yet another "Why do I keep losing?" thread

[eataTREE]

Okay, so I consider myself to be a reasonably intelligent fellow, but I've had an astounding lack of success at Civ3 so far. The game is very enjoyable, despite the fact that I keep having my butt handed to me over and over and over, but still, I wonder what I'm doing wrong!

Here are the particulars.

I play on Huge maps, 8 or 12 civs, Regent level. I have played as the Greeks, French, and Egyptians. I've noticed the same general trends in all my games:

1. I found my first city on Turn 1, wherever I happen to be. To do otherwise would mean falling behind in expansion.

2. I commence rapid expansion, churning out Settlers alternated with Warriors or Spearmen. I build no city improvements until the city has produced at least one Settler, when I'll usually build a Temple. I use my first Worker to build roads connecting my cities and may build a few more Workers in high-food cities to help.

3. At some point, I encounter a couple of neighboring civs, and seek to cement my borders to prevent settlement of my land. I notice that these civs have a signifigant technological advantage over me.

4. I seek to purchase the technologies I lack from these civs, but I run out of gold and other negotiable commodities before I can buy my way to tech parity.

5. Round about this time, I start clicking through pop-up windows telling me that AI civs have built every single Aincent-era wonder ever. Meanwhile, I am still building Settlers to back-fill the territory I've staked for myself, and am pretty much a non-starter in the wonders race.

6. My neighbours start demanding tribute from me. Being a prideful sort, I refuse their demands. Someone declares war on me.

7. I find myself looking at approaching columns of Knights while I still have no better than Spearmen. The resulting war doesn't go my way, predictably enough. I sue for peace to prevent the destruction of my empire.

8. By this time it is the middle of the Midieval period and my civilization is effectively a third-world nation.

9. Determining that I will not be able to bridge the yawning gulf between me and the AI civs, I concede defeat. Repeat at Step 1 for the next game.

Let me re-iterate that I enjoy the game very much, but I'd like to at some point experience that strange game state known as 'victory'. Do any of you learned and civ-succesful types have any advice?

r.

[Arrian]

Some questions:

1) What is your normal research path in the early game.
2) Do you use those warriors to explore, or only to escort settlers to their build sites?
3) What do you set your science rate at early on?
4) Do you buy tech from the AI civs when you meet them?

On Regent, I shoot straight for literature and try for the Great Library. Even if you don't get it, you're on the way to Republic, and you can build libraries, which will help your research rate.

I generally set my research rate as low as I can (no matter how high you set it, it's still gonna take you 32 turns in the beginning) and make cash. This, combined with huts, allows me to buy myself into tech parity with the AI (on regent... it's a little tougher on Monarch). They sell tech at pretty reasonable prices early on.

My Capitol build orders are similar to yours. I generally do something like warrior, warrior, settler, spearman, settler, spearman, settler...

My 2nd or 3rd city becomes the worker farm to punch out workers to build roads/mines, etc. When I have another city that is capable of taking over settler production, I start building other things in my capitol.

Grab luxury resources and trade them to the AI for cash or tech.

You may want to consider appeasement early on if you're a "builder" type player, like me. Often, early in the game, somebody demands 11 gold from me, or my territory map. I give it to them and snicker about their eventual destruction once I'm up and running.

Some other civs to try: Babylonians (good setup for peaceful builder), Persians (good hybrid civ - Immortals are great in the ancient era) and Iroquois (great if you want to kick butt early on - the Mounted Warrior is really nice).

-Arrian

[Haphazard]

First point: Swallow your pride and drop down to Chieftain until you get your first win. Try smaller maps to start with too. Have you played much Civ/Civ2 before?

Second point: If the starting place is hopeless, don't found straight away. Scout with the worker first. If you're one square from the coast, move to it. It's okay to spend a couple (no more) of turns finding the best place.

Third Point: Scout first, don't build settlers until size 4, because it takes a while to come back from pop1. Building a Granary in your capital means the pop recovers faster. You don't need too many workers early on.

Fourth Point: Build your first few cities around the capital - corruption is lower. These can help you expand. The AI doesn't expand everywhere straight away.

Fifth Point: Trade techs between the other civs straight away, don't let them always trade between themselves. Try not to trade comms with other civs. Act as a go-between (with commission!)

Sixth point: Don't be to worried about early wonders. Only Pyramids have a game long effect. To get a couple, mine lots in your capital and build after the first three/four settlers.

Seventh point: If the AI civ demands something, and they're bigger/better tech than you, give it to them.

Try some of those, but especially point one.

[mr.buddylee]

Check out this thread.
I posted in it and a lot of other people did as well. I picked up quite a few new ideas.

http://apolyton.net/forums/showthrea...threadid=36016

[eataTREE]

Originally posted by Arrian
Some questions:

1) What is your normal research path in the early game.
2) Do you use those warriors to explore, or only to escort settlers to their build sites?
3) What do you set your science rate at early on?
4) Do you buy tech from the AI civs when you meet them?

Hi Arrian! I have found your posts very informative and am glad you took the time to answer mine.

1. I research Wheel and Iron Working first, to locate resources. Then I usually head for Republic. If no one has built the GL by the time I am researching the techs leading to Republic, I will research Literature first.

2. I use warriors for exploration. I'm willing to send out settlers unaccompanied through explored terrain; even with Raging barbarians the gamble usually pays off. Having a road network for the settlers to use helps too, as there's less time while they are in transit (and therefore vulnerable). I often build a new city's garrison unit in another, higher-production city, then move it in after the new city is built.

3. I set science as low as possible. One problem I frequently encounter is that I take in so little cash, it doesn't matter how high I set the science rate, it still takes 32 turns (soon to be 40) per tech. So I set science low and save what cash I can.

4. Yes, but see point 3, I often run out of money after buying 2 or 3 techs and am not raking it in fast enough to buy more for a while.

In answering your questions it's occured to me that perhaps my civ's troubles are founded in a lack of cash. Anyone have some tips for a more prosperous civilization?

r.

[sophist]

2) Or move away from the coast. That way you will end up with more useful land in the long run (unless you're in a desert).

5) This is highly important. The AIs need to see you as a valuable trading partner. They'll be more friendly, plus you'll have more techs and money. It can help to go after techs that the AIs don't start off after. I've found Mathematics, Code of Laws, Republic, and a few others to be ones that the AIs take a while to get to because they're deep, but don't have many requirements so you can get there before them.

7) On the other hand, if you're out of reach, sometimes it's ok to make them go to war with you. They'll take forever to get to you in the Ancient/Medieval era, and once they show up, sue for peace. You generally don't end up paying much more, and it wastes AI resources.

To add: getting luxuries and trading them is essential. I've fought wars to get them.

Originally posted by Haphazard
Second point: If the starting place is hopeless, don't found straight away. Scout with the worker first. If you're one square from the coast, move to it. It's okay to spend a couple (no more) of turns finding the best place.

Fifth Point: Trade techs between the other civs straight away, don't let them always trade between themselves. Try not to trade comms with other civs. Act as a go-between (with commission!)

Seventh point: If the AI civ demands something, and they're bigger/better tech than you, give it to them.

[mr.buddylee]

I play on the second difficulty level (but I can't remember what the heck that is right now), so some of my advice may not be applicable to harder levels.

One thing to consider is which civ you're choosing. The stats that each civ gets really helps out. If you're having problems, you may want to try the civ that is scientific and commercial. That will help both your science status and cash flow.

One thing you could do is concentrate on one area of the tech tree. Try to get far advanced in one section and you SHOULD be ahead of a few civs in tech as long as you stay in one section of the tech tree. You can then sell/trade your new techs for multiple older techs that you've missed out on. (I once traded a new tech for 3 techs I skipped in my rush to get to this one.)

I keep my science at 50%. This allows me to stay close, but I'm never in the lead of the tech race until I get Monarchy or Republic (Then I up the tech to about 60% or 70% sometimes. By the end of the ancient era, I'm usually mid range on techs, top on money, mid on size, and low on military. I try to save money for protection if a neighbor gets frisky.

I dont know what else to tell you, but good luck. Try to take another radical approach to the game and see how it works. Try getting to just 10 cities, then build your military and try to influence your neighbors or destroy them and gain more land that way. Or try something that you wouldn't normally try. What are you going to lose?

[Arrian]

eataTREE - I'm glad to help. Haphazard brings up a good point. I'm a CIV II vet who was used to thrashing Diety consistently, but I fired up CIV III on chieftan at first. After winning twice there, up to warlord for a couple of games, then Regent (where I spent more time) and now Monarch, which still kicks my butt half the time.

Buddylee's point about picking a particular part of the tech tree and shooting up it in a linear fashion is a really good one. This way you can get ahead of the AI and trade some tech with them (I almost always pick the route to literature/republic, but I have gone with other routes with some success). Getting Iron Working and The Wheel is nice, but you can usually buy The Wheel fairly cheap. Iron Working is another beast... and I do sometimes research it myself (I do mostly buy it, though).

When playing the Persians, you can select Iron Working as your first advance. That is worth it, as you will beat the AI to it, and it gives you Immortals, with whom you really should waste one of your neighbors. If you're lagging in tech, one way to catch up is to attack a neighbor, beat up on them for a while (take some cities, etc.) and keep trying to contact them at the end of your turn. At first, they won't talk to you. Then they'll talk, but only accept a straight-up peace treaty. Hurt them badly enough, and they will give you all of their tech, gold, and cities other than their capitol. This last part is the way to go.

Keep in mind that starting location is CRUCIAL in Civ III. For your capitol, ideally you want a river (no need for aqueduct & increased trade), some bonus resource tiles (cows are the best, IMHO, but there are several others that are nice... gold on hills comes to mind). Corruption is brutal early on (despots aren't very efficient rulers), so definitely build a few cities near your capitol first and THEN charge off toward the important resources. If the AI builds near one, plunk down a city next door and try a "culture bomb." Concentrate on culture in that city and hope you get a defection.

-Arrian

p.s. about cities near your capitol - ideally you want one with high food production and one with high shield production (both, if possible, clearly, but most of the time the terrain won't cooperate). One builds workers and settlers and such, the other sets about building some military units. The AI smells weakness, and will demand things and attack you if it thinks you're weak (this is a simple computation of how many units you have vs. how many they have, so far as I can tell).

[Ralf]

1: Try, if possible, to found your capitol city near some super-food squares, like floodplains/grasslands with food-bonus resources. Sometimes, this let you build 2+ settlers in a row before you must build something else, in order to let the city-growth catch up.

2: Let your capitol city build at least 3-4 settlers, with city-protecting or scouting warriors and 1 worker in between. DONT build a temple or other city-improvements just yet. Dont even build more expensive units like archer & spearmen if you can avoid it. After that; switch your capitol production to some appropriate ancient Wonder of your taste. Be sure to improve the capitol city-area in the mean time in order to speed things up. Go for warriors until that first Wonder is built. Somewhat risky, but it usually pays of.

3: Dont bother to accompany settlers with warrior-protection for settlements closest to the capitol. Somewhat risky, but you gain momentum, and its usually pays of. Let city number 2, 3, 4 and possibly 5 also produce 2-4 settlers each + 1 worker for each city.

4: As long as you havent build any city-improvements, or to many combat-units, you can safely boost the science-allocation to 90%.

[andkal]

Just one thing that I don't think is mentioned yet, build roads around your cities. Since roads being worked on, earns you one extra cash, which makes you able to set your science rate higher.

[Velociryx]

Hey guy! I'd pretty much agree with everything that's been posted so far, and will try and add my take on top of what's already been said:

Initial Cities:
More important than a lack of corruption is a surplus of food, so make your goal (especially if your capitol isn't exactly swimming in good food tiles) to found a minimum of two cities adjacent to high food production tiles (think cows and wheat).

For your worker farm, a game tile in a forest is cool, but when you're looking to crank out settlers fast, you're specifically looking for cows and wheat (and esp. wheat + floodplain!)

Set two....maybe three cities aside to just crank out settlers. Even in Despotism, if you have either of the high food production tiles mentioned above, irrigation of those tiles will bring in an extra food....make that a top priority....at this stage in the game, there's no such thing as too much food@!

After you've got your settler farms up and running....the very NEXT thing you wanna look for (if you don't have it already), is some luxuries. Make that a high priority too....especially if there's a cluster of luxury items....that way, you not only make your people happy, but you have some trade goods to acquire techs from your neighbors (and to that end, have some of the workers from your worker farms go ahead and start building roads out toward your neighbors as soon as you find them.....you'll wanna get a trade network going asap! (even better if they're building roads out to city sites WHILE inching closer to your neighbors.

Next item on the list, set aside a city spot close to the capitol for wonder makin' and assign a worker to it to spike shield production at that city. That way, when you get a tech with a wonder you want attached, you've already got a city for it (and when the city has the basics (temple, garrison), start "building the palace" to give yourself a leg up on wonder construction time).

After that, 1-2 troop cities. These will get temples and barracks, and start cranking out the best troops you can afford.

Once you have a dozen or so....you can either drop to the defensive or find the weakest civ near you and relieve him of some cities, ESPECIALLY cities founded around more luxuries!

Do that...and take one more city than you really wanna keep, then offer it back to him in exchange for....pretty much everything (cash, maps, contacts and money) he's got.

And you're all set!



-=Vel=-

[Dis]

build more units than the ai.

since I have started doing that, I have never lost. my only loss was because of my weak military- and somehow the ai knew I was weak.

the game is just that simple- kind of sad when I think about it like that. But still a good game.

build lots of units to win.

[ICMB]

Dissident,

Would that be the "Army Ant" strategy? Just overload them with units?

- ICMB

[Matthevv]

Find the luxuries to start with and build your cities to get them. That will give you something to trade, and they also give good amounts of gold from the squares.

Concentrate on commerce. Make sure there is a road in every square your city is using. That will give you gold.

Find all the other Civs and trade with them for techs. That is how the AI Civs get so many so fast, so you have to join in. Buy communications. Sell communications unless you are the only one in contact. Watch constantly to see who has contact, the AI Civs are pretty quick to find each other, as soon as another has contact, sell communications to all the other Civs. Sell on all the techs you buy to anyone that doesn't have them yet. They will get them from another Civ anyway, so don't try and keep a tech to yourself unless you are the only one that has it. You can offset the cost of buying techs against the cost of selling them that way. Its usually also worth selling most of the techs you develop first, straight away you develop them when you get the best price. The gold you get allows you to research the next new tech, and if all the other Civs are paying you gold, that also slows their research.

Get horseback riding and a supply of horses as early as possible. Keep an army of horsemen in a central location. Horsemen (later Knights) are excellent, as they move so fast, they can both defend your cities moving fast along your roads, and also, if the AI declares war, just attack and capture one of their cities with a bunch of them and then get on the diplomacy screen and they will usually offer peace. The AI at the early stages rarely has more than a couple of units in its cities, so half a dozen horsemen will usually do the trick. Sell the city back to them for some gold if it isn't close you the rest of your cities, as it will be corrupt and always inclined to defect back. Or if you are feeling wicked you can sell it to another Civ, which will probably guarantee they end up at war with one another sooner or later.

Build all your cities on hills if you can. You will get two food for this square and a 50% defence bonus, which is very useful if you get attacked.

[Heliodorus]

All very good suggestions here for you. I just won a game on Regent and have moved up to Monarch, where I'm getting to eat my own butt off of the plate on which it's served to me by the AI, but I'm still having fun.

Wonders: I think there is a general consensus that most of the time you're not going to build many - certainly the higher you go in difficulty, the fewer you have (and this is more pronounced when you have more Civs to compete against).

Workers: Workers are great, but stay focused on what they do in the early game. Building roads for future settle paths helps (founding new cities sooner makes a big difference in the early game). I would say focus on building roads to where you intend to found cities. Don't irrigate unless you must (plains), and do mine when you have ample time.

Expansion: I agree that expanding to block off territory is good, but start with four or five cities on good ground (not perfect, just good) around your capital - a good core of cities is essential. Also, designating certain cities to ONLY produce settlers is a great help. With a cow square or a grain square, a city can crank out settles very well. DEFINITELY build a granary there, and don't forget to POP RUSH. Sacrifice population for settlers (and for granaries and temples early). You don't have to go overboard if you think it's cheesy, but it will help.

Defense: The AI will take advantage of you if you don't keep up numerically with military units (NOT quanlitatively - the AI seems to equate a warrior with a knight when it comes to thinking of attacking you). So keep around about 2 units in each city (helps with unhappiness, too).

Trade and Tech: You just won't keep up technologically in the ancient era. Make up for it by being a pimp in diplomacy. NEVER offer something for a tech. Say you want it and select the option that asks the AI "what can I trade you for TECH" and when they respond, sometimes it's as easy as a world map. Sometimes they want money. Whatever they want, HAGGLE. What they ASK for something you want is typically more than they will accept. I spend time reentering money quantities over and over until I find exactly how much the AI will accept (later on, when I'm rich, I don't micromanage that so much because I can afford it). Also, NEVER trade techs on the AI's turn. Do it on YOUR turn, and make sure to trade it with EVERY AI you know to get the maximum advantage out of it. It's okay to trade techs (generally speaking) in the ancient era and early medieval unless you think you can get a big rush on a wonder over the AI (which is hard to pull off). Expect to catch up in tech once you get to Republic (a key turning point in the game, which must be managed carefully).

Hope it helps.

[eataTREE]

Thanks to everyone who responded for my plea for aid.

I am beginning to think that my problems in my current game stem from the fact that my starting location is buttville... I have two squares each of two types of resources (Furs and Horses) on the entire subcontinent where my civ is based. Also, there's no Gold anywhere to be found. No wonder my civ is broke.

I was wondering how you succesful civvers set your objectives when going to war with your neighbors. Is it ever wise to wage a "limited war", or should you never stop until the other civ is eliminated or subjugated?

I had a bad experience with this while playing the Egyptians. The Americans were my neighbors, and they declared war on me when I was not expecting it and conquered a section of my territory. This annoyed me, so after I managed to reconquer my original lands, I went on a slash and burn raid deep into American territory, razing 3 or 4 of Abe's bigger cities and putting the captives to the sword (hey, I didn't want to deal with the logistical nightmare of escorting 15 American workers back to my territory without having them recaptured). Anyway, my raiding party was running out of steam, so I signed a peace treaty. However, the Americans held a big grudge! Several turns later, they declared war on me again, and this time it didn't go so well for me. I have the feeling I did the wrong thing by giving the Yanks time to get back up. Any thought on this issue?

r.

[Arrian]

If you're going to fight an early war, pick a neighbor who has luxuries you don't. Attack a neighbor, because the cities you gain may actually produce something, rather than lose everything to waste/corruption. Make your choice based upon luxuries, because these are incredibly important, and they remain so for the entire game (whereas the importance of strategic resources comes and goes). The best method is to attack, take anything worth taking, and then demanding all of their tech, gold and cities. This will leave them with one city which you can destroy at your leisure later on, if you want to. It will also gain you technology and money to go along with cities and luxuries.

Basically, if you can get tech out of them, then totally anihilating them is silly, because then you don't get the tech. Still, it is a bad idea to leave them strong. Beat them down to 1 city. That way, they can hold a grudge all they want, but they can't hurt you.

By the way, I didn't mention this before because I've done it right from the start: once a new city has a garrison, the next thing I "build" is a temple. I say "build" because it's actually by working some of my poor subjects to death that the project is completed in anything resembling a reasonable amount of time. This gives you solid culture early on, which helps prevent conquered cities reverting (diplomacy is a bit easier with the AI also, if they are in "awe of" or "admire" your culture). Plus, the temple makes unhappy people content, which deals with the fact that you just pissed them off by working them so hard.

-Arrian

p.s. Heliodorus - That's pretty much how my first few forays into Monarch went. I was always behind in tech, which was a new thing, and I couldn't deal with that. I have since gotten the hang of it, and have won my last two games. Starting location is even more important, of course... and I admit that I restart if it doesn't look good after a few turns of exploring.

[Haphazard]

I fight wars:

To expand early on.
To get resources/luxuries from backward civs.
To get Leaders for wonders (I hardly ever use armies).
To get tribute; gold/turn or tech.
To nobble a powerful, or potentially powerful rival.
To retaliate against an aggressor.
To claim a continent.
To win the game.

For the first and last two I fight to the death. The other are limited. The nobble approach eataTREE mentioned is very effective - if you're in the lead.



I try to sue for peace:

If their defensive units are better than my attackers.
If I'm caught upgrading/expanding my army.
If my population is screaming for peace.
If I'm fighting on more than one front.

To quote someone around here: 'Build an army, then start a war. Not the other way round'


(Monarch level is pretty tough. Even with only 3 civs on the board I still don't get the wonders I like. Still, I got a tech lead in the Modern age, and nuclear war is very effective - launch nukes at their capital and resources patchs!)

[eMarkM]

eataTREE, it would help a lot if you could post your save game to the forum. Zip it up and attach it. All the advice so far has been good on this thread, but kinda generic. If we can see your game we could give more specific advice.

e

[Velociryx]

There is a section in the strategy thread that talks about how the local geography of your starting position impacts your game. Part of that impact is as simple as

few resources = poor = high player aggression to succeed.

Essentially, if you find yourself in a resource-poor area, you must be agressive and you must be agressive *sooner* rather than later....that is to say, before the AI empires have a chance to firmly establish themselves.

Also consider the prevailing nature of the terrain you're in.

If it's rugged (low food production, lots of forests and hills, possibly tundra), then you won't be able to make much use of the pop-rush, but you'll have good shield counts, and if this is the type of start you're looking at, then keep your empire small and compact to minimize corruption and get the most out of those shields, augmenting with the occassional pop-rush (but I'd say no more than twice).

OTOH (and at the other extreme), you might have started in a desert along a long river/flood plain. TONS of food, but virtually no shield production. In this case, you can expand a lot more, and a lot more quickly, cos you can pop-rush more frequently to get around corruption, but you have virtually no "native" shield production (ie - once you switch out of despotism, you'll have giant cities that can't really MAKE anything).

Likely your start was not as tough as either of those (vast desert, or tundra), so you'll probably have a good mix of shields and rushing available.

Now...to your upcoming battles:

1) As has been mentioned...resources. Specifically Lux items in clusters, Iron, and Horses (ancient era). If you can deny a neighboring civ his only source of either iron or horses, you've all but eliminated that civ from the game.

* Eliminate his supply of horses, and your fast troops are nearly unkillable (retreat when wounded)

* Eliminate his supply of iron, and your opponent has nothing that can match your swordsmen as far as offensive punch goes (note too that in some cases - Persian, Roman), their UU *requires* iron, so you can hurt them in multiple ways by simply denying them that resource.

Individually, either of these are crippling to the AI. Both together are a blow that cannot be recovered from.

2) You won't need an overwhelming number of troops. In a recent regent-level game I'm making into a story, I fought a comparably sized civ with 14 units, and captured three cities with ease. Pop-rushing can net you that many troops inside of 5 turns, in all probability.

3) Stay focused. Pick some limited objectives and don't get distracted away from them....especially don't get lured into running down enemy troops! If you want, say, the town nearest their supply of spice, the town that controls their iron, and that other town that sits between you and the two you want, then beeline for them, hit them quick and get it over with! Don't get dragged into a centuries-long war of attrition with the AI.

The longer the conflict drags out, the higher the probability that something unforseen will come up, and that it will not be something in your favor.

So...get in....hit hard, and make peace to consolidate your gains.

Remember, you can re-declare war on the same civ every 20 turns without taking a hit to your reputation, so if there are a number of objectives you have, you don't have to hit them all at once (ie - don't over-extend yourself!). Nibble away at him, slowly strengthening your position while weakening your rival's....

-=Vel=-

[ledj]

I played 5 games so far, and managed to win in most of my games. The latest one was on deity. Here is my limited experience.
(1) At higher level, a militaristic civ is preferable.
(2) Always try to found the capital on a grass land with a cow near by, and on a river square ( great commerce. no need for aqueduct and allow hydroplant), even if you need to move your first settler one or two turns. If you can not achieve this, you may need to restart.
(3) Set your research low and commerce high for gold production. Buy all the necessary techs.
(4) Found the second and third cities on or near luxuries, if possible.
(5) Built barracks in most of the cities.
(6) In despotism, use pop rush to build the city improvements and the military units.
(7) Produce 10 - 12 veteran military units.
(8) Conquer one of your neighboring civ and take all his techs. Use the leader to build a forbidden city in the center of conquered cities. Sell remote cities for techs.
(9) No research until you build the infrastructure and change the government to rep.

[Harfang]

I am pretty much in the same situation as eataTREE, but I managed to get out of the ditch at Regent level--though I've admitedly restarted several games. What helped me out?

1. I read Vel's strategy notes (just the first of the three threads, and not fully). The problem is to pick out what will work for us. The advice is great, but to try to absorb and apply it all at once is just too overwhelming.

2. I played on small maps. I find that limiting the number of opponents somewhat simplifies things when you don't have a good grasp of the situation to start with.

3. I played the Persians and then the Iroquois. They have the best ancient special units. Not only are they great offensively, but, even more important, they have a very long shelf life compared to other ancient units; they basically don't become obsolete until the Middle Age, and then mounted warrior can be upgraded to knights. Use them liberally against neighbours to get some serious breathing room for your expansion.

4. I build roads like there's no tomorrow and leave the gold/research at 50/50. I never have gold problems in early game. I also aggressively build road to luxuries and to connect my cities. I've noticed there's almost always a clusters of luxuries close to my starting position. As an aside, I noticed on a couple of occasions that the AI can get seriously ticked off when you secure too many strategic resources while it is itself short of them.

5. I beeline to monarchy. True, I do get behind for a while, but then watch what they'll offer you for it! I can walk to the most advanced civilization I have contact with, demand four techs and all its gold for it. Just like that. Too bad we can't ask for their firstborn son too. . .