|
On April 28 2011 04:01 _Search_ wrote:Show nested quote +Tech is useless if the player lacks income and thus can't use it. Ignoring the pool might let the zerg make zerglings... which will do what? There's no more minerals, so no more drones. Zerglings can't attack air. Queens can. And you definitely do need a follow-up after killing 20 drones. I regularly kill half a Zerg's drones with fast hellions only to lose in the mid-game because the Zerg economy mechanics are so strong that the Zerg is able to replace those drones incredibly quickly. Yep, these situations are all theorycraft designed for a better understanding of the game. Also, positioning and micro will always have a significant effect. The situation was designed with the understanding that the various targets can definitely be killed. Really the question I wanted to provoke is that given the situation of a 2-port Banshee build, where is Zerg weakest? I'd rather point to pros who do. Show nested quote +2-rax is a valid strategy especially on maps with close spawns and open naturals. Actually many think it is too good. Many pros dont open hatch first because of it. Tech sacrifice sometimes brings the whole lot of advantages if used well. + Show Spoiler +Long rush distance and closed natural, but Select v Machine on Shakuras Plateau in the IPL showed one of the best executed 2-rax rushes I've ever seen...and Machine was ahead at the end of it.
Queens? Queens cost 150 minerals to build, are killed almost instantly by banshees and it takes ages to build them. If you kill 20 drones, the zerg can't afford to build a queen, and if they do, that's it. They are out of money. Even if the queen kills the banshees by a miracle, how will a queen win the game for them? In the time it takes a zerg to rebuild all those drones, in the situations that there's actually a few crystals left, the terran will basically have a 200/200 army.
You seem to think that after the 20 drones are dead, the zerg immediately recuperates and starts using their tech. That's now how it works. Having your workers demolated brings you several minutes back. Kill a pool, the zerg builds a new one, while maintaining the economic lead. Kill the drones and the zerg can't use the pool and certainly can't build a new one if you come back later with even more banshees later and destroy it.
|
On April 27 2011 21:20 parn wrote:
If you love that kind of reflection you should read The Art of War - Sun Tzu, you'll learn a lot of things like that.
Great book. I've never thought about applying it to SC2 though....hmm.
|
Very interesting, perhaps a tad oversimplefied in places, but it did offer a few things for me to think about.
|
Good read. It's not like it would be something of an amazing epiphany, I guess most people know this stuff when thinking about it. But it's really important to know these things the instant they happen and not to act on instinct then, which often is wrong.
Visualising all these situations in form of this article is very good.
Though for one detail I'ld like to add something:
"- The best responses to a 4-gate are cloaked units or units with high mobility, to hit the probe line and get out before the army arrives"
Make sure never to overcommit to cloaked units. Adding like 1-2 cloak banshees or going in with burrow roaches often is very good, but there has to be a followup to that. It's just perfect to have these cloak units to keep protoss in his base for a while to hinder him from following up with anything good though.
|
I guess i'm gonna have to ask if you have any replays to provide evidence for your conclusions.
Your making all of these wild conclusions like attacking tech buildings with banshees instead of attacking drones by saying that attacking drones won't damage the economy, even though that is what harassment is for.
"If the Queens are sniped the next target should be clearing the map of creep tumours."
How does this make any logical sense?
"he best responses to a 4-gate are cloaked units or units with high mobility, to hit the probe line and get out before the army arrives"
I'm pretty sure good gateway pressure builds are going to hit long before cloak is done
I can just go on and on, some of your conclusions are just kinda silly and you have not provided evidence for anything.
|
On April 28 2011 05:25 CookieMaker wrote:
Thanks OP, great read. Your examples were obviously oversimplified but you made some excellent points, especially the parts regarding recognizing advantages and exploiting them.
The "Attack Workers or Production" is a great example of decision making when you have the definite choice . What most people haven't realized is that the situation you described isn't in game-context and as a result they are including erroneous factors. The main clarification that should be made is that "You have the absolute choice of damaging/destroying one or the other". A better example would be:
"You are in the middle of a push and have two options to destroy without taking any losses: The mineral line or a zerg's spawning pool. Your choice"
When faced with this option the answer is obvious: the pool. Killing drones with undoubtably increase your chances of winning in the long-term, but killing the pool NOW means that after your two armies clash, you can reinforce and your opponent cannot. This is the equivalent of taking out pylons that are powering 3-4 gates at once. Assuming you and your opponent are both reinforcing, your advantage will be maximized by first destroying their ability to reinforce their army.
The Banshees bit has the same principle, but a more flawed design because the scenario surrounding banshees involves not having an army with which you can follow up your harass (or tech-kill). Unlike a tank/marine army, banshees aren't being reinforced with the intent of wiping your opponent out. In the tank situation, your army is at the doorstep and being constantly reinforced, so taking out production structures means they no longer have anything to fight you with. Banshees, however, are not normally going to win you the game outright, so halting army production momentarily is not as effective as it would be in the other scenario, where army is immediately required for defence.
ATTENTION NOOBIES
You are NOT furthering intelligent discussion by NOT READING the thread before posting. the critiques here are all the same. Once one person says it, every noobie with a keyboard regurgitates it as their own opinion as well. This is plagiarism btw, give credit where it is due (to the person who first intelligently criticized the OP).
So far I count only 2 unique critiques in the first 2 pages:
Paraphrased:
Aebriol: "Just Tech" is not usually the best, especially if your opponent has a chance to prepare because your advantage will be nullified after a strong defence.
Sotamursu: Toss doesn't have the weakest economy because of the ability to chronoboost probes. Also, tech advantages for protoss have more significant impact than the OP claims. Similarly, Terran's tech is not as strong as the OP describes, especially because of the strength of terran's low-tier army units.
@OP, possibly edit OP to reflect these comments, and encourage any first-time readers to read discussion critiques already posted
The simple fact is that this thread is pointless, mainly because it discusses situations that don't actually occur ingame. As many pointed out, the responses to an FE are blatantly wrong, so let's not go over that again but go straight to the OP's main point. Situations where you have a choice between destroying a production building or destroying a boatload of workers.
The only situation in which you should aim to destroy production are situations where you can capitalize on this advantage in the timeframe given, which either means you're already standing in the guys base or that you're pushing within, let's say a minute or two. Delaying certain units that hardcounter yours can be excellent, so if you see a spire in a marauder/hellion push, by all means go for it.
Tech is a slightly different story because you can snipe certain buildings to deny certain upgrades: say psy storm or the infestor glands upgrade. But even there, this is only effective if you can capitalize on in in the actual timeframe given.
The 2 port banshee build is a ridiculous example, but let's go with it anyhow. Assume you went 2 port cloackshees. You're heavy one basing, all your gas is used on the banshee's and all you will have next to it are marines and a couple of hellions. Your, let's say 6 banshee's reach the base. Now you have the choice mentioned in the OP.
You do what the OP tells you to do. You get those 3 buildings and retreat. Now what? You don't have the units to capitalize on your advantage, the zerg will be in a defensible position in the meanwhile with spores + queen + whatever army he has left. You will not be able to kill the zerg in the 2 minutes it takes to rebuild all 3 structures.
I'm terribly sorry, but if you manage to kill 20 drones with 3-4 blue flame hellions and you can't win the game after that, that's not because of the 'zerg macro strength', it's because of you not knowing how to follow it up.
As many others have said aswell, this guide actually points people towards an entirely wrong direction.
(The reason the 2 port banshee build is a ridiculous example is because it's a very heavy all-in. You need to kill with it, you need to get rid of all detection with them, then proceed with any AA, and then the idea of the build is to win straight away. Not to go into the lategame with it. Also, illustrating this with a TvZ example is rather weird because of the way zerg production and tech are seperated, unlike T or P. That's probably because it gets even more ridiculous if you have the choice between destroying a robo + a gateway or 20 probes, or a factory and a starport for 20 SCV's.)
|
Philadelphia, PA10406 Posts
Decent general advice, but you're oversimplifying nearly everything.
I know others have said the same thing, but it's an important point to make.
On April 27 2011 15:20 _Search_ wrote:
Hopefully it’s obvious that, even though they would be cheapest to replace, the Banshees should target the 3 Tech Structures. There’s no point in killing drones, against 2 hatcheries that would be like spitting into the wind. It would be like punching your enemy’s arm instead of kneeing their groin. Don’t hit your enemy where they are strongest.
Killing the zerg's tech is wrong for three specific reasons.
1. There is no way to know exactly how much you can kill with your army. This is why you should nearly always kill the workers at any expansion before trying to focus down the expansion itself. Buildings take time to kill, and if you commit to killing a tech structure, but can't, this is a huge waste of time.
2. Buildings can be replaced more easily than can workers, even for zerg. A zerg can start a second spawning pool even as you kill their first one. They cannot start twenty drones as you kill them. Replacing units takes a cumulatively longer amount of time the more you kill; rebuilding structures doesn't unless your economy is poor, which in your example, it wouldn't be. Furthermore, in killing the zerg's economy you can force them into the same defensive stance, while at the same time weakening their army. If you kill the zerg's tech structures, only, they might add another spine crawler and make more drones and queens while rebuilding their buildings. They may come out ahead, even. However, killing drones ensures that the zerg's larva will not be used on army structures, and that they will not be able to afford the army that they were planning to produce.
3. Econ is the be-all and end-all in starcraft. Economy lets you get an army and make tech. Economic damage is the most important kind of damage you can do, because armies can be rebuilt and tech can be rebuilt, but economy is a function of time, and an economic deficit is the most difficult to recover.
You gave the example of killing twenty drones. That's a 1000 mineral deficit without counting the effect of lost mining time, which brings gas into play. In comparison, you advocate going after the Spawning Pool, Baneling Nest and Evolution Chamber, which, should you successfully kill these structures will cost the zerg 275 minerals and 50 gas. The difference between these two is staggering and makes killing the drones the clear choice.
As a further variable, those three buildings have a combined health of 2600, with 1 armor. Drones have 40 health each and no armor. That comes to 2000 health in total, plus time spent chasing fleeing drones. That means that the banshees will have a much quicker time killing the twenty drones. Your estimates may be completely off then; in the times that the banshees could kill those three buildings, they may be able to kill every single zerg drone.
A second point that others have brought up, again in some detail:
On April 27 2011 15:20 _Search_ wrote:
Here’s another: Your opponent made a risky fast expand and you scout it just as it completes. Do you: fast expand yourself, mass a quick army and push or tech up?
You tech.
This is incorrect because it makes some shady simplifications.
In one example, if a protoss takes a fast nexus, that leaves a natural timing for the terran to take a fast third. This is because in taking the FE, the protoss has put themselves behind in army and tech. The terran can obviously take their natural in safety, and yes, they may be "behind". But however the protoss chooses to follow up their FE, they will be behind the terran in tech and army. That allows the terran to dictate the pace of the game, which is a tremendous advantage in and of itself. If terran FE's normally and thus does not stunt his army production, he often can gain enough map control* to take a faster third.
In another example, many builds risky builds, like FE's are designed to be safe in a majority of situations. Obviously, nobody would take three bases without first building a unit producing structure. That's just not safe. But often a build like a FE for zerg will compensate for the early vulnerability by specifically having failsafes against tech units. A protoss may delay their own tech to be ready for banshees, for example. And in starcraft, decent counters to units are always lower on the tech tree than the unit itself. So if a protoss FE's, and terran goes banshees off of one base, they terran has essentially put themselves all-in in the hope that the protoss will be unprepared. Teching takes time, and there is absolutely no guarantee that this is the case.
A perfect example of this is how a 12 nexus worked in Starcraft Brood War. EVEN if terran killed the nexus shortly after it finished, protoss was STILL usually even with the terran, because the resources that terran spent on the attack ate into their economy as well. Indeed, the only match-up in which it is ALWAYS much better to tech is PvP, in which getting more colossi or air is a better choice than a FE.
*Map control, or more generally what I'd call "Unit usefulness" is another huge part of SC, and one that you've completely ignored. For example, teching all the way to carriers is only useful if carriers are useful in your situation. Teching to siege tanks is map dependent. Sometimes, depending on the map and match-up, tech, economy, and army all vary in importance.
Finally, I'd like to bring up a point that I just alluded to. Weakness in starcraft is often illusory. The baseball pitcher Whitey Ford once said that he pitched to his opponent's strengths and not their weaknesses because they never expected it. The same is often true in starcraft. Any non-cheese or all-in build worth its salt has contingency plans against everything. No player who FE's is going to be surprised to receive pressure as a result. They will be prepared against it. So while yeah, you want to attack your opponent where they're not expecting it, it's often important to note that weaknesses are often not what they may initially appear to be. For example, the weakness of a protoss forge FE isn't necessarily army OR tech. Maybe it's economy. After-all, a protoss FE'ing cannot put on pressure or exert map control until much later. So in taking a fast third, zergs can simply retake the economic lead. Often this is the last thing a protoss has prepared for, as their primary attention must obviously be defending their FE.
Or a terran who goes for an aggressive early pressure build. Is economy their weakness? It surely is, but the protoss cannot hope to take a heavy economic lead because the terran attack will kill any expansion. Is teching the right move? It's a fine option. But there's also nothing wrong with simply committing even harder to army than the terran and breaking the back of the attack. How does the terran react to this? Undoubtably there will be only one player who prepared for that eventuality.
The absolute last point I'd make in response to your post is that your generalizations have crucially cut out the human dynamic to the game, which is easily the most important one. We don't play against the AI, who makes a bunch of units and attacks, and is obviously vulnerable to tech or nuanced builds. Human beings are more unwieldy creatures.
This is a good post for newbies, but it's plain wrong in some cases (banshees vs zerg) and naïve in most others.
|
On April 28 2011 05:25 CookieMaker wrote:
ATTENTION NOOBIES
You are NOT furthering intelligent discussion by NOT READING the thread before posting. the critiques here are all the same. Once one person says it, every noobie with a keyboard regurgitates it as their own opinion as well. This is plagiarism btw, give credit where it is due (to the person who first intelligently criticized the OP).
So far I count only 2 unique critiques in the first 2 pages:
Paraphrased:
Aebriol: "Just Tech" is not usually the best, especially if your opponent has a chance to prepare because your advantage will be nullified after a strong defence.
Sotamursu: Toss doesn't have the weakest economy because of the ability to chronoboost probes. Also, tech advantages for protoss have more significant impact than the OP claims. Similarly, Terran's tech is not as strong as the OP describes, especially because of the strength of terran's low-tier army units.
@OP, possibly edit OP to reflect these comments, and encourage any first-time readers to read discussion critiques already posted
You were in platinum last season, I don't think you should be calling anyone a 'noobie'. I'm also not sure how you find the critique that a spawning pool, baneling nest, and a drone have about as much HP as 60 drones irrelevant. The other, obvious, problem, is that if you try to kill a spawning pool with your 2port banshee and fall 10 hp short, you literally cannot win. If you try to kill 20 drones and fall 10 hp short, well, you killed 19 drones and that's still pretty good!
The thing that is not furthering intelligent discussion is the stunning amount of posters hailing this obviously flawed OP as the greatest thing ever. A general theory of Starcraft which makes sweeping claims backed up by obviously incorrect examples and for which there exist very simple counterexamples is not a good theory, sorry.
|
|
|
When will people get this. Starcraft is NOT simple and you can't break it into easy rules and theories. All of your assumptions simplify the game way too much, and most of them are uneducated and wrong. Btw, I bet you're a math student. (Nothing wrong with being a math student, just think you should stop thinking Starcraft is in any way similar too simple equations.)
|
Is it just me or does this post look like a Zerg player's ploy to get terrans to go banshees on him so he can just use his 3 queens and fancy new 6 second spore crawler root time to get free wins?
In all seriousness though... this post has a few soft truths in it, but its mostly fluff and generally a guide to failure if applied strictly. I think the main take home point for lower league players from this should be understanding that focusing on any of the three "pillars" comes at the cost of the two others, but thats as far as it goes imo.
|
I agree with this mostly, but Protoss is strong in tech, and Terran strong in army. MMM, which is mostly T1, rapes any combination of T1 gateway units, but Colossi and Immortals are hard counters to low tier units like roaches and marines. Also Protoss armies are not really mobile. Sentries and zealots are slower than probes, and Warp Prisms only hold 4 units. The only mobility advantages are warp-ins, blink, and maybe colossus cliffwalk.
|
I'm now curious to your actual skill level, as a number of people have pointed out why your banshee example is wrong, and you're still sticking to your incorrect choice. I feel like you managed to roll a few platinum or diamond players who responded incorrectly and you're basing this example off that.
|
Excellent post, will read again!
|
i think you need to give up on this 2-port example. because there are some glaring errors in it.
On April 28 2011 03:00 _Search_ wrote:
Let me reiterate the example, and we'll imagine it this way, basically you are using 2-port cloaked banshee and have the choice of wiping out either 20 drones or the tech structures before the Zerg defenses are adequate enough to deter Banshee harass. The attack isn't necessarily over after those are destroyed, but the Banshees are taking hits and it's more of a standard attack than free-kill harassment
buildings should be lowest priority and drones highest. buildings take a long time to kill and are cheap to replace. removing them is a temporary advantage. (a greater spire is the exception to this rule, but every rule has exceptions) drones both cost minerals and lava to produce and reduce your opponents capacity to gather minerals and use larva for production.
On April 28 2011 03:00 _Search_ wrote:
If you kill the drones you haven't gained an advantage, because 3 larva injects later all the drones are back. Sure, you killed 1000 resources, and, don't get me wrong, killing 20 Drones is still awesome, but what's the follow-up? The tech advantage has been lost, since the Zerg is now completing a Spire and still building Queens/Spore Crawlers, and the Terran is behind in economy
none of this is accurate, you have killed his gather units and cost him injects that could have been put to units on new drones. you cost him his income going forward becuase hes not gathering untill he replaces those drones, this is all an advantage, just one less obvious than the smoking crater that was once his spawning pool.
the tech advantage is not lost either, you have invisable death flyers, maybe they cant harras anymore but they can be easily intergrated into a standing army for a more standard attack.
and terran is the opposite of behind in economy if your opponet has less workers and therefore less economy.
On April 28 2011 03:00 _Search_ wrote:
On the other hand killing the tech wins the game. If you kill the Evo Chamber, no more Spores. If you kill the Spawning Pool, no more Queens. If you kill the Baneling Nest, suddenly a small number of marines (as always accompanies a Banshee build) becomes very dangerous.
Most importantly, there is a follow-up, which is to overwhelm with Banshees, which might not sound like a follow-up, because the Terran is essentially doing the same thing, but is, because the Banshees have taken a new significance. The follow-up is not to kill the army (Queens, Spore Crawlers), or to kill the drones. It's to continue killing tech, because that's where the advantage lies. Without question the next target after the Pool, Nest and Chamber should be the Lair and perhaps Spire if that's available. If the Terran kills all of that, even losing a significant number of Banshees, it's GG, because even though the Zerg has the economic lead the Terran has a much greater tech advantage. The Terran can keep producing Banshees and eventually the defenses will be overwhelmed, because they cannot be replaced.
this is all wrong. you need to find better zerg to play against if your opponent is letting you have your way with his buildings like this. either he is a bad zerg or your pure theorycrafting.
post some replays of these matches. based on your senario there are several things i can assume are happening:
- hes not scouting
- your not macroing
- your resources are building up
thats why for you banshee is a win now build instead of an early advanatge. and why you abandonded drone killing because you feel behind even after all his drones die.
|
hmm you have some interesting points. i don't agree with them all but most of them are quite valid. good read.
|
I disagree with the opinion that capitalizing on ONE of the three pillars is the best choice... for example if a Terran fast expanded, instead of teching on one base I would get robo tech, pressure ( with the army advantage maybe even win), and expand behind it. There are also certain points in the game where tech is a strange creature, certain comps beat others but that in no way means that tech is adapted to dealing with the situation. This guide captures the feelings of each race generally though, pays no special attention to Build Orders or specific Matchup Metagame. Good write up but a little too simplistic to be of much worth.
|
This is basically a twisted version of what Day9 says all the time: You can't get econ, tech and an army at the same time. But you can't make such generalized statements like "FE means they can't tech so teching will win". Going for econ means they can't tech AND make an army, not that they won't specifically not tech. Which one they sacrifice depends on the player, situation, race, etc. For example, forge expand protoss will sometime go 6-gate (army) or stargate/dts(tech). Only way to know is to scout. Basically your fundamental idea is right, but is already well-known. Many of your examples are too concrete and questionable.
|
Closing this because it's rife with misnformation. It reminds me of this old post: http://www.teamliquid.net/blogs/viewblog.php?topic_id=133609.
EDIT: Reopened by Hot_Bid's request. Take the OP and subsequent discussion with a grain of salt. There are many good points brought up, but there is a lot of incorrect advice as well.
|
- Take note of which tech structures you are able to pick off and adjust your army accordingly. Killed a Spire? Build 4-6 Vikings and go kill some Overlords. Killed a robo facility before the first Colossus was built? Get cloak for your ghosts and upgrades for your mmm. Emphasize Medivacs over Vikings. Do not expand. You haven’t killed their econ, you killed their tech, so make sure you recognize where the deficit lies and exaggerate it.
I like this quote a lot. I don't agree with everything, but you raise some very good points and I will think about these things when I strategise during games.
However, banshee killing drones = certain dmg. However if you go for a structure and you don't kill all of the queens/they kite around 1 spore, they can transfuse it, and it takes forever to kill it, and while you kill it, they can prepare their anti air and can mass their army for whatever they were planning to kill you with.
But, I will definitely reprioritise killing tech structures in this situation if there is a strong opportunity in the future.
|
|
|
|
|
|
EPT
