The Theory of Gaming

May 2, 2008

7 rules all video games should obey

Filed under: Game Analysis — Tags: , — spotpuff @ 0:01

This is a great article from that outlines some huge problems with current video game design that are really hampering the industry. is a humour site, but it’s a very good humour site, with good writing and actually funny jokes. Well, I think it’s funny anyways.

Rule #1 is definitely my favourite since it touches on all sorts of issues from casual gaming to the ending for FF VI, which is hands down one of the best RPGs ever made and one of the best video game endings ever made. It’s very true that many manufacturers don’t seem to spend time on the endings on their video games to make the ending satisfying. They can’t tie up loose ends, because hey how would they have a sequel then?


April 30, 2008

GG Valve

Filed under: Game Analysis, Patches — Tags: , , , , — spotpuff @ 13:22

Alright you rogue, I’ll play your game.

Valve released a new patch yesterday for Team Fortress 2. It’s the “Gold Rush” patch that includes a new map and 36 new achievements for the medic class.

For every 12 achievements you attain, you receive a new weapon. The first is a new needlegun that heals you when you damage oppponents, the second is a new medigun that gives 100% crit chance to your target instead of invulnerability, and the third is the ubersaw which converts bonesaw kills to ubercharge; 4 bonesaw kills = 1 full ubercharge meter.

After spending 3 hours grinding these out last night for Kodene, I found out there is a command to do it. If you open the console and type achievement_unlock_all, you get all the achievements and new weapons.

Ah Valve. Pitiful content patches for 6 months and now this. How are they going to rectify this without rolling everyone back? Maybe they can scan everyone’s console log for that line and if someone typed it they lose everything?

Time will tell, but this is a major screw up on what was supposed to be the first major content patch of TF2. It does not inspire confidence in their ability to either deliver new content or patch existing imbalances in the game.


Well, they patched it and rolled back everyone, but some of the cheaters still kept their achievements.

And Kodene did too, so yay! He is the proud owner of an Ubersaw and the envy of other medics with their puny regular saws.

April 28, 2008

How come the monsters in RPGs are never afraid of you?

Filed under: Game Analysis — Tags: , , , — spotpuff @ 2:29

I played Final Fantasy for the first time in 1992. I was 10 years old and NES was all the rage. The learning curve was steep; the beginning of the game was challenging and not a whole lot of fun. But a friend of mine insisted the game got more fun as you went along, so I persevered and indeed he was telling the truth: the game did get more fun if you played past the difficult early stages.  Final Fantasy was the first RPG I had ever played but it would almost certainly not be my last.

Fast forward 16 years and the Final Fantasy series is still going strong. Square’s still putting out arguably the best RPGs in the world, and Final Fantasy is on at least its twelfth iteration, not counting spin offs and releases in other genres like Final Fantasy Tactics. I recently got a PSP and have played through Crisis Core: Final Fantasy 7 and am replaying Final Fantasy 1 in its 20th Anniversary Edition.

FF1 is still great and you can certainly see the progression made since the early days of RPGs. Crisis Core is beautiful, no doubt about it, and Square Enix have continued to improve their CG animations from their disastrous movie venture because the FMVs in the game are absolutely jaw dropping. They are bar none the best CG animations I have ever seen in a game or on a movie screen. Maybe it’s because I haven’t played a Final Fantasy since number 6 for the SNES, before the days where every game had FMV sequences, and thus have no recent Final Fantasy visuals to compare these to, but even when compared to the FMVs in recent hits such as World of Warcraft these are some stunning visuals and hold up very well.

Nevertheless, there are still some nagging things which seem to creep into every Final Fantasy game even though there are ways around them.

One of these annoyances is encounters with low-level monsters in low-level zones. For perfectly legitimate reasons, many games will have you traverse back to an earlier location in the game. You might need to talk to someone, get an item, fight some monsters. Whatever. You have to go, so you go. Now in games like Final Fantasy, which are “booby trap encounter” games (that’s a very technical term, I assure you), you tend to walk around and randomly encounter monsters as you do so. Sometimes, you’ll walk around for what seems like hours without encountering any monsters, and sometimes you’ll finish one fight, take a step and then run into another group of monsters while shouting “Random encounters my ass!” at the game screen. The annoying thing is, even when you are 50 levels ahead of the monsters, you still run into them. And even though you can pretty much kill all of them in one hit, you can still only really attack one monster at a time and they still might get to hit you first. So combat ends up taking a lot longer than it should given the monsters should really be afraid of you at this stage.

Crisis Core has a similar problem. There are side missions in Crisis Core that you can do in addition to the regular storyline of the game. Now for whatever reason, Square decided that you should be able to do these missions and level your character up far beyond where you should be to progress in the regular story. The regular story wasn’t all that interesting to me because I never played the original Final Fantasy VII for Playstation of which this game is a prequel. On the other hand, every time you do a side mission, you get a piece of loot. It could be materia, it could be an accessory, it could be an item. 90% of the time it’s crap, but for whatever reason, I found myself compelled to complete missions and continue getting loot that I had absolutely zero use for. Probably for the same reason I put myself through two years of the same crap in World of Warcraft.

Anyways, what ended up happening in Crisis Core is I was about 15-20 levels ahead of where I was supposed to be in the game’s storyline, so when I went back to playing it instead of doing side missions I was one-shotting monsters and 3 or 4 shotting bosses. A boss my cousin took 15 minutes to beat I killed in literally 10 seconds. And it wasn’t hard either, I just button mashed my way to victory.

Crisis Core is a booby trap encounter type game as well, but from a 3rd person view rather than an overhead view. Because the monsters are invisible and the combat screen is essentially the same as the regular world-wandering screen, you need some sort of transition time for monsters and the combat HUD to fade into view. For Crisis Core the transition is a line of text along with a voice announcement that says “Entering Combat”. And when the fight’s done, you sheath your sword and you see another line of text, along with another voice announcement, “Conflict Resolved”. Now besides Square’s questionable definition of conflict resolution, I became hugely annoyed with these transitions. Why? Because actual combat was taking less time than entering and exiting combat. These transitions don’t take long, about 4 seconds per encounter, but when you’re running around killing everything in 2 seconds or less, should you really be spending double that time looking at text on the screen that you’ve already seen 500 times?

The most aggravating thing about this is that Square solved this problem at least as early as Final Fantasy 6, and possibly earlier. In FF6 you got a trinket called the “Moogle Charm” that made it so you didn’t encounter any enemies in dungeons or in the world map. If you were in a dungeon, you still had to fight boss monsters, but you didn’t have to deal with the fodder in between. I can see why Square didn’t include this in Final Fantasy 1. For one thing, there’s no “accessory” slot for you to equip this to, unlike FF6 or Crisis Core. They could of course make it some piece of armor, like a helmet, or a weapon of some sort that had the same effect, but for whatever reason they didn’t.

This wouldn’t be such a big deal if the game was an exact replica of the original game, but it isn’t. They did make some minor improvements which add to your enjoyment of the game greatly, such as party members auto-targetting a new monster to attack if the one you told them to attack at the start of the turn was killed along the way, or you not trying to talk to the air if you hit the talk button accidentally while wandering around town, or the use of mana to cast spells instead of the “x casts per magic tier” as in the original. Believe me, if you played FF1, you will understand and know that these are good things to have added to the game.

Given Square added other things to the re-release of the game, the lack of a “no monster encounters” (NME) item seems like a bit of an oversight. You need to go through monster battles to gain experience and level up; if not, you won’t be able to beat boss monsters and progress through the game, so having a NME trinket is not a huge advantage for the player to fast-track through the game and in fact makes progression impossible when you reach a gatekeeper monster such as a boss. It is however a huge time saver to bypass those annoying encounters that seem to pop up far more often than you would like when you are traversing through previously visited areas of the game.

I honestly have no clue why Square didn’t put such an NME trinket in either the FF1 remake or Crisis Core, but they didn’t. If you’re easily annoyed like I am, every time you run into a group of monsters you really shouldn’t be running into you get annoyed, and a 5 second battle feels like a minute and your 2nd encounter feels like your 100th. I hope that other RPG/adventure game producers consider this point and allow players to skip encounters that are wholly unnecessary to the progress of the game. If you want, you can give it to the player near the end of the game, where most of the backtrack-type quests are given anyways, but please put an item like this into all your games. Those goblins were scary at level 1, but not at level 50.

April 20, 2008

Creating Advantages in Warcraft 3

Filed under: Game Analysis — Tags: , , — spotpuff @ 3:01

This is a repost of an article I wrote several years ago and submitted to (which is now defunct). It won article of the week and I won $50. Woohoo!

Anyways, here it is. Enjoy.


Warcraft 3, as well as any other game, can be viewed as a homeostatic system, meaning that it begins in equilibrium (although that could be debated, until the next patch comes) and both players try to create disequilibrium in their favour, while responding to the disequilibrium their opponent creates. A player will try to create advantages in his favour while eliminating his opponent’s advantages, ultimately trying to destroy all of the opponent’s buildings.

This article is intended for Warcraft players who are of average skill level who would like to improve their play. Improvement will be reached hopefully by changing the reader’s mental perception of the game by formalizing some concepts and examining how those concepts affect the outcome of the game.

To be successful, you don’t have to understand what you’re doing; you only have to find someone else who is successful and copy them. While you could theoretically become a very good Warcraft player just by copying what other top players are doing, you will learn faster by understanding why it is those top level players do what they do.

Build orders and micro/macromanagement are important; however they can be practiced or copied from good players. After outlining some basic concepts, this guide will take a look at various strategies that can be used and what kinds of advantages they create.

The System

In Warcraft, advantages and disadvantages tend to fall under the following categories, in approximate order of importance (though this order may vary by race and strategy):

1. Material: possessing superior units to your opponent in number or function

2. Tech: possessing a superior technology level than your opponent

3. Hero: possessing heroes of superior level or function than your opponent

4. Resource: possessing a superior amount of resources than your opponent

5. Scouting: gathering more information about the map than your opponent

6. Micro: possessing superior micromanagement than your opponent

7. Macro: possessing superior macromanagement than your opponent

8. Space: controlling more of the map than your opponent

9. Time: decreasing the amount of time necessary to gain other advantages

10. Damage: being able to deal large amounts of damage quickly to a unit(s)

Games will revolve around gaining these advantages for yourself while preventing or countering the advantages your opponent creates for themselves. A common saying is that “Warcraft is a game of counters”, meaning that advantages in the game can be eliminated provided the correct actions are taken.

An important note is that while these advantages are distinct, they are not isolated. A resource advantage in gold and lumber by itself is useless unless converted to a material advantage. Similarly, a tech advantage is useless if not converted to a material advantage. There is no point upgrading to tier 3 if you do not plan on producing any tier 3 units or researching any tier 3 upgrades, as you have spent resources that could have been used elsewhere.

These advantages form the basis of any tactic or strategy employed during the game. The more of these advantages you can create for yourself, the greater your probability of winning will be. If you are not actively creating these advantages at all times, you are giving your opponent an advantage. The idea here is similar to that of chess, where players take turns moving pieces; do something every move. Every move is one that cannot be taken back or undone, so something productive must be done with each and every move or your opponent will most likely win the game (unless they too make similar mistakes).

In Warcraft, the same is true. If you are sitting at your base with your army and your opponent is creeping with theirs, then they are gaining a hero (XP (experience)) and resource advantage (gold, experience, items). If you are sitting on a stash of gold and lumber while your opponent is spending his resources to tech or produce units, then they are gaining a material advantage over you.

If you are a fairly novice player or very new to you may be overwhelmed at first by the dramatic increase in complexity that playing an actual person poses rather than playing against the computer. People can be unpredictable and don’t follow the exact same strategy every game. One thing you may notice from viewing replays of “top” level players (available at,,, and other sites as well) is that they are constantly doing something, be it creeping, harassing their opponent, or spending their money. They don’t hold a reserve of resources unless they need to for strategic reasons e.g. to tech or avoid upkeep. Good players are always actively creating advantages for themselves.

Now, let’s take a look at some strategies and examine why exactly they are effective at giving the player that employs them an advantage.


Harassment is one of the most devastating tools at a player’s disposal. Almost everyone has played a game and went creeping only to be interrupted by a Demon Hunter (DH) or hear that their base is under attack by a Farseer (FS) and his wolves. Depending on the opponent’s skill, they may be able to creep and fend off your harassment at the same time, although you are usually able to inflict some casualties on your opponent’s base.

Knowing the advantages outlined initially, the effectiveness of harassment may be a little bit clearer to you. Worker harassment provides the harassing player with an indirect resource advantage and gives them a time advantage. Your opponent must respond to harassment either by moving worker units out of harm’s way or by constructing buildings to help eliminate the harassment threat; both these options take up resources. Peons cannot gather resources while in burrows, nor can peasants while engaged in Call to Arms.

Night Elves (NE) and the Undead (UD) have less of a problem with harassment; NE by spacing wisps around the map (which provides a scouting advantage as well as being an anti-harassment measure) and UD because of their Nerubian (frost) Tower and the usually present lumber ghouls to attack any units threatening acolytes.

The UD frost tower is particularly effective as an anti-harassment tool due to its slow effect, making it easier to surround heroes and forcing them to either die or town portal (TP). An enemy hero’s death gives you a small resource (revival cost) and experience advantage (100xp + any xp gained while the enemy hero is reviving) and forcing them to TP gives you a resource advantage (350g for another TP).

As the harassing player, you should keep those things in mind. Level 1 heroes revive very quickly, leaving little time for your opponent to increase their experience advantage. Also, reviving a level 1 hero is cheaper than buying another TP; although if the time you save by TP’ing over reviving can be more constructively used through further harassment, then by all means do so. This is a judgment call based on experience and the countermeasures your opponent has taken to your harassment.

Should you choose to creepjack your opponent rather than harass their workers, you are attempting to gain a resource advantage by picking off injured units or creeps, giving you XP and possibly gold and items as well. This tends to work best with heros with a nuking spell (Death Knight (DK), Warden, Blademaster (BM), etc.) as it allows you to deal the final blow to a creep, after your enemy has done all the work and taken all the damage from fighting it.

Another function of harassment is to divert the attention of your enemy from your base. If you are trying to tech early, you will most likely not have adequate base defenses or troops. If your opponent is good they will scout your base and attack it, while relying on their natural base defense to hold off your harassment and hope they can use their material advantage to overwhelm your base defense.

Generally speaking, HU bases are the most prone to harassment, especially later in the game when heroes level up and area of effect (AOE) spells increase in damage. Large numbers of peasants may be dispatched extremely quickly, leaving the HU player’s resource inflow crippled.

Next weakest are Orc bases, whose primary defense, burrows, only have heavy armor until upgraded at tier 3 with fortified defenses. As such they take full damage from all damage types and extra damage from magic attacks (as many Orc players will attest from the dreaded destroyer rush, which was thankfully fixed in 1.16). Once burrows and watch towers are upgraded to fortified armor, however, they become extremely difficult to kill. Spiked barricades also prove extremely effective against light melee units such as footmen and ghouls, and totally useless against all NE tier 1.

NE wisps are prone to harassment early from heroes with a nuking spell and later in the game the entangled mine is also a target for harassment as it prevents the NE player from mining gold giving the harassing player a resource advantage.

UD bases are very difficult to harass as frost towers and ghouls tends to limit the amount of time a hero can spend in the base. A popular strategy is to kill acolytes while the UD player upgrades his Necropolis, as an upgrading Necropolis cannot make new acolytes. A frost tower, combined with moving acolytes around the base and forcing the harasser to run between buildings while taking damage, tends to make this rather difficult. Another option is to destroy the haunted gold mine, as without this the UD player is unable to mine gold and it requires considerable resources and time to rebuild. If you manage to kill a UD player’s mine and all their acolytes, and they didn’t save enough cash to build another acolyte, they will be forced to creep or sell items to afford an acolyte and a new mine.

Countering harassment is usually a matter of increasing base defense or diverting a few units to attack a harassing hero. For example, an AM and 6 footmen creeping may be harassed by a DH; create a new control group with 2 footmen and send them to attack the DH while continuing to creep with your AM and 4 remaining footmen. While most heroes have fast movement speed, this will still prevent them from hanging around too much while you go about your business.

If you are being harassed at your base early in the game, it may be prudent to leave 2-3 units there to attack any hero that may try to harass you.


Teching provides you with a functional material advantage over your opponent at the expense of resources. The units available at tier 2 and 3 offer much better functionality than tier 1 units for their cost. Knights and abominations are more generally more effective for their cost than footman or ghouls.

Generally units at tier 2 tech are support units and cannot function as the sole units in your army; the notable exception being wyverns, which can overwhelm opponents if timed correctly with harassment.

Tier 2 support units should be used to augment your army, not replace it. Everyone remembers the mass shaman days but those days are gone. Casters can augment your army, but are easily countered and shouldn’t be used as an army.

Tier 3 units will typically replace tier 1 units because of their increased damage/cost ratio, and several tier 3 units can completely dominate the game if obtained e.g. frost wyrms, tauren, chimeras.

Teching provides you with the advantage of having access to superior units than your opponent. To counter teching, you must nullify the advantage it creates by either teching yourself or not allowing your opponent to construct buildings that allow him to take advantage of his tech level. For example, destroying your opponents arcane sanctum, tauren totem, or slaughterhouse when almost finished constructing puts them even further behind. Having already sacrificed resources and time to gain a higher tech level, you can use those resources instead to build additional units, creep, and destroy his tech buildings. This nullifies his tech advantage, as he wasted all that time and money with nothing to show for it.

Hero Killing

Hero killing provides you with an experience advantage. Different races have different hero “nuking” strategies, with UD and HU typically being the most feared.

The premise is simple: if your enemy has no hero, you can trade units, even at a huge loss, and still have an experience advantage over them, since their heroes won’t be getting experience while they’re reviving. It frees up a temporary time advantage for you to creep and gain an even further advantage in experience over them; in addition to the gold advantage caused by them having to revive their hero. Killing your opponent’s altar will further cripple their response, as this provides you with a further time/xp advantage and they will be hesitant to trade more units with you for fear of allowing your heroes to level up.

UD players use this strategy a lot. They quickly tech to tier 3 to get frenzied ghouls, statues with healing and 3 heroes (typically DK, Lich, CL). The idea from there on out is to impale your army, surround your heroes with ghouls, and then nuke said heroes to death. Once that happens, the remaining ghouls and heroes mop up the priests/sorceresses/rifles sitting around, or the human runs away, and the UD creeps. Even if you do lose some ghouls, because they are tier 1 units they are worth less and less XP as your enemy’s heroes level up, and even less than the typical units you will be killing in the process (rifles, casters, grunts). Either way, the UD has a significant XP advantage over the human, and UD heroes are extremely hard to stop at high levels.


Expansion gives you a large resource advantage at the cost of time and a temporary resource outlay. If you set up an expansion and it goes unchecked by your opponent, you will have double their gold income, which leads to a unit/tech advantage. Expanding also puts pressure on your opponent, as he will want to take your expansion down quickly to minimize the economic advantage you gain.

Countering expansion follows the same ideas as countering teching: use the resources that your opponent spent on expanding to build additional units and then attack his expansion to gain an advantage yourself. If you are unable to do this, you can also expand yourself, or remove your opponent’s unit building structures such as barracks. This way you prevent your opponent from exercising his economic advantage (unless he decides to dump all that gold into a lot of peons and peon rush you). This gives you a time advantage to build even more units and continue attacking his expansion to take it down.

Expanding is important since resources on any given map are limited; if you control more resources and are of equal skill to your opponent, you will win. Typically if your opponent has not expanded and you have, you can defend your expansion while scouting the map and destroy any expansion he tries to build. Expanding gives you a huge resource and material advantage late in the game because you can continue building units while your opponent cannot; every unit you lose is replaceable (to some extent) while every unit your opponent loses without a source of income is gone forever; eventually this leads to your victory.

As a player looking to expand, put up a lot of towers at your expansion and wall them in with other structures, such as item shops, farms, or barracks. It is my opinion that putting up three or four towers at an expansion is a worthy investment if it prevents your expansion from being harassed. I cannot count how many replays I have seen where, in the face of constant harassment at an expansion, a player will refuse to build towers simply because they do not wish to allocate the resources to do so. Unfortunately, losing peons at 75 gold each and possibly the expansion itself should be enough reason to defend it, but other players disagree with me.

Defending your expansion forces your opponent to allocate more resources to destroy it. If you have 4 towers at your expansion, your opponent will not be able to kill all your peons with a few ghouls or dryads; they will be forced to send a larger group of units to it. Given the fact that a player’s micro resources are limited, most players will use their whole army to attack your expansion if it is well defended, as they cannot co-ordinate an attack on a sufficiently large scale as well as creep at the same time with their hero or another section of their army.


Scouting provides you with an information advantage which may lead to a unit advantage, a resource advantage, or a time advantage. If you see that your opponent is hard teching to tier 2, you can rush their base and destroy it before they even get there. If you see your opponent is investing heavily in casters, you can get anti-caster units and inflict heavy casualties on their army, giving you a material, hero and time advantage.

Further, scouting gives you control over the map. That is, you can creepjack your opponent, destroy his expansions, and time your attacks against his base if you scout well, as well as properly counter any units he may produce. If you see your opponent heading towards your base, you can also defend your base or attack your opponent’s base and destroy some structures before TPing back to your base to defend it.

Scouting is extremely important but oft times not done well. In the early game, it is well worth sacrificing a peon to gain information on where your opponent’s base is and what units they are building. Knowing your opponent’s base location gives you an idea of the creep order they may proceed in as well as giving you the opportunity to attack their base and destroy critical tech structures (sanctums, spirit lodges, tauren totems, etc.) before they can be used.

Later in the game, sending a peon to scout expansions (shift clicking) is simple and easy to do and provides you with lots of information about where your opponent is and is not. Often times late game victories depend on expanding successfully while denying your opponent an expansion, or destroying any peon lines that he may be using to get gold from a distant mine.

Constant Offense

In any game where you cannot determine the position and/or actions of your opponent’s army at any given time, the person who is constantly active and attacking will win. Constantly creeping or attacking your opponents base creates a situation where your opponent may be hesitant to leave their base for fear of it getting attacked. If this is the case, you are almost assured victory, as you are free to creep at will the entire map.

Coupled with adequate scouting, you can time attacks on a player’s base that inflict enormous casualties. If you destroy 4 burrows and then TP out, while your opponent was walking, then you have gained a huge material and resource advantage. If your opponent TPs you will generally have SOME time to destroy buildings and/or units and then TP out yourself; in which case you have traded TPs and the only way your opponent can come out ahead is if the resources and experience gained creeping (or whatever they were doing when you attacked their base) outweighs the resources lost in your attack.

Never get stuck in a defensive mode. In solo or AT games, if you are stuck defending, you will lose, as you are wasting time while your opponent may be creeping or expanding. Simply put, you cannot defend your opponent to death, you have to destroy their base. If you sit at your base and wait for your opponent to strike, they may simply send in a scout first (always a wise idea) and if they see your army is defending your base and not willing to leave, they can creep the whole map. In AT games, only one base may be adequately defended against a concentrated enemy force, forcing you to waste a TP and time which may cost units and/or buildings. If you do TP to help an ally defend their base, then you’ve spent a TP and your enemies can attack another base, forcing you to spend more money on TPs simply because they took the initiative to attack you.

You can see that being the aggressor in Warcraft holds many advantages, both tactical and strategical in nature.

That being said, this does not mean you should throw caution to the wind and attack your opponent’s base at every opportunity. Sending in a scout first to ascertain the situation is advantageous as if your opponent has a large force at his base you can spend time creeping rather than rushing in and losing units. If your opponent does not have units present, feel free to run in and kill some peons or important buildings. If you can catch them without a TP far from their base, you will have gained a huge advantage.


I hope you found this guide informative. I am not a great player and there are many top players who probably know all of the above information intuitively or through practice (that is after all why they are top players), but I hope this guide will help you increase your skill level and maybe think about the strategy you are aiming for and what advantages you hope to create or nullify in the course of a game.

Good luck and I hope this guide has helped!

March 30, 2008

Unit Costs Including Food Supply in Warcraft 3

Filed under: Game Analysis — Tags: , — spotpuff @ 6:38

Food isn’t free.

When looking at how much a unit “costs” in Warcraft 3, few people consider the costs associated with unit food supply into the cost of the unit. In most cases the most relevant data is simply the unit’s listed gold and lumber cost. Food usage does effect the unit’s final cost though since food supply is not free. Like any other resource in an RTS game, food costs you something to supply and that cost can be calculated.

This table illustrates the “actual” real cost of each race’s food supplying building.

Table 1: Food buildings with modified lumber cost

Building Food Provided Gold
Lumber Cost Build Time Lumber Cost+
Farm 6 80 20 35 55
Moon Well 10 180 40 50 90
Burrow 10 160 40 50 90
Ziggurat 10 150 50 50 50

The Lumber Cost+ column indicates the actual cost of lumber given the fact that a worker unit must construct the building and while they’re doing that they could have been harvesting lumber. This is of course assuming that you use a lumber harvesting worker to build a building and not a gold harvesting worker, which I think most would agree is a logical assumption.

Feeding the troops

So you’re ready to build a bunch of units to go crush your opponent’s base. Unfortunately for you, you have gold and lumber, but no food. What’s a player to do? Build some farms, of course.

Assuming you’re already at your pre-set food limit (all players start with a town hall which provides 10 food), you will be required to build additional farms (or town halls, which is less efficient than buildilng farms) to build additional units. Additional units do not just cost their listed gold and lumber price, they also incur an additional cost due to food costs. If you don’t believe the above is true, hit the food cap and try to build some more units. I promise it won’t work unless you spend some more resources to build some farms.

Now of course each race’s “farm” unit costs a different amount, and provides a different amount of food or function to each race. Burrows and Ziggurats can be upgraded at an additional cost to towers and Moon Wells double as health and mana restoration. But for the purposes of this article I’ll be looking specifically at the cost of the farm buildings and not their “value added” features, which at the end of the day don’t really matter in terms of food production. Regardless of how many static defenses your base already has, if you hit your food cap and need to build more farm units, a Ziggurat still costs the same amount even if you don’t need more defensive towers.

Given the food supplied by each farm building, you can calculate the additional cost involved in building any unit in Warcraft 3. That little food icon that each unit costs can be converted into an additional gold and lumber cost associated with each unit. Below is a table that shows the cost of each unit with the cost of their race’s farm building added in.

Table 2: Modified unit costs considering food cost.

Unit Gold
Lumber Cost Food Cost Gold Cost+ Lumber Cost+
Dragonhawk Rider 200 30 3 240 57.5
Flying Machine 90 30 1 103.33 39.17
Footman 135 0 2 161.67 18.33
Gryphon Rider 280 70 4 333.33 106.67
Knight 245 60 4 298.33 96.67
Mortar Team 180 70 3 220 97.5
Peasant 75 0 1 88.33 9.17
Priest 135 10 2 161.67 28.33
Rifleman 205 30 3 245 57.5
Siege Engine 195 60 3 235 87.5
Sorceress 155 20 2 181.67 38.33
Spell Breaker 215 30 3 255 57.5
Archer 130 10 2 166 28
Chimaera 330 70 5 420 115
Druid of the Claw 255 80 4 327 116
Druid of the Talon 135 20 2 171 38
Dryad 145 60 3 199 87
Faerie Dragon 155 25 2 191 43
Glaive Thrower 210 65 3 264 92
Hippogryph 160 20 2 196 38
Huntress 195 20 3 249 47
Mountain Giant 425 100 7 551 163
Wisp 60 0 1 78 9
Demolisher 220 50 4 284 86
Grunt 200 0 3 248 27
Kodo Beast 255 60 4 319 96
Peon 75 0 1 91 9
Raider 180 40 3 228 67
Shaman 130 20 2 162 38
Spirit Walker 195 35 3 243 62
Troll Berserker 135 20 2 167 38
Troll Headhunter 135 20 2 167 38
Tauren 280 80 5 360 125
Troll Batrider 160 40 2 192 58
Wind Rider 265 40 4 329 76
Witch Doctor 145 25 2 177 43
Abomination 240 70 4 300 90
Acolyte 75 0 1 90 5
Banshee 155 30 2 185 40
Crypt Fiend 215 40 3 260 55
Destroyer 0 0 2 30 10
Frost Wyrm 385 120 7 490 155
Gargoyle 185 30 2 215 40
Ghoul 120 0 2 150 10
Meat Wagon 230 50 4 290 70
Necromancer 145 20 2 175 30
Obsidian Statue 200 35 3 245 50

Of course food is not unlimited; there’s a cap of 100 food supply in Warcraft 3, so there is an even greater cost associated with using food than just looking at the resource cost. When you start getting close to either of the “soft” food caps at low and high upkeep (50 and 80 food, respectively), or the absolute food cap at 100, the cost of each unit becomes significantly more clear.

Perhaps you need to set up an expansion but you’re currently at 95/100 food used. Do you spend the last 5 food supply making 5 more peasants? Or do you build some more units to fight your opponent? Without food constraints you wouldn’t have to make that choice: you could make both. But food limits are in the game and so you have a decision to make. In the future I’m hoping to examine and quantify the costs associated with each unit more closely in order to look at effectively managing resources.

Generally speaking, the food limit will never be reached in a game since you and your opponent(s) will be trying to win. Low and high upkeep may be reached but both players usually do not allow each other to “build up” and hit the absolute food cap. Still, many players do not take into account that little food icon each unit costs is essentially a resource that costs gold, lumber, time and micromanagement resources as well.

That’s it for now, thanks for reading!

Rick commented that the cost of food supply is amortized over the lifetime of the farm building and this is true; I should have made it more clear that I am talking specifically about hitting the food limit and building more farms.  Assuming you’re going to build a NEW unit (which hopefully won’t die, but in all likelihood will) the costs are accurate.

Things can get even more complicated when you consider low and high upkeep, since in that case gold becomes even more scarce and costs are essentially inflated since you are making 7/4 gold per worker trip instead of 10.  I’ll hopefully look at resource management in a future article.

Thanks to Rick for the tip.

« Newer Posts

Blog at