|
So I spent some time practicing a specific build order, and I have it down pretty well.
First I watched a replay with a build I wanted to emulate. The replay is a PvP, Bisu Vs Stork, where Bisu uses a nice goon reaver build. I watched the replay, and wrote down the build. I watched it a couple more times just to double check. Then I played about 10 games vs a comp, where I just did the build, and quit when I got my second reaver. Then I would record the time and do it again, with the goal of doing the build as efficiently as possible. I felt like the build wasn't quite working the way it should, so I watched the replay again, and after having done the build several times, I could see that I had missed a couple details. Then I did the build probably another 30 times or so over the last two days.
This is the build:
8 - Pylon
10 - Gate
11 - Assim
13 - Zealot
15 - Pylon
17 - Core
18 - Zealot
22 - Pylon
23 - Goon
26 - Range
27 - Goon
30 - Robo
31 - Gateway and cut probe production
31 - Goon
33 - Pylon
34 - Goon
36 - Goon
38 - Shuttle
40 - Pylon
40 - Observatory
- Two more Goons
- Obs when observatory is finished
- Support Bay
- While your obs is building, your last two goons will pop out, then pause goon production for a moment to get out your first reaver.
- Add a third gateway
- Keep pumping goons and a second reaver.
The most interesting thing to me about practicing this build order is that I seemed to be making much more stuff than I normally would expect to be able to. The build is worked out so that several things that you need will finish just in time, at exactly the same time, which is agreat feel. Makes it a much more smooth operation that I normally have when playing by feel. I guess my "feel" needs to change. Next I'm going to play some ICCup games and use the build, and see if I can hold it together while I'm being harassed and distracted.
Here's the Bisu Vs Stork rep that I took the build from:
http://www.scref.net/replays/iefstoirk_vs_bisu1.rep
|
this is IMO, the smartest way to train
|
Appearntly F91 doesn't train like this. And he's doing pretty ok.
|
On March 09 2009 06:05 Navane wrote:
Appearntly F91 doesn't train like this. And he's doing pretty ok.
wow
I guess that argument is bulletproof
ok guys lets all pack up and go home
|
On March 09 2009 06:05 Navane wrote:
Appearntly F91 doesn't train like this. And he's doing pretty ok.
So what does he do? Just build stuff when he has money?
|
lol sorry, that was a bit short.
I think the op does a really good training method. But i think that while emulating progamers is a good way to take large steps skillwize, the style you are trying to emulate might be too good.
What Day[9] said in his talk about timings, creating your own. I think that way you create a style that fits yourself better and gives you more confidence and agility to adapt.
When you train for tennis, you don't try to emulate Nadal cause you just wont handle his style. YOu're better off learning from your local tennis instructor.
So again in short: OP's method is like x^(1/2) while studying for youself is more like x^2. Slow start vs quick start, but you're more likely to hit a roof.
|
On March 09 2009 06:16 Navane wrote:
lol sorry, that was a bit short.
I think the op does a really good training method. But i think that while emulating progamers is a good way to take large steps skillwize, the style you are trying to emulate might be too good.
What Day[9] said in his talk about timings, creating your own. I think that way you create a style that fits yourself better and gives you more confidence and agility to adapt.
When you train for tennis, you don't try to emulate Nadal cause you just wont handle his style. YOu're better off learning from your local tennis instructor.
So again in short: OP's method is like x^(1/2) while studying for youself is more like x^2. Slow start vs quick start, but you're more likely to hit a roof.
I'd have to disagree.
Tennis has the physical fitness factor that, of course, a beginner won't have when first starting tennis. So obviously a beginning tennis player can't copy nadal move for move.
This (I guess) can be seen as similar to skill in StarCraft. Sure, beginners can't multitask like bisu and stork or micro/macro at their level, but their goal is to match them, or even surpass them.
In both cases I'd imagine beginners to copy pros. No, you won't last in a long match like nadal can, but you can sure try to learn how to serve and return like he does.
Just because someone uses a build used by bisu doesn't mean he's using bisu's style. I can serve like nadal, but play a different game style.
Everyone hits a roof at everything they do. Musicians, gamers, writers, athletes. It's unavoidable. There is no way to learn that prevents anyone from hitting a roof.
|
Thanks for the comments guys.
I definitely want my own play style, but at this point I'm just trying to get a feel for what it's like to use a well worked out build that gets me up one side of the tech tree efficiently. I can't multitask like Bisu, but just because you can't hit it like tiger doesn't mean you can't get better by emulating him. The part of Bisu's game I'm copying is just the very beginning when there hasn't really been much combat yet. I don't seem to have much trouble doing this build while pressuring with goon reaver and producing from hotkeyed gates.
|
Lol @ the debate.
It can never hurt to emulate those that have had success.
|
I feel this method of training makes you extremely weak to the unorthodox and very bad at adapting.
|
If you have better ideas I'd love to hear them Thunk.
|
You seem pretty happy with so far and if works for you you should continue to do it. I think further down the road you're going to have to watch how this build stacks up against various cheese builds and how to scout them to find your counter.
It's also tough when somebody 2 gates you or takes out 1-2 probes your psi count is off + everything will come just slightly behind.
|
Makes sense. One step at a time though. =D
|
Oh, something to help you. If you use Chaos Launcher's apm alert, it'll help you deduce times when you don't have a rep (just a VOD). For instance, a DT drop (deduced from game 4 of Bisu versus Xellos) you cidatel @ 3:20, Gateway at 3:30, Robo @ 4:00, Archives @3:10, and get DT's at 5:30. Slightly imprecise, but very it's surprisingly useful.
Best of luck.
|
On March 09 2009 07:10 thunk wrote:
I feel this method of training makes you extremely weak to the unorthodox and very bad at adapting.
This method is and will always be the first thing a D player must do. As the first and formost thing in starcraft at D level is to learn a solid build order. Adaptation, timing, hotkeys, multitasking, micro, harass, ... that are all steps that come later in the proces. First things first, one step at the time.
But when you are busy practicing the mechanics for you build order (singleplayer or on iccup) it is also important that you understand the reasoning of your build. To understand why you build what you build when you build it. Because once you become a better player (and your opponents also), the next steps in your learning proces will require you to not only to be able to execute the build flawlessly, but to know why your executing it.
The rep you chose is actually i pretty awesome rep to learn something, compare the builds stork and bisu used. They are almost identical, but stork opens with 11-pylon and bisu opens with 11-assim. That can be a huge factor. Also when do you move out with this build and why? Do you expand or get a third gate? (stork and bisu made different decisions there, why?), what maps can or can it not be applied with succes, ....
Learning the build order is not only practicing the mechanics, but also understanding the reasoning behind it. If you get those both, you can move on to another build order and start thinking about adapting, timing, harass, etc...
EDIT: i actually would prefer the 11-pylon version of this build, especially at maps like longinus which have no ramp. Better safe then sorry.
|
On March 09 2009 07:10 thunk wrote:
I feel this method of training makes you extremely weak to the unorthodox and very bad at adapting.
as of now, sure, but if he uses it over and over on iccup it'll fix that
btw op if I had your dedication i would definitely use this training method, gj
|
I've been playing a bunch of ICCup games, but I can only practice this build if it's PvP. I think i made get down a PvT and PvZ order as well so I always have one to practice.
|
On March 09 2009 07:19 thunk wrote:
You seem pretty happy with so far and if works for you you should continue to do it. I think further down the road you're going to have to watch how this build stacks up against various cheese builds and how to scout them to find your counter.
It's also tough when somebody 2 gates you or takes out 1-2 probes your psi count is off + everything will come just slightly behind.
of course. So then you have a build you know by the back of your hand that doesnt respond perfectly to everything. Then you go play some games, get beaten by soething you've never seen/studied, go watch the rep, see if you can find pro reps that show how to deal with that situation, and if you can't try various adaptations until you can vary the build to properly respond.
But just because a 12 hatch can be bunker rushed, doesnt mean that you shouldnt try and study and perfect it. People don't just improvise the beginning because of that, instead they hold onto their build and learn to adapt it to deal with a bunker rush.
The same thing happens with longer build orders. Yes you will be orthodox until you learn all the different variations and responses, but its like laying a foundation to build a house. At least by having a consistent, well-executed build you know what is possible.
If I just improvise, then I dont know how to counter something I haven't seen before. Is my build inferior to it? Is my build good, but needs tweaks? Should I have more units than this?
By starting off with a well-executed build and then diverging you have the maximum _______ at a given period of time, where ____ is units/economy/tech/whatever the purpose of the build was. I then immediately know that I have the maximum resources to deal with this divergence and it then becomes much simpler to test and find out what is required to adapt. In other words, by making sure I'm executing a standardized build very well (hopefully perfectly), there are far less variables for me to test and consider when it comes to adapting.
In addition, every time I encounter that situation I'm always at the exact same starting point, which again, means its simpler to find a working adaptation.
|
I've thought about it ever since you've posted it and I think your argument is very valid but (keep in mind that these aren't points to take down your argument, but rather something to keep in mind):
a) You've now crossed the realm into training and getting better. It's no longer a game, it's something you do to get better. Not that this is a bad thing, you're now running to get better at tennis rather than playing tennis and getting better. It indicates nothing good, nothing bad, but the mindset you're in.
b) Some builds just don't adapt very well. I was trying this with Nony's 1 base reaver PvT (build stripped from his VOD) and I had a lot of trouble (even with the ramp) against an aggressive FD and there are some builds that don't lend itself to this training method. That said, I think it's very effective in some match-ups like PvP, TvZ, ZvT, but not to certain match ups like PvZ or ZvZ.
I have some funny anecdotes with this method but I'm not going to type them up right now.
|
There are several problems that you'll encounter if you're just trying to mimick progamer builds and are a newbie (I'm not saying I'm particularly good or anything either).
- if you don't actually understand the game enough to know why they did the build, what situation it's used in, etc... then you won't be able to get as much out of it.
- if you just copy a build and learn it by rote you will be shitty as fuck at adapting or making any changes. if you encounter something strange in the game you're going to be so baffled you won't know what to do and at that point you'll either: fuck the build up, OR continue it like a robot and get countered.
- if you have bad mechanics or whatever, you'll be focusing so much on doing the build order that you'll pretty much forget about everything like scouting, microing, etc... and overall it's going to hurt your game a lot more than it will help.
- if you are just copying build orders you won't be playing at a style that actually suits you. there's a reason bisu comes up with his builds, and a reason other players use different ones. if you just copy one guys shit you won't really know if you SHOULD be using that build, if it's even right for you etc...
all these things are reasons copying build orders without proper understanding is bad.
I'm sure there are tons of reasons why it's a good idea too, I guess it's a matter of preference.
|
|
|
|
|
|
EPT
