|
Besides playing a lot of games on ladder, unranked, etc. What kind of arcade games or unique mode games you would recommend to practice ?
For example, there is arcade mode called ``unit tester`` , there you can perfect different unit compositions, find out how casting affects armys. Instead of playing 100 full games where you need to build buildings, in unit tester you can create an army you want and put it against an army you want to, then see how it plays out 100 times without building anything you would need to create these forces. Obviously AI is not as good as other player controlling opposing force.
What kind of other modes you use to train ? I would like to find a mode where I could over and over practice early game aggression, for example there would be all the buildings built I just need to build specific units and attack.
|
Try SALT mod, it allows you to do exactly what you asked, restart without restarting the game as well as resuming from a save point (all your structures built)
To play it, just try to create a game, but instead of clicking create, click "create with mod" and search for "SALT" and there you go !
|
To play at the highest level you need to have responses ready for (nearly) every possible development in a game. Those responses will vary by map, strategy and taste, but you still need to have them so deeply ingrained that no time is spent on making the necessary decisions and you can keep up with everything happening.
I think extensively testing unit compositions is too inefficient for training. It's much more important HOW you get your stuff than WHAT you get. Any adjustments to super-late game compositions will take time, but then again it's much more helpful to understand the dynamic of units than what counters them, i.e out-maneuver Broodlords instead of building the anti-air of choice.
The aspects you're interested in are certainly important at the highest level, but in my opinion it's more worthwhile to master the concepts that allow you to get to GM first.
|
You need to master 6 skills in order to be great at Starcraft.
- Macro
- Micro
- Multitasking
- Decision making
- Preparation
- Composure
The first four you get simply by playing an extreme amount of ladder games every day. Over time your skills will automatically increase.
The last two you will get by participating in a lot of tournaments, trying to prepare for your opponents strategies and play style and practicing remaining calm in tournament settings.
Micro, macro and multitask maps can be good but it is typically much more effective to just play a lot of games.
|
Multitasking doesn't really exist, you don't truly do several tasks at the same moment, it's more :
you don't take all your army in a big group but you're able to divide it,
For example you send a small group to attack the other, while your main army attack another location, you hope to attract the army to the small group when your army arrive, so you can take a good position or just snipe a base and retreat.
With experience, players know the tactic and don't send all their forces to fight the small group but just enough army while the rest of the army is directed to the opposite direction.
Also it could aslo mean : it's not because your army is fighting you'll stop expanding, teching, producing, injecting even if you don't do that at the same time, you're fast enough that in let's say 10s you have done all of them.
|
I don't think it is something you can do alone. Take a look at how Korean pro teams trained, especially during the prime BW era.
If you really want to be top of the line, you need to have people who do stuff for you. Taking care of physical needs is one of those things (Every minute you spend buying groceries, cleaning house etc... is a minute you don't spend becoming better at SC.
But you also need people who just theorycraft builds, and you especially need competent players against whom you can train in a very specific way. Ladder is not actually a very good way of training. You need a guy to whom you can go and say "Ok, i want to practice a PvT build, can you play T using this build on that map against me for 30 games today?" And then you need a guy who goes over those 30 replays and tells you "at this point you could be weak against that, at that point x could hurt you a lot." Of course, you can do all of these things on your own, but it is going to be less efficient.
|
If possible, besides talking about the ways to become the best starcraft 2 players, you could share game mods that help to train please.
|
|
|
On February 11 2018 21:53 Saechiis wrote:
What race do you play?
I play with the Zerg.
|
Ok, I never practiced arcade games, but I did play a lot of custom games vs AI to streamline my builds. What helped me get better was to try and get 3 bases with queens, 40+ drones and lingspeed done at 4 minutes. This way you have a goal to aim for and a measure to compare your replays to.
Any time you play a ZvT/ ZvP and you don't have at least 40 drones at 4 minutes (when no-one is rushing), you'll know you can do better and need to work on your injects/ queen timings/ larva spending/ overlord timings.
|
On February 11 2018 23:01 Saechiis wrote:
Ok, I never practiced arcade games, but I did play a lot of custom games vs AI to streamline my builds. What helped me get better was to try and get 3 bases with queens, 40+ drones and lingspeed done at 4 minutes. This way you have a goal to aim for and a measure to compare your replays to.
Any time you play a ZvT/ ZvP and you don't have at least 40 drones at 4 minutes (when no-one is rushing), you'll know you can do better and need to work on your injects/ queen timings/ larva spending/ overlord timings.
I personally always go for full cheese zergling (+ 2nd hatchery and evolution chamber build due to resource surplus) attack on the enemy with ALL-MOST non stop ling with speed production. How I do it ?
First Step - I first attack and see if I can get in to opponents base, if I can do it its all over, the lings will pile up eventually, obviously with some micro I have to avoid direct combat if necessary against numerous enemy harvesters, besides in his base I harass the workers so much he can barely produce and must send harvesters to try tackle my lings loosing hundreds on minerals while my lings just come 2-4-6 at a time without stopping to reinforce the ones who are already in his base. When enough lings show up I will overpower harvesters and without them opponent cant defend or produce.
Second Step - If (most likely) the opponent manages to have a defense to his base I do instant drop to a side that is protected by fog of war and do super - over 9000 - harvester massacre in his main natural with approximately 3-4 ``DropoLords`` full with lings, that many lings overpower mineral line in seconds. At the same time if possible, I attack the main entrance with banelings and some additional lings. I may win then, I may not win after the drop, but my opponent is without some buildings and almost no harvesters, unless he has some surviving ones in his second base if it was created and protected.
Before the drop I focus on lings more than on drones after the drop I focus more on drones and roaches or banelings, some static defense to counter hellions, adepts, stalkers etc. whatever he may send at me. By the time I am done with this cheesy play my opponent was MOST LIKELY BUT NOT SUBJECT TO unable to open second base for long enough and has no harvesters in his main base mineral line, without some buildings too and damaged force.If I know he can further defend, then its macro game where I fill up 2 hatcheries, build 3rd have a force and attack him further based on mid-game army composition.
However the execution must be flawless as much as possible without mistakes. For example, things like supply depots or pylons on natural base corners will tell my opponent about the drop and he will be waiting for me on the corner, or my opponent just outplays me somehow with lets say his own cheese. Another thing to note is that when played right the opponent may be trappet in his main base but be warned he will focus on building reasonalby strong one time rush force with most likely 1/1 upgrades and will try to win against you and take out your bases, in this case its important to not only focus solely on economy after succesful drop, besides that your enemys counter attack may come quite quick with lets say 8-12 hellions just after the drop which is bad, real bad and must be countered so just as mentioned before dont focus full 100 percent on economy keep counter attack in mind.
Thats why I enjoy playing zerg the most, easy early game victory.
![[image loading]](https://i.imgur.com/NJDbuPb.jpg)
|
I watched a couple of documentaries about the greatest athletes of all time, most notably "winning" on netflix.
Nadia Comaneci was a gymnast who got 3 gold medals, 1 silver, 1 bronze and as the first gymnast ever a perfect 10.0/10.0 score (and in SEVEN events) at her first olympic games.
What I gather from interviews with her, her trainer and fellow athletes: what set her apart was the attention to detail. To not get bored doing the most basic hand movements a millionth time and paying attention as if it was the first time.
Edwin Moses does hurdling and he broke down every step he does in a race to optimize for milliseconds.
So I'd say, maybe training stuff isolated is not the best/only or even right approach. If you wanna play aggresive, play the next 1000 games very aggressiv, and see where you can shave off milliseconds. Where you can save clicks when macroing, moving, how to stay alert all the time. If you need better late game understanding and army controll, play a 1000 games and try to force the late game.
Splitting your workers at the start of a game, placing the spawning pool in the right position for the map and strategy you want to go for, rallying your queen when spawning to the perfect spot, over and over, faster and faster, shave off millisecons, then think how to use these for something, then make that even faster and repeat.
|
On February 11 2018 17:19 okright wrote:
Besides playing a lot of games on ladder, unranked, etc. What kind of arcade games or unique mode games you would recommend to practice ?
For example, there is arcade mode called ``unit tester`` , there you can perfect different unit compositions, find out how casting affects armys. Instead of playing 100 full games where you need to build buildings, in unit tester you can create an army you want and put it against an army you want to, then see how it plays out 100 times without building anything you would need to create these forces. Obviously AI is not as good as other player controlling opposing force.
What kind of other modes you use to train ? I would like to find a mode where I could over and over practice early game aggression, for example there would be all the buildings built I just need to build specific units and attack.
I'm going to let you in on a little sports psychology and coaching secret:
Grinding on ladder is a complete waste of your time if your only focus is to improve. It is like going to the golf course everyday with friends. Sure, you'll very slowly and incrementally learn things overtime and improve, but the main focus will always be to beat your friends, and that will hinder you. Eventually your improvement will slow to a crawl and you'll be so frustrated you'll hire a coach. The coach will then make you unlearn many of the bad routines you built up and reinforced overtime that got you easy wins. He will literally reshape you before you can actually become good.
That process of unlearning is not only wasteful and unnecessary, it can take a long time. Tiger Woods went through it with his golf swing at a time he was one of the best in the world; but the end result made him better.
So if you're entirely focused on improving, you need to focus on playing like a professional player. Shape yourself into the player you want to be now, don't learn and reinforce bad routines and mechanics in the first place. Like most games in the world, Starcraft 2 is a game where the real opponent is yourself. Laddering is fun, but becoming better often isn't. You'll need to schedule multiple times a week for practice and watching over replays. And you shouldn't laddering at all at first.
So how do you get better than during that practice time? Take a single early aggressive build that ends the game before 8 minutes and is used in Grandmaster tournament play (1 or 2 base all in, but nothing super cheesy like a Proxy 2 Rax/Gate or 12 Pool) for whatever race you play, and learn it to a Grandmaster level. Get replays of the build being done by Grandmasters, watch them and write down the timings for everything. Note every time a worker is built, every time a building comes down, everytime they scout, the exact moment they attack, literally everything.
Then go practice that by yourself or against the AI in a custom game. If you're 2 seconds too late on getting that structure down, you're too slow, if you have 1 less worker at X time, you're too slow. Do it again. Keep drilling that not until you get right, but until you can't get it wrong.
I fear not the man who has practiced 10,000 kicks once, but I fear the man who has practiced one kick 10,000 times. - Bruce Lee
Once you can't get it wrong alone, now practice it versus human opponents on the ladder until you can't get it wrong. It doesn't matter if you win or lose, it only matters if you do the build well or not (and if you do it well, you'll usually crush people). Keep doing that build until it becomes muscle memory. You won't even need to think about it, you'll become a machine. That is what gives people high effective APM. The best players never think about what they need to do, they already know, they just execute. Players that have to think in game because they don't know what to do, lose.
Overtime versus humans you'll learn how the build you're executing has to deviate in order to hold an early attack, or get around a certain defense. And you'll also know when you just have no chance of winning, when your opponent has countered your build. You'll master the build, and you'll be a masters level player. You'll even be able to point out mistakes professional players are making with that build. Then, you'll have a base and you'll understand the work it takes to be at that level. It is important to pick an aggressive build because the limits the time of the game and number of variations. You don't want to be having to learning everything at once.
We want to drill deep, not wide. A gold or platinum player probably has 5-10 gold or platinum level builds, and they all suck. But you'll have 1 masters build, which will make you a masters player. Then you learn another more advanced build that is similar or is a deviation off the one you already know. It will be easier this time because you know how tight the build order timings need to be, when you need to scout, ect.. And you drill that until you can't get it wrong and learn all the responses and deviations you'll need to know for that build. But always practice the first build you learned from time to time too, so you never lose it.
Then, keep learning builds as you see fit to build your game. That is how I went from high low Diamond to high Masters in 100 ladder games or so (months spent practicing builds previously). I was playing versus GM's at my peak, and taking out well known professionals in tournaments.
Winning in any sport is done on the practice field. When two people or teams show up for an event, the outcome has already been determined. If you deviate from you what you spent months training, you're almost guaranteed to lose. So if what you drilled doesn't beat what they drilled, you lose. If it does, you win. There should be no pressure. There never is for the prepared. Either your game is good enough, or it isn't. Over prepare, so you don't underachieve.
Good luck.
|
On February 11 2018 19:20 MockHamill wrote:
You need to master 6 skills in order to be great at Starcraft.
- Macro
- Micro
- Multitasking
- Decision making
- Preparation
- Composure
The first four you get simply by playing an extreme amount of ladder games every day. Over time your skills will automatically increase.
The last two you will get by participating in a lot of tournaments, trying to prepare for your opponents strategies and play style and practicing remaining calm in tournament settings.
Micro, macro and multitask maps can be good but it is typically much more effective to just play a lot of games.
i don't think these automatically increase. i've seen people with thousands more games than myself who won't do the most basic things.
like someone mentioned earlier, you don't truly multitask, but you can be thinking about multiple different things in anticipation. that is preparation and know-how.
i like to call it the priority system, and the biggest mistake i see any player ever make is controlling their army too much.
i'm not even talking about whether the control is good or not, i mean that there are so many things getting left unattended to while you're right-clicking away with your army.
some semi-allins become all-ins, and some regular builds become semi-allin all because the player is forgetting to make workers or buildings behind it. it is not as if they can't afford it, is that their attention is not divided.
your best tool in my opinion is to gather good replays; it's like a dog needing to watch another lift their leg to pee, to learn to do it themselves.
the goal in mind is to always do more with less. defend with less, deal more damage with less, and then you'll have more of everything.
you say you wanna practice early aggression? well luck is in store that early aggression ends in a fast game. so just keep playing and making mental notes.
|
Play the game. Watch the replays. Find the mistakes.
|
Literally best way to reliably improve is just playing the game with live opponents and watching replays. Of course you need to be very focused each game you play. Games against AI or Arcades won't teach you that much, if anything tbh...
Also I think this is pretty relevant to the thread:
+ Show Spoiler +
|
Replays are a pretty big tool, as is understanding how to analyze them, learn from them so that you can grow as a player. The unit tester is definitely used a lot and I would say is another key, though very small, part of the understanding that professional players have of the game.
Everything else revolves around playing on ladder a lot and practicing against a wide variety of players and playstyles (at the highest levels of play, of course) and, to various degrees, having solid regular practice partners.
Grinding games I argue is extremely important. However, in order for your grinding to matter, you have to be smart about it, you have to take it seriously. If you're grinding on the ladder using one to control your mouse and the other to control how many Cheetos are going into your mouth and you're only doing one strategy, then your grinding is most likely not going to get you anywhere unless you are absurdly extraordinary with one hand and that strategy.
Another "tool", briefly touched on, is other players and the community you're participating in the game with. If you surround yourself with competitive, positive-minded, intelligent and active people, that environment can be incredibly good for you. Likewise, who your opponents tend to be online can matter a lot, as can who you practice with and how often.
Quirky custom games or modes that are usually far from balanced but incredibly fun in small amounts are also very important imo, in any game. Or, if it isn't in the game itself, having a hobby outside the game that makes you happy can massively influence you. Raynor Party, Pudge Wars, that one Warcraft 3 mod where you all take turns running from a giant monster trying to make everyone else die first, I think those are great to play no matter what your skill level or goals as a casual player or serious competitor. Or something like Smash: Melee (or any Smash, they're all fine I guess) once a month or two, having the knowledge that taking breaks is a good thing is a powerful tool.
But, strictly talking about in-game skill, mechanics, it's doing the thing that you are supposed to do in competition over and over as best and intelligently as you can as often as possible. If it's hockey, play and watch and analyze a lot of hockey. If it's Mortal Kombat, the dojo is a great tool, but you still have to play a lot of 1-on-1 Mortal Kombat. If it's mounted archery, shoot a lot of arrows at a lot of targets while riding a moving horse. For StarCraft, same thing and even more so imo. Good equipment can help, a few other tools/mods can help for specific things for specific players, but playing full games, competitively, start to finish, is everything.
|
|
|
|
|
|
EPT
