On November 11 2012 13:06 Mavvie wrote: @Belial I just tried a fast muta opener (similar to Nestea's, but adapted because I forgot 4:30 gas) that gets a 7:50 spire if I don't delay it. http://www.twitch.tv/mavvie/b/339200763 1:18:00 is when the game starts, it's a ladder game and I just ultimately steamroll his army; I even defended my fourth. I skip roach warrens and evos, using my fast lair to scout his sentry count (and I even saw his robo) to know that I was safe.
I opened 11 overpool because I took so much damage from pylon blocks in the previous games
FPVODs are cool, much more so than replays. Anyways, here's the replay in case you don't wanna watch my FPVOD: http://drop.sc/274284
Despite being horridly out of position and forgetting ling speed (foreverbronze I know), I still defend. Mutas are great because they can't be forcefielded, so even if they suck you at least get some damage in. Also mutas are unaffected by immortals, so it's basically a delayed 7gate with sentries. Sounds like an easy defense, right?
I feel like at least one evo chamber is a good idea; there's plenty of time for +1/(+1). I dunno, I think fast mutas are a great idea ZvP. Maybe I only think this because it's the only game I won today, whatever.
Edit: If the sentry/immortal is like the 1/1/1, then Nestea's fast mutalisk build is like HerO/MC's (not sure) fast phoenix opener :D
I watched that VOD. I'm starting to understand the real power of mutas, which is that they are able to snipe all the sentries effectively...that really is their greatest use, I think you would only need around 16 to accomplish this before transition.
What i've been doing is standard gas timings (double gas at 6.00, 3rd and 4rth following) with a roach warren slightly earlier, like 6.45. I research speed with first 100 gas, dump all my gas into roaches once roach warren finish, then nothing but zerglings. Next 100 gas go to lair, next 150 to carapace (better for late game purposes than missile)
This means i'll drone cut around 54, which I think is better than 60. The plan is as follows: I've been noticing protosses tend to not make observers if they go for immortal all in. I'll wait outside the protoss base but far enough so that he cant see my army composition. Once he moves out at around 9.30 i'll launch a suicide attack on his army.
The goal is to kill every sentry. The roaches are there so that the protoss is in a dillema: either ff the lings away, but then roaches can shoot from behind forcefield, either ff the roaches away, but then the lings can kill sentries. He shouldnt have enough force field so early to do both if u catch him in an open area (remember his sentries should not have build a lot of energy). And most of the time you can catch the toss in an open area because most protosses feel safe at the time they move out of their base. This could be because most Z tends to make lair before speed and therefore they dont have lingspeed at that point, + they go roach heavy and Z dont want to engage with a big slowroach army.
Most of the time the toss army will be weak enough after this first attack that the toss will want to go back home. Most protosses will then make zealots because they need gas for new sentries and because they saw a lot of lings. That's why i follow this first attack with mainly roaches which will have speed by the time the second protoss attack hits. Bait a couple FF, engage from 2 directions and he will ran out of FF.
A word or two about the early drone cut: this drone cut lets u have roaches in time for the early zealot +1 shenanigans as well as having speed to shut down a variety of all ins like blink +2 etc. If the toss did a robo and a bunch of immortals but choose to take a 3rd, you' should have enough units to delay his third enough to drone back so that you get back in a worker advantage.
My winrate has substantially increase vs immortal all in since i did those adjustements. I tend to win toss with higher mmr than me most of the time.
Edit : now that I look back at those replays you could say my opponents screwed big time by moving like they did. But I think that under GM mmr, most protosses will do so because most Z lets them by having slowroaches only army at the time Toss moves out. As I said, Toss tends to not have observers because they want to maximize immortal count and if they send a zealot to scout, you can kill it with a bunch of slowroaches, hiding your lings.
On November 11 2012 19:33 Teoita wrote: That's like saying that the counter to 11/11 rax is 10pool into 1base roach.
If your point is that this build needs to deviate very early to counter the immortal all in that Z cant have scouted by the time he deviates, you are right. But honestly since 90% of my zvp includes immortal all ins, i'll do so mindlessly
I like the Muta Idea, but I don't have the multitasking to make muta work correctly in a macro game (either my muta are useless/dead or my macro goes awry).
I was thinking, we have a wonderful unit that has a long range, shoots up and down, and is quite sturdy. No, Hydras, Queens ! Could we defend the all in with ~7/8 queens (ie. the normal number + 1per hatch as soon as scouted) helping with the transfuse.
The complement would be putting the gaz in Hydras (short term solution) or Infest (long term). This is quite mineral intensive so I guess Hydras would make more sense, and they also have range.
The defense would then by handled by a few spines, 7/8 queen and ~10 hydras (and counting). With the range of all these units, either the protoss army will have little units hitting through forcefields, or many will be in range of the good dps of hydra/spines and the ok dps of queens.
Again, you would very likely have to commit to a build like that before you know it's an immortal all-in. What happens if he immortal expands? Do you just sit there with a bunch of useless hydras and later infestor tech, later hive, later infestor energy and 10 of the worst offensive unit in the game, while protoss turtles on 3bases with colossus tech? Do you try to push with hydras on modern big maps? Neither is going to work consistently.
That said yeah, creep spread is key to stopping this so i guess it would be interesting for Z to shift their openings a bit, maybe going up to 5 ish queens and keeping 2 to spread creep..not to fight. Modern openings don't really emphasize creep in zvp as much as they do in zvt and that's a pretty big fault.
The big problem isn't JUST holding off the immortal all-in, it's doing so while also accounting for every other possibility Protoss has, at least in some way (and having some freedom to do so), before having to commit to anything drastic like 2base play or hydras.
On November 15 2012 00:06 Teoita wrote: Again, you would very likely have to commit to a build like that before you know it's an immortal all-in. What happens if he immortal expands? Do you just sit there with a bunch of useless hydras and later infestor tech, later hive, later infestor energy and 10 of the worst offensive unit in the game, while protoss turtles on 3bases with colossus tech? Do you try to push with hydras on modern big maps? Neither is going to work consistently.
That said yeah, creep spread is key to stopping this so i guess it would be interesting for Z to shift their openings a bit, maybe going up to 5 ish queens and keeping 2 to spread creep..not to fight. Modern openings don't really emphasize creep in zvp as much as they do in zvt and that's a pretty big fault.
The big problem isn't JUST holding off the immortal all-in, it's doing so while also accounting for every other possibility Protoss has, at least in some way (and having some freedom to do so), before having to commit to anything drastic like 2base play or hydras.
Your best bet is scouting and doing a build that covers most of the possibilites your opponent has. I happen to like ling-muta OR ling-infestor-ultra. Which one of them I choose depends on whether I see an expand or not. If he expands, I macro up a little and go ultras and hold their push with ling infestor. If he has not expanded, mostly lings and then mutas tends to finish off things pretty well. Also, it's key to note that some protosses push with only 2 immortals at around 7:45-8:00... So you've got to activley scout for their exit, because if you're stuck with the 10:00 mark on your head, you will get caught with your pants down.
In a Day[9] video a while back, he showed TLO doing something really cool--basically, as the Protoss moved out with the all-in, TLO would mass spine at his main ramp. Then he nydus-wormed his entire army to attack the Protoss, either in the main or at the front.
You sac your third and natural to kill his main and natural. Bring your army back through the Nydus Worm if he decides to attack up the ramp. At the end of the fight, you should have one mining base to the other guy's zero.
I haven't seen anyone other than TLO try this yet, but it seems to me like there's some potential there.
On November 15 2012 02:21 Zozma wrote: In a Day[9] video a while back, he showed TLO doing something really cool--basically, as the Protoss moved out with the all-in, TLO would mass spine at his main ramp. Then he nydus-wormed his entire army to attack the Protoss, either in the main or at the front.
You sac your third and natural to kill his main and natural. Bring your army back through the Nydus Worm if he decides to attack up the ramp. At the end of the fight, you should have one mining base to the other guy's zero.
I haven't seen anyone other than TLO try this yet, but it seems to me like there's some potential there.
The problem is that if you cant destroy the main because the protoss doesnt suck, you are 1 base vs 1 base. Pretty much fucked.
I just found a replay where life hold an immo sentry all in with ling roaches. I dont know how he manages to do so... I think Brown failed his forcefields... I'm a right?
Well he loose this game due to a sentry drop after the immo all in. But thats another problem..
I am no longer active on EU the server but a few weeks ago, when playing for my team I designed a macro oriented counter to the imo-sentry all-in.
We know that without tech (mutas, hydras or infester) it's very hard to fight the toss army head-on. But getting the tech quickly enough to defend this very early push require a big cut on economy (late third), which can be easily scouted and exploited by the Toss.
I used to do a roach/ling/hydra defense on 3 bases using standard stephano gas and tech timing but I only defended barely and I was in a bad position if the Toss just took a safe third defended by sentry/imo once he forced enough hydras and followed by a pre-hive colossus push. As soon as colossus hit the battleground, supply and resource tied in hydras become a huge liability.
Counter attacking seem to be the solution because the Toss usually just wall off and doesn't keep any defending army to make the strongest push possible. But with usual timings only a small roach/ling army can be at the Toss nat when the WonWonWon train as left the station. To get a really theatening roach/ling army at Toss door around 9min, you have to delay either eco (less saturation) or tech (later roach speed and infestors).
Considering all this I got my idea of a reactive eco ling-baneling counter bust. This a reactive build and it should be able to defend this all-in but also put the zerg player in a good position for a standard macro game.
The first phase of the build focus on maximizing the mineral input and drone production during the first 7min of the game. We want to get a 3 bases fully mineral saturated (48 drones) ASAP. Delaying gas until 6:30 and not building any creep tumor should allow you to reach full min saturation around 7:10 and stop producing drone at 7:30. 7:30 is the turning point when you should have identified the build 100%, but against quick gateway pressure, you have to adapt earlier (if no gas at nat and no third building at 6:45, emergency roach warren and 2 spine at the third plus a few zerglings should keep you safe).
When the standard 7min overlord sac has seen many sentry and a robo, the build is clearly identified. So the reaction is as follow : ling speed with the first 100 gas from 3 gas at 6:30 (ling speed will be ready around 9:10), and then baneling nest (begin ~7:30, finish ~8:30) and then lair. Regarding mineral allocation, when you reach 57 drones (3*16+3*3) you should build a macro hatch, a fourth extractor, 5 more drones and a fourth queen. Any mineral left go to zerglings (zerglings production begins around 7:30 to get 2 injects rounds-60lings at the Toss door at 9min).
We can't defend the third, but with a quick mineral saturation it will pay for itself. If the all-in is certain, you should begin spining up the nat at 8:30. As you plan to sac the third, just morph 8 drones from the third to the nat. The 8 drones left will be transfered just before his army hit your third to make 4 additionnals spines (put the other on min). With the flood of 3 injected hatch building lings and 12 spines, you should have no problem defending your nat. You also have to be careful of your building placement : the macro hatch is safer in the the main, and don't build any extractor in the third before taking the first 4.
As soons as the baneling nest finish (8:30), you should have around 300 gas to build 12banes in a hidden position close to the Toss nat. This means that your counter attack will hit at 9:00, when the WonWonWon train is cruising in the midle of the map (according to Remarks guide). 10 banes is enough to destroy any wall (forge is 10 IIRC) so even if he left a single sentry at home you should make short work of the nat with zerglings amazing dps output, probes will fall to any baneling left.
Now your zerglings could get scouted and no doubt that he has enough sentry to defend if he decide not to push. In this situation we have a fall-back plan. At 9:10, your lair should finish and you should have just enough gas to build a spire. The threat of the counter attack and the army dance should buy enough time to have mutas before he can safely attack you or get his third. And from there the game is simple. You just have to threaten a ling/bane/muta counter attack while growing your econ advantage and teching. Given that you have 10+ spines already, any base race scenario is an autowin.
So here is the build order : Standard 3 hatch opening, don't build a third queen until the third hatch is built because you will need the extra min to spend it on the extra larvae given by not building a creep tumor, spend queen energy only on injects. 6:30 3 gas 7:00 send ov to scout and assert the WonWonWon build 7:20 ling speed, add an extractor 7:30 baneling nest, first round of lings stop drone production at 60 drones, add macro hatch, 4th queen 7:50 build lair in your main 8:00 2nd round of lings 8:30 build as many banes as you can in a hidden location don't be afraid to cancel if he chase your army, time is on your side. Spine up your nat, continue ling production if he is agressive. 9:00 destroy his naked nat if he pushed, ling speed will finish shortly 9:10 build a spire if he stayed defensive
Pros : - fully reactive build, doesn't sacrifice eco, slighly later tech but the first round of zerglings can delay Toss third for a long time to compensate. - no need to invest heavily in the defense before 8min (the 2nd round of lings). And if he doesn't look agressive, you don't have to morph banes. - a near bo win if he push and doesn't stop the counter attack, wich will often be the case on ladder given the surprise effect - no big tell until zergling production for the protoss to identify the BO - bo still very strong against stargate play (good eco, ling and banes to deny the third)
Cons : - weaker than standard build to +1 gateway pressure or just zealot/stalker harass (you have to be very active with the initial 4 lings to scout for that and be prepared in time) - if you fail to scout you can be caught with your pants down at a wrong timing (DT pressure, gateway all-in, zeal/archon all-in ...)
TLDR: Standard 3base late gas, 2 round of lings with speed and bane nest to treaten a timed ling/bane counter attack at 9 min, spine up the nat and sac the third if needed, buy some time for spire tech to punish heavy sentry and late third.
I don't have any valuable replay given that I don't play much ATM. But I think that given the current situation, any Zerg strugeling with the WonWonWon build should give this a try.
Yeah, ling/baneling openings are interesting and they would definitely do well vs immortal all-ins...i remember seeing a Z (most likely Dimaga) completely dismantle the all-in with it.
The downside of course is that the midgame isn't as refined as it is with roach/ling/infestor builds, but if would be really cool if someone ironed out the timings for that kind of playstyle, PvZ atm is pretty rigid.
I don't know if this has been said before but has anyone tried cutting drones earlier making a round of lings at around like 40 drones making drones then making roaches and lings? It seems everyone has a mindset of make 60 drones then make units. Why not make units earlier then make drones then make units? Sorry if this has been suggested before.
On November 16 2012 02:40 Abusion wrote: I don't know if this has been said before but has anyone tried cutting drones earlier making a round of lings at around like 40 drones making drones then making roaches and lings? It seems everyone has a mindset of make 60 drones then make units. Why not make units earlier then make drones then make units? Sorry if this has been suggested before.
On November 16 2012 02:40 Abusion wrote: I don't know if this has been said before but has anyone tried cutting drones earlier making a round of lings at around like 40 drones making drones then making roaches and lings? It seems everyone has a mindset of make 60 drones then make units. Why not make units earlier then make drones then make units? Sorry if this has been suggested before.
DRG has been delaying his lair in favour of early roaches against protoss. he still makes ~60 drones, but gets speed and 8-10 roaches before lair, and attacks the protoss. It hits exactly as the protoss would want to be moving out for this push.
But it seems to set him way behind when the protoss doesn't move out at 9:00.
I don't think that making lings at 40 drones would accomplish anything except lose money.
That push is really strong vs greedy macro builds like stargate into robo while only making phoenix, because he can put a ton of pressure on, delay the third and reduce the sentry energy. He even took a game off MC in the OSL. Obviously it's also really good vs any gateway timing, as he starts making units blindly when a Zerg should defend a gateway timing. The combination of the two makes it a great metagame move against MC's builds.
I think against anything else, timings and macro builds, including an immortal all-in, it's fairly bad as he has far less drones (and therefore production) to hold off the attack, and he delays his tech a lot. In their GSL match, Creator was able to even get a 4th and max out before his pre hive push because of how delayed everything from DRG was.
Unless the P gets caught horribly out of position (which can happen but all it takes is leading with a zealot or probe to make sure you are safe to move out), any delayed/slower all-in like immortal/sentry or even blink stalkers will be even stronger because the Zerg's drone count isn't optimal.
On November 16 2012 03:31 Teoita wrote: That push is really strong vs greedy macro builds like stargate into robo while only making phoenix, because he can put a ton of pressure on, delay the third and reduce the sentry energy. He even took a game off MC in the OSL. Obviously it's also really good vs any gateway timing, as he starts making units blindly when a Zerg should defend a gateway timing. The combination of the two makes it a great metagame move against MC's builds.
I think against anything else, timings and macro builds, including an immortal all-in, it's fairly bad as he has far less drones (and therefore production) to hold off the attack, and he delays his tech a lot. In their GSL match, Creator was able to even get a 4th and max out before his pre hive push because of how delayed everything from DRG was.
Unless the P gets caught horribly out of position (which can happen but all it takes is leading with a zealot or probe to make sure you are safe to move out), any delayed/slower all-in like immortal/sentry or even blink stalkers will be even stronger because the Zerg's drone count isn't optimal.
True. I just feel that if you let a protoss walk across the map with 7 sentrys 1 zealot 3 immortals and a stalker and don't at least bait out forcefields it seems a bit counter-intuitive. Your last point is valid but you can still threaten counter attacks with this smaller army and buy time for your re-droning to kick in or still bait out forcefields and look for proxy's. The only thing I think this is really bad against (like you pointed out) is just a passive macro build that takes a fairly late 3rd but even then you can sometimes catch units off guard but if he takes a 3rd off 7 gates and a robo without blink a muta switch can be devastating.
No any build that either gets a void ray or just makes units off 4gates and a robo should hold it off easily.
I agree on the necessity to delay the attack, but pumping units out a full minute faster than optimal is not smart when the attack is already so scary for Z.
In my experience, the best bet ATM at least for me is being super-reactive with a speedling-based eco build, then derive into something else depending on the composition and your econ after holding off his attack... It is necesary to delay the attack, and overlord possitioning and active scouting around the 7:00 mark seems key to me, because of the 2-immortal all in I stated earlier.
On November 16 2012 00:37 Teoita wrote: Yeah, ling/baneling openings are interesting and they would definitely do well vs immortal all-ins...i remember seeing a Z (most likely Dimaga) completely dismantle the all-in with it.
The downside of course is that the midgame isn't as refined as it is with roach/ling/infestor builds, but if would be really cool if someone ironed out the timings for that kind of playstyle, PvZ atm is pretty rigid.
EPT
