• Our friends from AlphaCentauri2.info are in need of technical assistance. If you have experience with the LAMP stack and some hours to spare, please help them out and post here.

Getting to know multiplayer

Gogf

Indescribable
Joined
Oct 12, 2003
Messages
10,163
Location
Plane Of Fish Sticks
Multiplayer is a completely different world from singleplayer, and while playing MP may improve your SP, the inverse is only true to a certain extent. Humans and the AI act completely differently, and it's wise to be aware of the many differences this causes before joining a game.

Wars: understanding when fighting will and should occur

Humans will rarely leave a war unless their objective has been reached. If you're invaded by a foe with rifles against maces, they're not going to accept a few gold for peace. Conversely, if a human suddenly offers you peace, they most likely have a very pressing reason to do so. If you've been on the defensive and just got your first cossack, it's better to re-conquer all of your old cities than let your opponent fortify them!

Unlike AI's, humans are usually very well prepared for wars if they chose to attack. If a competent player declares war on you, expect a sizable stack of well promoted units, not a few improperly promoted out of date ones. Humans are also much better at defending than the AI; you'll need a good stack to launch a successful attack, not just your one first tank.

Preparedness: knowing when an attack is coming, and being ready

It's important to always be prepared for an attack. Humans are much more ruthless than the AI, so they will attack you if they see an opportunity. When defending, there are three things that need to be protected:

1. Cities
2. Resources
3. The high ground

Cities

Properly dividing military resources is a game of prioritization. Cities are usually the most important target that must be defended - especially in games with city elimination on - but your one source of metal is more valuable than a worthless tundra city.

In order to protect cities, always have multiple roads running to and between them. Even before the discovery of Engineering, reinforcements will travel twice as fast. The 50% movement increase along roads is a compelling reason to research Engineering early. Keep a sizable garrison of units in your front cities, but only those you need for happiness in cities that will never be attacked. The AI likes to defend its core cities better because it values them higher. Humans are effect enough at attacking that you'll need to move all of the units to the front that you can.

Keeping your main force hidden behind a front city can allow for surprise "ambushes" of an attacker. If what was one turn ago a lightly defended city is suddenly armed to the teeth, the attacker may either move on, hoping for a weaker target, or attack the city even when it is impossible to take. Retreating is always a good option if an attack is doomed to fail.

Defensive improvements - walls and castles - are effective in forcing the attacker to either have overwhelming force, or bring siege gear in order to take down cities. Attacking and defending units will be roughly equal if they are similarly promoted, but a +100% bonus will swing the odds greatly in favor of the defender.

While siege gear is clearly enormously important for the attacker, it is equally important for the defender. Being able to soften up an attacking stack before it attacks will help to equalize the effects of collateral damage from the attacker. Beware, though, that if you attack with your siege gear without being able to kill of the enemy's units, you present the opportunity to heal and return at full strength. You will have more time to reinforce, yet your opponent's units will live. Using siege units when you cannot kill the stack is sometimes a necessity for survival, but it should be done only as a last resort.

Resources

There are a variety of wars to protect resources. Founding a front city in front of them so that they won't be threatened is a good idea if you can defend it. If that's not an option, the best defense is usually to keep a small responsive stack - including a fair number of catapults - in the area to kill an attacking stack.

The high ground

Unlike the AI, people think. They will have no trouble cutting one road into a city, and won't attack your City Garrison III riflemen with macemen. Similarly, humans won't walk up their stack across flatland if they can avoid it. Always take precautions to cancel out the easiest routes of attack, including those with the greatest defensive cover. Chopping your forests not only provides a quick production benefit to your cities than can help your expansion, but it removes necessary defensive cover from the attacker.

Forests and jungles, the best cover in the game, can luckily be removed. Hills, on the other hand, cannot. Hills next to cities provide valuable cover for your opponent, so it's helpful to position units of your own on them to hold the high ground. A similar approach can be taken to forests and jungles which you haven't been able to chop yet. It is imperative to chop all of your forests and jungles near the front, especially those on hills.

Always be mindful of the terrain in front city placement. A city on a hill will get a +25% defensive bonus that cannot be removed to units in it, and an additional +25% to archery units in it. A city placed next to a hill, on the other hand, is vulnerable to the enemy fortifying their stack on it while they bring in reinforcements and their siege gear pounds down your city's defenses. A city behind a river gets a defensive bonus against units attacking across it, but units attacking out of it will also suffer this penalty!

Knowing when an attack will come - the use sentries

Another important aspect of defense is knowing when an attack will come. Expanding your borders is an excellent way of seeing an incoming stack earlier, but positioning sentry units outside of your borders can have an even better effect. If you see a "surprise" attack coming ten turns early, you'll have amble time to whip defenders and fortify your position.

Sentry units are best positioned on hills, both for the increased visibility and the defensive bonus. Having a "sentry net" of units extending the length of your border will usually provide enough warning to respond to a potential attack.

Roading the front

Finally, always remember to "road your front." Having multiple roads leading to, between, and from your front cities will make life much easier for the defender. Roads extending forward will allow for the easier use of catapults and defensive stacks.

Attacking: taking down human-defended empires

Because of the way the system works, attacking in Civilization IV is somewhat difficult. An attacker must bring enough siege equipment to knock down the walls of a city and do collateral damage, enough attacking units to actually take the city, and a variety of other units to provide cover for the stack. If your opponent has not chopped their forests - a mistake often made by nonladder players - two Woodsman II units will provide excellent cover for your troops.

The attacker has a number of advantages when they launch their attack:

1. Knowing when they intend to attack.
2. Knowing where they intend to attack.
3. Knowing that they intend to attack.

Even if a defender knows that an attack is coming, their newest cities may not be properly prepared. If the attacker launches their attack as a surprise, often because of a double move and the lack of an effective sentry net, they will have a major advantage.

Similarly, many players are simply unaware that an attack is coming at all. The most common form of nonladder game is the FFA, or free for all. Some player expect a buildfest, and they can often be caught with only one units defending all their cities. While this tactic may hold in singleplayer, against a decent opponent, this person will quickly be annhilated in multiplayer!

Similarly, if the defender has a number of cities on the front, the attacker can pool all of his forces in one place while the defender's are split. Staying equally distant front two cities while cutting roads between them is an excellent way of keeping the defender's troops separate.

All of the aformentioned advantages can be neutralized through the use of a good sentry net. A devious attacker can, however, use a mounted unit promoted with sentry to "break" a sentry net. The promotion, which grants its unit an additional tile of visibility range, allows the attacker to skirt around the edges of a sentry net unseen. In this way, the cunning attacker can hunt for weaknesses they can exploit while remaining unseen.

Another benefit for the attacker is that he attacker can split his stack into more than one, while the defender of a strategic tile cannot. The attacker can thus protect himself against collateral damage far better than the defender. The downside of this tactic is that each mini-stack is provided with less defensive cover.

Units heal far slower in an opponent's territory - 5% per turn, compared to 10% per turn in neutral territory and 15% per turn in friendly territory. Bringing a unit with Medic I or II can often lead to a far improved rate of success. This is one of the reasons that aggressive civs are so popular for early eras: even with only one promotion, they can access either Medic or an anti-unit promotion on their melee units.

Remember that retreating is always a good option if an attack is doomed to fail. It's better to back off and heal your units than to provide your opponent with experience and Great General points.

Lastly, never attack an opponent armed with metal primarily with swordsmen. While the AI doesn't seem to build enough axemen, humans have no problem doing so.

Building: Balancing expansion and tech

Growing and expanding

Multiplayer is often about expanding as much as possible early on. While eight cities in BC times may not give you a sustainable economy, having four will lead to a stronger economy than three. Having more early cities than your opponents will allow you to get more military than them, and have a stronger economy later on once your cottages have matured.

Larger cities can work more tiles and be more productive than smaller cities, so growing them as quickly as possible is usually a good idea. Use Hereditary Rule and a large enough garrison to keep them happy, then grow them to the health limit. Building granaries early provides a drastic increase to growth, as well as a useful health bonus.

Do not bend your civ around getting a single wonder; waiting to get a second city in favor of stonehenge will only serve to criple you. The pyramids are a similarly often overvalued wonder: they're nice to have, but you'll often want to go with Herdditary Rule anyway, which comes early enough if you focus on it.

Because workers and settlers slow growth, chopping forests and slaving to speed these builds can provide a useful boost. Be wary of too much slaving though, as the happiness effects start to add up. In a free for all with good players, it's usually a good idea not to appear to be the strongest nation. Keep a sizable military, but slave enough to stay out of first place early on. My limited experience with FFAs with good players indicates that not being a target is extremely important.

Technology and productivity

Having a large force of workers improving the land around your cities provides a massive boost. Working improved tiles is far more productive than unimproved tiles, so it's important to get tile improvements up and running as quickly as possible. The most productive tiles - resources, cottages on floodplains, etc. - should be connected first. It makes no sense to irrigate a tundra tile before a grassland! Similarly, workers are required to "road the front" around any potential front cities.

Cottages are extremely important in remaining competitive financially. Many nonladder players inexplicably do not build cottages, perhaps because of habits learned in Civ3. Floodplains are an excellent candidate for cottaging, as the city gains both a commerce bonus and a food bonus by working them. While cottages are a necessity, it is also imperative to build improvements intelligently. Do not cottage every tile, as it will then be impossible to work mines and other high production tiles.

Expanding a city's borders at least once is very important in order to claim its "fat cross." Monuments, religion, the Creative trait, and stonehenge all offer a quick and cheap way to do so. Spreading religion is already very important because of the benefits of various religion civics, and so it's often a better choice than building a monument.

The traits

Aggressive

Very important early on, the Aggressive trait allows double promotion units early, and triple promotion units later. The bonus is more significant than an additional 10% strength bonus to each melee and gunpowder unit. Combat I unlocks both Medic I and the powerful anti-unit promotions.

Industrious

Industrious's faster wonder building effect can swing the race for a crucial wonder into the Industrious leader's favor. The bonus production to forges is also very useful on what is otherwise a quite expensive building.

Spiritual

Being able to often switch civics without a penalty - especially into Slavery or Nationhood in case of an emergency - is a great boost. The cheaper temples are somewhat less powerful.

Organized

Organized is often considered to be one of the worst traits. Financial is generally a better choice, but organized becomes more powerful later in the game when more high-cost civics are available.

Financial

A great trait, financial is the most powerful early on. Cottages on a river of hamlets get a +50% gold increase, while post-Printing Press towns only get a +20% increase. Like industrious, financial can swing to race to crucial technologies - including those which enable wonders - into the financial leader's favor.

Creative

Creative is a powerful ancient trait for free border pops. It provides additional score, allowing for access to the "fat cross" without religion or a monument, wreaks havoc with non-Creative leaders' borders, and provides a useful defensive bonus quickly.

Expansionistic

Expansionistic is another strong ancient age tech. Almost all cities will want a granary, so it's extremely helpful to have them at a lower cost. Since the happiness limit effectively vanishes with hereditary rule, a health boost is also always nice.

Philosphical

A very powerful trait, philsophical allows full use of great people. Whether they are for culture bombs, academies, or free techs, it's nice to have them faster and cheaper.

Protective

Protective is a useful trait for both the builder and the warmonger. The builder can use it to more easily maintain control of his empire, while the attacker is required to build less defensive units, allowing him to build a more powerful offensive stack.

Charismatic

Charismatic provides a number of useful bonuses. The happiness bonus can allow for slaving without regard to the happiness penalty, and allow cities to grow larger in the period before Hereditary Rule. The bonus to unit promotions is also incredibly helpful. With the change to barracks, Charismatic leaders are now the only ones who can access triple promoted units with the Pentagon.

Imperialistic

Imperialistic leaders have an enormous bonus to production of settlers. With less time spent on settlers - or more settlers built in the same time - Imperialistic leaders have a massive expanding bonus over leaders without this useful trait. The bonus to great generals provides a helpful boost in the quest for these powerful great people.

For further reading, I encourage people to read Fried's The Fine Art of the Siege. I learned many of these tactics from that article, which goes into greater depth about both attacking and defending.
 
Nice introduction, readed all of it. This will surely improve my multiplayer skills.
Any tips what to do in general if you get teamed 1vs2 and you can't get any allies?
That happens a lot, only once in my short mp career I've win those 2 civs alone.
 
Jowy said:
Nice introduction, readed all of it. This will surely improve my multiplayer skills.
Any tips what to do in general if you get teamed 1vs2 and you can't get any allies?
That happens a lot, only once in my short mp career I've win those 2 civs alone.


pray.

If you are playing competant players and are not equal to both of them put together then it will be very difficult for you to suceed.
 
You're a good writer gogf. :) Nice Job.

Anybody who is interested in the ladder can easily do so by visiting this site.

http://www.myleague.com/civ4players/

For those who don't know it's a free online ranking system that minimizes cheating and quitting. Some of the best Civ4 players are here and once you join you're practically guaranteed to become an extremely better player than when you started off.
 
Jowy said:
Nice introduction, readed all of it. This will surely improve my multiplayer skills.
Any tips what to do in general if you get teamed 1vs2 and you can't get any allies?
That happens a lot, only once in my short mp career I've win those 2 civs alone.
The short answer is to get a really high manufacturing. If your MFG = the combined MFG of your two attackers, you will be able to produce enough to hold them off. With catapults and the advantage given to defenders, you should be able to survive with about 2/3 of their combined MFG. Protect your defensive terrain and use catapults to minimize losses and gain more promotions for your "real" units.

Nice thread Gogf, by the way.
 
Any tips what to do in general if you get teamed 1vs2 and you can't get any allies?
That happens a lot, only once in my short mp career I've win those 2 civs alone.

A few important tips to remember that will save your ass....

1. Don't play Civ3 tactics in Civ4!! Building a super large SOD (stack of death) may seem like the right way to go, but they can fall apart in a snap with the right combination of seige and attack units. This is probably the most common error I see people make in multiplayer.

And that leads us to....

2. Never underestimate catapults! I once had a guy come at me with a stack of 30 horse archers and 5 axemen. He didn't waste any time smack-talking about how I should beg for mercy and all that. With just 10 catapults (relatively cheap to build prod-wise) and a few axes and spears, I pummeled his stack before it even had a chance to attack. A handful of his damaged units ran back to his territory, and about 10 I think remained for the attack. But since he didn't have any catapults of his own, all 10 of his attacking units died without doing much if any damage to my small garrison. Unfortunately, he didn't stay long enough for me to build a counter-attack force; he quit immediately after his last attacking unit died.

3. Humans can be manipulated in ways the AI cannot! This is one you'll have to think about on your own to grasp its full meaning. The AI plays by the numbers, so it probably wouldn't be fooled by a diversionary force. Humans, on the other hand, have a tendancy to send most of their units to the scene of the battle....

That leads us to....

4. Don't forget about the enemy cities BEHIND the front line! Even in this article, you'll notice that the author recommends that you leave your back cities virtually undefended. I've often found enemy capitals, which produce a bulk of the enemy civ's units and wealth, defended only by a lone warrior or archer as late as the industrial era! Make the enemy think you're attacking the front-line city-- stage a few smaller attacks to throw him/her off (also helps make them overconfident, another human weakness). While you're at it, casually pillage the roads behind the front city (about 1-3 squares radius should do the trick).

Then, create a moderately-sized stack and send it to the frontline city. For the sake of your sanity, assume that every unit in that stack will die a horrible death (you know what pawns are in chess, right?). While those units are attacking and keeping the defending garrison busy, if you time it right, you can sneak your super-stack from behind the lines right through (use a road) to the backside of the frontline city. The enemy will have already committed most of his/her units to what s/he thought was the main attack force.

At this point, the enemy may still believe you're planning on attacking that contested city from behind. Then next turn, put your super-stack on a straight course for the enemy capital. Depending on the situation, it may be better to split it up into a few smaller stacks and send a few toward different cities-- that way, the enemy won't know where you're attacking, and thus won't know where to send reinforcements. Also, with the roads pillaged, the remaining garrison units in the front-line city won't be able to give a very effective chase. And if they do, have a smaller backup stack ready to threaten that city in order to force the enemy to either send some of those chasing units back to defend the city or let it be captured.

If you make it to the enemy capital, throw everything you have at it (if practical). When you take it, well, that leads us to the next point....

5. Burn, baby, burn!! NEVER take an enemy city without raizing it to the ground if the enemy is still powerful enough to potentially take it back. Otherwise, your advance will ultimately be haulted, and you'll be put on the defensive. And if the enemy retakes the city, then the whole offensive campaign will have been a virtual waste.

But if you take the enemy's capital, and burn it to the ground (including some wonders hopefully), then it's gone forever. Do that with a couple more early cities, and the enemy is dead. Oh sure, he could very well destroy your super-stack after you destroy his capital, but the damage has already been done. S/he will be permanently weakened. Assuming the player doesn't quit (they usually do by this point), start on your next stack, and don't take too long! Keep the enemy guessing.

Once the enemy is down to a few weak cities and pretty much done for, THAT's when you can hold onto what's left.


Those are 5 little-heard tips that have won me more multiplayer wars than I can count. I have a few other tricks too, but I don't want to give away all my secrets! =)


--Kristopher
 
Great article, thanks, Gogf. A couple things contradicted other advice I've heard. I can't say much to argue either way on these, but if anyone can argue for or against them, it might be helpful here:
1. walls should be used judiciously, because the bonus doesn't stack with the cultural bonus.
2. cottages should be used judiciously, because they're too easily pillaged by other humans, for too much benefit to the pillager, and with too much investment lost.
 
Jet said:
Great article, thanks, Gogf. A couple things contradicted other advice I've heard. I can't say much to argue either way on these, but if anyone can argue for or against them, it might be helpful here:

Sure, I'll defend the points I made :)!

1. walls should be used judiciously, because the bonus doesn't stack with the cultural bonus.

That's a fair point. If you're not going to build a castle, it's pointless to build a wall in a city with a 60% defensive bonus. The higher bonus always applies, though, so you will get a minor benefit in a 40% city, a larger increase in a 20% city, and a dramatic bonus in a 0% city.

2. cottages should be used judiciously, because they're too easily pillaged by other humans, for too much benefit to the pillager, and with too much investment lost.

This is absolutely, totally, and unconditionally wrong. If you intend to keep up in tech with people who play well, you will need a vast amount of cottages. Cottages are no easier to pillage than any other terrain, and you should always be doing the best to prevent your terrain from being pillaged. If your opponents are running their economy on money gained from pillaging your cottages, that is a military problem, not a cottage production problem. In fact, if you don't build enough cottages and consequently fall behind in tech, you're more likely to have your land pillaged!
 
great article. :goodjob:

Is it fair to say that MP games usually don't reach the modern era?
 
Oh yea, I play nothing but MP and can count on one hand the number of games that have reached infantry and tanks.

That 's an excellent article which I largely agree with, and learned some things from. My one comment is that it needs to be split into two articles, one for solo MP and one for team play. I'm finding the latter to be quite fun, and in some ways rather different in strategy and tactics from solo MP.


P.S. I did enroll in the ladder and played a few games, however ladder players seem to only want to play on the one ruleset ("CTON"). Which isn't a bad set of game conditions but I look for a bit more variety.
 
Regarding cottages, maybe the way to put it is this: if you don't start building cottages pretty early, it had better be because you are going for a blitzkrieg military win.

With certain civs on some common multiplayer setups (e.g. small pangea FFA, and you are playing Rome or Japan), planning to be Hitler to your neighbors' Poland and France is entirely plausible. Of course you have to worry about somebody ending being the Allies: building a rich economy on the other side of the land mass whilst you're ignoring cottages, and later coming around to stomp your warmongering butt into the dust...But I've both won and lost MP FFAs where somebody got metal early, got a stack of praetorians or whatever built fast, and went on a rampage. Or another version is having ivory at my capital, getting to construction first, and surprising one or two neighbors with elephants/catapults before they are at all ready to defend. That can get me juicy new capital cities, a couple of wonders, some lovely new key resources, and the ability to power up further to finish wiping the floor.

Certainly on standard maps and larger, and on maps that don't put everyone on the same land mass, the above strategy is much less viable. In that case cottages are crucial for the reasons described here.


Also: cottages are far less important in MP team play than in MP solo play. Team games rarely last long enough for cottage-building to really pay off.
 
The single most-important thing to figure out about a multiplayer game is, WHEN will it be won and lost? Not many MP games ever get to the modern age; games on small Pangea maps are often decided before anyone even reaches gunpowder. So you have to adjust your strategic decisions accordingly: don’t invest in building cottages on a small crowded single-land-mass map where war will come early. On the other hand human players are generally less likely to be killable with early attacks across distances, so on a larger map (or continents or islands maps) you need to take a longer-term economic-growth approach or you’ll just get out-built.

The point is simply this: much more than in singleplayer games, take a minute early on to think about how what sort of game you’re in and adjust your strategy accordingly.
 
Excellent guide. I'd played some MP right after launch, but have been unable to try again till recently. Great addition with the Warlords traits as well.
 
Unless you have a ton of gold mines or gems, always build cottages early and often no matter what the map type is. Every key tech you need you will get more quickly with a strong economy. If you are fighting with Longbows, War Elephants, and Maces while your opponents are fighting with Axes and Spears, chances are you are going to win.
 
a) I've never won a multiplayer game in which I've built a wonder.
b) I've, won multiplayer games in which I've kept the a city with a wonder.

What I still don't understand is under what circumstances would one build wonders, rather than spending the hammers on units, barracks and courthouses? Stonehenge and great library, maybe...
 
When you want to go for a point win, or when you need a great person for a tech, or when you want to win a culture war so you have an easier time attacking a close by city.

But never build a wonder if it will mean you wont be able to defend well enough. Watch the powergraph every turn.
 
Back
Top Bottom