AU 100-A DAR 1: 4000 BC - 1520 BC

[Kidinnu]

Well, reading everybody else reveals plenty of errors. But here's what I did:

Built Washington on the starting square. {Wish I'd moved east.} Queue warrior, worker.
I don't have mysticism, so I won't try for an early religion. Thinking of the capital as a settler/worker pump, I'm not going for Pottery first off. Technology plan: wheel, mining, bronzeworking. {I was thinking road to the corn, but already knowing about all the flood plains I should have realized I'd hit happiness caps before health. In hindsight, the AI has apparently been really slow with religions - Isabella got Buddhism early, but Hinduism didn't come until after 2000 BC!}
My first warriors explore southeast; spotting the desert and ocean, they turn southwest and skirt the tundra. Second warrior, when built, heads north and west. {So I missed both the stone/ivory to the east and the hills by the wine to the north.}
Village in the southwest yields bronzeworking, just a couple of turns before I would have discovered it myself. (All the other villages I found yielded gold, about $200) This leaves me really unsure of a technology plan - I didn't have a chance to scan for copper nearby. I go for pottery and writing, thinking about alphabet - although I have my doubts, not having met anybody yet.
Washington queues a settler.
Meet Bismarck's scouts in 2960, Genghis Khan's in 2520.
I now see I have no copper nearby, and worry about defense.
Tech plan: hunting, archery, masonry, alphabet.
I send the settlers southwest to the coast of the peninsula near stone, founding New York. {In retrospect, this site was too food-poor to grow; I should have stayed farther east, as most others did. One of my goals in this game was to get out of the habit of a close spacing, but the resources I got access to - sheep, fish - aren't resources I really need, since happiness looks like much more of a problem.} Workers have irrigated corn and floodplains near Washington and built road; now they run a road to NY and start a quarry.
Genghis Khan has a warrior and three scouts around me, so I start worrying, and build a barracks in Washington. {Another mistake - it'll always be low in production. But New York's site won't be much better.}
Meet Isabella in 2320; she got Buddhism early, but no sign of other religions having been discovered. Saladin in 2280, Hapshepsut in 2000. I find Karakand about now
Gibbon's power assessment in 1960 ranks me #4 (Saladin, Genghis, Bismarck, me, ?, Hapshepsut, Isabella)
After Gibbon's announcement, I realize Washington's grown too much - it's into unrest because it's unprotected. {A common failing of mine is to build too little military. Luckily I can usually get away with it early on Noble, and I've had enough setbacks in games that I'm getting better about it most of the time. I'm not used to being so production-poor, though, and although I realized it early I haven't behaved consistently because of it. Sometimes I think my play suffers because I'm often interrupted. Here I built a barracks, not thinking that I should just churn out a warrior and then let some other city specialize in military.}
In 1760, I spot my first barbarian warriors. By 1600, there are several near me in the south, but they all tussle with the AI warriors and scouts rather than taking my undefended cities. At this point I have open borders with Saladin, Genghis, and Isabella, trying to be friendly, but still neutral (Cautious, +0) relations.

The screenshot's one turn too late, but shows what I've got to work with. Having met nearly everybody I'm hoping an early Alphabet lets me do some tech trading, or at least get a better idea of where the competition is. I need to: get two to three more cities out (N, E, and maybe NW), start producing great people in my capital (and a couple more workers), and get some military on the frontier. Plenty of errors already; AU's teaching me, and I just hope I can learn something from it.

[Arrian]

Ok, I'm really late to the party. My notes, which are really good for this DAR, are at home, but I wanted to comment on a few things. I played on Prince, btw.

First, like a few others (Cort, Lockstep, DeepO and Thror), I moved my settler east and settled there, so as to get the hills in the workable radius.

I researched Animal Husbandry first, but got it from a hut 4-5 turns before I'd have finished it, which was a little annoying (I'd rather have gotten something else and finished Animal normally).

My starting warrior was eaten by bears not long into the game, despite being in a forest and across a river. Toast.

I later built a scout that did ok after having several close calls (lions, wolves).

I built New York at Washington 2223 (coast, near cows, horses once I got a border expansion). Boston was to the SW, directly between the coast and the lake by the pigs.

I have no screenshots from ancient times (what is the hotkey to take/save the screenies I see people in this thread making??) and my first save is from 1400 AD (!). Typical me, eh?

Some things I noticed in other people's screenshots:

1) Many people went with a looser build than I did. I feel my somewhat tighter build helped me, but it *did* cost me the ivory site to the east. Isa-freakin'-bella got there first.

2) Lots of people seemed to prioritize grabbing the wines to the north/northeast. Why? I mean, wines are nice, but no more so than seveal other luxuries. I saw a number of New Yorks built up off the river on a hill to grab all the wines. Didn't seem worth it - you lose the health bonus from the river. You do gain +25% defense for that city, of course.

More later, once I can access my notes.

-Arrian

[chriseay]

Originally posted by Arrian
I built New York at Washington 2223 (coast, near cows, horses once I got a border expansion). Boston was to the SW, directly between the coast and the lake by the pigs.
-Arrian

Ok, just a quick question on this - what does Washington 2223 mean? I am I correct in assuming that is the numpad movement from Washington?

[Arrian]

Correct. My Washington, of course, is "6" of most people's, because I moved one tile east before settling.

-Arrian

[Kidinnu]

Arrian, I think the other reason to build on a (plains) hill is that they produce 2/2/1, instead of 2/1/1 like all other city sites. So you get an initial production boost, even without a worker, and long-term I think it's about equivalent to a windmill on that hill?

[Arrian]

Right, right. I didn't realize that was a plains hill.

Another thing I noticed: many people expanded aggressively north/northeast, whereas I expanded a bit more "organically" in that I kinda expanded equally in all directions. This clearly cost me ivory and the chance to really hem Ghengis and/or Isabella in. On the other hand, neither one attacked me for a good long time (Isabella eventually took a crack at me, too little and too late) and I still ended up bigger & stronger. Ghengis, however, is a power in my game.

-Arrian

[Aeson]

Originally posted by Cort Haus
Has anyone found that a policy of move-and-check on flat ground can preserve a scout? I haven't had enough scouts to try this with so far, It's the second move that tends to meet the bear, by which time it's too late. Single moves, unless a forest is available for the second, and retreating if needed, might be slow, but it's quicker than building another warrior and sending it out to where the ex-scout got to.

I don't think it's worth it vs animals, unless heading for somewhere specific and not in a rush. Like ending exploration phase, and moving to sentry. Once the Barb Warriors show up though, I tend to go exclusively move and check with Scouts if I want to keep them. Sometimes it's better to free up the unit support costs though, in which case full steam ahead.

[DeepO]

Originally posted by Aeson
Sometimes it's better to free up the unit support costs though, in which case full steam ahead.

Something more or less related:

I've thought before on going very worker heavy at first, disband them when every early thing is build, and later build workers again when lumber mills/RR appear (which will still eb around by the time Biology needs a couple of adjustment in terrain as well). I don't do that, though, as I figured that the support cost is not hurting me that much... so I simply tend to build fewer workers, and use them longer.

If you're very Dutch about your support costs, does that mean you tend to overbuild workers at first, disband them later, and rebuild (or conquer) them afterwards?
And, do you keep obsolete units around for happiness issues? I very rarely will disband a warrior when using hereditary rule, even if it won't give me a bonus at that very moment.

DeepO

[Arrian]

DAR I:

Prince Level.

4000 BC - Start looks solid. River, corn, floodplains, forest, etc. Popped hut w/warrior, got maps L This reveals some promising terrain to the east, however, including another hut near some ivory on a river. Mmmm.

Moved settler one tile east and founded (to get the hills in the radius).

Ah! Silks too. Nice. Work silks, begin worker. Set research to Animal Husbandry (I always think of Ani-MAL! from Sesame Street… ANIMAL!!!).

3880bc – our exploring warriors see stone! Mmm, Pyramids!

3760bc – 2nd hut popped. Well, I got good news and I got bad news. Good news: tech, yay! Bad news: I got Animal, which I was already researching. Still, not bad. Research set to the wheel.

3520bc – DISASTER! Our brave warriors are slaughtered mercilessly by bears. I had moved them so they were in a forest and the bears had to cross a river, but still the bears won. Damn.

3400 – Worker built, wheel discovered. Next up – mining. Build warrior.

3120 – Mining discovered… my exploration to this point has been terrible due to the death of my starting warriors. I therefore chose hunting, which I wouldn’t normally do yet.

3040 – warrior complete, Washington size 2, now using farmed corn and the silk. I decide to start on a settler, with the intention of sticking a scout in there once hunting is done in a few turns.

2880 – hunting discovered, build switched to scout. Writing next.

2600 – Scout complete. Washington now size 4 (farmed corn, farmed floodplain, silk, regular forest). Back to the settler.

2520 – Scout narrowly defeats lions. Well, at least they lived.

2440 – meet Ghengis Khan (oh, great), who has the same score as I do.

2360 – Our scout, still healing from his last battle, narrowly survives a wolf attack. Promoted him to hill defender type. Settler complete, warrior next. Writing complete, mysticism next.

2280 – New York founded on the coast by the cows (Wash 2223). Washington size 5.

2200 – Gold from hut (scout). Meet Bismarck. Open borders with him and Ghengis. I’m leading in score now.

2160 – Warrior complete. Start another settler… risky since I want the Oracle, but I’ll take the risk.

2040 – Warrior complete in New York, begin another. Meet Saladin. Open borders.

1720 – Boston founded 1144 of Washington (coast, pigs, lake). Research proceeding toward priesthood. Washington now up against its happiness limit, is using two (undeveloped, arg!) hills and a citizen laborer while building a library.

1600 – Priesthood. Start CoL & Oracle. Meet Hap of Egypt and Isabella of Spain. Open borders w/all.

HERE THERE BE DRAGONS. To put it another way, my notes trail off into oblivion for a while. The next event I saw fit to record was in 900 bc.

-Arrian

[Aeson]

I've disbanded or even given away Workers. I don't make a point of it, but especially after conquering a developed neighbor, and capturing so many Workers, I find I have way too many Workers. (I've also given away entire armies to the civ I just made peace with, so they can fight my ally... )

Unit support costs can strangle an economy. City maintenance and civic upkeep have to be dealt with, but unit support costs don't. I'd rather have fewer units, and more potential to build/rush/upgrade/draft units, than the other way around.

[DeepO]

Originally posted by Aeson
I've disbanded or even given away Workers. I don't make a point of it, but especially after conquering a developed neighbor, and capturing so many Workers, I find I have way too many Workers. (I've also given away entire armies to the civ I just made peace with, so they can fight my ally... )

Unit support costs can strangle an economy. City maintenance and civic upkeep have to be dealt with, but unit support costs don't. I'd rather have fewer units, and more potential to build/rush/upgrade/draft units, than the other way around.

If so, at least this doesn't change over levels. I did the same thing gifting units away from noble onwards (the moment you don't know what to do anymore, gifting or disbanding is a logical choice). The need for that seems to have lowered in 1.09, as Soren tweaked the AI, it seems. You won't capture workers so easily anymore.

and unit costs: the first time you're losing part of your attack force because you're running a deficit while at 0 gold, you are learning a lesson. A single look on F2 can learn a lot of things.

One thing I don't know, though, is how the different unit costs are calculated. It seems easy: one part for all units in general, one part for those outside your borders. You get a couple of freebies in every category. However, I can't ever figure it out precisely. Have you?

DeepO

[Cort Haus]

Originally posted by DeepO
Is it just me, or did most of the players got a scout at some part in their early expansion? I've seen at least two scouts at 4000 BC (Aeson and my own), but plenty report scouts later...

DeepO

I don't recall getting any scouts - I think 2 warriors, 2 maps, 2 gold caches and then the 7th hut surrounded me with barbs, almost Civ-1 style.

Over and over in these DAR's I see the barbs founding a city on the exact same square in the SW tundra - with lots of resources around. That's consistency and quality! We all kept it. I think many of us also had AI settling crap spots down there that even the barbs wouldn't touch!

[Cort Haus]

Originally posted by DeepO

One thing I don't know, though, is how the different unit costs are calculated. It seems easy: one part for all units in general, one part for those outside your borders. You get a couple of freebies in every category. However, I can't ever figure it out precisely.

DeepO

That F2 unit cost display has been driving me mad - I can't work it out, I can't plan the costs for a campaign or defensive boost, and I can't even predict whether 1 more or less unit wil make a difference. That 'handicap' number seems to make itself up as it goes along.

[Arrian]

I got maps from the first hut, tech from the 2nd, then my unit died. When I built a scout I popped 2-3 more huts and IIRC I got gold from them all.

-Arrian

[Arrian]

Over and over in these DAR's I see the barbs founding a city on the exact same square in the SW tundra - with lots of resources around. That's consistency and quality! We all kept it. I think many of us also had AI settling crap spots down there that even the barbs wouldn't touch!

Yeah, I noticed that furs city was in many games. Did many people get the northwestern barb town I did (Vandal, in my game)? I didn't see it... but that's a solid spot. It's got a sea resource and a land food source (corn or wheat, I think).

-Arrian

[chriseay]

Originally posted by Cort Haus Over and over in these DAR's I see the barbs founding a city on the exact same square in the SW tundra - with lots of resources around. That's consistency and quality! We all kept it. I think many of us also had AI settling crap spots down there that even the barbs wouldn't touch!

It seems to me that the barb cities must be based on something that is present in the initial scenario. It would be an interesting comparison to look at where outlying AI cities are in different games and see if the same thing occurs, or if it is just the barb city decision that is hardwired from the start of the game (land being available of course).

I would hope that though the barbarian AI might be locked in to certain spots, that the rest of the AI have some decision process built in to their city placement. Hopefully they can base it on how well the player (and other AI's) are doing in the game, and what the threats and goals are both in the short and long term. I have no idea how the AI city placement process works, so this might be way out there, but.....

[Cort Haus]

To be fair to the AI, it has to pick from the few lonely tiles outside the borders, while barbs usually have a large area to pick from. I'd expect the barbs choose city sites using the same criteria as the AI would.

[Fosse]

I played the game on Prince, using version 1.00 (haven't updated yet because of some long running MP games).

I'm putting my notes in the approrpiate DAR threads. I have over 150 pictures to sort through, and if I don't at least put these up I might never make a showing here... so forgive the lack of illustration for now (if anybody as questions I might be able to dig up relevant screenshots).


4000 BC
Map from barb hut shows a big desert to the east, and fertile land NE. I plant settler where he is, and see a food heavy, hammer poor starting location.

I start building a Settler... I want to stretch out quickly, and a worker wouldn't be able to do much for a while here... until cottages probably I don't need him. Start The Wheel to head for Pottery.

3880 BC
My warrior explores close to home (I decide not to run straight for the hut in the NE... might not get it anyway since it's so far, and might be another closer to home. I'd rather know the best spot for the first Settler. He finds cows and horses SE of Washington! Happy day.

3600 BC
Meet Isabella, scout appears NE of Washington.

3400
Discover POttyer, was going start mining. I'd like to see copper as I plant my first settler. Probably should have gone for bronze first since I wanted to see it. But, since I know I have horses I decide to go Hunting --> Husbandry. I could go straight for husbandry, but I'd like to send a scout or two out because it looks like a big world. The worker will be busy enough with a floodplaines cottage before he could build a pasture anyway.

3280
I pop a hut. Spain SHOULD hvae... and could have, with a scout. But didnt'. I get a scout from the hut... so decide to switch to mining. I'll finish the 2 turns of Hunting later on. get mining now, then husbandry... then...????

3080
Hut pop gives Mysticism. Nice. Cool mountain/valley configuration in the North. Warrior is heading toward home, uncovering bits of fog along the way. He'll stand on a hill to keep fog of war off of potential barbarian areas. The settler will walk one at a time, with no escort, to the SE to avoid barbs.

3000 BC
I'm 1000 years old. I discover mining, head to Husbandry. My settler finishes, and Washington starts a scout. I now pause to draw on the strategy layer where I want all of my cities. This will almost certainly not come to pass.

I plot three cities. One SE to get cows, horses, coast. One in the far NW, that's for GPP later on. Three food sources all in one radius, tons of coast. Merchants galore!

One straight North. It has lots of useless land, and overlap. I choose it though to take advantage of food and commerce in the area. It'll run specialists, so it won't ever need all of its tiles anyway. Late game it won't be my best city, but midgame it'll be a commerce powerhouse. What I don't have mapped out is a high shield city. The Cow/HOrse one will have to do that until I set up a proper town for it.


2920
Build New York, it starts a worker. Two workers being made. I met Germany the turn before, with his scout being NE of Washington. I need to get scouts and warriors out to uncover all of this fog... it's scary how vast this landmass seems. No sign of foreign cities yet.


2600

Still waiting on workers. The warrior is SW, heading toward the black area. Both units fight animals almost every turn. There are two lions east of washington. The scout defends in a forest tile against a bear! The warrior fights two wolves in two turns. Both units get Woods I. I discover Husbandry, and go for Masonry. I decide I can probably dedicate New York to the Pyramids soon. Stonehenge would be nice on this vast continent, but Pyramids will do more longer. I'm not going for an early religion at all.... probably shoot for Code of Laws soon, use artists to pop borders, and spread that religion out.

2280

Discovery masonry... need to get stone soon for Pyramids. New York builds Obelisk... I hate doing it but working that mine and those forests will matter. Meet Saladin... NE of washington again... Must be crowded up there.

Warrior in Washington... at least three barb warriors to my SW. Scary stuff.

Nothing too eventful until I discover Alphabet... so on to the next thread...

[DeepO]

Originally posted by Cort Haus


I don't recall getting any scouts - I think 2 warriors, 2 maps, 2 gold caches and then the 7th hut surrounded me with barbs, almost Civ-1 style.

Maybe better to distribute some inside info here, after discussions Aeson had with Soren.

Scenarios are possible in 2 varieties. A worldbuilder save is different from a python scenario. Simply taking the save without any editing means that every player will start with his initial units saved. Start the original game on Noble, and the AI won't get any free units. Use this save to start a new game on higher level, and the AI's units are already saved: they will start with the same units then on Noble. Only the bonusses they get productionwise are like they supposed to be.

Thus, if you play Prince, you expect the AI to start with a worker. It didn't in AU 100 A. Which explains the slow starts we're seeing: no early rushes, the players expanding faster than the AI, early research being hampered (the AI will probably start building either a worker or a settler, thus does not grow), early wonders falling late on high difficulty, AI scouts only meeting you in 1500 BC, etc.

This also partly explains why people on higher levels are doing so well: normally they should have gotten a higher initial threshold to take to get even. On Prince and above, the AI starts with a worker (and a scout?). On emperor and above (may be immortal), the AI will start with an extra settler. On deity, the AI will start with several scouts, 2 settlers, 2 workers, and several military units. Except in this game: none of all that happened. On noble, the game is probably the same as normal, and on prince there won't be much difference (although it will be noticeable). On Monarch and above this becomes quite apparant, and Aeson on Deity saw some very strange things compared to normal games.

To fix this, normally it would be enough to delete the units, and put only the starting locations in the WBS (I have no idea what I'm saying here, mind you, just rephrasing Soren. But I'm sure there are people who will understand what he means)

--------

For the scouts thing: It's been properly identified, in a reproduceable bug. Which makes solving it easy. On higher levels, going to 'Custom game' first before going to the scenario (as you were looking for the file, or simply because you made a mistake and clicked the wrong menu option) will give you an identical seed. As luck would have it, if you do this with this map, you will be getting a scout in 4000 BC. Stay away from custom, and it's random like it supposed to be. This also has implications for the identical-map bug people have been reporting but wasn't catchable.

Thank Aeson for figuring this one out. It might not be apparant from the very small comments posted here by him, but one of the bigger issues this game still had just got identified properly.

PS: noble players won't get a scout in 4000 BC, as they never get a scout from huts (it seems).

DeepO

[DeepO]

Originally posted by Cort Haus
To be fair to the AI, it has to pick from the few lonely tiles outside the borders, while barbs usually have a large area to pick from. I'd expect the barbs choose city sites using the same criteria as the AI would.

Blue dots algorithm... I suspect it gets reused for about anything. If you defog one blue tile, barbs will go for the next one. I didn't get the SW fur city, simply because I was patrolling the area, and wanted to build my own city on the spot I liked, and not the one the AI liked...

AIs are similar: they will go for the blue spots too. By the time they can run out of place to settle in this game, you will have expanded as well... leaving only the more crappy locations to pick from. With their penalties (on Noble), the AI are not always making a bad choice either: pure fisherman villages are in most cases worth it. It's because we're getting penalized that we will never consider building there, but for the AI it won't be that bad

(there are always exceptions: tundra with ice will get settled as well, and takes hundreds of turns to become viable. But the settling algorthim gets constant attention for what I can tell. Improving this will never stop.)

DeepO

[Aeson]

I'm not sure of the exact formula for unit support costs, but it's tied to population. (As is the bonus from Feudalism.)

[DeepO]

Originally posted by Aeson
I'm not sure of the exact formula for unit support costs, but it's tied to population. (As is the bonus from Feudalism.)

That's the handicap bonus, right?

DeepO

[Aeson]

The handicap is in addition to it. You definitely get more than 1 free unit on Deity for instance. How many is dependent on population (and Feudalism).

[Makahlua]

I'm still breaking about 50/50 on Noble, so that's where I'm staying fro this one

4000 BC - I popped the hut, actually hoping for a map, but humm, hills. I decided to settle one E of the start. I picked up Vel's settler first strat from that thread, and it's been wotking nicely for me We'll be starting with an early religion and then see if any one's close, needing bw/iw/mc beeline.

3280 BC - no sign of neighbors yet, but I did get hinduism Nothin' but coast to the N and W and SE - looks like I can use culture blocking to grab a chunk to backfill. Going to grab mining/wheel for now, unless I meet someone. Maybe hunting after if this continent looks much bigger - need map data to plan with!

3240 BC - Hello! I spoke too soon - here's bismarck coming from the ne germans= panzers = die bleeper, die! will chung up the bw/ iw path asap

2920 BC - Well here's the initial expansion plan - swine and wine! And although it's part hidden I have plotted a city on the N coast too, right in that right angle turn


And oh joy - Saladin (also probably coming from the ne). Aw crap. Lions ate my settler ;_;

(somewhere in between) And Isab****a. rrr. ... -And- Ghengis? ;_; Time for a quick diversion to hunting - think I need a scout or 2

1600 BC - built Napa w/o incident, and iron? You'll have to wait for the next DAR ^^

[Arrian]

Aw crap. Lions ate my settler

Unescorted, I presume?

-Arrian

[Makahlua]

Yes, I was too busy poking about with my lone warrior I figured they'd be fine, only going a short distance from the border, but noooooooo!

It's worth noting that for the dozen or two games I've played and used Vel's "Settler First" start, that's the -first- time I've had that happen. Come to think of it, it's never happened in earlier games either.. AU games hate me!

[Rhothaerill]

It almost happened to me too. I had two lions and a bear surrounding my unescorted settler right before I settled. It's definitely a risky strat. Of course I've only tried it once though.

[Arrian]

This game was definitely animal-heavy. Bears ate my starting warrior, and I saw lions & wolves all over the place. America was a wild kingdom in the beginning.

-Arrian

[joncnunn]

In Civ IV, it's a really bad idea to have a unescorted settler outside your cultural boundary for more than one turn.

But if you slow down your settler builds a bit, the capital culture will grow and wild anamals never cross into player owned terriory and in fact get an instant kick out if you found a city that places them within your territory.

Not to mention that a size 3 to 4 city can spit out a settler much faster than a size 1 or 2 city can.

[Dubhghlas]

Another reason that "settler first" sucks.