Yup, it's true what they say. I just got done with 18 games in three hours, winning 16 of them. I used nothing but 4 Warp Gate. One loss was to 8pool on Delta Quadrant, and the other was to a fast-expanding Terran on Lost Temple who managed to defend his expo with SCVs repairing bunkers. Kudos to that Terran player. But I'm not sure what I could have done against the Zerg. I happened to scout their location last, so by the time I knew what was going on it was probably too late. I guess I'll have to start sending a second scout against Zerg.
Anyway, 4 Warp Gate is stupidly amazing. (Here's an example replay.) I think I'll try adapting it to include taking my natural expansion. Sometimes it can be difficult to break my opponent's defenses, but then I can usually contain them in their main base. And once my second base is up and running while they're still on one base, it's only a matter of time before I can overrun them with pure numbers.
Games played: 395
League/Points: Diamond/1039
Saturday, September 25, 2010
Thursday, September 23, 2010
Greetings, Prophet. The great work begins! The Messenger has arrived.
This is where the real work begins and I start deciding how dedicated I truly want to be. After almost 400 games, I've finally stopped coasting. I don't just automatically win most of my games anymore. I can't just expand when I feel like it. Harassment isn't guaranteed to be profitable. Sloppy wall-ins get broken down, and sloppy macro gets outdone. And most importantly, just because I have a big army doesn't mean it's going to be a good idea to attack move into the opponent's main. Here's a good example of me losing to brazenness (ZvZ on Delta Quadrant).
It's time to start thinking a little more and polishing my game up a little more. I think I'll focus on just one race for a while, probably Protoss, my weakest of the three. Perhaps I'll start with the typical 4 Warp Gate push and see where that takes me. I like the idea of starting by using only one strategy. Eventually, you perfect it, and then you can look at the games you're still losing to see where that strategy's weaknesses really are, at which point you modify it as little as possible to accommodate encountering a similar problem in future games. As you continue to modify it over time, you'll in theory eventually end up playing a fully versatile, many-faceted game that may or may not even incorporate the original strategy.
Games played: 372
League/Points: Diamond/806
It's time to start thinking a little more and polishing my game up a little more. I think I'll focus on just one race for a while, probably Protoss, my weakest of the three. Perhaps I'll start with the typical 4 Warp Gate push and see where that takes me. I like the idea of starting by using only one strategy. Eventually, you perfect it, and then you can look at the games you're still losing to see where that strategy's weaknesses really are, at which point you modify it as little as possible to accommodate encountering a similar problem in future games. As you continue to modify it over time, you'll in theory eventually end up playing a fully versatile, many-faceted game that may or may not even incorporate the original strategy.
Games played: 372
League/Points: Diamond/806
Wednesday, September 22, 2010
O_O
I'm finally feeling better, so I've been playing a little more. First impressions of Diamond League: Oh my god, people actually know how to make units! Like, as many as I do! Am I going to have to employ... strategy?
Sunday, September 19, 2010
Starcraft a-slow-slow
Haven't been playing much recently due to being seven kinds of sick. Just now I tried some games and ended up winning 3 and losing 6. I kept making ridiculous macro mistakes, immediately falling behind like I did in the initial weeks. I guess my mind isn't as sharp right now. I'll get some rest and come back when I feel better.
Saturday, September 18, 2010
Tuesday, September 14, 2010
Lessons in the Fundamentals of Starcraft
There is a famous book about Go called Lessons in the Fundamentals of Go, written by Toshiro Kageyama, a top professional player. Instead of using lots of variation diagrams, individual joseki analysis, etc., the book takes a small number of simple examples and uses them as a platform from which to speak about how you should think about the game. I remember a line in there somewhere about how the way to improve at go is to start by studying the fundamentals vigilantly, building a solid foundation for your game, and then, when you begin to progress to the higher stages, to study the fundamentals again. Kageyama seems obsessed by this idea in his writing, as though frustrated that more people don't understand and appreciate it.
So does Day[9]. Just look at some of his recent Dailies, especially #177 "Simple Winning Techniques", which I am now halfway through watching. His basic idea is that people become too concerned with intricately countering the opponent's strategy and/or unit composition and lose sight of the fact that you can just win by getting a huge army and rolling all over them. And it makes perfect sense. Sure, you can beat those Colossi by getting Vikings, but you can also beat them with more and more Marine-Marauder. You will lose more army value in the fight, but that's value that would have been spent on the tech anyway. I'm speaking in broad theoretical terms, of course. Continuing the Colossus/Viking example, it seems the main differences between the two approaches are... 1. You only have to pay for tech once, but you have to pay for army size repeatedly. If the game goes longer (that's an important "if"), you will still have access to Vikings as you build and rebuild your army. But every time you rebuild with huge numbers of MM, you have to pay again for that bulk. So teching to a counter could be seen as an investment just like expanding. 2. Specialized units are just that--specialized. After your Vikings take care of all the Colossi, what are they going to do? If you end up using them a lot in ground assault mode, you basically just have a bunch of slow, overpriced infantry that can't stim. Since you cannot exactly predict what units your opponent will get, this phenomenon is essentially guaranteed to occur every time; the only question is to what extent.
It's ironic that Day[9] has started harping on this subject recently, because it's exactly the stage where I find myself. I am still just hideously awful at micro and harassment--both using it and countering it. It's because I instinctively interpret macro (for better or worse) as the initial requirement that you must master to at least 90% or so before you can even begin to play the real game. If I let a Queen get to 50+ energy because I'm forgetting to inject, and then my opponent shows up at my base and kills my whole army, then as far as I'm concerned, that fight didn't really happen. It was totally invalid, imaginary. I've heard of some Go teachers who despise being asked questions about how move 50 could have been improved in a game where move 40 was some fundamental error. "It's a meaningless question anyway! That position is impossible." To these people, playing a game is like doing math. Who cares what the square root of x equals if x was obtained by dividing by zero? Logic says that if you assume a falsehood, you can prove anything. I will now prove that I am a god. Let us assume that pigs fly. Clearly, since pigs don't fly, it is true that if pigs fly, then I am a god. But I have assumed that they do. Therefore, I am a god. Now let us show that that wave of Stalkers my opponent is sending to my base is going to utterly annihilate me. First, let us assume the absurdity that I, for no particular reason other than forgetfulness, did not inject larvae. Clearly, if I do not inject larvae, then my economy and army will be inferior. Thus, as we now see from the mass carnage on screen, I lose the game. Well, so what? The situation was impossible anyway. Let us not, oh please for the love of Aiur, let us not begin postulating some new strategy to deal with a similar wave of Stalkers in a future game. The fact that I made a series of dumb macro mistakes and then got killed tells me hardly anything at all about whether my build/strategy was any good. Would you think that Hellion harass was bad because you got distracted and accidentally ran your Hellions straight into a ball of Roaches? That, to me, seems just as silly of a conclusion. But here's the difference between those two things: Hellion harass is something you try from time to time, and macro-management is something that you have to do every damn game. So how can I know if, uh, getting units and attacking with them is any good until my macro is reasonably close to ideal?
I've now played over 300 games with this attitude, and I still see myself as a total macro noob. (Oh, but did I mention I'm back in Platinum, rank 2 in my division, my most played race is now listed as Zerg, and I usually play Random? Yes, things are a-changin'.) I recently played a game (here's the replay) as Protoss where I totally blew my own mind by managing to maintain four bases and then using my massive income to overrun my 3-base Zerg opponent. Being able to easily and continuously produce off of ten Warpgates, two Stargates and a Robotics Facility gave me a childlike glee like what I felt when I played my first 1v1 on 7/27 and wreaked havoc with a fleet of Battlecruisers. It made me think back to when just the idea of expanding to my natural was a lot to handle, and it reminded me of how much of a way I still have to go before I gain the skill to actually go around the map and do what I want to do. Strategy is all fine and good, but I'm still working on moving my pieces across the board.
Here are my main macro points to work on:
1. Understand the progression of your income, and anticipate the immediate future. Don't wait to get to 1000 minerals before you realize you need to build more production buildings. Understand that as you get more workers and expand, you will need to have things in place to be able to spend the added income. Always know approximately how many workers are at each base and how many are collecting gas.
2. Don't panic. The speed of the game can make it feel like you'll never have the time to build enough stuff to spend all your money, while making sure you've got three workers on each gas and minerals aren't over- or under-saturated, etc. But that's just not true. Really, how much does it take to start spending your money? Maybe building, say, four more production buildings and adding them to your hotkey groups? Is that really so complicated?
3. Similar to #2, don't get sidetracked by less important but more practiced activities. If you are gas blocked, stop figuring out how many Overlords to make and start making Extractors and putting three Drones on each. Especially in a panicked state, it is easy to fall back on the most ingrained habits. But if the habits are ingrained, that means you're already doing them well and don't need to devote as much attention to them! It's a vicious circle. By definition, you have to put more effort into things that are more difficult.
4. Don't stop producing workers! No really, don't stop! Ever! I mean it! Keep making them! Do you have a lot now? Oh, that's cool. Keep making them!
Some other points:
1. The enemy is not the units attacking you. The enemy is the guy clicking on those units and telling them to attack you. And that guy is just as limited as you are. Gaining an understanding of what that guy is doing and why is fundamental to understanding the progression of a game. As you play, imagine casting your own game as a spectator with full vision of the map. If you don't know enough to be able to do that, scout! (Or, as Day[9] suggests, just reason out his limitations.)
2. Chokepoints are death chambers. The DPS of a Roach that can't get close enough to the enemy to attack is 0. It is perfectly normal to have two comparable armies on either side of a choke in a standoff situation. That's why it's so difficult to attack the opponent's main.
3. Don't get chained to army hotkeys. Box off units and issue orders appropriate to that chunk. In particular, don't repeatedly press 3a-click (I use 3 for my main army) on the same spot. That's like trying to make Mario jump higher by squeezing the A button as hard as you can. A waste of energy.
4. Have fun! Remember how awesome this game is and what an opportunity it is to be a part of such a large, active community.
Games played: 317
League/Points: Platinum/887
p.s. Fun fact: Glancing at the analysis in Sc2gears (highly recommended app, by the way), I see that since September 1 when I started using the program, Random has landed me on Terran 16 times, and I won 15 of those games. XD
So does Day[9]. Just look at some of his recent Dailies, especially #177 "Simple Winning Techniques", which I am now halfway through watching. His basic idea is that people become too concerned with intricately countering the opponent's strategy and/or unit composition and lose sight of the fact that you can just win by getting a huge army and rolling all over them. And it makes perfect sense. Sure, you can beat those Colossi by getting Vikings, but you can also beat them with more and more Marine-Marauder. You will lose more army value in the fight, but that's value that would have been spent on the tech anyway. I'm speaking in broad theoretical terms, of course. Continuing the Colossus/Viking example, it seems the main differences between the two approaches are... 1. You only have to pay for tech once, but you have to pay for army size repeatedly. If the game goes longer (that's an important "if"), you will still have access to Vikings as you build and rebuild your army. But every time you rebuild with huge numbers of MM, you have to pay again for that bulk. So teching to a counter could be seen as an investment just like expanding. 2. Specialized units are just that--specialized. After your Vikings take care of all the Colossi, what are they going to do? If you end up using them a lot in ground assault mode, you basically just have a bunch of slow, overpriced infantry that can't stim. Since you cannot exactly predict what units your opponent will get, this phenomenon is essentially guaranteed to occur every time; the only question is to what extent.
It's ironic that Day[9] has started harping on this subject recently, because it's exactly the stage where I find myself. I am still just hideously awful at micro and harassment--both using it and countering it. It's because I instinctively interpret macro (for better or worse) as the initial requirement that you must master to at least 90% or so before you can even begin to play the real game. If I let a Queen get to 50+ energy because I'm forgetting to inject, and then my opponent shows up at my base and kills my whole army, then as far as I'm concerned, that fight didn't really happen. It was totally invalid, imaginary. I've heard of some Go teachers who despise being asked questions about how move 50 could have been improved in a game where move 40 was some fundamental error. "It's a meaningless question anyway! That position is impossible." To these people, playing a game is like doing math. Who cares what the square root of x equals if x was obtained by dividing by zero? Logic says that if you assume a falsehood, you can prove anything. I will now prove that I am a god. Let us assume that pigs fly. Clearly, since pigs don't fly, it is true that if pigs fly, then I am a god. But I have assumed that they do. Therefore, I am a god. Now let us show that that wave of Stalkers my opponent is sending to my base is going to utterly annihilate me. First, let us assume the absurdity that I, for no particular reason other than forgetfulness, did not inject larvae. Clearly, if I do not inject larvae, then my economy and army will be inferior. Thus, as we now see from the mass carnage on screen, I lose the game. Well, so what? The situation was impossible anyway. Let us not, oh please for the love of Aiur, let us not begin postulating some new strategy to deal with a similar wave of Stalkers in a future game. The fact that I made a series of dumb macro mistakes and then got killed tells me hardly anything at all about whether my build/strategy was any good. Would you think that Hellion harass was bad because you got distracted and accidentally ran your Hellions straight into a ball of Roaches? That, to me, seems just as silly of a conclusion. But here's the difference between those two things: Hellion harass is something you try from time to time, and macro-management is something that you have to do every damn game. So how can I know if, uh, getting units and attacking with them is any good until my macro is reasonably close to ideal?
I've now played over 300 games with this attitude, and I still see myself as a total macro noob. (Oh, but did I mention I'm back in Platinum, rank 2 in my division, my most played race is now listed as Zerg, and I usually play Random? Yes, things are a-changin'.) I recently played a game (here's the replay) as Protoss where I totally blew my own mind by managing to maintain four bases and then using my massive income to overrun my 3-base Zerg opponent. Being able to easily and continuously produce off of ten Warpgates, two Stargates and a Robotics Facility gave me a childlike glee like what I felt when I played my first 1v1 on 7/27 and wreaked havoc with a fleet of Battlecruisers. It made me think back to when just the idea of expanding to my natural was a lot to handle, and it reminded me of how much of a way I still have to go before I gain the skill to actually go around the map and do what I want to do. Strategy is all fine and good, but I'm still working on moving my pieces across the board.
Here are my main macro points to work on:
1. Understand the progression of your income, and anticipate the immediate future. Don't wait to get to 1000 minerals before you realize you need to build more production buildings. Understand that as you get more workers and expand, you will need to have things in place to be able to spend the added income. Always know approximately how many workers are at each base and how many are collecting gas.
2. Don't panic. The speed of the game can make it feel like you'll never have the time to build enough stuff to spend all your money, while making sure you've got three workers on each gas and minerals aren't over- or under-saturated, etc. But that's just not true. Really, how much does it take to start spending your money? Maybe building, say, four more production buildings and adding them to your hotkey groups? Is that really so complicated?
3. Similar to #2, don't get sidetracked by less important but more practiced activities. If you are gas blocked, stop figuring out how many Overlords to make and start making Extractors and putting three Drones on each. Especially in a panicked state, it is easy to fall back on the most ingrained habits. But if the habits are ingrained, that means you're already doing them well and don't need to devote as much attention to them! It's a vicious circle. By definition, you have to put more effort into things that are more difficult.
4. Don't stop producing workers! No really, don't stop! Ever! I mean it! Keep making them! Do you have a lot now? Oh, that's cool. Keep making them!
Some other points:
1. The enemy is not the units attacking you. The enemy is the guy clicking on those units and telling them to attack you. And that guy is just as limited as you are. Gaining an understanding of what that guy is doing and why is fundamental to understanding the progression of a game. As you play, imagine casting your own game as a spectator with full vision of the map. If you don't know enough to be able to do that, scout! (Or, as Day[9] suggests, just reason out his limitations.)
2. Chokepoints are death chambers. The DPS of a Roach that can't get close enough to the enemy to attack is 0. It is perfectly normal to have two comparable armies on either side of a choke in a standoff situation. That's why it's so difficult to attack the opponent's main.
3. Don't get chained to army hotkeys. Box off units and issue orders appropriate to that chunk. In particular, don't repeatedly press 3a-click (I use 3 for my main army) on the same spot. That's like trying to make Mario jump higher by squeezing the A button as hard as you can. A waste of energy.
4. Have fun! Remember how awesome this game is and what an opportunity it is to be a part of such a large, active community.
Games played: 317
League/Points: Platinum/887
p.s. Fun fact: Glancing at the analysis in Sc2gears (highly recommended app, by the way), I see that since September 1 when I started using the program, Random has landed me on Terran 16 times, and I won 15 of those games. XD
Subscribe to:
Posts (Atom)