EPT+ Show Spoiler +
Ever since Day9 talked about Frisbees and Baseballs, people have been humming and hollering about Broodwar’s superiority to SC2’s pathing system: mostly talking about how spread out the armies were and how each unit have this “micro potential” inherent within it.
I would like to break that illusion with this blog post.
Specifically, I would like to clarify *what* exactly Day9 was talking about, in order for people who have never played Broodwar to understand what we miss about the game.
Day9 brought up the idea of baseballs and Frisbees.
The metaphor, to an extent, makes sense. Baseballs, for the most part, fly straight (they don’t really, but that’s nitpicking) while Frisbees fly in curves (sort of, but close enough). Since baseballs fly straight, being good with a baseball is about speed and power; which is why baseball players wow each other with how fast they can throw a ball and how well they can slow it down, speed it up, as well as creating illusions of how far it can go and causing it to create slight curves in its trajectory. Because Frisbees fly in curves, being good with a Frisbee requires finesse and accuracy; which is why you see Frisbee players being able to toss a Frisbee across a field with just a flick of the wrist, or be able to get a Frisbee to curve from the perimeter of a park back to your intended target.
By and large, it is harder to hit targets with a Frisbee than with a baseball because Frisbees are delicate and glitch out with the slightest gust of wind. Baseballs, on the other hand, requires near perfect precision and strength—because unless you throw a baseball hard enough, it won’t gain enough momentum to go any distance worth-while.
In short, Day9 attempts to reveal the dynamic nature of Broodwar’s units by likening them to a Frisbee while also showing the difficulty and yet all-innish nature of Starcraft 2’s units by likening them to a baseball.
However, the metaphor doesn’t actually reveal *what* it is about Broodwar units that feel like a Frisbee and *what* it is about SC2 unit’s that feel like a baseball. Overall, it simply paints a picture that Broodwar is obviously more dynamic and that SC2 is obviously more rigid. This might have been the intent—I don’t believe it is, but I’m willing to accept the possibility that Day9 is more biased towards Broodwar. + Show Spoiler +
Understandably so, it’s a great game.
First, I will talk about the dynamic nature of Broodwar unit control. I will then talk about the precision based nature of Starcraft 2 unit control. I will not have a conclusive summary at the end, because I do not believe this is the kind of topic to have a conclusion.
Broodwar is Speed-Chess, on Crack, with Aliens, at the same time.
By and large, the Broodwar pathing could be oversimplified into that very simple description. More specifically, Broodwar pathing is a grid based system of open squares and closed squares. Open squares allow units to move through them while closed squares do not. Sound’s simple right? Except it isn’t.
![[image loading]](http://www.codeofhonor.com/blog/wp-content/uploads/2013/02/Starcraft.jpg)
As can be seen in the image, the grids are not evenly split apart. While some are half filled, others are barely filled, while some only have their corners filled. Each unit that will pass by those grids, due to their varying unit sizes, will treat each grid differently.
In Patrick Wyatt’s own words
Because the project was always two months from launch it was inconceivable that there was enough time to re-engineer the terrain engine to make path-finding easier, so the path-finding code just had to be made to work. To handle all the tricky edge-cases, the pathing code exploded into a gigantic state-machine which encoded all sorts of specialized “get me out of here” hacks.
If you read Wyatt’s blog, he will talk about the many tricks and coding practices he implemented into the game, the many stories of how those codes came about, etc… But that’s not what I want to talk about. The how and why of the coding process is irrelevant to the physical act of actually playing the game itself. What this post is about, first and foremost, is a discussion on why we play the game the way that we play it.
When people talk about fighting the unit AI what they are talking about is fighting the desire of a unit to neatly fit into a single square on the game grid; the problem being that not all units take up the same amount of space. Let’s look at the initial image again.
Notice how the zealots fit neatly into the 8x8 boxes. Now look at this image.
![[image loading]](http://www.thinkdigit.com/FCKeditor/uploads/Brood%20Wars%20-%20Zealots%20and%20Dragoons%20make%20their%20way%20to%20take%20on%20a%20Zerg%20player(1).jpg)
And notice how the Dragoon is double the size of the zealot? The dragoon wants to be as centered on a square as the zealot is: but when you’re a unit that big, which of the 4-8 squares that you’re standing in do you decide to walk into?
The answer is all of them:
This is the where a large majority of a unit glitches come from in Broodwar, but it is also where the micro potential comes from as well. Each unit in Broodwar has a different size, a different shape, and a different orientation. So while a Dragoon is a square that expands and contracts as the legs move, the Vulture is a tight rectangle that doesn’t change in size or shape. The marine is tiny, while zerglings double in length as they run, their stride stretches them as their legs first expand outward and then contract back inward. Units rotate, and as they do, they change their shape and their spatial relationship with the squares that they are trying to fit into. A vulture moving is a flat rectangle while a goliath moving is a vertical rectangle. A Goliath never changes his vertical orientation but a Vulture could be a vertical or horizontal rectangle depending which direction he is going. Each of these shapes interacts with the grids differently depending what direction they enter a square, and which parts of the square is open and which parts are closed.
So let’s go back to the picture of the bridge.
Zealots will tag these incomplete boxes and then correct themselves into the open boxes, adjusting themselves as they traverse the bridge. Notice how the zealots clump into their own respective squares as they maximize the space in which they are present in. The same thing occurs when they are in an open space, but instead of a bridge chopping up the partially filled squares, it will be other units doing it.
Imagine a dragoon and a zealot in an open field. The Dragoon will take up at least 4 squares if it stands at a corner of a square and up to 9 squares if it stands at the center of 1 square with its unit box spreading to the 8 squares around him. The zealot, much like it did in the bridge, will attempt to fit inside its own isolated square causing the zealot to create a space between himself and the dragoon. Assuming a completely flat field without obstacles, they will try to maintain this formation since it allows them to be isolated in their own squares.
This causes the Magic Box effect we all know and love from Broodwar. For reference:
http://wiki.teamliquid.net/starcraft/Magic_Boxes
When the units are far enough apart from each other, there will be enough empty squares for them to fill. But otherwise, they will be content to staying in their own relative squares from each other barring obstacles such as terrain, other units, etc…
The micro potential with units happens when you want them to traverse a non-open field. Because although they want to correct themselves and fill an ideal space, the truth of the matter is that even squares that have been cut smaller by terrain/units are still open areas that a unit can theoretically fit itself into, assuming you as the player force the unit to fit into it.
Remember the earlier video of the glitching Dragoons? Notice how during the first 2 seconds of the video, before the attack command is given, the Dragoons inside the ball kind of dance and jiggle in place while the dragoons forming the outer wall of the ball stand still as stone. This variance between the inner Dragoons and the outer Dragoons came about because each was given a different unit command.
The outer dragoons are given a hold position, forcing the Dragoons to stand in place despite their desire to fill an empty square. The dragoons inside the ball have not been given a hold position and so they want to fit into an empty square. Because they are unable to find an empty one, they jitter and wiggle as their pathfinding continually reroutes them over and over again looking for that empty square.
If these dragoons were in an open field, the jiggling of one dragoon would force the dragoon beside it to also jiggle as the first dragoon walks into the private space of the second dragoon. If no terrain or hold position units were there to stop the correction, the slowly spreading jiggling of dragoons would spread the tight dragoon ball into a spaced out army that fills 2-3 screens.
What does this mean? It means you have to use the whole command card.
![[image loading]](http://img600.imageshack.us/img600/3933/gridye.jpg)
This means that with the proper use of Move, Attack, Patrol, and Hold Position commands, you can choose to either have a tight ball of units moving across the map, or a spread out army of units moving across the map.
This means that when you have a Vulture running around the back of a mineral line, you can tell it to stop while being pointed at the perfect position to fire a shot without it trying to adjust itself, and then begin to move it the second the shot has been fired creating the illusion of a never stopping unit doing a “move and shoot” maneuver and in doing so it never loses momentum as it never is given the chance to decelerate allowing your microed Vultures to always be faster than unmicroed vultures. If the vulture is not pointed in the right direction, it will need to adjust, once it tries to adjust, it will then try to fill an open square, and only once it fills the open square will it attempt to fire or move, in which case it has to accelerate to top speed all over again. This "glitching" means that an unmicroed vulture harass literally moves slower and shoots less often in the same amount of time as a microed vulture harass even if the opponent does not react.
But more specifically, since you have to manually tell the unit when to stop, when to shoot, and when to move—the unit’s ability to perform basic tasks rests solely on your ability to make the unit do those tasks. When people talk about Starcraft 2 being simply “A-Move” units, this is what they mean. Design wise, almost all the Broodwar units are A-Move units and in comparison are much less dynamic and much less interesting than the SC2 units they are matched with. However, because of the limiting nature of the pathing system in Broodwar, all units in Broodwar required attention. You not only needed to perform micro tricks with them, but you also needed to tell them when to move, tell them when to stop, how to stop, what direction to face, and in what formation to move or not move in. All these commands required not only different buttons, but different button combinations depending on direction of movement, terrain present, and activity performed.
If you simply attack move a ball of Dragoons, they will first spread themselves out trying to stay clear of each other and then attempt to surround the targeted area. But if you use a move command to override the desire to isolate themselves, you move them as a tight deathball . If you then use the hold command, they will maintain their tight formation to maximize damage against melee units like Zerglings and Ultralisks. But if you need to dodge storms you could use an attack command or a stop command instead of a hold command in order to cause the dragoons in range to start attacking while the dragoons that aren’t in range will start spreading. As the dragoons in the back start to auto spread, you have time to manually spread the front dragoons. During this time the enemy could then send a flanking waze of zealots to hit your rear dragoons that are currently glitching, getting free hits while your ball breaks apart.
The choice of a deathball or a spread out army becomes that, a choice, instead of a norm. What allows players in Broodwar to gain an identity is the fact that you can’t master all the different ways to command all the units in the game. There will be people who spend their energies mastering how to move dragoons and there will be players who master how to move Vultures. And it’s not even just about moving vultures or moving dragoons. Some will be very good at moving marines in a tight ball. Others will be good at moving marines as a spread out composition. The tight ball will destroy mutalisk flocks but die instantly to lurkers. The spread out marines, because they’re too far away from each other to focus fire, will be picked off one by one by mutalisk packs; but are better able to break lurker lines. And since moving tightly or loosely had pros and cons, no player would be punished by their choice to master one instead of the other. This means that you can have two players of the same race, who specialize in the same unit compositions, still play completely differently from each other. Fantasy’s vulture control was great at harassing, but Flash’s vulture control was great in open field combat. This allowed Flash to win with his suffocating map control play style while Fantasy was able to abuse his highly mobile terrorist squads to be as equally devastating.
A lot of people talk about units having better AI or about units having more micro potential and wanting that micro potential to be programmed into the SC2 units—but there never was anything programmed into the AI to do that. What they are referencing is the limiting factors of the pathing system being circumvented through precise unit control. It is also these same limiting factors that give Dragoons a hard time when walking up a ramp: the size of the dragoon in relation to the squares in a tight ramp causes the algorithm to seemingly glitch when all it is doing is trying to find an ideal square for the dragoon to stand in and being unable to find it.
Imagine a dragoon trying to find the empty square on this ramp.
![[image loading]](http://www.teamliquid.net/staff/Hot_Bid/3MSLsemis/Free_Nada_24.jpg)
Not only do the rocks on the sides of the ramp chop up the squares, but the marine at the top and the dragoons in the bottom also fill up squares. Without babysitting, the dragoon will glitch out trying to find the perfect empty square that doesn't exist except for the center of the ramp. This causes a lot of Dragoons to either get stuck in the middle of a ramp, or to believe the entire ramp is blocked off causing them to "reroute" and find a another way around the cliff. Without babysitting the Dragoons at the bottom won’t even begin to climb up until you manually force them up the ramp 1-2 at a time.
Knowing that, could you imagine going up this ramp?
![[image loading]](http://img264.imageshack.us/img264/4096/scscrnshot111310203539.png)
Getting large units up this ramp requires babysitting and lots of move commands. As you gently nudge units up this ramp the cannons gets dozens of free shots. If you don’t babysit the ascent then 1-2 units get stuck going up the ramp and the 3 cannons will only be facing at most 2 units at a time allowing them to increase their effective hit points by the hundreds. In Broodwar this defense would be able to stop large pushes long enough for reinforcements to arrive especially when supported by either a templar or a Reaver.
Now imagine if instead of Photon Cannons it was lurkers?
Imagine having a hard time getting units to go up a ramp, and imagine if the pathfinding forced your units to clump in ramps: now imagine trying to move command up a ramp with AOE landing on you?
1-3 Lurkers
1-3 Siege Tanks
1-3 cannons and a High Templar
Defenses like that could stall entire armies. It wasn’t the high ground miss chance, or the better defensive units, or higher siege/storm damage, or even the existence of lurkers that gave true defenders advantage. Lurkers happened to be the most cost effective at it but 6-10 Hydralisks could hold the top of a ramp just as easily—just not as effectively.
Many times the high ground was not just an advantage due to the miss chance, it was literally much more difficult to fight high ground units than low ground units from a micro perspective.
And it’s not just ramps. Tight chokes whether from buildings or from terrain created this dynamic as well. It was physically more difficult to get units walk through a valley let alone to get them to fight effectively in a valley. It’s not just about who has the better concave, but also about the attacking force being more likely to have a unit glitch out due to the clumping.
A large part of Broodwar AI came about due to the core problem of an archaic grid systems being used with variant sized units. People keep thinking that its bad unit design, or missing broodwar units, or extreme unit clumping that is the problem with Starcraft 2 when in reality none of those things is what made Broodwar work. Many of the things being asked for are not things that was initially given in Broodwar. The development team did not think “Oh, we need to make sure these units could be microed well” or “We need to make certain units spread out.”
They simply had an archaic system repair a problem that wasn’t even conceived of yet. However, since it wasn’t conceived yet, it wasn’t understood as a problem.
Starcraft 2 came into the scene and unlike the dev team of Broodwar, this Dev team was not crunched into thinking that
Because the project was always two months from launch it was inconceivable that there was enough time to re-engineer the terrain engine to make path-finding easier, so the path-finding code just had to be made to work. To handle all the tricky edge-cases, the pathing code exploded into a gigantic state-machine which encoded all sorts of specialized “get me out of here” hacks.
Having all the time in the world, they fixed the previous issue by implementing a more streamlined pathing system and in doing so created the so called “baseball” feel of the game.
![[image loading]](http://i.imgur.com/9PxLP.jpg)
That it is shaped like a baseball is beside the point.
Starcraft 2 loses the clunky features of the old systems and replaces it with a style more akin to Chess than it does to Broodwar. Don’t mistake the usage of the word chess to mean strategic (because both games are very strategic) but, more importantly, understand that unlike its predecessor, Starcraft 2 emphasizes strategic choices more so than it does strategic execution.
In chess you simply had to decide on a move, and then execute it. You want the pawn to move? Then move it. Think it should be the Queen’s turn? Then it’s the queen’s turn. By the simple act of deciding an action, you can, for all intents and purposes, simply perform the action.
Do your Stalkers need to go up the cliff? They’re now up the cliff.
Siege Tanks need to come down the ramp? They’re now down the ramp.
This is not to say that there are no mechanical limitations in Starcraft 2, but the mechanical limitations in Starcraft 2 comes from your brain/hand speed matched against your opponent’s brain/hand speed. Because it is easy for you to position your units in Starcraft 2, it is also equally easy for your opponent to position his own units in Starcraft 2.
Did you forget to keep track of his army? Too late, he’s doom dropped your base.
Did you notice that blip that passed the edge of your observer's vision? Too late, they’ve sniped your Templars.
This is a laughable defense in Starcraft 2. Without the high ground miss chance, the effective health of the cannons does not increase. Without bad pathing, units simply stutter up the ramp dealing too much damage for the cannons to handle.
Units are not slowed down by terrain in Starcraft 2, they do not get hindered by high ground, chokes, or each other’s company like they do in Broodwar. This makes drop play even more devastating in Starcraft 2 than it is in Broodwar. In Broodwar, unmicroed Vultures will hit about 1-2 workers at a time while simply right-clicking beside mineral patches with a pack of Hellions can roast more than that in 1-2 volleys. And while well microed Vultures can wreck a worker line, one well placed Hellion shot can evaporate a mineral line.
![[image loading]](http://www.gosugamers.net/starcraft2/images/2011/may/bombermvp1.JPG)
This means workers die in 2-6 clicks of the opponent as opposed to the 2-6 clicks it took just to fire a shot with vultures. A Terran player can send three dropships into three different mineral lines, not micro any of them and still wipe out the grand majority of workers in all three areas. Smartcasting means that if your army is out of position storms can and will hit your entire bio force without quarter all while zealots and stalkers charge forward decimating your retreating forces.
Starcraft 2 is faster, much faster, than Broodwar—and I'm not talking about game/movement speed. Tactically speaking, mistakes are punished harder, and making a positional mistakes is much more difficult to recover from. This is because, like chess, since it is easy for a player to make strategic decisions, it becomes very difficult for the player who makes the first strategic mistake to recover from the initial setback.
Imagine you are playing chess. The board has moved about, as tends to happen in chess, and suddenly your opponent captures your queen. Maybe it was a mistake, maybe he made a good play, whatever the reason: your queen was taken. As the opponent moves his piece and takes over the queen’s space—what fanfare and action occurs?
Nothing.
Because it is easy to execute moves, once you are caught in a bad position, more often than not the result is already decided. Broodwar, on the other hand, is not like chess; Broodwar is more like streetfighter.
In Broodwar, positional play is possible because the clumsy controls slows down the pace of the game. Units are guided through chokes, they are babied up cliffs, and you try your best not to pull a Jaedong.
+ Show Spoiler +
![[image loading]](http://i.imgur.com/RpUZK.jpg)
All of these feel very chess like because of their methodical nature: you move your army around the map, you place key pieces on the board, you set up feints, traps, etc… But once an engagement happens, that is where Broodwar legends make their names.
In fighting games, you need very precise button combinations to initiate individual moves. Simply wanting to throw a fireball at the opponent does not mean the fireball will fire. Simply knowing that a combo chain requires X, Y, and Z buttons to be pressed in a specific order does not mean that you will get it to happen. Majority of the skill in a fighting game is not about knowing when to produce a move—it's about having the ability to execute the move and having the reflexes to capitalize on it.
Because of Broodwar’s pathing system, many maneuvers that are currently normal in Starcraft 2 required certain button combinations to be able to achieve. Between hold position, patrol, move, stop, and attack commands you could sometimes be tapping 2-3 different buttons just to move around 1 unit. Once you’ve mastered these moves it becomes as normal as watching Ryu throw a fireball—but because it requires some level of dexterity the players best able to execute these moves the fastest quickly become recognizable.
![[image loading]](http://i865.photobucket.com/albums/ab215/jlee9201/jangbi1.jpg)
Above is a Jangbi storm. On paper, it is relatively boring. Two storms are cast on a line of tanks; units in the back are caught by a Stasis spell, and a line of zealots that was there is no longer there. In Starcraft 2 this would be considered a bad engagement for the Protoss. But this is Broodwar, to simply be able to cast two storms side by side like that on an empty field is a trick most players can’t perform. I’m not joking, when I say that many players cannot even cast those two storms side by side in that manner on an empty field with no enemy units killing your High Templars.
Jangbi is not only able to cast two perfectly spread storms; he can also aim to hit the maximum amount of tanks despite EMPs, Siege Tank fire, and Spidermines getting in the way. It required fast hands, good timings, and a complete understanding of not only how protoss and terran units behave, but also an understanding of what type of play style the enemy player has specialized in.
In Broodwar, it takes a lot of clicks to properly move an army from point A to point B. Ramps can sometimes take up half your APM, but that same amount of hand speed is even more needed in combat and whoever is more effective during the engagement is usually who comes out the victor: and this is what Broodwar players miss the most. In Broodwar, if a piece tries to capture your queen, your queen still has a chance of making it out alive. And since the tactical maneuvers in Broodwar are not as quickly decided as it is in Starcraft 2, every engagement is exciting. In Broodwar, even a strategic checkmate does not mean the king is captured—with proper micro even a bad position can be won.
In place of this, Starcraft 2 has emphasized what most people describe as “positional play,” “Good Concaves,” and “deathballs.” These are misused terms that is attempting to describe a new game using the parameters of an old game.
The reason for this is that combat in Starcraft 2 is largely logistical and compositional in a nature. This is due, in part, to the smooth pathing system implemented into the game in conjunction with the removal of the grid system. If you bring X Colossus to a fight against Y Vikings, the viewer already knows the results because they already know that a 1 to 3 ratio of Colossus to Vikings is bad for the colossus. For the most part, the excitement is most present about 3-5 minutes before the engagement when both players are still scouting each other.
People who watch poker knows the tense feeling of watching two players reading each other over the course of 10-20 hands. Once all the flops, turns, and rivers are in play the excitement is over, but it’s the raising, the folding, the bluffing, etc… those are what make the game fun to watch.
![[image loading]](http://graphics8.nytimes.com/images/2007/12/12/nyregion/12poker.span.jpg)
“I know I have Blink Colossus, but does he have pocket Vikings? Should I Raise?”
The difficulty of Broodwar’s controls shapes how we perceive it as a game. The ease of Starcraft 2’s controls has a similar effect. In Broodwar, seeing 10 Dragoons engaging 10 Dragoons can have many different outcomes and the better Dragoon player will always come out on top. However, in Starcraft 2, 10 stalkers meeting 10 stalkers will produce heavy casualties for both parties involved unless it’s a pure blinkstalker war like the homestory cup match between JYP and MC.
Which looks very similar to this Bisu vs Jangbi game. + Show Spoiler +
Despite the lack of Blink
Both result in a back and forth fight between even numbers of troops where you aren’t sure who will win in the end. So it isn't that Starcraft 2 is unable to produce the same types of micro intensive games and fights as Broodwar, it is that the emphasized aspects of Starcraft 2 as a whole does not stem from its unit control. It instead stems from positional play, concaves, and deathballs.
![[image loading]](http://www.gosugamers.net/files/images/StarCraft%202/August%202013/taeja-innovation.jpg)
Positional Play
Because armies are so much more mobile in Starcraft 2, it is much more efficient to block army movements with your own army instead of depending on small clumps of troops stalling large pushes. You can't afford to forget the minimap because it only takes a few seconds for the enemy army to suddenly wipe out your main. This creates a very tense back and forth tempo based game that is not as interested in engagements and is more interested in good scouting and proper positioning.
This is why base races happen much more often in Starcraft 2. Since you can't slow down enemy advances with pockets of troops, being unable to put your main army in front of their main army puts you in a disastrous position where all of a sudden you are stuck in the middle of the map and have to choose between retreating to your ravaged home or charging to attacking their undefended structures.
![[image loading]](http://3.bp.blogspot.com/-UdhF6fHaUNM/TdUofC2cwVI/AAAAAAAAAWs/yLqfEzv5AN4/s1600/1.JPG)
Concaves
The smooth unit movement creates fewer chances for in-combat micro. Because there is less unit glitching, you don’t need to babysit your units as much as you did in Broodwar. It also means that your opponent creates less holes for you to take advantage of.
In Broodwar, stray units glitching away from the main army can be sniped for free. Pulling back your army and then reengaging with it causes the enemy units to stop attacking and instead they begin walking towards your troops causing the bad pathfinding system to make them fumble and glitch out of formation and hence easier to kill with your tighter formations. A good player will see this happening and grab hold of his straying units in order to keep them in line, usually pulling them back as well. This creates the back and forth play you see in the Bisu vs Jangbi video posted earlier.
Because Starcraft 2 does not have any this embedded into its pathing, the micro comes about before the fight starts since most of the time you do not have the speed necessary to keep up with the units on the board. Prepositioning becomes important and is usually in the form of concaves, flanks, and surrounds. This links back with the positional play nature of Starcraft 2 since you want to constantly have preset formations, but at the same time, if you fail to constantly move your army in front of the enemy army, your perfect formation will get you stuck in the middle of the map. A game thus forms between two armies constantly bobbing and weaving from each other while trying to setup the better formations before a fight ensues. The player who gets to setup the concave/flank faster than the other player will win the fight but all that micro happens mere seconds before the fight actually begins. Once the fight starts, it’s much like in Chess—we already know which piece will win the engagement.
![[image loading]](http://i.imgur.com/0eAeh.jpg)
Deathballs
This is the reason unit movement is so efficient. By getting rid of the archaich grid system of Broodwar, Starcraft 2 units are always comfortable exactly where they are. Terrain doesn't bother them, units do not bother them. There is no longer a desire to fit into a square box that is usually smaller than the unit trying to fill it.
But to really understand what I’m talking about, we need to look back to the Broodwar pathfinding system.
Recall this initial image:
Notice how those incomplete squares remain naturally unfilled and notice how blocky the compositions of the zealots are on the bridge itself. They don't clump evenly since each zealot is attempting to isolate itself into a single square even during the act of moving across a bridge.
Go back to the deathball image, notice the circular formation of the units? Without a desire to isolate themselves into a single isolated square, the units compress as close as they can to the last command order—the mouse click. They all converge into the singularity and evenly spread themselves around it. Since there is no longer partial squares scaring off units, movement across terrain is fluid, like water flowing over smooth rocks.
Because the positional nature of Starcraft 2 as a game, the bobbing and weaving of armies so that they are always facing each other head on forces them to always move, this movement clusters them into the spherical deathball shape we all recognize. When a lull occurs, the army that wins is normally the army that is able to properly form a concave—however, since armies move so fluidly, most players avoid enemy concaves and instead attempt to walk around entrenched positions. This constant mobility causes units to clump up again and again despite the desire for players to spread out the units into preset formations.
Since terrain does not slow them down, ramps do not provide the defenses that they used to provide, chokes do not frustrate the attacking force in the same way; army movement in Starcraft 2 will always be faster and more sudden than in Broodwar. However, due to the smoother interface, you will also, for the most part, never have this problem in Starcraft 2:
+ Show Spoiler +
These three aspects of Starcraft 2 create the game dynamics we see every time a game pops up on Twitch.
If you do not see where your enemy is, isolated units will not stall them long enough for you to run back home and engage. Broodwar is tactically slower since each move you make takes more time to execute than it does in Starcraft 2. Once armies are in position in Starcraft 2, most fights become decided much like in Chess. Broodwar is less punishing, tactically, than Starcraft 2 because of this. Which is why Starcraft 2 feels like it is over in one fight; because once the dealer flips down the river, there is nothing left to do but show your hands.
+ Show Spoiler +
![[image loading]](http://upload.wikimedia.org/wikipedia/commons/1/14/Holdem.jpg)
“Final round of bets gentlemen”
Starcraft 2 is about everything that happens before the fight occurs. It’s about mini-map blips on the screen; it's about realizing you need ghosts despite spotting 2 colossi in production; it's about pre-splitting; it’s about knowing fully and perfectly where your opponent's army is, and how to contain their position with tempo based attacks. Much like in chess, engagements are predetermined and hence one must rely on perfect strategic play at all times. Strategic mistakes will cost you your entire army with you being unable to do anything about it.
This is why we get stuck with terms like concaves, positioning, and deathball. Because when it is easy to execute commands, we do not become impressed by the results of the commands themselves. We can only describe the physical actions we see in a macro scale. This is why I have compared Starcraft 2 to chess and poker. No one is impressed by how well a player moves a chess piece. No one is impressed that you are good at commanding Bishops while your opponent is good at commanding Knights. In chess, like in poker, the beauty of the game comes from the steady reading of each other's movements and decisions. The execution of the movement matters little; it is the culmination of the movements made that is exciting. But when we use terms that only define the end results of Starcraft 2’s foreplay, we end up describing seemingly bland things such as deathballs and concaves, ignoring the beautiful dance that occurred before the fight happens.
That is not to say that Broodwar is purely executional in nature and Starcraft 2 is purely strategic in nature. Strategic play happens all the time in Broodwar while intense micro play happens all the time in Starcraft 2. ST_Life can charge a line of Widow Mines and lose few to no zerglings or banelings due to his strong unit control in Starcraft 2 while Savior can pre-position Lurkers to setup concaves and traps like no one else could in Broodwar. Jangbi’s hit and run movements with Carriers is an amazing feat to watch while Rain’s seemingly effortless Oracle control makes the unit seem broken. The difference between the two games is not the absence of strategic or tactical play in one or the other, the difference is on the emphasized aspects of strategic and tactical play inherent within both games.
In Broodwar, you can set up the perfect trap—but if you don’t have the mechanical capability to execute the trap you will still lose.
In Starcraft 2, you can have the best micro in the world—but if you failed to see the trap there is almost nothing you can do to recover.
You need to be able to perform strategic play in both games, and you need to be able to execute tactical play in both games as well. A terran who can’t split marines will never win the GSL much like a Protoss who can’t baby a shuttle will never win the Global Seol Guji League. The point of I’m trying to make is that the two games are the way they are due to how we as players respond and interact with the game board presented to us.
The units in Broodwar were not better designed than the units in Starcraft 2. The limitations pressed upon the units in Broodwar forced a play style that rewarded fast hands and, as Day9 put it, twitchy personalities. Due to this, before you are able to play strategically in Broodwar, you need to be able to execute certain maneuvers that allow you to have strategic options that you can actually use. Knowing where to move your army is not as important in Broodwar as being able to *actually* move your army. It is all too easy for a better player to show up at your doorstep with fewer units than you and still win the engagement due to his better control leading to an early loss.
You can get a lot out of a single unit in Broodwar.
+ Show Spoiler +
The opposite is true for Starcraft 2, since the interface is easy to execute, the pace is increased from a marathon into a sprint. Since units already do what you need them to do in Starcraft 2, getting caught off guard is something that will lead to an immediate loss. Unlike in Broodwar where a smaller enemy army can still kill you—Starcraft 2’s logistical nature makes being able to hit benchmarks the most important aspect to the game. No matter how terrible of a player you are, if you are at least able to keep up in unit production as your enemy you will survive.
You don’t necessarily need perfect marine splits to break a ling/bane army with pure macro.
+ Show Spoiler +
So when people talk about the better “designed” broodwar units or the more “spread out” nature of Broodwar units—take their statements with a grain of salt. Not because they are wrong, but because they are asking for something from Starcraft 2 that is outside of Starcraft 2’s abilities.
The lurker was not what allowed Zerg to hold terrain on the map—it was simply the best tool for the job.
The siege tank was not necessarily stronger in Broodwar—but it did severely punish bad unit pathing.
The circular clumps of Starcraft 2 do not look any less “natural” than the blocky nature of Broodwar.
Etc…
A lot of what was loved about Broodwar came from the fact that one needed to overcome the game itself. This need for proper execution meant that players would always make mistakes creating holes in their play that could be exploited by the opponent. The best players were the ones who either made the least mistakes (like Flash) or the ones who punished mistakes the most severely (like Hyvaa).
Players, for the most part, do not make execution mistakes in Starcraft 2. They do make strategic mistakes; and it is these strategic mistakes that are most taken advantage of. When you scout wrong, when you position wrong, or when you become too predictable—you suddenly get routed. This means that the best players are the ones who either make the fewest strategic blunders (Innovation) or are the ones who punish strategic mistakes the most severely (Soulkey).
This is what makes these games so different from each other despite their name and despite the fact that they are both RTS games.
Broodwar loyalists don’t understand why one fight wins the game and Starcraft 2 loyalists don’t understand why your units sometimes wander off to some random cliff face.
People who miss Broodwar don’t enjoy watching the army posturing, and the back and forth displays of dominance that Starcraft 2 armies employ like birds flaring their feathers at each other while people who prefer Starcraft 2 don’t understand why you have to right click a mineral patch for every worker you build.
People who dislike Broodwar dislike its tedious nature while people who dislike Starcaft 2 dislike its lack of granular control.
The reason for this is because they are not the same game at the most fundamental level; as different as Age of Empires is from Total Annihilation. Abstractly they are both RTS games, they both require the construction of units, and they both require your direct control of units. But in the end its more akin to Go vs Chess, Soccer vs Football, and Frisbees vs Baseballs.
The reason that taking small aspects of Broodwar and transferring them into Starcraft 2 will not work is not because Broodwar concepts are bad or even because Starcraft 2 concepts are better. The reason is because aspects of Broodwar that people keep referencing to only worked in Broodwar due to its holistic interaction with the totality of the Broodwar experience. It all comes back to the bad pathfinding. The micro potential of units did not come from their design, the tendency for units to spread was not an implemented feature, etc… the twitchy and granular nature of the Broodwar experience came from the fact that you had to overcome the limitations set before you. Without these preset limitations there is nothing to differentiate your successful implementation of a random unit with another player’s implementation of that exact same unit.
In Starcraft 2, my hellions will almost always kill units as efficiently as Bomber’s Hellions. Bomber will always be better at making hellions, dropping hellions, and preserving hellions—but once the flames are spewing Bomber and I are on even ground. Until this equalization of unit capabilities is stripped from Starcraft 2, none of the changes that people are suggesting will have the desired effect of bringing back the Broodwar feel of the game.
Broodwar was not defined by the Reaver.
Broodwar was not defined by a high-ground miss change.
Broodwar was not defined by Goliaths, by Scourge, by clump patterns, or any of the other things constantly spewed out by many posters in TL.
Broodwar is more than just 1 or 2 of the aspects that people remember here and there. At its base level, Broodwar was about being able to control individual units better than the other guy could. Why? Because at its core Broodwar was just the Warcraft development team trying to make Orcs in Space and accidentally creating a masterpiece.
As Patrick Wyatt said about Warcraft:
Later in the development process, and after many design arguments between team-members, we decided to allow players to select only four units at a time based on the idea that users would be required to pay attention to their tactical deployments rather than simply gathering a mob and sending them into the fray all at once.
This statement right here is where the Broodwar pathfinding code came from. Square boxes filled by oversized units chopped up unevenly in an isometric shape while using code for a top-down game.
And please don’t misunderstand this post as asking for Broodwar pathfinding to be put into Starcraft 2, I am simply pointing out that the reason Broodwar had granularity was due to its faulty pathfinding. This does not mean Starcraft 2 needs bad pathfinding it simply means that Starcraft 2, due to its streamlined pathfinding, is a different game than Broodwar and hence can’t be measured with the same metrics as Broodwar.
![[image loading]](http://i.imgur.com/Ic58bJw.jpg)
