[G] The Ultimate Dark Templars Abuse Guide

First let's see how it fares against FD. Stork cleverly hides his citadel of Adun and templar archives in the back of his base as to not alert his opponent of what's coming; to the first scouting SCV, it looks like just another one gate core build. Meanwhile ForGG opts for an aforementioned FD build and denies scouting to the first zealot. Observe that Stork builds the minimum number of units to keep him safe and deny scouting; when the second SCV tries to pass through, all it sees is a warping nexus at the natural. So at this stage it looks like just another fast expo build, likely fast observers or reavers.

But even as the first Terran push advances, so does the first dark templar; they meet at the center of the map. This first one's goal isn't to kill those units but get past them, to ForGG's natural, so it advances with the move command (don't attack move! you will spoil everything). It avoids the first mines, which had been laid before the ramp instead of before the bridges, and denies the expansion for quite some time. ForGG attempts to do the same with his tank and vulture but a second dark templar is used to defend the base.

Notice how both dark templars destroy mines as they are being laid. It has become popular lately to open with fast mines as they can take down dark templars but with proper micro the mines are rendered mostly ineffective. First, a dark templar can withstand a single mine hit, so you have a margin of error. Second, it can swipe a mine just before it burrows, so they can hardly be used in an offensive manner. Third, a dark templar can destroy a mine in flight, jut before it detonates. So when you see vultures, keep cool and stand your ground; they're not as threatening as they look.

Back to the game. While ForGG struggles against dark templars, Stork transitions to carriers. A strategy now doubly suicidal, you might think, but it worked! A fast dark templars, fast carriers build defeated an FD Terran! It shows that nearly any opening can work, assuming the surprise effect and decent micro.

Not only is it possible to warp in dark templars in 4:45, it's also possible to drop them just as early. This even more reckless build requires not to make any units save for a zealot to block the ramp (and vitally deny scouting!), as Much does in this game, and every single drop of gas one can gather. The robotics facility starts to warp in sometime between the citadel of Adun and the templar archives. It is then possible to make a shuttle in time and drop the first two dark templars right into the enemy base for an even faster rush. When they land into Mind's base, the first turret isn't even complete yet, and only brillant micro saved the day for Mind. As a bonus, this build allows for fast observers! Mind keeps fighting valiantly but never recover.

Fine, it works fine against an FD build, but surely it cannot work against Zerg players, right? Since they have overlords, the dark templars will be promptly taken down, won't they? And isn't that build awfully vulnerable to early zerglings? All these are pertinent objections, which this game shall put to rest.

First the early zerglings. The secret is to cheat a little with our build and open with a forge. Jangbi does and withstands the zergling rush, if only barely. He suffers significant damage but don't count him out of the game yet. His dark templars are obviously delayed, and significantly so, but even then they shall prove surprisingly effective. Verify for yourself that this isn't the Bisu build, for there is no stargate, let alone corsairs. There is nothing to chase away overlords.

Yellow is relentless with his zerglings, but Jangbi manages to hold with a cannon in his mineral and dark templars for defense; his buildings are to tight that the zerglings cannot take down anything without being exposed. Meanwhile, a single dark templar sneaks past Yellow's natural, to his main, kills all the drones and starts cutting into the spawning pool. At this stage of the game, overlords don't have either speed or sight range upgrades, so if you sneak past them the zerg player cannot defend effectively. The spawning pool is saved in extremis, but the dark templar lives, hiding past the overlord's sight. See how it survives even within a closed area merely by eluding the overlord. Later it shall come back to finish the spawning pool, the moment Yellow is busy elsewhere.

We transition to middle game. Jangbi opts for archons to defend against Yellow's mutalisks. Yet even then dark templars still wreak havoc in Yellow's base; his natural is guarded by a sunken colony and an overlord but the templar simply sneaks past it and kills drones. The one that remains trapped in Yellow's main keeps threatening to come back even though there's an overlord and many zerglings, a quite wearing type of harrassment even before it makes any damage; since whatever a dark templar kills in a single attack (likes drones and zerglings) does not alert his opponent that he is under attack, Yellow has to be extra watchful. On the other hand, mutalisks don't fare well against cannons and archons, so Jangbi is very safe, and soon after claims victory.

We shall conclude that this build is quite safe even against a very aggressive Zerg player, as long as you micro properly.

Since this map's natural defenses makes sneaking inside the enemy base very difficult, Bisu has little choice but to abandon his initial plan and instead give his dark templars secondary roles. GGPlay attacks with mutalisks but Bisu wisely made archons, therefore they do little damage.

As the game progresses into mapwide mayhem, how can dark templars help? Bisu uses them in two ways. First he mixes some with his attack groups, so that they can make free kills during encounters, for dark templars are difficult to target in the midst of a battle. Second, he uses them for defense, as rarely do overlords closely follow attacking Zerg forces, so dark templars can even kill ultralisks! We shall elaborate about this different role in the next section. For now just keep in mind that just because there is no early opening for your dark templars doesn't mean you have made them in vain.

Beware of defilers; plague can reveal cloaked units. This is why GGPlay uses it in desperation late game, and shows how badly the dark templars hurt him. Another example game which features this technique (fast forward to 26:05 video time, 24:45 game time):

2008/08/17 2008 WCG Korea Finals 3rd Place Match Much vs Luxury Set 1 on Andromeda
+ Show Spoiler [VOD] +

This PvZ game on Zodiac during the Ascension tournament (VODs) featured Fenix who tried this build but was countered by Squall's hydralisk push. Still, one dark templar barely made it past the blockade and managed to kill every single drone of his opponent's two mining bases before the Zerg player even noticed! As mentioned above, whatever a dark templar kills instantly does not trigger an under attack alert, so if you micro properly as to not attack buildings prematurely, you can kill ridiculous numbers of drones at unprotected expansions, even if there are overlords, long before your opponent notices!

Can we devise a counter even more dreadful than a zergling rush followed by 3-hatch mutas? How about proxy gateways, plus a forge, plus fast observers? Now it's impossible to pull out, isn't it? It looks tailored to defeat any fast dark templar build!

Think again; in fact, it works even if the dark templars are late, as in the previous game, even without complete surprise, and even if they're microed lousily. This is how Bisu manages to defeat Free, who builds three proxy gateways right next to his base. Free closes in with dragoons as Bisu attempts to expand. As in the first game, the first dark templar isn't used to defend but moves past the attacker, to the opponent's base, but Free notices it anyway (his dragoons bumped into something, and dark templars blur the space around them as they move), therefore builds a forge. The cannons warps in slightly late, but they're well positioned, at each end of the mineral line, and the dark templar is trapped between them, amid the probes.

Here's the bad micro part. In such a situation, don't try to take down the warping cannons; do as much damage as possible and pull away right before the cannons finish. Usually you're fighting a losing battle against multiple incoming defenses, like cannons or turrets, while not doing any actual damage to the opponent. In this case, Bisu should have just taken as many probes with him as he could, instead of trying to destroy the cannons, in vain; not only did he fail but his first dark templar went down with only a few kills.

So that leaves Bisu microing with probes against dragoons, while his opponent's base is protected by cannons. Hopeless? Not so fast. Two more dark templars come to the rescue and wipe out Free's dragoons before they do crippling damage. While Free warps in a robotics facility for observers, Bisu takes down whatever he can outside of the cannons' range, then with two dragoons moves in to kill one of the cannons and wreak havoc in his opponent's base. An observer comes out to protect Free's units crossing the map, having emerged from the proxy gateways. Free's plan is to defend and take down Bisu using troops from his proxy gateways but Bisu simplifies, taking down Free's nexus while suffering only minimal damage in return back home. Free gives up.

Some will argue that the above game was unorthodox. Fine then: how does a fast dark templar rush fare against a standard fast observer build? It all depends on timing: if the dark templars are on time, they can win the race against observers, if only barely, but if they are late, they won't make it through.

Here's how it happened in that game. Draco denies scouting to White-ra and builds a manner pylon inside his base, but that throws off his build order timing, so that everything is delayed by as much as 45 seconds, while his opponent's fast observer build goes smoothly. So when Draco's first dark templar leaves his base around 5:30, White-Ra already has an observer underway, therefore the dark templar is unable to slip through and must turn around.

But if the build had been executed properly, the dark templars would have been ready by 4:45. At 5:30, they would already have been in White-Ra's base killing workers. So against a fast observer build timing is key: don't waste time, and don't waste gas, or you're in trouble. And indeed Draco lost that game shortly after, far behind in tech, units, and economy.

Is there nothing then that can counter this build? Let's be devious by fighting fire with fire, and pit dark templars against dark templars; this way one side is sure to fall! Are we not brillant?

Bisu opens with a proxy gateway in Free's natural. For a couple minutes it turns to zealot fight, until the proxy is taken down. So far Free has build only one gateway and rushes for templar tech. Bisu technically has built more than one gateway (there's two in his base) but the beginning of the game explains it. Both players also build a forge and robotics facility, anticipating that the other player has also opted for dark templars. Their respective dark templars meet in the middle of the map.

Free's templars reach their destination first, before Bisu has any cannons or observers at home, therefore Bisu has to block the ramp to prevent their incursion into his base. But he only buys some time for they get in anyway and he has to micro his way out of that situation; one barely survives (but dies soon after) while the other one is trapped amid probes and dragoons. Bisu isn't so fortunate in his attempt to annoy his opponent because there are already cannons awaiting.

This reciprocal build transitions neatly to middle game as both players make high templars, while dark templars take the back seat. So who won the game? It is irrelevant, as neither side's dark templars proved decisive! So our dark templars versus dark templars contest gives... a draw. So much for our clever mirror matchup.

Let's revisit this mirror build idea. In this game both players open fast dark templars, although BackHo does so with a small variation, researching the dragoon range upgrade, delaying his dark templars a bit. One of Much tries to get through BackHo's ramp, which is guarded by two dragoons, but his opponent's own dark templars come to the rescue just in time and block the ramp. So the two players block one another on that ramp in a rare standoff! We shall conclude, despite this amazing deadlock, that Much's build is better because it allows him to expand to his natural, whereas BackHo is forced to research observers ASAP, all due to his dragoon range upgrade blunder.

The game continues as follows. Both players make observers. BackHo attempts to come back from behind by dropping on Much, but his shuttle is spotted by an observer long before it arrives, so the drop fails. BackHo's last hope is to break his opponent's defenses with an all-out attack before getting outmacroed due to his economical disadvantage but it is already too late. gg.

The above notwithstanding, there is one factor that is essential to this build's success: surprise. If your opponent sees your tech buildings before your dark templars are out then it is easy to counter.

In this game Much opens fast dark templars. He hides his citadel of Adun yet neglects to hide his templar archives at the same location, therefore BackHo's second probe sees it. Then BackHo has plenty of time to make observers and arrange a reception committee at his base. Now Much is behind. Ironically, it is BackHo that surprises Much butt naked with a dark templar drop before the first cannon warps in! The game goes normally from there but Much remains behind until he concedes. The cause of defeat? Not hiding a tech building by pure negligence.

Brillant play by Horang2, who builds proxy gateways next to his unsuspecting opponent's base, who opens FD; as you must have figured by now, opening FD versus dark templars means death. Flash sends a marine to scout and sees the proxy, but too late! The first dark templar tags along as the marine comes back, then cut through the enemy troops.

The first vulture lays mines on the ramp but there are two dark templars now, so one sacrifices himself and brings down nearly every remaining unit with him in the mine blast, including the tank! Then Horang2 uses a zealot as a minesweeper to clear a path through Flash's main base. Soon there are three dark templars wrecking havoc.

The micro here is impeccable. Horang2 only takes down the building turrets he can and leaves the rest to focus on workers and buildings, then spreads them out as turrets complete and reinforcements come out. Flash resists in vain and soon gives up.

Sorry VOD not uploaded to Youtube yet! Try TL.net's Torrent Tracker.

It is trivial to wall in on this map but a dark templar rush remains doable, at least with a proxy. Horror sees it long before even the second proxy gateway warps in and predictably turtles in his base, bracing for the impact.

A zealot and dragoon keep him in his base repairing his wall and rushing for siege tanks. But little does he know that dark templars are coming! Faced with two of them trying to breach through, he's left with no choice but to buy time with three repairing SCVs, a significant drain to his economy.

The dark templars don't make it through the wall however, and eventually they have to pull out as Horror builds turrets and pushes out. This isn't the end of his problems though, as IefNaij drops two more inside his base, where there are no turrets! IefNaij initially made a micro mistake by attempting to take down the comsat station, but he only wastes time as the SCVs gather around to repair it, but rapidly rectifies and attacks workers instead, the right thing to do in most scenarios. With nice shuttle micro he succeeds at avoiding comsat scans, since Horror has nothing that shoots in the air! He's forced to make turrets all over, and builds a wraith just to take down the shuttle!

All this harrassment allowed IefNaij to pull out so far ahead that at some point I'd swear he has more probes than Horror has units. It degenerates into a protracted but dramatically uneven match in which IefNaij control three fourths of the map and is just happy to suicide macro and recall until Horror is forced to concede.

This game shows that even if your dark templars make few kills, they can be so annoying that they keep your opponent in his base while you comfortably spread over the map.

In this old game Anytime makes proxy gateways facing iloveoov's two factories build. If you need any evidence that dark templars can overcome nonstop vultures laying mines, this is the game to watch.

One note about the micro in this game, something that also applies to other games on this list. When a mine springs up from the ground, every player has the reflex to pull away, which is the wrong thing to do. You must instead go towards the opponent, in the hope of bringing down enemy units or other mines with you; spider mines are a double edged sword, as they damage both friendly and enemy units indiscriminately.

Once again we face an FD Terran, but the build order differs as outlined above. By the time the first dark templar comes out, around 6:00, Draco has already begun to pressure Strelok with zealots and dragoons, and pushed him back into his base. Draco exchanges his zealots for mines, clearing a path for the dark templars.

Meanwhile Strelok harasses at Draco's warping natural expansion but more dark templars are ready to welcome them. See how easily the mines are destroyed before they have a chance to burrow, merely granting the vultures time to escape. Draco's micro here is impeccable; his dark templars stick to the vultures instead of running away from them, so if they lay more mines those can be cleared as well. On the other hand they do not follow when the vultures escape or they would be led into a minefield. Thus they hold until the observers arrive.

As we transition to midgame, Draco mixes his dark templars with his main force and engages the enemy with them. See, instead of sneaking dark templars inside the enemy base, Draco attacks with them, using the confusion of the battle to do extra damage, and this highlights the difference in aim between this build and one gateway builds.

Here's an interesting variation. Bisu opens with a standard 2 gateways cybernetics core dragoon range upgrade build and expands, defending against the FD push with a sizeable dragoon force. Then he makes templar archives and a robotics facility simultaneously to drop dark templars in his opponent's base. While this is reminiscent of the fast dark templar drop build, its goal and execution are instead those of reaver drops: pick up whatever you can and keep the Terran player in his base as long as possible. There's just too many turrets and units to do crippling damage as in a fast dark templar build, so don't make kamikaze moves with your templars and shuttle. Think pressure instead. Dark templars may also be left behind to lurk in corners, providing extra awareness of enemy movements and granting further harrassment opportunities, like expansion denial, when the opponent no longer expects it.

Another advantage of this build is that you can get observers and high templars quickly, and also smoothly transition to arbiters afterwards, all this while expanding. This clever all-around build has been designed with macro intensive games and early high tech in mind, just what you need to defeat prudent Terran players like Hwasin.

In this game both Stork and Kal opt for two gateways dragoon range build. However, whereas Stork follows with a robotics facility, Kal in secret goes for dark templars. The first dark templar comes soon enough to sneak inside the enemy base before his first observer arrives, while another harrasses Stork's troops, temporarily forced to withdraw.

At this point Stork has the macro advantage. Kal prepares for the upcoming siege but Stork instead has a surprise for him: he's making dark templars himself, which he drops at Kal's expansion in the back, nearly taking down the nexus, then finishing the job the second time with zealots.

The remaining of the game is more standard. Both mix their remaining few dark templars with their main fighting force, with little effectiveness given their meager numbers and since both sides have observers.

The course of the game could have been quite different if Kal had opted for a single gateway fast dark templars build, and not had harrassed Stork's units. This way, not only would the templars have reached Stork's base 45 seconds earlier, they also would have without warning, but instead Kal tried to play safe and instead ruined his chances of doing significant damage, and his victory proved more difficult to earn. But suppose he really meant to play this build; what then would have been the correct play? To pressure Stork with the rest of his units as well, which would have made a huge difference, instead of merely sending dark templars alone against a fast observer build. This is the difference in purpose and execution between the two builds.

The above game's conclusions notwithstanding, sneaking dark templars inside your enemy's base remains an option even with this build, if your opponent is careless enough not to get fast observers.

In this game both players opt for two gateways builds, but Cloud's scouting probe gets taken out at the entrance to IefNaij's natural, instead of at his main base's, which hints at a very fast expansion. Therefore Cloud chooses to make quick dark templars, betting that his opponent's observers will be significantly delayed. He sneaks the first one inside IefNaij's base, while the next one is only kept out by a wall of units. IefNaij's observers are so late, as he even made a robotics facility before his observatory, that by the time the first one gets out, the game is essentially over.

If Cloud had run into his opponent with his units and used his dark templars to attack, it might not have worked so well. The first dark templar might not have been able to slip inside the enemy base, at least not so quickly. Both approaches have pros and cons, so you have to use your judgement to decide which option is more likely to work.

This variant of the Bisu build delays templar tech in favor of reavers and the shuttle speed upgrade.

No further analysis provided; watch for your own enjoyment.

In a grudge rematch, Savior makes a proxy hatchery at Bisu's third mineral spot up the ramp, then pumps out hydras and attacks his opponent's underprotected natural expansion before the dark templars come out, the first one as late as 7:10. These barely manage to stop the first push with the help of corsairs, but in the end Bisu just gets outmacroed with tons of hydralisks followed closely by overlords. All because he neglected to scout properly.

This game is similar to the previous one, except there is no proxy hatchery. July judiciously denies scouting with zerglings guarding the ramp across the center of the map. The push comes even sooner than in the previous game, around 6:15, so July's hydralisks have plenty of time to wreak havoc to Bisu's natural and beyond. The game ends in victory for July by decision because Bisu dropped.

This shows once more than relying solely on corsairs to scout is a terrible idea, for by the time they come out doom may already be knocking at your door. Send a zealot or two by the time your natural expansion is operational, and if they don't make it through, cannon up ASAP.

Sorry VOD not uploaded to Youtube yet! Try TL.net's Torrent Tracker.

Horror gets horribly beat up by a fleet of carriers, but decides in desperation to ignore that threat and simplify, besieging his opponent's base. This game could have lasted a lot longer should IefNaij not have summoned dark templars to clear up the remains of Horror's force, since he had just deprived his opponent of comsat stations. Horror attempts to build turrets instead but it proves ineffective.

Bisu wears down Flash with arbiters, denying him expansions. Resources-starved, his opponent opts to take out his army for one desperate push, which may very well succeed since Bisu's army melts before his tanks. But his science vessels are taken down and he exhausted his comsat stations' energy, so Bisu cleverly calls on dark templars to finish the cleanup job and then harrass expansions. With no means to remake such an force, Flash surrenders.