mitchum · May 7, 2010

Irgy said:
I must be missing something here. Why is whipping wonders a less efficient food to hammers conversion than whipping something else? Is it that you don't get the marble bonus on the whip? If so, are you sure about that? I'm pretty sure I've confirmed getting the forge and OR bonuses on whips before, but I must say I've never checked the resource bonuses explictly.

I understand the axeman trick though. I used a similar trick once to build the globe theatre in a city with 3 food resources and no production whatsoever (literally 1 hammer). I couldn't get enough population to whip the theatre on its own, but I built it with whip overflow from a wide variety of buildings and units until I could get enough hammers invested to whip the theatre itself.

You get fewer hammers per citizen when you whip them into a wonder. I don't know the exact numbers, but a world builder test could prove the point and provide the data if you feel inclined to verify it yourself.

There may also be something in the Civilopedia...

Dhoomstriker · May 7, 2010

BLubmuz said:
No need to do so.
Even if i think that Mitch stressed a bit the test (well done, BTW) and even if in my test game (see below) no barbs appeared, i'll stick to the original plan to keep our fogbuster fortified in the forests.

Well, it seems that you aren't convinced 100%, due to your test game, but as I said, it's really hard to prove that the theory works--one test game wouldn't have been enough to do so. Anyway, the good part is that we've got a test that tells us the best way to go and you've agreed to follow that method.

I got a rule while playing TSs, a rule i hope everyone of us will follow:
- Open the F10 and make a round of all the cities (no need to do this with just one city and one blocked, but to give the idea)
- Open F4 and look for any trade opportunity (too soon for this, but...)
BEFORE hit enter every damn turn!
I need to open the save, i'll let you know.

I do look at each city and each of the F4 screens (Glance, Resouces, and Techs) on each turn, so that's at least one more player who will do so.

I'm not sure what's on the F10 screen... Demographics? Is there anything on the F10 screen that we should be writing down from turn to turn?

I often open up the Event Log, as sometimes the game events don't appear clearly on the screen. For example, before whipping a Wonder, you should always check the Event Log in order to make sure that it wasn't already built by someone else, meaning that you would have just whipped away the population just to achieve a very poor Food -> Gold conversion ratio.

Dhoomstriker · May 7, 2010

Irgy said:
I must be missing something here. Why is whipping wonders a less efficient food to hammers conversion than whipping something else?

Because the game design made this situation happen. Whipping is a powerful thing to do. The game was designed to limit the value of whipping on Wonders. It's either 50% Hammers from whipping or something limited like that just as effective as whipping for anything else.

Why did I suggest building an Axeman and whipping him? Well, the method that I suggested, as long as we don't have a Forge, will allow you to whip 2 population points' worth of Hammers (90 Hammers) into a 52 Hammer Axeman. As long as you don't whip him on the first turn (which also gives you a bad conversion ratio), and as long as you have few enough Hammers (more than 45 Hammers remaining needed to complete the Axeman), then you'll get 90 Hammers from 1 whipping, for a cost of 2 population points.

OVERFLOWED Hammers do NOT get any kind of a penalty on them, so you will want to overflow as close to 45 Hammers as possible, without going over. Note that if the Axeman completes with more than twice its Hammer cost, you will hit another overflow limit. I.e. if you have more than 104 (= 52 * 2) Hammers invested in the Axeman, than any Hammer above 104 Hammers will not overflow and will simply be lost.

So, on the turn that you produce the Axeman and on the next turn, you have to be careful about how many Hammers you are making, but since you are whipping, you'll probably focus on the Food squares anyway, so you're more than likely not going to have to worry too much about the total overflow--you'll need to worry more about the first number (producing 6 or less Hammers on the first turn of Axeman production).

Let's say that you can overflow 44 Hammers into the Wonder (you might get more, you might get less), then that's much better than the 22 Hammers that you would get per population point whipped that you would get by directly whipping the Wonder.

I understand the axeman trick though...

Whoops, I guess I didn't need to explain it again, then!

Dhoomstriker · May 7, 2010

Mitchum said:
There may also be something in the Civilopedia...

Ahahahaha! Now that's a funny statement! If it were true, I wouldn't be writing so many messages in this thread, as we'd have all understood the game's mechanics ahead of time just by reading the in-game documentation. Yes, that's a funny joke there, Mitchum. I don't blame the documentation team, though. Nope, just the bad game designers and programmers that couldn't make up their minds, and the project managers who didn't budget fixing things like a Civilopedia and an Instruction manual after they let the game testers get their hands on things...

Mitchum · May 7, 2010

Mitchum said:
I did keep warrior 1 in place while defending in one of my test games to verify that the space bar was equivalent to fortifying a unit; something I never knew before.

Now I know where my confusion was coming from. I was toying around with Warlords today and found out that the space bar does NOT give you the fortification bonus like it does in BtS. I assume that the same is true with Vanilla, which is where most of my experience lies.

SGOTM 11 - Fifth Element

Deity

Deity

HoF Quattromaster

HoF Quattromaster

Deity

Deity

Girlie Builder

Girlie Builder

Girlie Builder

Girlie Builder

HoF Quattromaster

Girlie Builder

Deity

Emperor

Emperor

Deity

Girlie Builder

Girlie Builder

Girlie Builder

Deity

Similar threads