|
Final Edit: I've come to realize that this analysis could go on forever, so the below benchmarks while easy to reach with some practice should be understood as a setting of pace. Reaching these benchmarks requires quick execution, and to fully understand this way is to launch the pace set by these benchmarks into the game till its completion.
/
As you progress in SC2 skill whether you know it or not you will be able to hit certain benchmarks. A benchmark is related to the game state versus time. Hitting benchmarks is all about playing at proper speed. A player training with benchmarks can eventually begin employing properly timed attacks. A timing attack is often effective when your opponent is too slow to hit a benchmark.
I'm working on converting to relative timings because scout timings affect benchmark dramatically.
-----------Terran Benchmarks 
+ Show Spoiler +Benchmark #1+ Show Spoiler +Goal: The second SCV is queued before the first SCV is finished.
Execution:
1. Queue first SCV.
2. Send initial six SCVs to mine.
3. Set Command Center rally point.
4. Queue second worker ASAP.
Goal: 1st Supply Depot finishes before 11th SCV finishes.
Execution:
1. Benchmark #1
2. Continuous SCV production.
3. Build Supply Depot when 10th SCV has four seconds of training or less.
Benchmark #3.0
Goal: 1st Gas finishes before 2:24GT. 1st Barracks and 15th SCV finished before 2:35GT.
Execution:
1. Benchmark #2
2. Continuous SCV production.
3. Barracks ASAP.
4. Refinery ASAP.
Benchmark #3.1
Goal: 1st Marine finishes, and Command Center starts before 3:01GT.
Execution:
1. Benchmark #2
2. Continuous SCV production.
3. Barracks ASAP.
4. Orbital Command ASAP.
5. Continuous Marine production.
6. Command Center ASAP.
7. Depot ASAP.
-----------Protoss Benchmarks 
+ Show Spoiler +Note: Each Chrono Boost knocks 10 seconds off of 30 seconds of production.Benchmark #1+ Show Spoiler +Goal: The second Probe is queued before the first Probe is finished.
Execution:
1. Queue first Probe.
2. Send initial six Probe to mine.
3. Set Nexus rally point.
4. Queue second Probe ASAP.
Goal: 1st Pylon and 10th Probe finish at 1:12GT.
Execution:
1. Benchmark #1
2. Continuous Probe production up to 9th Probe.
3. Pylon ASAP.
4. Probe ASAP.
5. 1st Chrono Boost on Probe.
Benchmark #3.0
Goal: 1st Gas before 2:27GT. 2nd Pylon, 1st Gateway before 2:47GT. Add 8s if scout after Pylon.
Script:
1. Benchmark #2
2. Continuous Probe production.
3. 2nd Chrono Boost when 12th Probe has 13 seconds of training.
4. Gateway (1:42).
5. Assimilator.
6. 3rd Chrono Boost when 15th Probe has 13 seconds of training.
7. Pylon (Gateway[40]).
Benchmark #3.1
Goal: 1st Gas before 2:22GT. 2nd Pylon and 1st Gateway before 2:37GT.
Exeuction:
1. Benchmark #2
2. Continuous Probe production.
3. Gateway (1:32).
4. Assimilator (1:52).
5. Pylon (2:12).
Benchmark #4.0+ Show Spoiler +Goal: Against Terran, Expand after three Gateways at N:NNGT.
testing:
1gate > zealot 38> stalker 42 > sentry 37> sentry37 = 154s = +2:34GT 1CHRONO = 2:28
warp research: 160s =2:40 1 chrono = 2:34, 2 chrono = 2:28
-Warp finish 6:00GT (2 Chronos total)
-gateways ready at 6:00GT.
Execution:
0. Backwater Gulch.
1. Benchmark #3.0, scout with 10th Probe.
2. Cybernetics Core (2:48).
3. Zealot
N. 2 Gates (4:55)
N. Stalker (3:38)
TBC: Go contain with a few units, while expo, if overaggro comes, can always cancel nexus, but the goal is to get it out asap. use sentry contain if cannot break. Prepare for cloackshees if sense tech.
Goal: Against Protoss have four Gateways and Warp Gate technology and five Pylons finished by 5:20GT.
Execution:
1. Benchmark #3.1
2. Continuous Probe production.
3. Cybernetics Core ASAP (2:37GT).
4. Zealot ASAP, rally to map awareness.
5. Warp Gate ASAP.
6. 3rd Chrono Boost Warp Gate technology.
7. Stalker ASAP.
8. Stop Probe production after 20 Probes.
8. 4th Chrono Boost when Warp Gate research is at 30 seconds.
9. Three more Gate Ways ASAP.
10. 2nd Stalker ASAP.
11. 5th Chrono Boost when Warp Gate research is at 60 seconds.
12. Proxy Pylon.
13. 6th Crhono Boost when Warp Gate is at 90 seconds.
14. Two proxy Pylons (5:05GT).
15. 7th Chrono Boost when Warp Gate is at 130 seconds.
-----------Zerg Benchmarks 
+ Show Spoiler +Benchmark #1 (with Replay!) + Show Spoiler +Goal: The second Drone is queued before the first Drone is finished. Replay: http://www.mediafire.com/?855xb35zuaxwdai (watch the production tab the drones in production will show the number 2.) Execution: 1. Queue first Drone. 2. Send initial six Drones to mine. 3. Send Overlord to first destination. 4. Set Hatchery's rally points. 5. Select Larvae closest to open mineral patch 6. Queue 2nd Drone and set its rally. Note: It's much easier to check speed with either Terran or Protoss. While practicing even as a Zerg, it's easier to see how fast the 2nd worker is queued with a Nexus or Command Center. Goal: 2nd Overlord and 10th Drone finish before 1:13GT.
Execution:
1. Benchmark #1
2. Continuous Drone production up to 9th Drone.
3. Select Larvae closest to 2nd Overlord's desired scout location.
4. Overlord ASAP, and rally to desired location
5. Drone ASAP
Reasoning: Increase awareness with faster 2nd Overlord. In Losira v. Haypro on Metalopolis at MLG, Haypro's expansion is two seconds faster. Also easier to manage mineral stacking sense the Drones are coming out at a more steady rate. Goal: Natural expansion completes at 3:41GT with Pool and Gas. Execution: 0. Xel Naga Caverns with close Larvae. 1. Benchmark #2 2. Continuous Drones to 14. 3. Hatchery at natural expansion ASAP (2:01GT). 4. Spawning Pool ASAP (2:22GT). 5. Two Drones. 6. Gas ASAP (2:38GT). Reasoning: From Liquipedia. Hatchery First. Goal: Have 60 Drones, 4 Queens, 66 Roaches, Metabolic Boost, Glial Reconstitution, Tunneling Claws, Burrow, +2 Missile Attacks, and Pneumatized Carapace finished by 13:00GT.
Execution:
0. Xel Naga Caverns with close Larvae.
1. Benchmark #3.
2. Overlord after 15 supply.
3. 3 Drones to gas.
4. 2 Drone
5. Queen (3:27).
6. Zergling (3:27).
N. Queen (4:07)
N. Evolution Chamber (7:35)
Note: This benchmark contains a unique task which requires eye to top right at or before 88 Gas, as well as intricate micromanagement of the Drones mining from the Extractor. Start with the Drone that brings Gas 92 and send it to mine minerals. Continue with the next Drone you can use shift + click mineral right as they pop out so you can continue with another task sooner.
I'll update with more benchmarks as time permits.
If you happen to hit any of these benchmarks sooner than the time I've posted, feel free to send me the replay and I'll update.
Note: Game Speed: Faster
Note: If your internet connection slows unit response time, try to hit these benchmarks in offline mode first.
|
As zerg I send my drones to minerals first and build a new drone afterwards. This should provide me slightly more minerals.
|
I thought for zerg it was more efficient to first send the 6 drones and then queue the 7th drone, contrary to the other two races.
|
I always worker first, tri split workers, then if zerg send out olord.
|
I just watched an Idra replay and he builds a drone before he sends.
And actually if you think about it if it's faster for Terran and Protoss, it will be faster for Zerg as well.
Edit: Larvae are not that magical!
|
Wouldn't you include a time that you want your benchmark completed by, thus you can actually see if you're improving or not.
+ Show Spoiler [example] +Bench Mark #1 Goal game state: The second worker is queued before the first worker is finished. Execution: 1. Queue first worker. 2. Send initial six workers to mine. 3. Set Command Center rally point. 4. Queue second worker ASAP. Time: 00:03
|
On July 28 2011 17:29 Komsa wrote:
I just watched an Idra replay and he builds a drone before he sends.
And actually if you think about it if it's faster for Terran and Protoss, it will be faster for Zerg as well.
Edit: Larvae are not that magical!
I'd imagine that for Zerg, you want to send your drones first, and then build the 7th, and the opposite if you're Terran/Protoss.
Assume it takes 1 unit of time to send your drones off to mine, and 1 unit of time to start building a drone. If you build the 7th first, it will pop 1 unit of time earlier, giving you 1-unit-drone worth of extra mining. However, during the 1st unit of time, your initial 6 drones will be sitting idle. If you had sent them off to mine instead, they will begin mining 1 unit of time earlier. Thus giving you 6-unit-drone worth of extra mining. This might even allow you to start building your 8th earlier. Either way, your 8th larva isn't bottlenecked by your 7th drone morphing.
For Terran/Protoss though, your 8th worker is indeed bottlenecked by your time-to-completion for your 7th. By sending your workers first then building your 7th: You're getting 6-unit-worker extra mining with your initial workers, but every single consecutive worker will be delayed by 1 unit time. So if you were to build 20 workers, each of them will be delayed by 1 unit time, thus losing you 20-unit-worker of mining. Hence the difference between Zerg & others.
|
On July 28 2011 17:58 whacks wrote:Show nested quote +On July 28 2011 17:29 Komsa wrote:
I just watched an Idra replay and he builds a drone before he sends.
And actually if you think about it if it's faster for Terran and Protoss, it will be faster for Zerg as well.
Edit: Larvae are not that magical! I'd imagine that for Zerg, you want to send your drones first, and then build the 7th, and the opposite if you're Terran/Protoss. Assume it takes 1 unit of time to send your drones off to mine, and 1 unit of time to start building a drone. If you build the 7th first, it will pop 1 unit of time earlier, giving you 1-unit-drone worth of extra mining. However, during the 1st unit of time, your initial 6 drones will be sitting idle. If you had sent them off to mine instead, they will begin mining 1 unit of time earlier. Thus giving you 6-unit-drone worth of extra mining. This might even allow you to start building your 8th earlier. Either way, your 8th larva isn't bottlenecked by your 7th drone morphing.
Golly gee wilikers what a great theory. Contact Idra immediately!
Benchmark #2 for Terran and Protoss is up! Still deciding on what Zerg's 2nd benchmark should be.
On July 28 2011 17:48 AnxiousHippo wrote:Wouldn't you include a time that you want your benchmark completed by, thus you can actually see if you're improving or not. + Show Spoiler [example] +Bench Mark #1 Goal game state: The second worker is queued before the first worker is finished. Execution: 1. Queue first worker. 2. Send initial six workers to mine. 3. Set Command Center rally point. 4. Queue second worker ASAP. Time: 00:03
Where the timing isn't implicit I'll add a Game Time (GT).
|
This is actually a awesome idea. Low level players don't realize just how bad their mechanics are and comparing to strict timings is a great idea.
To all of the protoss player that I coach, I teach them the standard 4 gate, then tell them that I won't be coaching them anymore until they can consistently get 6 stalkers + 1 zealot at under 5:45 vs the AI. (consistently = 3 times in a row while I am obsing)
I believe execution is the primary skill to have when learning to play starcraft.
|
For zerg there is a difference in sending drones/queueing another one. Zerg is different in building drones, because terran or protoss is bind by building time (you can build one worker per nexus/command center) but zerg can boost drones by big batch not like the others. So you are bound by harvest time, not by build time, when you want to eco +. splitting first is more effective.
|
On July 28 2011 18:31 Komsa wrote:Show nested quote +On July 28 2011 17:58 whacks wrote:On July 28 2011 17:29 Komsa wrote:
I just watched an Idra replay and he builds a drone before he sends.
And actually if you think about it if it's faster for Terran and Protoss, it will be faster for Zerg as well.
Edit: Larvae are not that magical! I'd imagine that for Zerg, you want to send your drones first, and then build the 7th, and the opposite if you're Terran/Protoss. Assume it takes 1 unit of time to send your drones off to mine, and 1 unit of time to start building a drone. If you build the 7th first, it will pop 1 unit of time earlier, giving you 1-unit-drone worth of extra mining. However, during the 1st unit of time, your initial 6 drones will be sitting idle. If you had sent them off to mine instead, they will begin mining 1 unit of time earlier. Thus giving you 6-unit-drone worth of extra mining. This might even allow you to start building your 8th earlier. Either way, your 8th larva isn't bottlenecked by your 7th drone morphing. Golly gee wilikers what a great theory. Contact Idra immediately!
Not sure why you're being snarky... I'm just responding to a discussion you started.
Anyway, looking forward to your benchmark #2
|
I could see this thread going places and being really helpful. Good luck and keep working at it.
|
For Zerg I am pretty sure it's ideal to go send-build-split with your initial workers (although I understand not everybody thinks they can do this quickly, a little practice makes it fairly easy). In broodwar this was my default for all races
For the other races I always build-send-split because the 1st worker comes out ASAP and the 2nd worker is still queued up for a second or so (and hence there would be no benefit to having 50 minerals earlier).
EDIT: The reason I say send-build-split with zerg and not send-split-build is because that actually just takes too long.
|
Assume it takes 1 unit of time to send your drones off to mine, and 1 unit of time to start building a drone. I build the worker first and then I send. The idea is that I don't need to look at anything to build the drone. I place my cursor in the middle of the screen as the game loads, and click+s+d blindly as soon as the game finishes loading. During that time, I will recognize where my drones are (which is variable) and move my cursor to where I want to start the selection box.
The net result is that both actions finish faster. I can do the blind automated action (selecting hatchery + build drone) immediately while I figure out how I want to do the non-automated action (selecting the drones, which depends on spawn position).
|
Russian Federation63 Posts
A very advanced benchmark would be an ability to control more than one group of units simultaneously in different parts of the map.
Best i saw so far was ~3-4 of IMMVP.
|
On July 29 2011 03:22 Anfi wrote:
A very advanced benchmark would be an ability to control more than one group of units simultaneously in different parts of the map.
Best i saw so far was ~3-4 of IMMVP.
The situation you describe is not a benchmark as I defined it in the OP. To be a benchmark you must describe a specific game state at a precise moment in time.
Your situation describes a player manipulating the game state over many moments. I do agree however that one should also gauge their ability to handle an ever increasing number of groups if they have a practice partner.
I've reworked the format of the OP, I think this will be easier to navigate.
I also added Zerg Benchmark #2 !! Give it a shot.
|
I just wanted to point out you should always build your worker first and then send your workers. mining first and building second as p/t actually slows your worker production down because if split correctly there is no delay between 1st and 2nd worker building.
as a zerg it is doubly important because the faster you build your first worker the faster the cooldown for the next larvae pop will be.
|
I wouldn't consider those benchmarks to matter. Improving your drone split by 0.8 seconds isn't going to make much of a difference in your game
|
On July 29 2011 07:28 gdot5 wrote:
I wouldn't consider those benchmarks to matter. Improving your drone split by 0.8 seconds isn't going to make much of a difference in your game
Yes it is. People who care about those things also get more precision later in the game. It's the difference between a master execution and a plat execution which gets everything out 1 minute later.
|
But basic goals that you should be able to meet with your race further in into the game will really let you know how good your mechanics are. I've found goals such as "get 48 drones by 6.30" to be very motivating and helpful for both training and build order design 
|
based on these some of these comments, benchmark 1 should be changed to the following:
Know that you make your 1st drone before sending workers to mine (same as other races). Timer for the next larvae to spawn will not start until that first larvae is used.
|
|
|
On July 29 2011 03:22 Anfi wrote:
A very advanced benchmark would be an ability to control more than one group of units simultaneously in different parts of the map.
Best i saw so far was ~3-4 of IMMVP.
Thats best done practicing against AI or something else. Dry runs via just making the units and then trying to control them while continuing o make macroing adjustments, and casting at air spots, controlling unit groups back and forth, etc.
Its just like anything else in life. you get the mechanics down apart, do each thing on its own, then string it all together at the end to consolidate your learning into a complete whole that you win with.
On July 28 2011 18:42 Geiko wrote:
This is actually a awesome idea. Low level players don't realize just how bad their mechanics are and comparing to strict timings is a great idea.
To all of the protoss player that I coach, I teach them the standard 4 gate, then tell them that I won't be coaching them anymore until they can consistently get 6 stalkers + 1 zealot at under 5:45 vs the AI. (consistently = 3 times in a row while I am obsing)
I believe execution is the primary skill to have when learning to play starcraft.
Yeah, the biggest problem forpeople who play 500 games and still bronze is macro, or in other words, the ability to expand, produce more units/control unit production multipliers such as Inject or Chrono. I had a 500 win bronze zerg ask me for coaching but onlly had 15 minutes, so I taught him 14 pool 14 hatch as a standard, and he still came out with 2000 unused minerals by 8 minutes. I had to tell him to be more proactive with injects and constantly using larvae. I got him to extractor too, and I never has as many minerals as he did, because I just do so much macroing intuitively. It was a little effort to try to intellectualize the process I use in order to explain to him how to keep his macro up.
The truth is that the key to getting into plat is INDEED macro. Thats how I got there from playing SC:bw so much and having 10 years of macro under my belt. (fyi i was plat for a long time when I was supposed to be higher league due to lock).
|
|
|
While i expect this thread could be good, the first 2 benchmarks for T and P at least are retarded. It makes fuck all difference if you are say 2 seconds behind. Its poor yes, but not game changing in any way.
Something like get X Stalkers and Y zealots by Z Z time would be good. This is pointless
EDIT: haha at the unintended emoticon
|
These are silly benchmarks for getting better. You could tell anyone who has never played a video game before to make an scv at a certain time and a supply depot at a certain time. It's remarkably simple. I was expecting these benchmarks to be more like never miss a single worker throughout the game until you reach your target number, always have your money under 100/100 unless you're saving up for something in particular, etc. Those are things that you can actually shoot for without achieving them once you realize you're supposed to do them.
|
On July 29 2011 07:35 Geiko wrote:Show nested quote +On July 29 2011 07:28 gdot5 wrote:
I wouldn't consider those benchmarks to matter. Improving your drone split by 0.8 seconds isn't going to make much of a difference in your game Yes it is. People who care about those things also get more precision later in the game. It's the difference between a master execution and a plat execution which gets everything out 1 minute later.
No it's not. People put way too much stock in these things. You look at someone like July who goes 10 drones then overlord, doesn't use extractor trick, then builds all 3 drones when overlord pops... it seems like he should be way behind, yet he's still dominating.
Not saying it doesn't help, but personally I'd rather spend two hours of my time trying to improve other parts of my game.
|
On July 29 2011 09:49 saus wrote:
These are silly benchmarks for getting better. You could tell anyone who has never played a video game before to make an scv at a certain time and a supply depot at a certain time. It's remarkably simple. I was expecting these benchmarks to be more like never miss a single worker throughout the game until you reach your target number, always have your money under 100/100 unless you're saving up for something in particular, etc. Those are things that you can actually shoot for without achieving them once you realize you're supposed to do them. I completely agree. As a Zerg player, I'm not really concerned with how quickly I build my first drone and split my workers. Sure, it might make 1-2 seconds difference in the early game, but it's significance is basically nil.
More meaningful benchmarks for Zerg, if you are looking to improve macro, would be something like these:
1) Each queen has < 50 energy at the 10 minute mark.
2) At no point in time does a hatchery have 3 idle larvae sitting around (unless you are maxed or out of money) before the 10 minute mark. This would be due to either forgetting to build stuff or getting supply blocked.
I find it silly to be focusing so hard on the first 30-60 seconds of the game, when many masters players (including myself) have ridiculously sub-optimal injections, which matter 100x more than your drone timing at the beginning of the game. Every 25 missed seconds of injection is 4 larvae, which could be completely game changing.
|
Don't see how the timings can be precise. You got change your depot and pylon placement depending on the map and the match-up.
|
On July 29 2011 10:13 FairForever wrote:Show nested quote +On July 29 2011 07:35 Geiko wrote:On July 29 2011 07:28 gdot5 wrote:
I wouldn't consider those benchmarks to matter. Improving your drone split by 0.8 seconds isn't going to make much of a difference in your game Yes it is. People who care about those things also get more precision later in the game. It's the difference between a master execution and a plat execution which gets everything out 1 minute later. No it's not. People put way too much stock in these things. You look at someone like July who goes 10 drones then overlord, doesn't use extractor trick, then builds all 3 drones when overlord pops... it seems like he should be way behind, yet he's still dominating. Not saying it doesn't help, but personally I'd rather spend two hours of my time trying to improve other parts of my game.
to be fair, I have done 10 OL 13 drone a lot, but I did 9 Ol once while watching the clock.
At 10 ol, the Ol comes out at 57-58 seconds. This is actually regardless of whether u float up to 70 minerals during drone 8 or 9 or not. this means, with OL being a 25 second build, that your next three drones start at 1:22.
At 9 ol, your next two drones after 10 supply block start at ~1:10. So 2 more drones sooner, or three more drones 10 seconds later.
|
On July 29 2011 07:35 Geiko wrote:Show nested quote +On July 29 2011 07:28 gdot5 wrote:
I wouldn't consider those benchmarks to matter. Improving your drone split by 0.8 seconds isn't going to make much of a difference in your game Yes it is. People who care about those things also get more precision later in the game. It's the difference between a master execution and a plat execution which gets everything out 1 minute later.
I disagree again. "gets everything out 1 minute later"
Although I do agree that a master level timing attack can come out a full minute before a platinum league one, it is not because they split their drones 0.8 seconds faster.
That would make their timing come out 0.8 seconds faster, not a minute
|
Speed.
Speed is paramount if you can shave one second off of your build every twenty seconds, you will be ahead of an opponent who is lazy. Economy of motion with regards to thinking, mouse movement, and timings; it's all to get a timing edge on the opponent.
The first timing (benchmark) is just as important as the 27th timing. These benchmarks set a foundation for perfect economic play.
To hit these benchmarks you must be focused and your coordination and hand speed must be dialed in. If your hands are moving fast enough to hit these timings they'll be moving fast enough to hit that inject timing while micro'ing.
These are the basic first baby steps, to neglect these initial benchmarks is to accept a shaky foundation. I say baby steps, because it's easy for a player to look at these and think of how insignificant they are in the grand scheme of a game.
If there is one thing I could hope for a student of the game to get out of the OP it's that every single action in SC2 should be looked at as insignificant in the grand scheme. But, each individual insignificant baby step can be scrutinized and perfected one by one. And at some point you'll be sitting across from an opponent who has scrutinized the game almost as much as you have, until you both reach benchmark #1999. Benchmark #1999 is the mark your opponent stopped his analysis. B#1999 is the moment in time when you start leading.
Why not start scrutinizing from moment one, rather than some moment nine minutes into the game?
And one direct reply to a player with perhaps little tunnel vision. He says, "Although I do agree that a master level timing attack can come out a full minute before a platinum league one, it is not because they split their drones 0.8 seconds faster."
The master's attack does come out a full minute before a plat because she splits her drones .8s faster, and makes her 1st overlord .8s faster, and makes her spawning pool 4s faster, and has her 16th drone 12s faster. Keep adding all these perfect little baby steps up. Do you understand now?
|
On July 28 2011 18:58 Pamposek wrote:
For zerg there is a difference in sending drones/queueing another one. Zerg is different in building drones, because terran or protoss is bind by building time (you can build one worker per nexus/command center) but zerg can boost drones by big batch not like the others. So you are bound by harvest time, not by build time, when you want to eco +. splitting first is more effective.
Wow, never thought of that, or knew that zergs [should?] split first. Pretty neat thanks for explaining!
|
And one direct reply to a player with perhaps little tunnel vision. He says, Show nested quote + "Although I do agree that a master level timing attack can come out a full minute before a platinum league one, it is not because they split their drones 0.8 seconds faster." The master's attack does come out a full minute before a plat because she splits her drones .8s faster, and makes her 1st overlord .8s faster, and makes her spawning pool 4s faster, and has her 16th drone 12s faster. Keep adding all these perfect little baby steps up. Do you understand now?
This is terrible logic, because that is NOT what seperates a master league player from a platinum league player. The difference is not that they are faster at splitting and doubling up drones. Not at all, in the slightest.
At MAX. The difference in time you could save on a 6gate by mastering the art of splitting probes/sending probes out to build at perfect times/doubling up on close mineral fields/laying new chronos immediatly is 6-8 seconds, that being the difference from someone playing for a week, and someone playing for a year. Although you will justify with "but 7 seconds can make all the difference," it is still a stupid, idiotic way to look at practice.
When korean pros practice, they don't sit and practice their drone splits for 8 horus, nor do they play the first 3 minutes of the game over and over to ensure they can get their 3rd ovy out by 2:54 and not 2:56. Why? Because it is an utter waste of time.
The major issues that seperate a platinum leaguer from a master leaguer are timing related, but not how well they split drones. Spending resources, and taking gases at proper times, scouting properly to see when it is just to cut corners to bring a timing out earlier, army control, and the game sense required to decide when to move out/pull back with a timing attack are what seperate platinum from masters
You call me the one with tunnel vision? Because I view the game as a broader entity? Because I figure that starcraft skill entices more than who can hit a timing attack faster than one another? Go for it. If you want to spend all your practice time splitting your probes to save 2.1 seconds on a 4gate, be my guest. But please don't advertise this method of skill progression as a legitimate one. This enhances no army control, no decision making, no general macro skills, no scouting skills, and no understanding of the flow of starcraft. This just teaches you to become like bitbybit.Prime.
As fun as he is to watch, he is by no means a good starcraft player. because he progressed his skill in the same way this thread suggests to
|
On July 29 2011 13:14 gdot5 wrote:Show nested quote +On July 29 2011 07:35 Geiko wrote:On July 29 2011 07:28 gdot5 wrote:
I wouldn't consider those benchmarks to matter. Improving your drone split by 0.8 seconds isn't going to make much of a difference in your game Yes it is. People who care about those things also get more precision later in the game. It's the difference between a master execution and a plat execution which gets everything out 1 minute later. I disagree again. "gets everything out 1 minute later" Although I do agree that a master level timing attack can come out a full minute before a platinum league one, it is not because they split their drones 0.8 seconds faster. That would make their timing come out 0.8 seconds faster, not a minute
Well it's the butterfly effect, man. Building a later depot means a later rax, which means later so on and so forth. The slight delays are compounded until they coalesce into a significant setback. This is why pro zerg players can max out much faster than some random nabs. Of course, optimizing drone splits is absolutely useless for the majority of players, who simply just need better macro, unit control, and scouting.
|
Thanks for this. I like Protoss bench #3. I never knew that.
|
I've tested different ways and orders of sending your drones out and building the 7th, then kept mining without doing anything else to see how many minerals I'd have at the 1 minute mark. The difference was so insignificant that I didn't bother to write any of it down.
Methods I've tested:
- Build drone, then send drones to mine;
- Build drone, then split drones to mine;
- Build drone, use the idle worker key to send each drone to mine individually;
- All 3 of the above, but building the 7th drone after sending the initial 6 to mine.
The method I personally prefer using is building a drone first, then split drones. My reasoning for building a drone before sending the others to mine is the consideration of larvae being a resource. If I spend that larva quickly, it will respawn earlier than when I'd do it the other way around. It'd be a different story if doing it the other way around would give you a huge advantage mineral-wise, but that's not the case.
|
All of the benchmarks displayed so far (#'s 1-3) are designed to keep the economy on track. I'm not sure where BitByBit enters into this discussion.
For people who are fast already these benchmarks can be understood and practiced in a few minutes. People with slower hands however have an applicable goal to work towards. It's the hand speed that helps the novice player improve more than a single benchmark.
The nice thing so far is that all these are standard and can be practiced every game. There is no special time to set aside, just play your normal amount of games and try to hit these benchmarks as you go.
|
On July 29 2011 16:20 gdot5 wrote:Show nested quote +And one direct reply to a player with perhaps little tunnel vision. He says, "Although I do agree that a master level timing attack can come out a full minute before a platinum league one, it is not because they split their drones 0.8 seconds faster." The master's attack does come out a full minute before a plat because she splits her drones .8s faster, and makes her 1st overlord .8s faster, and makes her spawning pool 4s faster, and has her 16th drone 12s faster. Keep adding all these perfect little baby steps up. Do you understand now? This is terrible logic, because that is NOT what seperates a master league player from a platinum league player. The difference is not that they are faster at splitting and doubling up drones. Not at all, in the slightest. At MAX. The difference in time you could save on a 6gate by mastering the art of splitting probes/sending probes out to build at perfect times/doubling up on close mineral fields/laying new chronos immediatly is 6-8 seconds, that being the difference from someone playing for a week, and someone playing for a year. Although you will justify with "but 7 seconds can make all the difference," it is still a stupid, idiotic way to look at practice. When korean pros practice, they don't sit and practice their drone splits for 8 horus, nor do they play the first 3 minutes of the game over and over to ensure they can get their 3rd ovy out by 2:54 and not 2:56. Why? Because it is an utter waste of time. The major issues that seperate a platinum leaguer from a master leaguer are timing related, but not how well they split drones. Spending resources, and taking gases at proper times, scouting properly to see when it is just to cut corners to bring a timing out earlier, army control, and the game sense required to decide when to move out/pull back with a timing attack are what seperate platinum from masters You call me the one with tunnel vision? Because I view the game as a broader entity? Because I figure that starcraft skill entices more than who can hit a timing attack faster than one another? Go for it. If you want to spend all your practice time splitting your probes to save 2.1 seconds on a 4gate, be my guest. But please don't advertise this method of skill progression as a legitimate one. This enhances no army control, no decision making, no general macro skills, no scouting skills, and no understanding of the flow of starcraft. This just teaches you to become like bitbybit.Prime. As fun as he is to watch, he is by no means a good starcraft player. because he progressed his skill in the same way this thread suggests to
you are correct. the benefits of more workers scales linearly, and as such a 3 second difference in the beginning means a 3 second difference at the end. The only issue is when worker accelerators come into play (inject,chrono). with these, you can constantly keep getting ahead of the clock if you started at :01 seconds with all your splitting and droning perfected. A multiplier such as speed boost of unit throughput/production is what causes it to be bigger. Linear production does not multiply, and thus magnify, the difference like an actual multiplier.
However, yes, the gain is so small in early to mid that this should be the LAST thing anyone learns. theres way more important things before this.
On July 29 2011 16:49 KimJongChill wrote:Show nested quote +On July 29 2011 13:14 gdot5 wrote:On July 29 2011 07:35 Geiko wrote:On July 29 2011 07:28 gdot5 wrote:
I wouldn't consider those benchmarks to matter. Improving your drone split by 0.8 seconds isn't going to make much of a difference in your game Yes it is. People who care about those things also get more precision later in the game. It's the difference between a master execution and a plat execution which gets everything out 1 minute later. I disagree again. "gets everything out 1 minute later" Although I do agree that a master level timing attack can come out a full minute before a platinum league one, it is not because they split their drones 0.8 seconds faster. That would make their timing come out 0.8 seconds faster, not a minute Well it's the butterfly effect, man. Building a later depot means a later rax, which means later so on and so forth. The slight delays are compounded until they coalesce into a significant setback. This is why pro zerg players can max out much faster than some random nabs. Of course, optimizing drone splits is absolutely useless for the majority of players, who simply just need better macro, unit control, and scouting.
the butterfly effect only occurs when the disparity becomes so great that it can lead to one person being set back significantly. in a mirror matchup, with identical player skill levels, one will lead and the other will follow, based on who droned perfectly and who was lazy about it at the very beginning, and they will continue to keep on track, one just a little behind the other, but neither actually deciding the fight as a win or loss. What comes into play is whether one player is less skilled/knowledgeable than the other and makes a mistake in countering, thus having caused that butterfly effect of putting them in the chase position instead of lead position to make tthem lose as well.
if you have to players using the exact same builds and units, the one with +1 extra unit every fight will win, if both players A move and get the exact same luck of unit firing killing the units etc etc.
If player B gets a little luckier, he will be able to take the lead from that point. If player A gets a little luckier, he wins decisively.
|
On July 29 2011 07:40 Island wrote:But basic goals that you should be able to meet with your race further in into the game will really let you know how good your mechanics are. I've found goals such as "get 48 drones by 6.30" to be very motivating and helpful for both training and build order design
High master zerg here.
I could hardly imagine the level of play at which these little benchmarks could matter. Maybe in some PvP between players at the same caliber as MC and Alicia, but these little early game timing things are really getting out of hand. I play against top masters and GM's all the time, but I don't ever feel hugely disadvantaged because I didn't get my 13hatch down 2 seconds quicker or my overlord timing was half a second off. I guess these little things can help you since one should take advantage of everything possible in a game, but considering how most players/readers on TL are not at super-high level of play, I don't see how this can be a crucially important factor.
It's almost detrimental for players at lower level to even worry about this sort of stuff. They should focus on more fundamental basics like scouting, proper macro, accomplishing game plan and building placement. Macro becomes harder to perfect as bases are added and unit count gets larger, so if anything, lower-level players should focus on maintaining acceptable macro continuously during the game. For the people who are arguing "these sort of things separate masters players from ______ players," I'm at a loss for words.
The above quote shows just how ridiculous these little "benchmarks" can get. At no point should a player ever set a goal like "get x workers at x:xx." Starcraft 2 is a reactionary game, and ESPECIALLY if you're zerg you should learn to cut drones according to the situation. Creating benchmarks under the assumption that everything will go perfectly smooth is almost useless. A simple 12/12 2rax pressure, 1 rax poke, zealot stalker harass, 2 stalker harass, ling harass will all alter your outputs. It's much more important to UNDERSTAND what you need to have at a certain point in the game WITH RESPECT TO what the enemy player has.
|
+ Show Spoiler +On July 29 2011 23:25 jhk0219 wrote:Show nested quote +On July 29 2011 07:40 Island wrote:But basic goals that you should be able to meet with your race further in into the game will really let you know how good your mechanics are. I've found goals such as "get 48 drones by 6.30" to be very motivating and helpful for both training and build order design High master zerg here. I could hardly imagine the level of play at which these little benchmarks could matter. Maybe in some PvP between players at the same caliber as MC and Alicia, but these little early game timing things are really getting out of hand. I play against top masters and GM's all the time, but I don't ever feel hugely disadvantaged because I didn't get my 13hatch down 2 seconds quicker or my overlord timing was half a second off. I guess these little things can help you since one should take advantage of everything possible in a game, but considering how most players/readers on TL are not at super-high level of play, I don't see how this can be a crucially important factor. It's almost detrimental for players at lower level to even worry about this sort of stuff. They should focus on more fundamental basics like scouting, proper macro, accomplishing game plan and building placement. Macro becomes harder to perfect as bases are added and unit count gets larger, so if anything, lower-level players should focus on maintaining acceptable macro continuously during the game. For the people who are arguing "these sort of things separate masters players from ______ players," I'm at a loss for words. The above quote shows just how ridiculous these little "benchmarks" can get. At no point should a player ever set a goal like "get x workers at x:xx." Starcraft 2 is a reactionary game, and ESPECIALLY if you're zerg you should learn to cut drones according to the situation. Creating benchmarks under the assumption that everything will go perfectly smooth is almost useless. A simple 12/12 2rax pressure, 1 rax poke, zealot stalker harass, 2 stalker harass, ling harass will all alter your outputs. It's much more important to UNDERSTAND what you need to have at a certain point in the game WITH RESPECT TO what the enemy player has.
mid master zerg here, and I fully agree with the above stated.
Mechanics are great, but they come with practice anyway. Learn WHAT to do, before you learn how to do it best. There is really not a lot of use in improving your larvasplit, if you don't know how to react to an opening.
about the "drone or mining first" thing: It doesn't have a lot of impact. You get larvablocked at 13supply anyway, and larva pops every 15sec while workers take 17seconds, so there is hardly any difference for zergs, terrans and protoss players around that time in terms of building up.
Using an early queen build (11p/18h) or an early hatch build (14-16h) is like a million times stronger than any early game mechanics you can do, to build up zerg production.
One last exapmle to show you how little a lot of progamers care about those little things:
IMNesTea, arguably the best player/zerg in the world uses mostly 10 supply OL with or even without extractor trick, while there have been a lot of calculations that have shown 9OL is straight up better than those two.
|
I think another tip would be to make sure you are sending workers to the closest mineral patches first and if you're good at it you can double up on the close mineral patches early on.
|
On July 30 2011 00:46 EvilZergling wrote:
I think another tip would be to make sure you are sending workers to the closest mineral patches first and if you're good at it you can double up on the close mineral patches early on.
It's not even that important. Personally I do it out of habit, but I've watched a lot of pro games where the workers are rallied to occupied patches and they bounce around (which is inefficient).
At a GM level I'm sure it helps a little, but even then you have to wonder if it's worth practicing. Anything below that (I'm high masters) seems to be just a waste of time. Personally I'd rather learn how to defend 15 hatch vs 11/11 rax than try and get an extra 5 - 10 minerals from an efficient split.
|
I'm starting to realize why this forum is such a filthy cesspool.
I present an objective way to train hand speed in an applicable manner.
Negative Nancys start moving their lil fingers about how it's a useless method and the way they like to train is obviously superior.
I hate visiting this forum because it's usually so sad. But I do still frequent it in the hopes some actual strategy discussion will break out. I can see now why it doesn't, too much ego and negativity.
I'm not going to work on the OP any more, this discussion is so disgusting to me.
|
On July 30 2011 04:03 Komsa wrote:
I'm starting to realize why this forum is such a filthy cesspool.
I present an objective way to train hand speed in an applicable manner.
Negative Nancys start moving their lil fingers about how it's a useless method and the way they like to train is obviously superior.
I hate visiting this forum because it's usually so sad. But I do still frequent it in the hopes some actual strategy discussion will break out. I can see now why it doesn't, too much ego and negativity.
I'm not going to work on the OP any more, this discussion is so disgusting to me.
Your "benchmarks" had nothing to do with training hand speed. If you're gonna post something that's [D][G] you better be ready for some counter arguments. Why post at all if you're expecting people to agree with everything you say? Don't post if you don't want feedback.
It's not about ego or negativity. It's about viability. You presented an idea and people disagreed. Extending this single case to say "this forum is such a filthy cesspool" goes to show how close-minded you are. If you feel like your arguments are correct, then respond to the "negative nancys" with an organized set of reasons why you are right.
If anything, people being strict about posts on TL is a good thing because it keeps the level of posts high. What's the point of having a strategy forum if everyone posts [G]'s?
|
Komsa, why do you say that 3 chronoboosts with contiguous probe production nets you one extra probe when a probe takes 17 seconds to build and each chrono boost shaves 10 seconds of build time? 3 chrono boosts will give you slightly less than 2 probes.
|
On July 29 2011 16:49 KimJongChill wrote:Show nested quote +On July 29 2011 13:14 gdot5 wrote:On July 29 2011 07:35 Geiko wrote:On July 29 2011 07:28 gdot5 wrote:
I wouldn't consider those benchmarks to matter. Improving your drone split by 0.8 seconds isn't going to make much of a difference in your game Yes it is. People who care about those things also get more precision later in the game. It's the difference between a master execution and a plat execution which gets everything out 1 minute later. I disagree again. "gets everything out 1 minute later" Although I do agree that a master level timing attack can come out a full minute before a platinum league one, it is not because they split their drones 0.8 seconds faster. That would make their timing come out 0.8 seconds faster, not a minute Well it's the butterfly effect, man. Building a later depot means a later rax, which means later so on and so forth. The slight delays are compounded until they coalesce into a significant setback. This is why pro zerg players can max out much faster than some random nabs. Of course, optimizing drone splits is absolutely useless for the majority of players, who simply just need better macro, unit control, and scouting.
No it doesn't. If I split 1 second later than you, my supply depot will be 1 second later, my rax will be 1second later, my marine will be 1 second later.
It doesn't magically get worse and worse
|
On July 30 2011 04:03 Komsa wrote:
I'm starting to realize why this forum is such a filthy cesspool.
I present an objective way to train hand speed in an applicable manner.
Negative Nancys start moving their lil fingers about how it's a useless method and the way they like to train is obviously superior.
I hate visiting this forum because it's usually so sad. But I do still frequent it in the hopes some actual strategy discussion will break out. I can see now why it doesn't, too much ego and negativity.
I'm not going to work on the OP any more, this discussion is so disgusting to me.
Why come to a forum, and post a [D] topic if you're not willing to discuss it. You presented us an idea, and we unanimously determined it to be a bad one. If you really were in search of "some actual strategy discussion," you wouldn't come in with such a narrow frame of mind.
Actually strategy discussion would be you presenting an idea, and changing it based on the feedback you receive. You presenting a bad idea, and raging when people disagree is just as productive as practising a drone split.
|
On July 30 2011 04:20 jhk0219 wrote:Show nested quote +On July 30 2011 04:03 Komsa wrote:
I'm starting to realize why this forum is such a filthy cesspool.
I present an objective way to train hand speed in an applicable manner.
Negative Nancys start moving their lil fingers about how it's a useless method and the way they like to train is obviously superior.
I hate visiting this forum because it's usually so sad. But I do still frequent it in the hopes some actual strategy discussion will break out. I can see now why it doesn't, too much ego and negativity.
I'm not going to work on the OP any more, this discussion is so disgusting to me. Your "benchmarks" had nothing to do with training hand speed. If you're gonna post something that's [D][G] you better be ready for some counter arguments. Why post at all if you're expecting people to agree with everything you say? Don't post if you don't want feedback. It's not about ego or negativity. It's about viability. You presented an idea and people disagreed. Extending this single case to say "this forum is such a filthy cesspool" goes to show how close-minded you are. If you feel like your arguments are correct, then respond to the "negative nancys" with an organized set of reasons why you are right. If anything, people being strict about posts on TL is a good thing because it keeps the level of posts high. What's the point of having a strategy forum if everyone posts [G]'s?
thats exactly what it IS about for him. He believed he had a sound idea, and having it crushed makes him feel inadequate, hence his comments about how horrible the forum and all of us are.. He might be onto something if he sees whats wrong with his idea and reworks it to do as he intends and succeed at that. we'd all agree with him then. Thats the problem with ego, it causes you to shut down and stop doing anything productive.
|
We really need a liquipedia page on the split/build first discussion. Its actually quite simple to figure out.
The tl;dr of it is that you do whichever option you can do faster. This will typically be build->split, because it involves less total mouse movement than split->build.
However, optimal play for a Zerg in particular, and P/T in certain situations, is to split->build. Theoretically, if you had infinite time to train, you'd train yourself to perform both sequences in as little time as possible.
I can re-do the math (for like the 4th time) if people are interested.
|
On July 30 2011 04:03 Komsa wrote:
I'm starting to realize why this forum is such a filthy cesspool.
I present an objective way to train hand speed in an applicable manner.
Your training of hand speed is potentially the worst method I've seen to date, maybe after 5+ hours you will have your first zealot 2 seconds faster whoop-de-freaking do.
If you want people to increase their hand speed they should do micro challenges. Not retarded shit like planting buildings down faster. Even the best of the best players don't care too much to share 1 second off planting all their shit down faster.
|
On July 31 2011 23:33 kzn wrote:
We really need a liquipedia page on the split/build first discussion. Its actually quite simple to figure out.
The tl;dr of it is that you do whichever option you can do faster. This will typically be build->split, because it involves less total mouse movement than split->build.
However, optimal play for a Zerg in particular, and P/T in certain situations, is to split->build. Theoretically, if you had infinite time to train, you'd train yourself to perform both sequences in as little time as possible.
I can re-do the math (for like the 4th time) if people are interested.
I do basic click to mine (all 6 probes), build probe, the split off second(2 probes) and third group(1probe).
|
On July 29 2011 23:25 jhk0219 wrote:
The above quote shows just how ridiculous these little "benchmarks" can get. At no point should a player ever set a goal like "get x workers at x:xx." Starcraft 2 is a reactionary game, and ESPECIALLY if you're zerg you should learn to cut drones according to the situation. Creating benchmarks under the assumption that everything will go perfectly smooth is almost useless. A simple 12/12 2rax pressure, 1 rax poke, zealot stalker harass, 2 stalker harass, ling harass will all alter your outputs. It's much more important to UNDERSTAND what you need to have at a certain point in the game WITH RESPECT TO what the enemy player has.
Yes, absolutely right. Almost never will you be allowed to just "drone away" for 7 minutes of game time totally unchallenged. But before I realized that it was possible to have 48 drones by 6.30 I played games where I had 30 drones by 7-8 minutes and thought that I was doing great. I wanted to try and shift the focus of the thread away from "doing X instead of Y saves you 0.5 seconds" and more towards "you should be able to do X in a solo game or your macro is simply not good enough". I would like the benchmarks more if they were goals you could aim to achieve regardless of which build/s you're using.
A similar and perhaps less idiotic benchmark would be "max on 60 drones/140 army by 11 minutes" or something such (totally arbitrary numbers); a benchmark you can compare yourself to in order to see how your macro is compared to the "objective best". Or for terran, "Kill 20 banelings with 10 marines" etc.
That's what I thought it would be about when I read "benchmarks" and I liked the thought, so I imagine that there are other players who would also enjoy testing themselves to get an objective read on how good they are. Like I said, I used to think 30 drones by 8 minutes was great. 
|
I have an idea. Let's start a poll about what league you're in and what benchmarks you have beaten:
Poll: What league are you in and what benchmarks have you reached?Bronze/Silver - All Benchmarks (1) 33% Gold/Plat - Benchmarks 1+2 (1) 33% Dia/Master - All Benchmarks (1) 33% Bronze/Silver - Benchmark 1 Only (0) 0% Bronze/Silver - Benchmarks 1+2 (0) 0% Gold/Plat - Benchmark 1 Only (0) 0% Gold/Plat - All Benchmarks (0) 0% Dia/Master - Benchmark 1 Only (0) 0% Dia/Master - Benchmarks 1+2 (0) 0% 3 total votes Your vote: What league are you in and what benchmarks have you reached? (Vote): Bronze/Silver - Benchmark 1 Only
(Vote): Bronze/Silver - Benchmarks 1+2
(Vote): Bronze/Silver - All Benchmarks
(Vote): Gold/Plat - Benchmark 1 Only
(Vote): Gold/Plat - Benchmarks 1+2
(Vote): Gold/Plat - All Benchmarks
(Vote): Dia/Master - Benchmark 1 Only
(Vote): Dia/Master - Benchmarks 1+2
(Vote): Dia/Master - All Benchmarks
If the author of the OP wants to add it to the OP feel free to do so 
|
Platinum Zerg here, interesting benchmarks, given this won't make us perfect, but it is worth considering, despite whatever others are saying. I mean, brushing your teeth each day is important, but probably not as important as eating and sleeping, though i doubt anyone would be in their right mind to say, Fuck brushing your teeth, being healthy is able to occur as long as you simply eat and sleep well. Seriously guys? Seriously? It's all important, given maybe pedantic in realtion to the big picture, but it is a part of the picture, nonetheless.
Well, I have always heard that splitting drones and then queueing production of drone #7 is optimal for Zerg, (while queueing for Terran and Toss then splitting is optimal for them) so the OP first made me skeptical as to the knowledge base of its author. However, one post (near top of 2nd page i think it was) had me thinking, one which claimed that the next larvae doesn't begin producing naturally until there is <3 larvae on the hatch, which I believe is in fact true (or is it <4?). Regardless, this is interesting.
However, it does seem detrimental to Zerg to drone first then split, because we seeming must click-s-d, while toss and terran just click-s or click-e... So I had a thought: What if we could click the larvae, and then d? then split? That shouldnt be any longer than it is for T or P, given we can aim our mouse to that tiny target. Then I had another thought: the hatch is always center screen, and the larvae are always directly beneath... Assuming there's no doodads right where a larvae initially spawns, and there's not much lag, shouldnt there be a pixel just south of center screen where once the game begins, we can click and be guaranteed to hit that middle larvae?
I don't know, maybe... But if so, its just click, and hit d, then go to splitting, and then send olord, then route your drones building to a min patch b4 the drone #7 is built, such that we're not wasting time right clicking minerals b4 splitting...
Am I missing something, or does this sound reasonable? I would test thismyself, but unfortunatelyI'm not able to today due to my life's schedule...
|
Good benchmarks for early game, its a little harder after your opening build orders to be able to set most benchmarks however just because of the pure situational feel of builds. All in all a good OP cant wait to see more ^^
|
On July 29 2011 02:11 Qxz wrote:Show nested quote +Assume it takes 1 unit of time to send your drones off to mine, and 1 unit of time to start building a drone. I build the worker first and then I send. The idea is that I don't need to look at anything to build the drone. I place my cursor in the middle of the screen as the game loads, and click+s+d blindly as soon as the game finishes loading. During that time, I will recognize where my drones are (which is variable) and move my cursor to where I want to start the selection box. The net result is that both actions finish faster. I can do the blind automated action (selecting hatchery + build drone) immediately while I figure out how I want to do the non-automated action (selecting the drones, which depends on spawn position).
Ctrl +F1 to select all idle workers.
Removes a variable.
Works for me as toss, but might be a bit cumbersome for you as zerg have more button presses to make that first drone.
|
To be honest, I voted Dia/Mas all benchmarks, but as zerg I've been doing all the benchmarks since gold+. I came from brood war so I guess its second nature to move as quickly as possible to get things done. However this does NOT equate to hand speed, as macro with one or two lazy right click rally points =/= a super strenuous exercise.
Now a good benchmark for zerg would be:
1. Create multiple hotkey groups of only 1 type of ground unit.
2. use these hotkeys in conjunction with regular move to get positioning and surround on protoss ball.
3. kill said protoss ball with only these forces when protoss ball has colossus.
4. Repop and recreate hotkey groups within 10 seconds of remaxing.
5.Repeat from step 2.
|
are you mineral stacking scv's to get the 1st depot finished before the 11th scv finishes? I don't have enough minerals to build it soon enough for it to finish before 11th scv.
|
|
|
|
|
|
EPT
