|
EDIT #1: Osiris does a nice analysis of a couple of variations on 11pool/19hatch vs 14pool/16hatch on page 2 of this thread:
http://www.teamliquid.net/forum/viewmessage.php?topic_id=370042¤tpage=2#26
EDIT #2: From the discussion so far, it seems that some pros still do 11 or 12 pool openers, even at the highest levels of play. Furthermore these openers are safer than 14pool/16hatch vs any cannon aggression, so it's a good ladder opening. 14pool/16hatch is easier to execute in tourney play due to the supply depot at the ramp. Last, the eco hit taken by the 11pool/19hatch opening is slight and can be made up for by pressuring the toss player with the first 4 lings after securing the expo.
EDIT #3: Added section on how to roach rush vs 3-pylon block, with replay. See bottom of post.
Hi everyone,
I was not able to find a definitive answer on the +'s and -'s on the different openings in ZvP. I know most pro's/high-level players nowadays play 14pool/16hatch, but I am curious as to why, exactly... is it just the flavor of the season, or is it the best opening? As such, let me present the pro's and con's from my perspective on each of the "standard" ZvP openers:
1. 15hatch (easy one):
Pros:
a. Most economical opening.
Cons:
a. Very succeptible to ocannoning, and requires many drones to stop (and still many times results in loss).
Although ocannoning is very strong vs 15hatch, there are alternatives to dealing with this, the main one being some combination of cancelling the hatch into expanding to 3rd and proxy-hatch, a la Catz. The main issue with this is that it tends to be gimmicky, though can be pretty effective if not scouted in time.
2. 11pool/19hatch:
Pros:
a. Allows for easy securing of natural with fast 4 lings.
b. Once expansion is secured, 4 lings can be used to pressure toss. If the lings make it into his base (with nexus-first openings it can be close), then it can pretty much be GG due to probes running around/3+ probes getting picked off that early in the game. This is of course gimmicky and unreliable, but does force some early response from the protoss player and results in no loss to the zerg player, other than the cost of originally going 11pool.
c. Requires no drone scout.
d. In case of any aggressive ocannoning (pylon block or just early cannoning before hatch goes down), allows for a fast transition into 7RR, since the initial BO for both openers is the same. This can be scouted with 2nd overlord. Replay and
discussion/analysis of replay can be found at the bottom of this post.
Cons:
a. All else being equal, the least economical of the three.
3. 14pool/16hatch:
Pros:
a. Easier to deal with many ocannon variations due to pool-first.
b. More economical than 11pool.
c. I see most pros get only 2 lings instead of 4 once they feel that toss will not cannon them.
d. Can transition to a fast 5/7RR build which starts 14pool in case of some kind of ramp block before hatch starts.
Cons:
a. Succeptible to 3-pylon block without patrolling drone.
b. Due to early hatch, drone scout is pretty useful in case of aggressive cheese (so loss in mining time).
c. Can get pylon/probe blocked until lings come out.
So the questions I have are these:
Does anyone have any data on the economics of the three openers? I remember there was a graph that basically said 11pool/19hatch was better than 14/14/21 (and not too much worse than 15hatch), but 14pool/16hatch should be much better since you're not mining gas.
As a followup question, what's the economic impact with/without drone scouting? 11pool/19hatch requires no drone scout (your first 4 lings are sufficient), but 14pool/16hatch seems to lend itself to 2 non-mining drones: one to patrol the ramp, and one to scout. The scouting drone may be less important but on non-tourney maps it's very hard to stop a pylon block without a drone patrolling the ramp. If there're 2 non-mining drones at the start of the game, it seems like 11pool/19hatch would be roughly equal/better, but once again, I do not have, nor do I know where to find, data on this. I do think it's possible that most pros switched to 14pool/16hatch mainly because tournament maps do not allow for the 3-pylon block, so no defense is required against this.
Last question: off of pool-first openers, what are the pros and cons of getting 2, 4, or even 6 lings? 4 lings seems to be safe vs pretty much all cannon aggression (except 3-pylon block), lets you take both watch towers + ling outside his base, and have 1 left over to scout for proxies. 2 is a little low for my tastes, but lets you get another drone instead. 6 is used just to be aggressive.
Dealing with 3-pylon block:
The basic idea here is that when opening 11overpool, you can scout your natural with your OL for a 3-pylon block. If you see this, you can immediately transition into 7RR without losing a beat. I played this game with a masters toss opponent here (only the first 8 minutes or so are worth watching):
http://drop.sc/256301
Now there are many problems with my play:
1. My 7RR build was slightly off. The correct build can be found here:
http://wiki.teamliquid.net/starcraft2/7_Roach_Rush_/_Expand_/_Lair
Also, looking at it if I was to play macro like I did, 5 roaches with the remaining larva turned into drones would have been better.
2. I should have taken all my roaches, killed the cannon, then killed the pylon. While killing the cannon I should have sent my 2 drones out to double expand. From my analysis, this would have allowed me to put down each expansion 30 seconds earlier than I did in the game, even with a suboptimal build. Furthermore, I should have not bothered with the two other pylons and sent my roaches immediately, forcing faster cannons. He would not have had warpgate in time to do anything cute with the pylons anyway.
3. My macro after 3rd is put down is pretty bad. 3rd queen is significantly delayed, meaning my money piles up absurdly high before he even attacks. This is probably the difference in about 7+ drones and should have put me in a comfortable position economically.
Now, despite all these issues, I still play a pretty even game with my opponent. Drone/probe count is very comparable at the 6-8 minute mark despite lackluster macro on my part.
A similar build can be done opening with 14pool:
http://wiki.teamliquid.net/starcraft2/Roach_Expand
The main drawback is that the toss player is likely to start the pylons after you throw down the hatch at 16 (assuming no drone patrol), which means you have to cancel your hatchery, losing 75 minerals. Also, the 14pool 7RR variation is a few seconds slower. Neither of these are deal-breakers but they are worth considering.
Toss can also aggressively cannon your nat (3+ cannons), at which point you would probably want to switch to nydus or muta. The extra cannons sets him back a good amount, and you should be able to see this with your OL before committing to roaches (or you can use roaches with nydus). 7 roaches can handle 3 cannons after the pylon is killed however, not sure about more than that.
|
Although ocannoning is very strong vs 15hatch, there are alternatives to dealing with this, the main one being some combination of cancelling the hatch into expanding to 3rd and proxy-hatch, a la Catz. The main issue with this is that it tends to be gimmicky, though can be pretty effective if not scouted in time.
That's an outdated video. The only time hatch first is safe is if Toss is going nexus first, which is common on certain maps or against certain players. I've also heard that gateway openers that reactively add a forge, and go zealot+cannon rush will kill hatch first. I'm not sure if that's an absolute, but I've seen a few GSL games where this happened and zerg died.
2. 11pool/19hatch:
Pros:
a. Allows for easy securing of natural with fast 4 lings.
b. Once expansion is secured, 4 lings can be used to pressure toss. If the lings make it into his base (with nexus-first openings it can be close), then it can pretty much be GG due to probes running around/3+ probes getting picked off that early in the game. This is of course gimmicky and unreliable, but does force some early response from the protoss player and results in no loss to the zerg player, other than the cost of originally going 11pool.
c. Requires no drone scout.
d. In case of any aggressive ocannoning (pylon block or just early cannoning before hatch goes down), allows for a fast transition into 7RR, since the initial BO for both openers is the same. This can be scouted with 2nd overlord.
a. not worth the economic cost of the quicker hatch with way less drones. a pylon block still takes a bit to wear down too.
b. If you want to pressure Toss, you are better off opening 10 pool or 6/7/8 pool. 11 pool really doesn't get lings out much quicker than 14 pool, map size might be a bigger factor than 11p vs 14p. And going 11 pool puts you down about 5 drones from 14-15p, not including how many lings you make. And, you will have to really delay your third.
c. Any opening can transition into a 7RR easily. There are multiple build orders out there for roach rushes from all sorts of pool timings. 3RR is done with a 13 pool. If Toss cannon rushes you they will hold their natural easily as they'll see you bust out with roaches, and just put cannons down at home. If they are the pylon/forge ramp block I'm not really sure, the game works out differently, but I don't see you holding a standard 13 forge FFE who does a cannon rush.
Also, 11 pool is highly susceptible to ramp blocks, just as much as any other opening zerg does. If you dont patrol the ramp when going 11 pool you will get ramp blocked. 1 cannon behind 2 pylons > 6 lings.
So the questions I have are these:
Does anyone have any data on the economics of the three openers? I remember there was a graph that basically said 11pool/19hatch was better than 14/14/21 (and not too much worse than 15hatch), but 14pool/16hatch should be much better since you're not mining gas.
Search "Zerg openers" , jdseemoreglass did some analysis on the economics of some zerg openers. It's a bit outdated, in that he doesnt test some openers, but he does test a lot of openers.
Assuming you bring a drone back immediately after seeing the main of an opponent, a drone scout costs about 80 minerals, and will be gone for 2 minutes, regardless of when you send it. A drone scout in general ends up costing you about 1-2 workers around the 6:00 mark or so. Not a very large cost.
You have to pull 2 drones on blizzard maps no matter what opening you do. What opening you do doesn't really effect the impact of this (ie 11 pool vs 14 pool which is hurt more by this). It's a flat rate they are both hurt by doing this. Just know that with 11 pool you have to pull 2 drones too, so it's not like 11 pool might be better than 14 pool because you dont have to pull a drone.
Pros never used 11 pool as a macro build even when there was no neutral depot and back when 2 pylons could wall a ramp.
Last question: off of pool-first openers, what are the pros and cons of getting 2, 4, or even 6 lings? 4 lings seems to be safe vs pretty much all cannon aggression (except 3-pylon block), lets you take both watch towers + ling outside his base, and have 1 left over to scout for proxies. 2 is a little low for my tastes, but lets you get another drone instead. 6 is used just to be aggressive.
you need 4 lings to take down a pylon block at your third (or natural for that matter) in a timely manner. The less lings you make at 3:05 when pool pops, the much better it is for you since it's a drone made so early on. There's not much reason to ever make 6 lings.
You don't make lings to be safe against cannon aggression. You need to pull drones against cannon aggression, if you don't, you die. As for your third, you have to have an overlord spot over it, or have a ling over there, in order to stop cannon rushes. Sending 4 lings to your third when a cannon is going up will be too late, so you have to have an overlord spotting so you can stop a cannon rush before the cannon is made.
I'd recommend making 2 lings, and only make more if toss looks like he might pylon block you. I usually get 4 eventually.
Most of this stuff has already been answered in this forum. My guide talks about some of this, as do the zerg openers threads.
|
Generally speaking, the characteristics of doing an economical Zerg opening are the following:
- You make as many of your larvae into drones as possible.
- You minimize larva blocks (3 larvae at a time at a hatchery), especially early in the game.
- You make more mineral patches accessible by expanding timely, so you don't have over-saturation in your main.
- You don't take gas needlessly.
- You make a queen asap after your pool finishes.
I have experimented a lot with all these openers recently, and have found that sticking to these points is actually much more important than the exact order of drones and buildings. In other words: all of the openers you mentioned give you a lot of minerals and drones if executed properly. Things get complicated, though, when the probe arrives at your base.
You and Belial have already mentioned a lot of the technicalities involved: blocking the hatch with the probe or a pylon, cannon rushes, or even blocking the ramp. The three openers you mentioned all deal a bit differently with those. 15 Hatch I haven't even tried, since I suck at dealing with cannon rushes in general. 14/15 Pool gives you excess minerals quite early without the incentive of spending those on a queen (since the pool isn't done at that point). So you can try to take your expo from ~2:30 on, and more often than one would expect Zerg gets the hatch down before 3:00, meaning earlier mining, additional larva production, and the possibility to make only two lings instead of four. Protoss can, however, put a pylon down and either delay the expansion even further or force the third to be taken that way, which both hurts your economy or is, at least, inconvenient. This has become less popular recently though, maybe because Protoss prefers going Nexus first now and doesn't want to delay the forge after that.
11/12 Pool is different from 14/15 Pool in the way that you decide to get your queen before your expansion no matter what---since if you don't get the queen immediately after the pool finishes (and only lings to clear the natural), the earlier pool is a waste, in my opinion. Since this build is much less common than 14/15 Pool, I'll point out what I am talking about:
+ Show Spoiler +9 Overlord
12 Pool
16 4 Lings, Queen (double extractor trick)
20 Overlord
20 Hatchery (natural)
Rally next drone to third
21 Queen (goes to natural and injects there)
23 Hatchery (third)
24 Overlord
26 Queen -> main queen to natural (after inject), natural queen to third (after inject)
31 Overlord
36 Overlord
Notice that I am droning up until 16 after the fairly early pool before doing anything else. Compared to 15 pool, one drone is by ~30 seconds delayed, one by ~25, and one by ~20. With an average mining rate of 42 minerals/minute, the immediate mineral loss is thus approximately 75/60*42=52.5. Since 3 Hatch ZvP generally involves getting two hatches off of one base with ~16 drones on minerals, and 16 drones harvest about 700 minerals per minute, the third base is hardly delayed. With the build above, you should be able to take the third around 4:15-4:20. Notice you will have up to 5-6 idle larvae in your main, don't feel obliged to spend these right when they pop! In my opinion, this larva block doesn't hurt much since you can spend them after your putting down your third really quickly.
Edit: I forgot to add that a 12 pool consumes a drone something around 20 seconds earlier than 15 pool, so the early mining loss is more around 60-70 than 52.5. Points made still hold though, of course.
The main point of this is that most Protoss players will go forge first once they see that your pool is already building when they scout your main. And even if they go nexus first, you can send your four lings to your opponent and force him to wall off his natural completely. On most maps, this should be possible; however, in my experience, a cannon after 17 nexus forge will not complete in time to stop your lings from killing the forge! I have a replay of that, but I am not sure if I can easily find that after today's patch. I'll have a look, though.
Honestly, I don't know how much better nexus first is compared to forge first, but I would assume it is bigger than the difference between the 12 pool I described and the standard 15 pool into moderately delayed expo. This is controversial though since almost no pro does it (YuGiOh did it recently in the GSL, and there was a daily where ret did the similar 11 Overpool). If I have made any errors, I'd have them gladly pointed out. I am not a good player after all, and only posted this because this is basically a pre-5:00 numbers' game; when the actual game has started, the difference between how the builds play out becomes negligible.
|
i remember nestea did 12 pool 19 hatch in a GSL match. It actually timed out perfectly when I played it. I think 12 pool 19 hatch will be best cause you get an early queen and that way you make the pylon block basically useless and deny any cannon rush, force an earlier cannon from the protoss, AND you get the three hatches super easy.
THERE ARE 0 LARVAE BLOCKS WHEN DOING 12 POOL 19 HATCH! (this is important that's why I'm yelling at you :p). There is a small one if you do 11 overpool 18 hatch though.
|
Okay, if you are really that much into yelling, then I don't mind it, NOOBALOPSE.  But what you said is wrong. A 12 pool queen can do as much about any pylon or cannon block or cheese as a 15 pool queen can: nothing. Same goes for the lings as Belial pointed out. Those are bad arguments for a good thing.
And when I was mentioned the larva block, I was referring to the situation when your first inject finishes around 23 supply. At that point, you cannot spend all your larvae right away and get a third at the time.
When I tried 11/12 pooling way back, that was actually the mistake I would always do: spending all the larvae from my first inject right away. But that delays the third too much, as said, and you couldn't get much use from making those drones earlier anyway because your natural isn't done when these would hatch (I think so, at least, not sure if I recall that correctly).
|
The numbers you guys have provided seem pretty useful. So let me highlight the relevant ones:
Assuming you bring a drone back immediately after seeing the main of an opponent, a drone scout costs about 80 minerals, and will be gone for 2 minutes, regardless of when you send it.
With an average mining rate of 42 minerals/minute, the immediate mineral loss is thus approximately 75/60*42=52.5.
From these numbers it seems like this is how they should be ranked, from best to worst economically (assuming equal numbers of lings):
1. 14/15pool 16hatch with no drone scout and no ramp blocking. From games I've seen, this is the most common opening.
2. 14/15pool 16hatch with one drone patrolling ramp. Assuming one minute of patrolling, loss of 40 minerals vs opening 1.
3. 11overpool 19hatch with no drone scout. Loss of 52.5 minerals vs opening 1.
4. 11overpool 19hatch with one drone patrolling ramp. Loss of 92.5 minerals vs opening 1.
5. 14/15pool 16hatch with one drone patrolling ramp and drone scout. Loss of 120 minerals vs opening 1.
So basically, if you feel safe opening 14/15pool 16hatch without drone scouting, this is the best economic opening, even with a patrolling drone. If you open 11overpool 19hatch (an opening I can do where I feel safe against any toss opener), you lost about 50 minerals (or a drone). Drone scouting loses you more, but as Belial said it's a loss of 1-2 drones at around the 6-minute mark. This is not going to be entirely accurate because you do get your 10th and 11th drone slightly earlier due to building an OL at 11 instead of 9, so the 11overpool opener suffers less than what's indicated here, but it's useful for comparison.
One note about the 3-pylon block: if toss does this before you throw down a hatch, which is what is necessary in an 11pool opener (since you don't get a hatch until you have lings), then he's completely stupid and is relying on you not knowing how to react. Proper reaction to this kind of thing will put you ahead of the toss player. If this was not the case then in every single high-level game, you would either see a patrolling drone by Z or a 3-pylon block by P if they had nothing to lose (I haven't seen either). I'm not sure how all pros deal with this (I have my own set of responses that have worked very well for me), but I'm sure they have a way to do it off of 14pool/16hatch without a patrolling drone or it would be much more common than it is. However if you do not know how to defend against this with 14pool/16hatch without a patrolling drone, 11overpool is slightly better due to basically equal economy and the other advantages it affords you (earlier lings, earlier queen), assuming you know how to react to a pylon block.
Back to the point that velvex raised... it seems that if you execute all the openings well, long-term it makes very little economic difference as to which opening you pick. If the above numbers are correct and people are still drone-scouting, then 14pool/16hatch does feel like more of a "flavor-of-the-season" build than necessarily the best economic build. If you don't drone-scout though and build only 2 lings instead of 4, Then 14pool/16hatch should be the best opener.
|
There is something on pros and cons the OP doesn't mention:
14 hatch 16 pool:
con - susceptible to 1 pylon block on the hatch @ natural followed up by 3 cannons @ natural; forces 12 drones to be pulled & cancels cannons, putting protoss ahead economically
11 pool ~21 hatch:
pro - no need to pull drones to deal with above proxy cannons @ natural, lings will arrive in time
The difference between a perfectly executed macro for 11 pool 21 hatch vs 14 pool 16 hatch build is 10 food @ the 10 minute mark (I tested this multiple times), and having about 500 less total resources mined (5 roaches). Considering how quickly you are mining resources at the 10 minute mark, and you've mined over 10k resources by this point, it is a significant but not unjustifiable difference.
So determining whether this is economical will depend on whether pressuring the protoss results in reducing his total accumulated resources by 500 @ 10:00; and also considering the potential for the proxy cannons @ natural & how that may potentially set the 14 pool 16 hatch opening behind.
I suspect it is economical but I am not intimately familiar with the protoss openings enough to know for certain. It may be potentially advantageous depending on how safe the protoss opponent opens. It will vary with how easily the map may be walled, too. How much they must delay probes to block the lings from entering, etc. And what I can do if my lings do enter his base, can I drone up higher in different circumstances? Probably.
It will take somewhere around 2.0-2.5 probes worth of delay to make it approach economical. Since your pressure will begin around 3:30 (estimation), and full saturation of the protoss base will occur around 9:30 (estimation), 6 minutes of 2 probes not mining = 400 resources, 2.5 = 500.
BUT you still have to factor in whether the protoss gets an advantage from those earlier buildings (possibly an earlier core). That may or may not prove to be important. I'll have to look at the finances of the protoss build to know, and identify which builds I am concerned about.
Afterwards... dont forget to factor in the potential 3 cannon proxy rush @ natural.
It will also be slightly more economical on maps with distant 3rds, by ~30 minerals or so. : D
The 11 pool is promising. Whether you can get those 2.5 probes will depend entirely on the wall you force him to construct.
If they can be gotten, then I think it is a better opening since it runs no risk of the natural pylon block & 3 cannons. I will commit myself to analyzing the protoss response & calculating its economics at some point.
|
|
|
On September 20 2012 06:33 osiris17 wrote:
There is something on pros and cons the OP doesn't mention:
14 hatch 16 pool:
con - susceptible to 1 pylon block on the hatch @ natural followed up by 3 cannons @ natural; forces 12 drones to be pulled & cancels cannons, putting protoss ahead economically
11 pool ~21 hatch:
pro - no need to pull drones to deal with above proxy cannons @ natural, lings will arrive in time
The difference between a perfectly executed macro for 11 pool 21 hatch vs 14 pool 16 hatch build is 10 food @ the 10 minute mark (I tested this multiple times), and having about 500 less total resources mined (5 roaches). Considering how quickly you are mining resources at the 10 minute mark, and you've mined over 10k resources by this point, it is a significant but not unjustifiable difference.
So determining whether this is economical will depend on whether pressuring the protoss results in reducing his total accumulated resources by 500 @ 10:00; and also considering the potential for the proxy cannons @ natural & how that may potentially set the 14 pool 16 hatch opening behind.
I suspect it is economical but I am not intimately familiar with the protoss openings enough to know for certain. It may be potentially advantageous depending on how safe the protoss opponent opens. It will vary with how easily the map may be walled, too. How much they must delay probes to block the lings from entering, etc. And what I can do if my lings do enter his base, can I drone up higher in different circumstances? Probably.
It will take somewhere around 2.0-2.5 probes worth of delay to make it approach economical. Since your pressure will begin around 3:30 (estimation), and full saturation of the protoss base will occur around 9:30 (estimation), 6 minutes of 2 probes not mining = 400 resources, 2.5 = 500.
BUT you still have to factor in whether the protoss gets an advantage from those earlier buildings (possibly an earlier core). That may or may not prove to be important. I'll have to look at the finances of the protoss build to know, and identify which builds I am concerned about.
Afterwards... dont forget to factor in the potential 3 cannon proxy rush @ natural.
It will also be slightly more economical on maps with distant 3rds, by ~30 minerals or so. : D
The 11 pool is promising. Whether you can get those 2.5 probes will depend entirely on the wall you force him to construct.
If they can be gotten, then I think it is a better opening since it runs no risk of the natural pylon block & 3 cannons. I will commit myself to analyzing the protoss response & calculating its economics at some point.
Thanks for the response Osiris =) Very insightful... if there's a 500 mineral difference at the 10-minute mark, that is a factor worth considering in the opening, although as you stated 11pool is safe vs any pylon/cannoning of the natural. Also, do you know the tradeoff between 11pool 19hatch vs 11pool 21hatch? I have not heard of that variation. This is the opener I typically use:
+ Show Spoiler +
11OL
11pool
drone to 15
at pool completion, extractor trick to get queen + 4 lings
When lings finish, kill probe at expansion/secure expansion, take 4 lings and pressure toss front
In my experience I will get in if the toss goes nexus-first and is not conscientious about walling off, or if they go forge-first and forgo a cannon for some reason. At this point, it's very easy to either stop mining for a significant amount of time or pick off 3-4 probes. This early-on in the game that gives you a huge advantage, but vs a decent opponent you can only really count on them being forced to temporarily build an extra building until their cannon finishes, at which point you back off and take map control. This seems to put me on even footing with my opponent in general, and it does force toss to make some small concessions in his build (nexus-first with extra building/cancel seems optimal). In this case I doubt it would make up for the 500 minerals lost by the 10-minute mark.
However as to your other points, as you stated it's much easier to harass a 14/15pool 16hatch opening, even with the probe just running around to deny the expansion for a little while (and throwing down a pylon). Small factors like this seem like they could hurt the build enough to make it equivalent/worse than 11pool/19hatch, not sure. Still I'm surprised I do not see this more often in high-level games... it seems like zergs are just allowed to put down their hatch in this opening without any real probe harass other than a few seconds of hatch-blocking.
|
decemvre (zerg progamer) does 12pool 16queen, then 2 sets of ling with double extractor trick, then 20hatch
Advantages:
- You can do anything u can do with 14pool, you are not hindered in any way (he did Stephano 200/200 build to demonstrate and I remember it being under 11:30)
- The game isn't random depending on whether protoss pylonblocks your nat or not - which is good practice and its nice to not rely on coinflip
Disadvantages
- 14pool ends up ahead if protoss doesn't pylonblock natural
edit: never assume your 4lings do any damage to protoss, they shouldnt and you should play the game out as if they never would do damage
|
On September 20 2012 05:15 NOOBALOPSE wrote:
i remember nestea did 12 pool 19 hatch in a GSL match. It actually timed out perfectly when I played it. I think 12 pool 19 hatch will be best cause you get an early queen and that way you make the pylon block basically useless and deny any cannon rush, force an earlier cannon from the protoss, AND you get the three hatches super easy.
THERE ARE 0 LARVAE BLOCKS WHEN DOING 12 POOL 19 HATCH! (this is important that's why I'm yelling at you :p). There is a small one if you do 11 overpool 18 hatch though.
He did it once, a long time ago, as an aggressive build, i believe. You don't deny any cannon rush either, if Toss cannon rushes you and you don't react exactly the same as if you were going 15 pool or even hatch first, you lose. You also can still be ramp blocked.
Please do not post bad information like this, it can misinform a lot of people who may be looking for answers.
The point about larva blocks makes no sense and is inconsequential. Who cares if there are 0 larva blocks. Doesn't really help if your economy is much further behind. If you really cared, you can time out things like queens and overlords with any type of opener to avoid 'larva blocks' rather than do a radical build.
decemvre (zerg progamer) does 12pool 16queen, then 2 sets of ling with double extractor trick, then 20hatch
http://www.twitch.tv/quanticdecemvre/b/332051855
He does it as an aggressive opener, not as a macro opener. If he does not do damage because toss responds properly, then he would be way behind. As a macro opener, 12 pool is terrible.
|
i think we have to stop thinking and go away from the mindset that pylon block = death, which is not true. However, the true response to a 3 pylon block is not to pull drones, but teching up to roaches and put him on the defensive side while you double expand behind the pressure. Therefore ultimately it evens out.
|
14 pool is called more economical than 11 pool, but if you gas trick 11 overpool (not sure what timings you were talking about) that should not be the case. 14 pool will have better short term economy, but due to the additional larva, 11 overpool should be equal or better economy in the long run (as long as it doesn't make lings or something to put it "behind")
|
Dude I think you are making something that is really, really simple, way too complicated.
- Patrol a drone at ~16 supply to prevent ramp blocks. It doesnt matter what your opening is, you need to patrol a drone to deny a ramp block.
- Pull drones to stop a cannon rush, no matter what opener you do. The quicker lings come out just means the less time toss has to cannon rush, but they still won't be out in time when a cannon rush starts.
- the later the pool (up to about 15) the better you are economically and quicker you can expand to nat and third. earlier pools dont leave you dead because of the early queen, but it's still behind.
|
14 pool is called more economical than 11 pool, but if you gas trick 11 overpool (not sure what timings you were talking about) that should not be the case. 14 pool will have better short term economy, but due to the additional larva, 11 overpool should be equal or better economy in the long run (as long as it doesn't make lings or something to put it "behind")
jdseemoreglass specifically stated in that thread that this is only about 1 base. If you ever plan to expand, 14 pool is way better economically. This was back when it wasn't viable to expand in ZvZ earlygame (or rather, that was the metagame).
In both the short run and long run, 11 pool is way behind 14 pool. 11 pool is better if you are planning to do a 1 base ling play, but otherwise 14 pool is better if you ever plan to expand (or make a roach warren, for that matter).
|
What does the term ocannoning mean?
When you play 14(15)pool/16hatch protoss should almost always pylon block your nat. With 11overpool that is not possible. Does anyone know how the ecomics play out if this happens?
I have seen 11 overlord 19 hatch a few times in GSL. I believe one of them was Nestea doing it. Don't remember which games those were though.
|
On September 20 2012 08:43 Belial88 wrote:
jdseemoreglass specifically stated in that thread that this is only about 1 base.
Which thread? there's more than one that he's made, and the one I'm referring to is with regards to expanding at 18. As far as I know all the comparisons made with 11 overpool were for pool-first builds, not just 1-base builds (perhaps you confused them?). 14 pool works well as a 1-base build because it gets good economy early on; the 11 overpool takes time to catch up because of the lower mineral count but higher larva count
|
On September 20 2012 08:53 ulfryc wrote:
What does the term ocannoning mean?
Ocannoning = Offensive cannoning
|
On September 20 2012 07:27 Defenestrator wrote:Show nested quote +On September 20 2012 06:33 osiris17 wrote:
There is something on pros and cons the OP doesn't mention:
14 hatch 16 pool:
con - susceptible to 1 pylon block on the hatch @ natural followed up by 3 cannons @ natural; forces 12 drones to be pulled & cancels cannons, putting protoss ahead economically
11 pool ~21 hatch:
pro - no need to pull drones to deal with above proxy cannons @ natural, lings will arrive in time
The difference between a perfectly executed macro for 11 pool 21 hatch vs 14 pool 16 hatch build is 10 food @ the 10 minute mark (I tested this multiple times), and having about 500 less total resources mined (5 roaches). Considering how quickly you are mining resources at the 10 minute mark, and you've mined over 10k resources by this point, it is a significant but not unjustifiable difference.
So determining whether this is economical will depend on whether pressuring the protoss results in reducing his total accumulated resources by 500 @ 10:00; and also considering the potential for the proxy cannons @ natural & how that may potentially set the 14 pool 16 hatch opening behind.
I suspect it is economical but I am not intimately familiar with the protoss openings enough to know for certain. It may be potentially advantageous depending on how safe the protoss opponent opens. It will vary with how easily the map may be walled, too. How much they must delay probes to block the lings from entering, etc. And what I can do if my lings do enter his base, can I drone up higher in different circumstances? Probably.
It will take somewhere around 2.0-2.5 probes worth of delay to make it approach economical. Since your pressure will begin around 3:30 (estimation), and full saturation of the protoss base will occur around 9:30 (estimation), 6 minutes of 2 probes not mining = 400 resources, 2.5 = 500.
BUT you still have to factor in whether the protoss gets an advantage from those earlier buildings (possibly an earlier core). That may or may not prove to be important. I'll have to look at the finances of the protoss build to know, and identify which builds I am concerned about.
Afterwards... dont forget to factor in the potential 3 cannon proxy rush @ natural.
It will also be slightly more economical on maps with distant 3rds, by ~30 minerals or so. : D
The 11 pool is promising. Whether you can get those 2.5 probes will depend entirely on the wall you force him to construct.
If they can be gotten, then I think it is a better opening since it runs no risk of the natural pylon block & 3 cannons. I will commit myself to analyzing the protoss response & calculating its economics at some point. Thanks for the response Osiris =) Very insightful... if there's a 500 mineral difference at the 10-minute mark, that is a factor worth considering in the opening, although as you stated 11pool is safe vs any pylon/cannoning of the natural. Also, do you know the tradeoff between 11pool 19hatch vs 11pool 21hatch? I have not heard of that variation. This is the opener I typically use:
That is the correct build - 11 pool 19 hatch. If I remember I used to get my 3rd hatch @ 36 after the ovi @ 36. Not sure when I got my 3rd queen.. whenever was ideal for not sacrificing drones. Havent done this build in a while, but I may revisit it.
On September 20 2012 07:27 Defenestrator wrote:+ Show Spoiler +
11OL
11pool
drone to 15
at pool completion, extractor trick to get queen + 4 lings
When lings finish, kill probe at expansion/secure expansion, take 4 lings and pressure toss front
In my experience I will get in if the toss goes nexus-first and is not conscientious about walling off, or if they go forge-first and forgo a cannon for some reason. At this point, it's very easy to either stop mining for a significant amount of time or pick off 3-4 probes. This early-on in the game that gives you a huge advantage, but vs a decent opponent you can only really count on them being forced to temporarily build an extra building until their cannon finishes, at which point you back off and take map control. This seems to put me on even footing with my opponent in general, and it does force toss to make some small concessions in his build (nexus-first with extra building/cancel seems optimal). In this case I doubt it would make up for the 500 minerals lost by the 10-minute mark. However as to your other points, as you stated it's much easier to harass a 14/15pool 16hatch opening, even with the probe just running around to deny the expansion for a little while (and throwing down a pylon). Small factors like this seem like they could hurt the build enough to make it equivalent/worse than 11pool/19hatch, not sure. Still I'm surprised I do not see this more often in high-level games... it seems like zergs are just allowed to put down their hatch in this opening without any real probe harass other than a few seconds of hatch-blocking.
When I tested the 14 p 16 hatch against the 11 pool I built my 2nd hatch @ the 3rd for consistencies sake.
I don't count on getting in to the base. If I do get in, I expect to capitalize on what I scout to make it financially worthwhile. Getting into his base is a major victory and I think worth the resources.
If the toss is conscientious about walling off, they sacrifice probes to do so. I'm not sure how much they sacrifice. Smaller maps with wider ramps more, larger maps with smaller ramps less. 1 probe is constructed every 20 seconds. So to delay him 2.5 probes worth of construction, you must delay him by a total of 50 seconds construction time considering both nexus's: the delay in building the natural nexus added to the halting of production at the original nexus, in order to save up money for the wall; not counting delayed chrono accumulation & some other factors like having to deal with more early zealots, and all this. I am fairly confident this ~2.5 probe delay happens on some maps. With nexus first it will be more difficult, but nexus first will also delay the cannon. Depending on the map size you may even be able to kill a building & get in to his base before a cannon completes. But I will examine it closely, including nexus 1st & all the variations.
As for the earlier core / gateway / etc., still needs considering whether its relevant. I will get back to this thread, since I wana know for sure...
Off the top of my head the map this would be the strongest on would be Antiga with a very wide natural, 4 base scouting distance, and relatively short rush distance.
One thing which just occured to me though, a soft counter to this, is the protoss has the option upon scouting your 11 pool to save up his chrono boost until after he has completed his wall, his probes building slower & thus minimizing the down time on his main nexus... So that also demands consideration
|
Okay so I am going to keep testing this and post my results in here, so that I can finally get to the bottom of this. Because it's bugged me for a while but I've never put in the effort to figure it out with absolute certainty which is better.
I don't have time to test this extensively right now, that will be for tomorrow.
But there is this scrap of information. In a preliminary test, the time period from the protoss initial delay of probe production (for beginning building placement of the wall) until 44 worker saturation (2 on each mineral patch & 3 on all 4 gas) is about 4 minutes long.
After that mark I am willing to count the ~3 minutes left before 10:00, while the player approaches absolute saturation, as part of the loss in probe mining efficiency. But this also requires I consider the Protoss priorities. If they are staying at 44 probes and attacking me, this period should not be considered since no further saturation is being pursued, no potential gain is lost through delay. Thus if the Protoss is 2 base allining you, he has more of an edge, but if he is 3 basing you, his advantage is dulled.
For this problem I will assume he is taking a 3rd; at least probing up to 60 probes in some fashion, since I also assume I can hold all 2 base builds which stay at 44 probes. even if I am minorly 100 resources more behind. It takes 2:40 to build the 16 more probes, moving from 44 to 60 for absolute saturation of 2 bases, and the extra probes are mining at half efficiency during that time.
So 4:00 delay plus 2:40 / 2 = 5:20; 5:20 / (40 minerals per minute x # of probes protoss is delayed) = total minerals protoss loses, not counting a few other minor factors which will be considered later (changed chrono boost, earlier core & zealots & so fourth).
So rounding it to 5:00, a 2.5 probe delay loses 500 resources.
Also, in the preliminary test (this is just sort of precursory) the total amount of time probe production was delayed in order to construct a wall was just about 1:00 (which amounts to over a 2.5 probe delay). From a quick glance this appears to remain the same regardless of the map, since following a 3 building wall (2 gates a forge & a cannon) a pylon must be built immediately in order to facilitate probe production; on a larger map this pylon may be part of the wall but this is irrelevant. This is true however the opponent opens, whether it be with nexus 1st or forge 1st.
TLDR: 2.5 probe delay is indeed the goal, and the current analysis shows this is achieved; infact exceeded even against a nexus first, IF the opponent full walls (which he should, and you should make sure he does). The build looks extremely promising to me now but a more thorough analysis is still needed.
|
|
|
|
|
|
EPT
