Light's Blog Translation

1. Go to the original blog post (link provided in each title)
2. Click "첨부파일"

3. Click "내PC저장"

4. You should then be linked to the replay download

Using a Factory Expand Against a Barracks Expand in TvT

Right now is League of Legends's time!

However, there are still many people who love StarCraft: Brood War. So in order to help those who want to learn more about StarCraft, I will share some of the know-how, strategy, and tactics that I gathered while living as a progamer.

Now we will start the lesson.

The opponent spawns as yellow at the 1 o'clock position on Fighting Spirit


I spawn as purple at the 11 o'clock position



I go 12 barracks, 12 gas, and scout at 13.5 supply

Generally speaking, factory builds take two basic forms: the 11-11 (11 barracks, 11 gas) or the 12-12 (12 barracks, 12 gas). The 11-11 is often used by 2 starport or 2 factory builds, or other such strategic plays. It is a little short on economy compared to the 12-12. There's also a matter of personal taste, but this is the general difference.




The opponent skipped gas after barracks and expanded immediately


We lift off the barracks when the scouting SCV has almost finished scouting two of the starting positions. Thus we can quickly send the barracks directly toward the enemy base. (It is okay to build a marine after the barracks finishes.)


The vulture has already killed the scouting SCV and waits here to cut off any additional scouts.



One advantage of the factory expand is that because of the SCV-barracks search timing, the barracks arrives early to scout the enemy base.
The opponent took the gas at his natural expansion early so 3 factory vulture or two factory follow-ups are unlikely; we can almost be certain that he is going 1 starport or 2 starport.


Adding a machine shop after the first vulture, we upgrade mines and reinforce with more vultures, planting mines outside of the opponent's base. It's also good to plant one or two at the expansion for visibility.



The opponent followed up his barracks expand with a 2 starport build


Earlier, our barracks scout was able to confirm an early second gas from the opponent. Since we are almost certain the opponent is going with a starport-centric strategy, we get a fast armory.


The enemy tank engages the minefield.


The opponent adds more factories following his two starport wraith build


The wraiths arrive with cloaking researched, but we have anticipated properly and our goliaths (with booster upgrade) defend easily.


Even the one remaining wraith is shot down as it retreats.


The opponent transitions to a ground army.


After building a reasonable number of tanks and goliaths we switch production back to vultures.


The opponent expands to a third base.


Because we blocked the wraiths so easily, we attempt a frontal attack with the vultures.


It was a somewhat risky attack, but we lowered the opponent's tank count..


The opponent counterattacks with his vultures.



In TvT, factory count is certainly important. However, abandoning your strategy will put you at a disadvantage in the late game. So it is good to add on more factories only after you have finished executing your build system.

By build system I mean getting the upgrades, expansions, and tech which are central to your build.

Usually with 5 factories I will have 3 with add-ons building tanks, and the other two building vultures.
2 bases: 5 factories
3 bases (3 gas): 6-7 factories is reasonable

Later on I will elaborate on this part with another post.



We have split the map and are engaging at both north and south positions. (Engagements are crucial in TvT.)



The opponent has abandoned his tech (starport + upgrades) and attempts to attack


But we defend it. (An important point for TvT.. in an even situation, you should not blindly attack into enemy tanks which are in the process of sieging.)

Positioning is extremely important in TvT..
To explain simply, we started at 11 o'clock and 1 o'clock, right?
Then you can imagine splitting the map down the middle.

Can you visualize the line?
Securing that line would be nice, right?

If the enemy pushes the north line and manages to secure the 12 o'clock expansion, it's good to push the south line to take the 6 o'clock expansion. Just attacking head-on in such risky situations is disadvantageous.

Using the starting positions as reference, being the first to take the rest of the map is important.



Here we are both trying to gain vision of the opponent's position, while also denying the opponent vision of ours.



The opponent has decided on a vulture-tank-wraith composition.


Gathering dropships one at a time, we continue fighting in the center of the map and attacking the opponent's vulnerable points. In most TvT games, we usually end up with two key areas (north and south in this case). While engaging at these two positions, it's important to use scans, mines, and buildings to check for the opponent's weak points and attack them quickly.




We attack the opponent's expansions and cut off his resources.


Since things are going poorly, the opponent takes a risky position...


But we break it easily.




Continuing to push our advantage, we attack all of the bases to force the gg.

That's it for the first lesson!!..

Replay available here.

TvZ 1-1-1 Build

Hello, this is retired progamer Lee Jae-Ho (Light).

For the second lesson...we will cover the 1-1-1 build as requested by Maran.

I used this build often in Proleague matches. The 1-1-1 build involves building 1 barracks, 1 factory, one starport, and then reacting according to your opponent's moves.

First I will explain how to use the build against a 3 hatchery strategy.

Before we begin, I will note that I played this game against an acquaintance of mine, whom we will label as the opponent. Then let us begin!


The opponent spawns at the 11 o'clock position


He opens with a 3 hatchery


Seeing that the terran has walled off (barracks + factory + supply depot), the Zerg starts to prepare defense.

Some of you may already know this, but there are many ways to wall off the front (useful for blocking zerglings and zealots). I will cover this part separately in a later post.


It's a matter of personal preference, but when I use this build, I open with 11 barracks, 11 gas.
This is because I feel that speeding up my tech even a little has a singificant impact on the effectiveness of vulture and wraith timings.

But there are also people who go 12 barracks, 12 gas. It's up to individual tastes.


This is the wall-off simcity I mentioned earlier (the barracks was originally to the left of the factory).
At the 5 o'clock position on Fighting Spirit, if you wall off in this way, zerglings will not be able to get in.

I will cover walling off in depth in a future post.


When the first vulture comes out, I use it to guard the natural expansion. You need to do your scouting carefully with an SCV. If it dies...there might be some unpredictable variables, right?


I place a marine here in order to block the scouting overlord. It's a valid move on other maps, but for now I will describe the standard play on Fighting Spirit.

Right now we are at cross starting positions, right? 11 o'clock Zerg versus 5 o'clock Terran.
The Zerg will generally send the overlord scout to a safe area at a close by natural expansion.

For example, let's consider a standard case (of course, there are exceptions).
If a Zerg spawns at the 11 o'clock position on Fighting Spirit, he will send the first overlord to the 1 o'clock natural expansion. Then, the drone scout will go to either the cross position or the 7 o'clock position.

If the Zerg spawns at 1 o'clock, the overlord would go to the 5 o'clock position.
Does that make sense? The second overlord will usually stay at the Zerg's own natural expansion (in order to prepare for a possible 8 barracks rush against a 12 hatchery).

But for cross positions on Fighting Spirit, for the overlord to reach a place to see the Terran's tech (cliffs outside the natural expansion, for example), the path is difficult and takes a long time.

From the Zerg's point of view, scouting that way requires more care, and the overlord must move at a timing when zerglings can prevent marines from moving freely about the center of the map.


The second marine moves out in order to deny the overlord.


Terran's vision


Zerg's vision


We punish the overlord for being too ambitious with its scouting.


Bam! Here I will give a tip for Zerg users. Please see the next image.


(subtitle in picture: After scouting the opponent's position with a drone, the overlord immediately changes direction straight towards the Terran base)
Ack...I don't know why the picture is buggy.

If you look at the red circle in the minimap, after scouting the Terran's position with a drone, the Zerg put the scouting overlord on a direct course 'as the crow flies' to the Terran base. If you do this you will end up seeing the same terrible scene from above.

Then what does the Zerg need to do?!
See the next picture.


(subtitle in picture: If the overlord moves to the red circled area it has space to escape the marines)
For a Zerg in cross positions, the overlord must hide in a safe location like the red circle in the minimap above, and then scout carefully while anticipating the opponent's movements.

In this case, I went for a factory build instead of a fast expansion, so it is difficult for zerglings to do damage, but in the case of a fast expand, 4-6 zerglings can limit the area that the marines can cover, so that is a good time to move the overlord.



Now we will look at the build again.

The build order is
First vulture -> command center after 3rd supply depot starts building -> 3 supply depot finishes -> wraith -> spider mine upgrade


The Zerg went 3 hatchery into lair and then hydralisk den.


When the mine upgrade finishes, use your first vulture to plant mines (such as at the Zerg's next expected expansion).


Use the wraith to hunt overlords and scout (in particular, it's important to see if the Zerg gets a spire once the lair finishes).

In this match, since the marines were able to kill the first overlord, we send the first wraith directly to the enemy base. If you did not kill the first overlord earlier, use the wraith to go around the likely places it to be, and kill it before proceeding.


(You can get up to four vultures depending on your preference. I personally get three.)
With the additional two vultures I built, I make a defensive line in front of my base (leaving some space in between the mines as shown).


The wraith scouts the spire building (Try this timing for the engineering bay in order to block mutalisks. When the spire hits 500/600 health, build the engineering bay to get turrets on time).


Against a 3 hatchery opponent, after getting one wraith there are two courses of action.
One is to go directly to science vessel tech.
The other is to get a dropship and upgrade wraith cloaking. I chose this strategy in this game.


Cloaking being researched.

At the natural expansion, build a refinery when you have roughly one worker per mineral patch (this build is very gas intensive~~).


Get an engineering bay at the proper time.


Three cloaked wraiths is a good number (including the one we built earlier) and if you start the science facility right after the third wraith, you will be able to start building science vessels as soon as the wraiths finish.


We put vultures which have used their mines in the dropship to go harass (at this time, it's good to use the first wraith to fly within sight range of the natural expansion in order to draw the opponent's attention).


The Zerg takes a third base.


A drone dies in two (simultaneous) hits from a vulture, right? Since it's hard to guarantee simultaneous hits, we use three vultures to one-shot drones.


Using the vultures to kill drones, we also confirm the presence of mutalisks.




After confirming mutalisks, we build turrets.

We don't need many turrets because we can use the three cloaked wraiths to help keep mutalisks in check plus in just a little time we will have irradiate for science vessels, so two turrets per base is sufficient (plus, being in cross positions buys us a little more time).

If the starting positions are close, you will need to build more turrets!


The Zerg is still busy rebuilding drones.




Due to anxiety from seeing the cloaked wraiths, the Zerg deploys his mutalisks.

When mutalisks come like this, fight them off with your wraiths. Since cloaking is upgraded, you can use them to help secure your next base or go around the map and scout.


The mutalisks take a lot of damage.


The opponent is panicking, even upgrading hydralisk speed.


At this time, upgrade weapons at the engineering bay and gather tanks and science vessels. Also, build an academy and add some more barracks.


Diligently proceeding


The Zerg is also teching up after securing a third base.


However, in this match, since the vulture drop was successful, the mutalisks were blocked, and the Zerg spent too many resources upgrading hydralisk speed, blocking the hanbang rush (vessels, tanks, and marine-medic) will be very difficult.


All production facilities complete.


With 3-4 tanks, 8 marines, 4 medics, and 2-3 science vessels, we move out to take the center.
This is not the attacking rush; instead we eliminate Zerg's vision (zerglings) around the Terran base.

This way we build our advantage (the actual rush will happen when we have roughly a group and a half of marines!).


As you move out for the rush, build vultures to reinforce instead of tanks (because we have mines upgraded, we can use them to defend against returning Zerg forces).



The opponent falls to the attack without incident.

To use this build, you need to practice reacting to the Zerg's play.
Depending on the opponent's strategy, your strategy will also change.

To simplify, specifically against a 3 hatchery build, the possibilities are...
1. No lair hydralisk push (with move speed)
2. Lair with hydralisks (range upgraded) and overlords (speed upgraded)
3. Lair, small number of hydralisks, then spire and mutalisks
4. This is a bit rare, but no lair speed zergling rush (if you have good intuition, you can block this with bunkers and simcity..)

These are the most common.

In case 1, we can verify it is happening by scouting with the SCV and seeing a hydralisk den building with no lair. It is easy to block with a bunker, some mines, and tanks (and you can harass with a dropship as shown above).

In case 2, it's a little difficult, but..
This strategy is possible even with a spire, so verifying it may take longer.
In the case where you anticipated mutalisks and built a lot of turrets, but the Zerg attacks with hydralisks and overlords, all you can do is put many SCVs on repairing your bunker. The tech path is more or less the same (reaction time is particularly important here).

Case 3 is similar to what I showed above.
Case 4...you can just block with simcity and vulture control...

Going 1-1-1 against a 2 hatchery build is similar the going against 3 hatchery, but your timings will be a little bit faster.

To start there are the two tech paths I described above:
1. 1 wraith -> vessels
2. 1 wraith -> dropship -> cloaking -> 2 more wraiths -> vessels

If you go with the first strategy, you will need to build a lot of turrets. You will be strained for resources...
It will be particularly difficult against good players, no?

If you go with the second strategy...just omit the dropship part.
In this case, you also need to build more turrets than you would against a 3 hatchery, although it will obviously depend on mutalisk count!

Aside from that, everything else is similar to what I showed above...you will get it quickly when you try it.

---

In the later stages, your main choices are light mechanic or pure bionic.

For those who go wraith-mechanic, don't build more than five barracks, and when you move out to attack, build vultures to reinforce and add two factories and an expansion. It's also okay to expand first.

For those who want to go bio, take good care of your vessels and add on another engineering bay (keep using vultures and build them one at a time out of the factory).

Since I used both images and text to explain, there are many things I wasn't able to cover but I hope this helps all of you.

This is the end of the second lesson.

Replay available here.

TvP Barracks Expand Into 1 Factory 1 Starport

This post was requested by dmsgod9533 and LeuBeu.

Things can vary according to personal taste, so please use this as a reference.

This post is on a game I played with 4ever`-_-. Thank you.

Then let us study the barracks expand into 1 factory 1 starport build.


Protoss spawns at 11 o'clock on Fighting Spirit.


I spawn at 5 o'clock.



The opening is a barracks expand~ We simcity with the barracks in order to defend against possible zealot pushes by the opponent.


Scout at roughly 13~14 supply.



The Protoss did not get a zealot after cybernetics core (we often call getting a zealot after cybernetics core a "14 zealot").


The object of the simcity is to minimize damage from an early zealot push

The simcity is pretty right? Because the marines can pass between the barracks and supply depot, but zealots must go around the top of the command center, this simcity is very effective in minimizing damage from zealots.
In order to do this, you should study the map in advance.

(I will make a separate post on this topic later; if anyone has pictures of TvP simcity and walling, please send it to me and I will make good use of it!)


The Protoss expands after getting the first dragoon.


We send out a second worker early.

Generally speaking, this is a move to increase the safety of a barracks expand.
It's called feeding the fish(t/n: not sure about this). In order to counter the opponent's strategy, we hide a worker outside before the dragoon arrives in order to scout hidden buildings, expansions, or tech.


After a single dragoon, the Protoss builds a robotics facility.


After the factory completes, we build a starport in a hidden corner of the base.


At the same time, build a refinery at the natural expansion.

In this game, I saw all of the Protoss's tech and build my academy late.

If the SCV fails to check the opponent's tech, you need to build an academy at this time in order to scout the Protoss's tech.

*To get a general idea of the Protoss's tech, check the gas at the natural expansion.

If the natural expansion's assimilator is built early, you can expect high tech units like dark templar and reavers, but if there is an early third expansion but no gas, you can expect 3 base dragoon + observer.

Fast tech -> get comsat and an engineering bay (or wraiths)
Fast third expansion -> skip academy and engineering bay (use those resources to build another factory or to expand)


The Protoss takes a third base.



The general build order is 1 tank (+ siege mode) -> 2 vultures (+ spider mines).


When you get a dropship, navigate over cliffs and water (in order to minimize chances of being intercepted).


As the dropship moves out, send a single SCV towards the enemy's front in order to scout the location of the enemy's dragoons and the possibility of a dragoon push.


The drop has the best chance of doing high damage at the Protoss's third base.
The reason is because the forces in the dropship do not have that much offensive power.
Thus they are best used to delay the opponent's third base, even just a little.

If you try to drop the main or natural expansion, there is a good chance of there being 6+ dragoons there, which lowers your chances of doing significant damage.


Taking advantage of the narrow entrance, we lay mines at the top of the ramp and place the tank behind.




After sending the dropship, we continue building tanks out of the first factory while building two more factories.
Then, when money allows, build a fifth factory plus an armory and science facility.
Also upgrade vulture speed at that time.

Then the order goes like this:

1 tank (+ siege mode) -> 2 vultures (+ spider mines), dropship -> continuous production of tanks out of one factory -> 2 additional factories -> 2 more factories (total of 5) -> armory, science facility

You can upgrade vulture speed at either 3 factories or 5 factories.



When the science facility finishes, build one science vessel and one wraith.





The Protoss abandons arbiter tech, instead opting for mass gateways with templars and storm.


If the dropship is still alive, continue to harass the Protoss. It's good to go between the main and the natural expansion.


Using a contain, we can safely take a third expansion.


As for the reason we got a wraith... It was in order to cut down observers... Of course it is good to get goliaths, but the wraith is more maneuverable and can kill observers in hard to reach places (highly recommended).


We contain from two directions using vultures, dropship, and tanks.


Proceeding on, continue to build up your factory count.


Protoss has also stopped teching to focus on army production.


Using the terrain, we contain the Protoss and deny expansions.


And as I have always emphasized... Move in two places at once (we call it multitasking; it's necessary in order to gain advantageous positioning).



After pulling the opponent's army and attention to the 9 o'clock side, we attack the Protoss's base at 12 o'clock with a small group of vultures.
We attack the opponent's natural expansion with our dropship forces at 9o'clock (although we don't do much damage it is a significant gain for us).


The Protoss finally takes a late expansion.


While we had fun with harassment, we built a large army and upgraded +2 attack, see?
When we fill up 200 supply, use the remaining money to invest in some additional factories. Usually we will have around seven factories at this time, so maybe 2~3 more.


The reason for building so many factories is that when we fight a 200 vs 200 supply battle, we will often find ourselves with 2-3 thousand extra resources (in the event the the Protoss does not engage early). In the case of such a war of attrition, being able to produce out of ten factories versus seven makes a significant difference.

So it is important to have many factories in order to not get overwhelmed in the late game.


When we have roughly a group and a half of tanks with +2 attack and just over a group of vultures, we can move out.
(In this game, since the Protoss skipped arbiter tech in favor of templar and storm, it isn't as important to have a high science vessel count.)

*For science vessel timing, use comsat scans to keep track of the opponent's arbiter tech and match it accordingly


The Protoss sees us moving out and attacks quickly, but...


The attack is completely crushed by the entrenched tanks and vultures.


Also, in order to scout possible counterattacks on your bases, lay many mines in advance.


With our tech, production, and upgrades, we have a significant advantage.



We receive the opponent's surrender.

This is it for the TvP barracks expand into 1 factory 1 starport lesson.

As for this build's weaknesses, it is weak against strategic plays by the Protoss, such as a hidden robotics facility build.

So use two workers for scouting, and if you see suspicious signs of this type of play, switch to a 2 factory build.

The advantage of this build is that because it already gets all of the necessary tech structures, it is very flexible (reacting to fast arbiters with fast science vessels, for example).

If you have any questions please leave a comment and I will try to answer.

Replay available here.

Terran Simcity and Hotkeys Part 1

Hello, this time I will be going over important subjects on playing as Terran.

For simcity...we've all probably had a tank come out of a factory and get stuck, right? Or maybe run out of space while building supply depots?

Terran users will be able to sympathize. Therefore, being able to place your buildings well is important!

Also, I will be going over how to group and manage your army hotkeys.

Finally, I will go over a way to quickly change the rally point for Terran and Protoss production structures (it's easy for a Zerg to change rally points).

Now then, shall we begin?

First, we will go over how to place factories in mech-based play such that we have efficient use of space.


Above is a method for placing factories as close together as possible.
(For factories with addons attached, you can leave a half space or full space to their left if you wish)

The advantage of this placement is that space usage is minimized.

The main thing to watch out for is the empty space to the top left of the factory formation (northwest corner). If you build another factory in that spot, the center factory will be completely surrounded by the other factories, so it will be unable to output units. You will get the message "Building exit is blocked" and the unit will not appear (though you will get the resources back).

So then one factory will be rendered useless. When building in this formation, just be careful to not completely surround any factories to block them in.

Taking the map Fighting Spirit as an example...you can use this formation at the 11 o'clock and 5 o'clock positions.


This is a variation for use at the 1 o'clock position.


This is a variation for the 7 o'clock position.


As for the armory...it's location is not that important in Terran versus Terran

However, in Terran versus Protoss, always build the armory in the natural expansion. This is to prevent too much damage from recalls into the main base.

The reason for building in the natural expansion is that generally speaking, the factory rally point will be closer to the natural expansion, so the unit response time will be faster to defend against the recalled forces. Therefore the armory will have a better chance of surviving in the natural expansion versus the main base.


Think of this picture as being located at the 1 o'clock position on Fighting Spirit.

Now I will go over building placement for bionic play.

Unlike the factories above, we can just stick the barracks together without worry.

Not only do we need to consider space for addons with factories, mechanical units are big enough such that they can get stuck easily if we do not pay attention to proper building spacing.

Bionic units are the opposite. Marines, medics, firebats, and ghosts are small units, so no matter how closely you build the barracks, they will not get stuck.

I personally prefer building everything close to my command center.

Generally, I will build 4 barracks, and then add on additional buildings to the bottom or to the left. You can build the engineering bay near the barracks such that it lines up properly. Then, with the remaining space, build the factory and starport.

Looking at the picture above, we can see that this building placement allows us to make use of space without much stress. When you have the time, try it out for yourself in single player cheat mode.



Think of these as being the 7 o'clock position on Fighting Spirit.

This is all I will cover on building placement for now~

Next, I will summarize how I group and manage my army hotkeys.

To start, I will go over managing mechanical forces (using a hanbang rush as an example).


Hotkey 1 and 2 are used for vultures.


3 and 4 are used for tanks.


Group 5 is goliaths.


Group 6 is science vessels.


Using groups 1-6, I can command the army (get into the habit of always using attack-move).

Next is management of bionic forces.


Groups 1 and 2 are marines.


Group 3 is medics.


Group 4 is tanks.


Group 5 is science vessels.


I command the army using groups 1-5 (using attack-move, of course).

In Terran versus Zerg, the most important timing is when the Zerg takes a second starting position as an expansion. At this timing, the Terran will usually have 3-4 control groups of marine-medic, and depending on how well you control them, you can destroy the expansion. Depending on how well you manage your marine-medic forces at that timing, you can take control of over 60% of the games..

At that timing, I split the marine-medic forces into 4 control groups. I will go over it in detail in a future post.

In the next post, I will go over how to change rally points.

Terran Simcity and Army Management Part 2

So finally, I will go over how to change rally points.
(This is easier to explain in action, so I have recorded some short videos)

First is how to change rally points using hotkey settings.


This is not a commonly used method, but there are times when it is useful. In particular, in the late game when there are a lot of things to manage, you can use this method in the few moments you have free (good when you are in a rush).

The way to use it is to hotkey the production buildings to keys 1-8, then move to the rally point and cycle through the hotkeys and right click to individually set each rally point in succession.

More specifically, you hold down CTRL while selecting each factory and hitting keys 1-8, then hit F3 to center the view at the rally point, then 1 + right click, 2 + right click, 3 + right click, etcetera.

Second is using a hotkeyed building and a location hotkey (this is what I personally use).


If you watch the video, you will see a green circle appear next to the center-right factory when the view changes. This factory is hotkeyed to 5 and so I am able to quickly move to that factory by double tapping the 5 key. Then I can use a location hotkey to go the rally point and right click to set the new rally point.

For example, with a factory hotkeyed to 5 and the new rally point location at F3...
5 5 -> F3 -> right click (set rally point)
5 5 -> select a neighboring factory -> F3 -> right click
Repeat for all neighboring factories as necessary.

That is the idea.

Do you know how location hotkeys work? For those who don't, I will go over it briefly.
Holding down left Shift while hitting F2-F4, you can set up to 3 screen location hotkeys. By hitting F2-F4, you will be instantly transported to that screen location.

Personally, I use F2 for the main base, F3 for the front entrance or rally point, and F4 for the natural expansion.
These settings are different depending on personal preference.

Lastly, the third method is to use only location hotkeys to change the rally point. This is probably the easiest method to use.


There's a noticeable difference in speed, right? This is probably the easiest to use.

The method is similar to the second one listed above.

First, set the screen location of your production structures to F2.
Then, set your desired rally point to F3.

Finally, just change the individual rally points.

Key entry would go something like this:
F2 -> select a factory -> F3 -> right click
F2 -> select another factory -> F3 -> right click
(Repeat for all factories)

That is how you do it.
For those who have difficulty setting rally points, this advice will be useful for doing so more quickly and comfortably.

You can choose whichever of these three methods that you think fits your play the best. This is the end of the lesson!

BONUS Tip
I will teach you about using Shift + P (patrol)
Usually, people use the single patrol command quite often. But there are not that many that know that you can also use patrol commands in multiple directions.

For example, generally we will use patrol to go between two points. This could be vultures going between the main base and the natural expansion, zerglings patrolling around the map, scourge covering airspace; these are all common uses for the patrol command. The Shift key plays a reserve function.

If you move to point A and then hold shift while issuing an attack command to point B, units will move to point A while ignoring enemy attacks, then move to point B while fighting enemy forces that they encounter (for example).

The function of patrol between points A and B is such that units will endlessly move between A and B while engaging enemy forces that they meet (this is the common usage).

Zerg in particular use it very often with scourge against Terran and Protoss.

Question: If we want to make some scourge at point A move to point B and then patrol between points B and C, what do we do?

Answer: Issue a move command to point B, then hold Shift and make a patrol command to point C. As for why Zerg use this technique so often...

In a real time competitive game, the player who can to more actions in a given time span is advantaged, right? By queuing up multiple commands in advance, we can save time which can be used for other tasks.

In game terms, if we were to move the scourge from A to B, then watch them go, then make them patrol to C once they arrive, wouldn't that take too much time? That's why we use queued commands.

Generally speaking, Zerg use this technique the most often. So there are many Terran and Protoss users who don't know about it.

Now I will go over an application: using multi-directional patrol.


You can see from watching the video, the unit moves to each patrol point in succession in an endless loop, as patrol implies.

There aren't that many places to use this technique, but here is one:
When checking for enemy expansions, Zerg uses zerglings, Terran uses vultures or marines, and Protoss uses zealots.

For our scouting scenario, consider a Terran at 11 o'clock playing a Protoss at 1 o'clock on Fighting Spirit.

The Protoss wants to use zealots with speed upgraded to check for a hidden Terran expansion. But if the zealot patrols between the 5 o'clock and 7 o'clock expansions, the 6 o'clock position will be left unscouted.

In this case, you can use the infinite loop Shift patrol.

Now I will go over how to execute the command.
To use the infinite Shift patrol, there is one condition. Under normal circumstances you can move from A to B and patrol to C easily, but with multiple directions, it becomes a little more complicated.

The condition is, all patrol commands must be issued before the unit finishes moving.

For example, on Fighting Spirit, you have a zergling that you want to patrol between the 11 o'clock, 12 o'clock, and 1 o'clock bases.

Input would go like this: Issue a move command to the 11 o'clock position, then, while holding down Shift, hit the P key, then left click on the 12 o'clock position, hit P again, then left click the 1 o'clock position. Then the zergling will go between the bases in 11 -> 12 -> 1 -> 12 -> 11 order. If you do not finish issuing the patrol commands before the zergling reaches the 11 o'clock position, then only those commands that were issued before arriving can be carried out. If you issue another patrol command while the unit is already patrolling the new command will overwrite all previous ones.

If you issue a move from A to B, then patrol to C, then to D, and then patrol back to B, then you will get a similar result to the video above. It reminds you of a moebius strip, right?



The unit will infinitely patrol B -> C -> D -> B and so on.

Even if you don't do a repeating loop like this, you can still use the B -> C -> D -> C -> B version as described with the zergling example above. The explanation is a little confusing, but it will make sense once you try it out.

TvZ 3 Barracks Timing Rush

Hello, Light here. Seeing all of the visits to the blog lately give me a good feeling inside..^^
Thank you(_ _).

The topic for today is an extremely deadly build in TvZ, the 3 barracks timing rush.

You have probably seen it a few times on the broadcasts.
It is a timing build which starts with getting 3 barracks on low economy and then attacks with a strong early push.

Note that this build is particularly effective against a 3 hatchery Zerg.

Now, I will go over when it is good to use this build.
Since many people still play on Fighting Spirit, we will take it as the standard example.


Picture of Fighting Spirit

Let's say the Terran spawns at 11 o'clock. We send the SCV to scout the bases in counter-clockwise order.
So first we scout the 7 o'clock position. If the Zerg is at 7 o'clock, we cannot use this build.

Conversely, if the Zerg is not at the 7 o'clock position, we can use the build.

The key difference between the starting positions is the initial overlord scouting distance.
If the overlord is able to see the timing of the Terran's attack, the attack's probability of success drops significantly.

But if we are not in the first overlord scout's path, the chance of success is higher, right? It's a 2 in 3 chance so the odds are in our favor, no? Under the assumption that we are indeed using the build, we will continue on. (Luxury.Episode helped me with this game. Thank you.)


Terran spawns at 1 o'clock


Zerg spawns at 5 o'clock. Judging from the starting positions, we are not in the first overlord search path, right?


We build a barracks at 11 supply.


The Zerg takes his natural expansion at 12 supply.


The scouting SCV and drone pass by each other.
(I only remembered my point about scouting direction after this game, so please forgive my mistake in sending the SCV in the wrong scout direction..^^)


After the barracks, we build a command center.


The Zerg is going 3 hatchery.


After scouting our base with the drone, the initial scout overlord changes direction.


The second overlord watches the entrance to the Zerg's natural expansion.


After expanding, we temporarily cut SCV production at 21~22 supply to build two more barracks.


Building a single marine from our initial barracks, we also build a refinery and academy.


At the same time, we use the scouting SCV to gather information about the Zerg (specifically the position and number of zerglings) and also move our marines to deny the scouting overlord access to our natural expansion. (If you look at the minimap, the overlord is still in the center, right?)


When the natural expansion finishes I transfer about six SCVs. Furthermore, I only build additional SCVs at the natural expansion. (This is in order to fool the opponent into thinking you are going for standard play.)


See how the main base does not have that many workers?


In order to hide the number of marines we have, rally one of the three barracks to the top of the ramp.


With careful micro we deny the zergling scouts in front of the base.


The overlord struggles past the center of the map.


When the academy finishes, we build three medics. (Also start upgrading stimpack.)
Then build a comsat station at the natural expansion. (Build SCVs at the main)


The three medics that just finishes group up with the marines. (If you hit the build timings well, you will have around twelve marines at this time. In this game I had eleven.)


When the three medics finish, send one SCV to the center of the map to lure the zerglings.
(The difference this maneuver makes is huge. Not only can we see the number of zerglings, because the Zerg has difficulty scouting with overlords, he must depend on zerglings for reconnaissance, but if they are lured out of position by this SCV the game skews heavily in our favor.)


The marines advance far in front in order to hide the number of medics.


Zerg's vision.


Terran's vision.


As the zerglings pass, they caught a glimpse of a single medic, right? (At this point it is too late to make a difference.)


After the three medics, build two firebats and a marine (to defend against zergling counterattack) after which you can build whatever you want.


The overlord finally arrives, but..


In reaction to the recon, the Zerg builds sunken colonies and walls off with evolution chambers, but..


The Zerg is helpless against the three medic, marine, and firebat attack.


Easy, right?


Thus we receive the gg.

As you advance with the marines, you can upgrade their attack range and build an engineering bay.

Since it is such a strategic build, if the attack is unsuccessful it is difficult to follow up into the later stages of the game.
However, even if the attack fails, if you kill a significant number of workers, you can prevent a 9 mutalisk transition which is a slightly better situation.

Well, this is the end of the lesson.

p.s) The replay is attached. If it bugs out, try restarting it and keeping the speed at Fastest for 1-2 minutes and then raising the speed as desired.

Replay available here.