|
On the other hand, this is pretty worthless as you will die to any early aggression. A minor sacrifice, and you can be in position to deal with anything:
15 extractor
14 pool
17 ov
17 hatch
16 speed
17 queen
22 roach warren
23 lair
ov
extractor
22 queen
drone til 44 drones.
You can replace roach warren with baneling nest if needed, or skip it altogether if you don't need it. In any case, this gets you fast speed (which gets you map control early on), followed by pretty fast lair, with little sacrifice to economy. This is the strongest (safe) mutalisk opening, because you have two fully saturated bases, and can double expand with the extra minerals once the mutas pop.
|
14 hatch, 15 pool comes out the winner. Not too suprizing given thats what like all the pros do (and me!  ). Still the 11 pool 18 hatch is very intresting. Definatly will be using that build on steps/DQ, where having that earlier pool can give you a lot more flexability that you need on those maps.
|
On December 01 2010 13:23 imbecile wrote:
Interesting as a framework, but not really for practical use. Because no build can do without a scouting drone and without gas and without any defence or creep spread until the 6:00 mark.
You have a rather appropriate user name. Obviously, the timing of the extractor or scouting drone or zerglings and the order we build them in will effect the build's economic fitness. But as long as the order and timing of whatever buildings or units we add to these skeleton builds stay the same, they are only delaying the next drone or overlord or queen or hatchery. So their economic efficiency without scouting drones or creep spread would translate into the same relative efficiency with it.
For example, if we were to compare 2 builds for the purpose of economic efficiency that included say, 4 zerglings, we wouldn't want to compare builds with 4 zerglings at 4:00, to other builds with 4 zerglings at 5:00. It would be trivial to prove that the best (in terms of economy) build with zerglings at 5:00 would be better than the best build with lings at 4:00. Since a scouting drones, zerglings, spine crawlers, creep tumors always slow down the economy by a constant (creep tumors could help if you're transferring a queen, but that's a really special case that's reasonable to ignore to reduce complexity), it's fine to ignore them and then just add them in uniformly to all the builds that we have tested.
The extractor is a different story, because it's actually an investment in the economy, so would be interesting to see how adding that in effects the economy. One thing about adding an extractor is that it complicates the research by a lot because we will need to qualify how much vespene gas is valued versus drones, queens, and minerals. However we weight the value of gas will change the number of drones we want to put on gas, and the time we want to build the extractors. This adds a lot of complexity to our model, and as we add more options to these test builds, the number of possible builds grows geometrically. Since we are using people and man hours to test these builds, it's reasonable to limit them to just drones, queens, overlords, and hatcheries. If we were to use an AI to test these builds, then it would be feasible to add in more components. But it would still be reasonable to start out with these 4 components from a efficiency standpoint.
|
11 pool, eh?
Perhaps I will toy around with this in ZvZ. Seems like it would be a great answer to 6 pooling shenanigans.
|
On December 01 2010 16:21 Mainland wrote:Show nested quote +On December 01 2010 13:23 imbecile wrote:
Interesting as a framework, but not really for practical use. Because no build can do without a scouting drone and without gas and without any defence or creep spread until the 6:00 mark.
You have a rather appropriate user name. Obviously, the timing of the extractor or scouting drone or zerglings and the order we build them in will effect the build's economic fitness. But as long as the order and timing of whatever buildings or units we add to these skeleton builds stay the same, they are only delaying the next drone or overlord or queen or hatchery. So their economic efficiency without scouting drones or creep spread would translate into the same relative efficiency with it. For example, if we were to compare 2 builds for the purpose of economic efficiency that included say, 4 zerglings, we wouldn't want to compare builds with 4 zerglings at 4:00, to other builds with 4 zerglings at 5:00. It would be trivial to prove that the best (in terms of economy) build with zerglings at 5:00 would be better than the best build with lings at 4:00. Since a scouting drones, zerglings, spine crawlers, creep tumors always slow down the economy by a constant, it's fine to ignore them and then just add them in uniformly to all the builds that we have tested. The extractor is a different story, because it's actually an investment in the economy, so would be interesting to see how adding that in effects the economy. One thing about adding an extractor is that it complicates the research by a lot because we will need to qualify how much vespene gas is valued versus drones, queens, and minerals. However we weight the value of gas will change the number of drones we want to put on gas, and the time we want to build the extractors. This adds a lot of complexity to our model, and as we add more options to these test builds, the number of possible builds grows geometrically. Since we are using people and man hours to test these builds, it's reasonable to limit them to just drones, queens, overlords, and hatcheries. If we were to use an AI to test these builds, then it would be feasible to add in more components. But it would still be reasonable to start out with these 4 components from a efficiency standpoint.
This is tempered by the fact we cannot build early units using late minerals.
One particularly severe flaw is the fact that many openings simply will not be able to afford constant production from 2 Hatches + 2 Queens by the time it's available. I believe this fact favors Hatchery first builds.
Another oversight is that middle pool FE builds walk a very thin line on being able to afford everything -- sending your 10th Drone to scout, I am quite sure, would far more seriously impact a middle pool build than it would a hatch first build.
|
On December 01 2010 16:29 Klamity wrote:
11 pool, eh?
Perhaps I will toy around with this in ZvZ. Seems like it would be a great answer to 6 pooling shenanigans.
I do believe 11 pool that starts putting pressure on the opponent with the first zerglings to be an excellent ZvZ opening. Of course, this strays pretty far from the topic of the thread.
|
On December 01 2010 16:29 Klamity wrote:
11 pool, eh?
Perhaps I will toy around with this in ZvZ. Seems like it would be a great answer to 6 pooling shenanigans.
I really love this build. It prevents your opponent from playing hatch first. It auto-wins all 6 pool rushes. Most importantly, I think people will overreact when they scout it. They may think "ok so he has to commit to attacking or he is behind in economy." From there you simply continue producing drones and fly past your opponent who has over-adjusted toward perceived early aggression and stopped playing pure macro.
I will definitely start testing this on the high-diamond ladder. I will let you guys know how it holds up and maybe provide some replays.
|
On December 01 2010 16:36 jdseemoreglass wrote:
I will definitely start testing this on the high-diamond ladder. I will let you guys know how it holds up and maybe provide some replays.
I already do since beta. Never wanted to change to 14 pool. Works just great.
I just dont understand why on the graph 11 pool gets so much behind on drone count at later stages. Both queens must be already up, why being behind on production? A mistake is possible.
|
I've been toying with an 11 overpool opening in 2000+ diamond. It seems to work well with the enormous versatility it offers.
|
Here's what I took away from this: 14 Hatch 15 Pool is the preferred economic opener if one can defend early aggression with it. 14 Pool 16 Hatch is a fantastic substitute to 14 Hatch 15 Pool if your hatch is blocked. 11 Pool 18 Hatch is the most flexible, easiest to defend/apply early pressure with, and overall the safest build for less experienced players.
Assuming a standard FE opener from the zerg player, I don't understand why anybody would want to use anything other than these three builds (assuming no other builds are posted).
Any build where the player makes both a pool and hatchery after 13 and before 18 (16 Hatch 15 Pool, 15 Hatch, 14 Pool, 15 Hatch, 16 Pool, ect) will not be far off. Any player below 2.4k will not win/lose games due to this minimal difference. However, if you know 14 Hatch 15 Pool is slightly more economical than the others, why not use it...
|
Gonna 11 pool on small maps forever for sure
|
how does 11 pool work, dang
pooling that early has to some kind of huge consequence, right? whats the catch to this build >_>
|
On December 01 2010 17:24 Let it Raine wrote:
how does 11 pool work, dang
pooling that early has to some kind of huge consequence, right? whats the catch to this build >_>
Almost nothing, it's so counterintuitive but still so awesome. Mind = blown, for sure
|
Yeah seriously, mine too. Lol.
I'm gonna test this out in some games I guess. BBL!
|
I used the 11 pool build yesterday in a few games.
I am platinum only, but the early pool puts quite some pressure on the opponent, while you still get a very decent eco going when not making 10-12 lings as threatened with first batch of bonus larvae.
|
So... what's the difference between single extractor trick 11pool and double extractor trick 12pool? s the 11pool really that much better?
edit: also try making 6 lings with both 11pool and 14h 15p. It might be that the early advantage of 11pool means more because you'll have more drones in comparison and so on... I'm not sure. Might be useful for a real-life scenario.
|
On December 01 2010 21:20 Shikyo wrote:
So... what's the difference between single extractor trick 11pool and double extractor trick 12pool? s the 11pool really that much better?
The difference was talked about in this thread: http://www.teamliquid.net/forum/viewmessage.php?topic_id=133917
IIRC, it had something to do with minerals lost on canceling both extractors making the build only break about even with the 11pool build, and more difficult to pull off perfectly.
|
Hey ! Thanks for the topic !
9 ov
16 Hatch
15 pool
17 ov
18 queen
21 queen
23 ov
31 ov
35 ov
is quite good for a macro game...
|
I like doing the 11 pool 18 hatch as people get confused by it. Also fun if toss puts down a pylon at nat, because you have plenty of time to take it down
|
On December 01 2010 16:21 Mainland wrote:
But as long as the order and timing of whatever buildings or units we add to these skeleton builds stay the same, they are only delaying the next drone or overlord or queen or hatchery.
That's the catch. If you remove a drone from mining, you delay timings, and at some point you won't have enough minerals to make use of a larva. Every second you have 3 larva lying idle is lost production time, akin to not building anything from your cc or nexus.
Anyway, did a quick and dirty test as described without scouting on caverns:
4:30 2985m 22.75d 56g
5:00 3400m 26.d5 116g
5:30 3885m 30.5d 172g
6:00 4460m 41.5d 232g
pool and hatch finish @3:44 and the two queens pop accordingly a@4:35
gas is planted @3:16, which is the same time I'd fit gas into the current leading 11pool build
So, how do the 11pool an my build compare? @6:00 I have 200 minerals and 2 drones less, but I also have 232gas. I'd say if gas is fit into the 11pool the numbers would be pretty much the same, unless it screws up some timings later, which might as well be, but needs to be tested.
So what matters then is the order you get things. Early pool is great, having your first queen @3:30 is too, against cheeses. But that is nothing that can't be addressed by scouting and adapting. I get my 2nd queen 20s earlier though, which s good for blocking ramps if need be. And it also means I can immediately start a 3rd and then lair a few seconds later.
Earlier hatch also means better creep spread, just as earlier second queen (I vomit at the nat and plant a tumor at the main to connect the bases).
So, I'm gonna stand by my initial statement: it's good to have a framework to compare against, and I think it's good that I can now. But once you adapt those pure economy builds to be usable in real games, there is little difference anymore, and your build choice depends on what you like better personally.
|
|
|
|
|
|
EPT
