|
Hello TL community!
I am following the advice of numerous newby advocates (like Day 9 or Cecil) to just pick ONE build and practice it. You can hear all about attempts to follow this advice in my blog. My question is whenever I am having trouble with something and ask for advice it usually is use a different build. How should I reconcile these comments with the advice that I should only be doing one build. I have a few ideas on the subject.
1. The build I pick should be solid enough that it works in any match up and for almost all situations.
Concern: Are there builds like that out there?
2. When my opponent "hard" counters my build I should consider it a GG as I am playing to practice and not to win.
Concern: If the solution to that game is do a different build what can I learn from that game to improve next time? Obviously my newbness will not allow me to use a different build, yet.
3. Most solid builds should work even if they are "hard" countered at lower skill levels. Implying that with enough practice I should be able to beat my "hard" counter due to skill discrepancy
Concern: At about what skill level will that no longer be the case? Moreover where do I go for advice on those situations because mostly I get use a different build framed advice and not you need to this better and you can beat that counter from some one at your skill level.
Also what are some of the things I should be looking for to signal that it might be a good idea for me to throw in a different build? Just smooth timing, getting into masters and still winning, only losing games where I get hard counters, etc.
Any ways any advice would be welcome. For more details specific to me and what I am doing I will reference you to my blog as not to clog these forms with too much newbness.
http://www.teamliquid.net/blog/BroTalk
|
On April 14 2011 01:09 phone33 wrote:Hello TL community! I am following the advice of numerous newby advocates (like Day 9 or Cecil) to just pick ONE build and practice it. You can hear all about attempts to follow this advice in my blog. My question is whenever I am having trouble with something and ask for advice it usually is use a different build. How should I reconcile these comments with the advice that I should only be doing one build. I have a few ideas on the subject. 1. The build I pick should be solid enough that it works in any match up and for almost all situations. Concern: Are there builds like that out there? 2. When my opponent "hard" counters my build I should consider it a GG as I am playing to practice and not to win. Concern: If the solution to that game is do a different build what can I learn from that game to improve next time? Obviously my newbness will not allow me to use a different build, yet. 3. Most solid builds should work even if they are "hard" countered at lower skill levels. Implying that with enough practice I should be able to beat my "hard" counter due to skill discrepancy Concern: At about what skill level will that no longer be the case? Moreover where do I go for advice on those situations because mostly I get use a different build framed advice and not you need to this better and you can beat that counter from some one at your skill level. Also what are some of the things I should be looking for to signal that it might be a good idea for me to throw in a different build? Just smooth timing, getting into masters and still winning, only losing games where I get hard counters, etc. Any ways any advice would be welcome. For more details specific to me and what I am doing I will reference you to my blog as not to clog these forms with too much newbness. http://www.teamliquid.net/blog/BroTalk
Your 3 questions are essentially the same thing. There are builds for each matchup that don't have hard counters. Assuming you're P based on your icon, do these builds:
PvP: 4 gate
PvT: 2 gate robo
PvZ: 3 gate expand
Keep in mind that build orders are strictly for early game, and there are many transitions from those builds.
|
Out of curiosity, why is PvT 2 gate robo? With the stim pack nerf (+30 seconds), a lot of protoss have started doing 1 gate FE.
|
Italy12246 Posts
|
A lot of protoss were doing 1 gate FE before the stim pack nerf, but I think it's a little unstable and depending on what you scout you'll need to get more buildings before you expo. Personally I find a 1 gate robo FE to be sufficent against a terran.
That said, I think day9's advice on sticking to one build, doesn't apply to Starcraft 2 because the 'standard builds' haven't yet been developed. And it may never, since SC2 sometimes requires extreme reactions depending on what you scout (like 4 gating in PvP). It would just be ideal if you COULD stick to one build.
|
On April 14 2011 01:28 GoldenH wrote:
That said, I think day9's advice on sticking to one build, doesn't apply to Starcraft 2 because the 'standard builds' haven't yet been developed. And it may never, since SC2 sometimes requires extreme reactions depending on what you scout (like 4 gating in PvP). It would just be ideal if you COULD stick to one build.
I think that at a similar point in SC1, less than a year after "real" competition began, we could have said the same thing. I believe that it's possible that the standard builds will become clear, and I would not be surprised if they are similar to the ones stated above.
As to your statement about reacting, this is part of the game of course, but does not mean that standard builds cannot be established. A standard build is simply what you plan to do in the absence of relevant scouting info. If you're planning to do a 4gate in PvP, which is almost certainly the standard, and you see your opponent building a cannon in your base, nobody is advocating that you stick to your build and build your second gate when you are half way done with WG. No, you throw down a forge yourself immediately. SC1 had the same mindset.
As a final note, I'm a protoss player and I would consider the recommendation of 4gate PvP, 2gate robo PvT and 3 gate PvZ to be the safest possible given current knowledge.
|
i don't like the practice one build order approach, well its good for learning how your race works (bit faster i would say). Also its gives better success at first so the motivation is higher, but at some point you will have to switch from it and lose alot, where you first would have easily won, which is a terrible demotivator. Anyway if you don't get demotivated by losses, its probably a good thing ^.^ .
But playing ladder i often encounter people that once thrown off their bo which is pretty easy (especially tosses), don't recover from it. (even if i consider their mechanics way better then my in the replay)
|
So, you want to pick a build that's solid in most matchups, and try to execute it every game?
ok...
How about DT rush --> expand ---> Blink Stalker & Colossus?
From my perspective as a random player, I find it difficult to handle DTs. This is mostly due to my love of tier 1 units with most races. It generally causes me to get slower detection.
Focusing on getting DTs every single game would probably get you a hell of a lot of wins, or at least allow you to expand and transition into that scary stalker colossus ball. You'll have the twilight council for level 2+ upgrades, as well as blink.
This build is generally weak to any kind of forge opener, or very strong, early, aggressive attack.
Also, it counters itself...
Two protoss going DTs against one another can be a pretty funny situation.
EDIT: I don't agree with doing this kind of thing (doing 1 build all the time and praying it works). I prefer the scout and react approach, but that can be useless to newer players who may have no idea what they're scouting.
|
Step 1: pick a build.
It should either be:
very agressive, so that you can make it work almost no matter what the opponent does
very safe, so that you can make it work no matter what the opponent does
very reactive, so that you can make it work no matter what the opponent does
Step 2: do the build.
When you lose, do the build again, but do it better.
Step 3: Ignore random "build counterunit X!" advice
If for example you play a ZvT game, open with speedlings, and lose to hellions, 90% of the advice you will get will probably be something like "build roaches!". Ignore this kind of advice. Instead, see step 2, and do the build better. for example, better scouting, better macro, better micro, less missed injects, less missed tumors. experiment with the same thing you did, but a slight variation. 4 more lings. an earlier tumor. different building or queen placement to block the ramp, and so on. Do your build, stick to your build, dont take counterunit advice.
Step 4: Forget about "Hard" counters.
There is no such thing. Do your build better, and you will be fine. Going mass roaches, having trouble against mutas? instead of randomly deciding to add in hydras or something like that, do your build better. scout better, and push earlier, crush him right before/as his mutas pop. things like that.
Going mass marines against mass tank-hellion? you arent hardcountered, or anything like that, just be more agressive, expand more, drop all over the place, micro better to avoid splash, whatever.
the most important is just to ignore the "build counterunit X!!!!" advice.
Unless your build is absolutely shit, you will be fine. And even if the build you decide to do and stick to is 100% terrible, doing it well enough will get you into mid masters easily, probably a lot higher. Once you are in grandmaster league, you can feel free to ignore my advice, and consider that perhaps a different build would be more appropriate. Until then, stick to your build.
|
It's not so much that you should pick one build for ALL situations with no deviations (though you should pick a single stable, standard, robust build to OPEN with), but you should stick with the same gameplan. Practice one way of playing; learn its good points, its bad points, what you need to be scouting for, and the adjustments you need to make in certain situations. The idea behind it is that it's better to gain experience and become good at one thing, than to constantly change how you play your race and are mediocre at everything.
MY GAMEPLAN:OPENING: As a zerg player I always open hatch first vs terran. I want to play defensively and economicallyIf I suspect hellion I'll skip zergling speed for a bit in favour of getting some roaches first.
If I suspect banshees I'll keep making extra queens.
If I suspect 2-port banshee I'll ALSO make an early evo chamber for spores
If I suspect a big early marine push I'll get my baneling tech faster than usual. MIDGAME: I get fast melee & carapace upgrades, infestors, and mass ling/bling. I want to harass and counterattack as much as possible.If I see that he's making a lot of hellions I'll make less lings and throw in some roaches instead.
If I see big mech play I'll also research neural parasite.
If he's doing a lot of drop harass I'll get a spire and make a few mutalisks to discourage him. LATEGAME: I get a fast hive, ultralisks, and max out on upgrades.
That's what I do. It might not be the "best" (it's a very rough outline, and I'm only a Diamond scrub), but that's how I like to play my zerg. Note that my gameplan is simply the bits in bold! All those deviations seen in the code blocks have been learned because I've practice doing this style a lot and I've found they were necessary, or at least helped quite a bit. If you keep changing the core style you play a matchup that learning is going to be slower. By keeping with the same plan it is easier to make small iterations and adjustments and see how they pan out in subsequent games.
|
When people give that advice they mean 1 build per matchup. So 3 total builds. And yes there is a build for each matchup that can adapt to anything. That means there is no hard counter to the build, since if there were, that would make it a cheese.
The point of following a single build per matchup is to get a good feel for the match, develop game sense of timings, gain an understanding of racial dynamics and how different army compositions work and feel, and, above all, to hone your mechanics.
Once you bring your macro and micro to a decent level (high plat to mid diamond), you can begin playing with other builds to get a feel for how they work. Anything below this level and you should be able to win with just about any crisp build, regardless of "hard counters".
Anihc had a good list of builds for learning a solid macro style. In the interest of being a complete player, once you hit diamond I'd recommend you learn a few early cheeses and also experiment with stargate openers and adelscott style play (focus on upgrades and gateway units). Doing the same build every game will get boring after a while, and the build you choose at first may not match your personal style  .
|
1. Yes, this has already been answered in the thread and since i'm not a P player i don't have good inputs for you here but what you have been told looks pretty solid.
2. It really depends on your mindset and what you are trying to accomplish with this game, even when a game looks lost you can still try to train your micro and make it hard for your oppenent to finish you.
3. Until you are fairly high master/kind of a competitive player
Bear in mind that while training the same build over and over you have to be looking for key stuff.
The #1 for me would be timing, good thing with playing the same over and over is that you can get your timing downs (i.e. when i do THIS my oppenent is probably doing THIS, if he's not then i gotta worry about something else).
Also what you don't seem to process is that builds are for the early game only, you have to choose the right transition to the midgame by scouting.
|
On April 14 2011 01:21 Griffith` wrote:
Out of curiosity, why is PvT 2 gate robo? With the stim pack nerf (+30 seconds), a lot of protoss have started doing 1 gate FE.
OP seems low-level, and 1gate expo takes very good execution and micro to hold early (pre-stim) attacks.
|
On April 14 2011 01:09 phone33 wrote:
My question is whenever I am having trouble with something and ask for advice it usually is use a different build.
This is a good sign that the guy giving you advice has no idea what he's talking about and can ignore his post. If you were working on a PhD and needed advice on finishing your dissertation, would you listen to advice that says "just become a doctor instead, that's what I did and I make tons of money"? Likewise if you're working on 3 gate expo vs mutaling, you ignore the people who say "just go 15 nexus instead and mass phoenixes every game, that's what I did and I beat mutaling every time". Each time you switch your build, you reset your learning process and have to start from scratch. It's no different from switching your race or switching to another RTS game.
As for your questions, here are some answers I can give you.
1: Unfortunately there is no good build that works in every matchup, except maybe 6 pool.
2 and 3: Good builds don't have counters. If they did, people wouldn't use them.
If you want a good set of builds to practice, here are the most standard ones that don't have counters.
PvP: http://www.teamliquid.net/forum/viewmessage.php?topic_id=177569
PvZ: http://www.teamliquid.net/forum/viewmessage.php?topic_id=194376
PvT: http://www.teamliquid.net/forum/viewmessage.php?topic_id=196385
|
On April 14 2011 02:13 FreezerJumps wrote:OP seems low-level, and 1gate expo takes very good execution and micro to hold early (pre-stim) attacks.
Indeed. As a fairly newbie player myself, 1-gate FE (or 1 rax FE or hatch first, for that matter) is really hope-based, and probably a bad idea in general; because I can't defend it, it is basically a huge gamble that will either pay off big (massive economy allowing me to crush my opponent midgame), or fail spectacularly if my opponent goes for some early pressure.
Because I am throwing the dice every game, it is only somewhat of a better idea than cheesing.
As others have said, 3 gate expo into gateway/colossus PvZ, 2 gate robo expand into gateway/colossus PvT and 4 gate PvP are all great "standard" builds to work with.
If the OP is still having trouble, I would also suggest an "economic" 4 gate (i.e. don't cut probes) in every matchup until he is good enough at probes'N'pylons™ to make the other builds work.
|
Your plan for improving should carry you to Master's on 1build for each matchup.
#2: If you're not in Master's league, there is always a ton of little things you can do better that will let you win with almost any strategy because your opponent will make a lot of mistakes. Scout better, make probes better, refine your build order as you get on 2-3 bases so that your buildings are going down at the best times, get your assimilators at the perfect time, build units at the right time so that you have more stuff, harass more, split your army better, build your buildings in better locations. I've been a diamond league player for awhile now and I suck at doing all of these things.
#3: the number of people who play unexploitably (i.e. they don't miss macro when microing, etc) is incredibly small. you will know when you hit the point where you can't get mechanically and tactically better than the people you're playing against and need to use different strategies, but by the time you get there you'll already have become bored with your primary strategy because you will have refined it so well that you want to work on something new. If you don't know your build and why and when you build different tech buildings and how the changing tech trees effect your gas timings and # of gateways and what unit compositions you're shooting for with your different tech options and... (insert everything about your current strategy) then you're probably not at that point. Basically, by the time you reach that point you will understand the game a hell of a lot better and you won't be asking for this advice on TL.
|
Personally as Toss, at Masters level I use one to two openings per matchup.
PvP: off-4 gate or def-gate if the other guy is 4-gating as well.
PvT: 1-gate expand or 3-gate expand if I sense early aggression. If I see 2 gases I'll rush a robo instead of the other 2 gates. 15 nexus in extremely rare situations.
PvZ: usually 3-gate expand, sometimes forge FE. 1-gate expand in very rare situations.
Basically I feel like if there's one build you should learn it's 3-gate expand as it's probably the most versatile. Transitions are really easy into almost all other styles, i.e. you can drop a robo + gates or 2 gates or 3 gates after your expo is up. Once you master 3-gate expand and how to do it safely, you can start learning other builds easily since you'll have a great understanding of warpgate unit dynamics. You can in some sense see the other builds as variations of the 3-gate expand.
PvP is obviously different since the 4-gate is such a staple of the MU.
|
On April 14 2011 01:19 Anihc wrote:Show nested quote +On April 14 2011 01:09 phone33 wrote:Hello TL community! I am following the advice of numerous newby advocates (like Day 9 or Cecil) to just pick ONE build and practice it. You can hear all about attempts to follow this advice in my blog. My question is whenever I am having trouble with something and ask for advice it usually is use a different build. How should I reconcile these comments with the advice that I should only be doing one build. I have a few ideas on the subject. 1. The build I pick should be solid enough that it works in any match up and for almost all situations. Concern: Are there builds like that out there? 2. When my opponent "hard" counters my build I should consider it a GG as I am playing to practice and not to win. Concern: If the solution to that game is do a different build what can I learn from that game to improve next time? Obviously my newbness will not allow me to use a different build, yet. 3. Most solid builds should work even if they are "hard" countered at lower skill levels. Implying that with enough practice I should be able to beat my "hard" counter due to skill discrepancy Concern: At about what skill level will that no longer be the case? Moreover where do I go for advice on those situations because mostly I get use a different build framed advice and not you need to this better and you can beat that counter from some one at your skill level. Also what are some of the things I should be looking for to signal that it might be a good idea for me to throw in a different build? Just smooth timing, getting into masters and still winning, only losing games where I get hard counters, etc. Any ways any advice would be welcome. For more details specific to me and what I am doing I will reference you to my blog as not to clog these forms with too much newbness. http://www.teamliquid.net/blog/BroTalk Your 3 questions are essentially the same thing. There are builds for each matchup that don't have hard counters. Assuming you're P based on your icon, do these builds: PvP: 4 gate PvT: 2 gate robo PvZ: 3 gate expand Keep in mind that build orders are strictly for early game, and there are many transitions from those builds.
This is pretty much it, I believe. Play a well-known build that can deal with anything. If you play any of those builds, there are no hard-counters to them. There may be some soft counters to be build, but you can adapt to them.
On the other-hand, for example, if you decide in PvT to open 3-gate 1-star and try and bust the choke, be prepared to get hard-countered by bunkers + cloaked banshees.
|
On April 14 2011 02:00 morimacil wrote:
Step 1: pick a build.
You should have some understanding of your build's strengths and weaknesses before you choose it.
some builds are:
-very agressive: These are usually built regardless of what your opponent does because you are attempting to put on pressure/kill your opponent before their tech tree develops and has no need to be reactive. Very agression-focused rather than economy focused.
-very safe: These are usually also buiult regardless of what your opponent does but rather than focusing on agression, you are focusing on a strong defense
-very reactive; These builds are based on scouting information and follow several paths based on what you scout. Most often are economic-defensive-style builds but the defenses aren't built until you feel an attack is imminent.
Step 2: do the build.
When you lose, do the build again, but do it better.
-Watch your replay. Focus on where you slipped up. Focus on making that better next time
Step 3: Ignore random "build counterunit X!" advice
If for example you play a ZvT game, open with speedlings, and lose to hellions, 90% of the advice you will get will probably be something like "build roaches!". Ignore this kind of advice. Instead, see step 2, and do the build better. for example, better scouting, better macro, better micro, less missed injects, less missed tumors. experiment with the same thing you did, but a slight variation. 4 more lings. an earlier tumor. different building or queen placement to block the ramp, and so on. Do your build, stick to your build, dont take counterunit advice.
-99% of the time your loss can be attributed to a sloppy build (building placement, supply block, non-continuous unit production), unit control, or poor scouting information. If you get 4gated and lose to a really strong push that no amount of micro/unit warping could have saved you from, it means you didnt scout well enough.
-stick to your build.
Like this post, but disagree with some points. I've changed them above to reflect my opinion.
|
The point of following a single build per matchup is to get a good feel for the match, develop game sense of timings, gain an understanding of racial dynamics and how different army compositions work and feel, and, above all, to hone your mechanics.
Thats kind of true, but also very misleading.
The real point of doing 1 build a lot, is to get good at it.
If you know 1 build inside out super well, with incredibly crisp timings, knowing exactly what to do in every situation and so on, well you can be a pro. Idra spent most of his time until recently doing pretty much 1 build per matchup. He was predictable, but even when ppl knew what he was going to do, he made it work.
On the other hand, if you know how to do 50 different builds/strategies/openers/whatever, but you arent really particularly good at any of them, well that means that you arent really good.
Its the same with everything really. This also applies to most other things in life.
If you are the world's leading brain surgeon, and you dont know anything else but brain surgery, but you are the best brain surgeon out there... well gz, you probably have a great job.
If on the other hand, you know a little about brain surgery, but not really more than the average guy, and youve also tried painting a couple of times, but it didnt work out too well, and you tried to do a little programming one time, and kinda had a little success with it, and you also can cook a little, but not all that well, and ..... Well really, that makes you kinda useless, and you will probably get a shit job
|
|
|
|
|
|
EPT
