|
I’ve been using this build for at least 3 months now and I kow it’s nothing revolutionary (it’s possible that day 9 even did a show about it) but I searched and haven’t found anything on the forums that details this so I decided to write up a guide since I’ve used this enough to get the timings pretty well I think.
A little about this build: this will not work well in the higher leagues. This build is great for Bronze-Gold and will easily get you into Platinum. After that this will work about 50% in platinum and way less in diamond (which is where I’m at now and struggling when using this build). Also depending on how you define cheese/all-in, this build can be considered cheesish/all-in since you’re staying on one base as Zerg for at least 7 minutes so if you don’t like that then certainly do not use this build. This is only a ZvT or ZvP build. I think I’ve tried this maybe once against a Zerg and clearly failed horribly since you’ll get attacked way before you have anything to defend with.
The nice thing about this build is at the lower levels it’s not too timing dependent and is actually semi flexible to mistakes/ or variations. Sometimes I vary my BO slightly and it still works out pretty well (extractor trick instead of 9 overlord etc…) . Also at the lower levels it doesn’t require micro since once you have your roaches you can A-Move them and usually destroy your opponents army.
+ Show Spoiler +9 Overlord
14 Pool
14 Gas (send three drones as soon as gas is done)
17 gas (wait until pool is almost done to get a little more minerals)
16 queen
18 overlord
after the overlord pops send the next 3 drones that you build on the second gas
21 As soon as the queen pops lair and roach warren
26 overlord
26 overlord
26 overlord
As soon as the lair is finished put down a nydus network
26 roaches the rest of the way until the nydus worm pops
As soon as the nydus worm is finished put down the nydus worm in his base
Overlord positioning: In the very beginning if it's a 2 player map send your first overlord to the edge of your opponents base. Every map has a spot that is relatively close to the mineral line but still in the FoW (I'll talk more about this later on in the maps discussion). If it's a four person map send the Overlord to the closest air position first to see if that's where your opponent is. If it is then send your overlord back to the spot in the FoW. If your opponent didn't spawn in the close by air position then right away send you overlord towards the other bases. You will figure out which of the two bases your opponent is at with your 14th drone. Basically once your pool is down just send a drone to the closest of the remaining two bases. Also always send your second overlord to their base as well. Again almost every map will have an alternative good place to put a second overlord in case the first spot gets scouted before you put down your worm. For Shakuras and Delta Quadrant the way to scout and send your overlords is actually different. See below
MapsThe best map for this build is Scrap Station. I actually think the designers of the map were thinking of exactly this strategy when they were making this map. Send your first Overlord clear accross their base to scout and leave him on the other side. Send the second overlord to the edge directly "above" the mineral line. This is the ideal spot to drop a nydus, however if your opponent has built a depot/pylon there then just place it on one of the sides. I will also send a third overlord to the other side so that I completely surround his base
In Blistering Sands there is a perfect little outcropping that is perfect for a nydus (top left corner for the 2 o'clock position bottom right for the 7 o'clock position). I always send the second overlord to the opposite end of the base it's both a good place to drop a nydus and a good place to watch the choke through which you can see whether or not he's moving out with an early rush.
For Xelnaga caverns I send the overlord to the extra rectangular area that each base has. In Xelnaga there really isn't a good place for a second overlord so I usually only send one.
In Steppes I just send the overlords to the two sides. One right over the big statue and the other on the opposite side of it.
In Jungle basin I send one to the back corner of the base that is by the natural. On this map it's important to first send your OL to the 12 or 6 oclock position of the map first and then send him to the back of the base hugging the upper or lower edge. Otherwise he will get spotted and taken out (especially if it's a terran)
On Lost Temple there is a huge area in the back of each base that your first OL should go to. The second OL should be on the same edge but much closer to the center of the map.
On Metalapolis the first OL should be in the wall of gas corner, the second should be in the corner closest to the center of the map.
Shakuras plateu is so huge that if you spawn on opposite corners the it's hard for the OL to arrive there in time. I usually don't nydus on Shakuras, however it is still possible just scout earlier and send your OLs to the two back corners of the base.
Delta quadrant is the worst map for this because if you spawn in the clockwise close position to your opponent it will take forever for hte overlord to get to the right spot (the edge on the other side of the rock expo) therefore I just avoid this map in general.
Other notes about this build In general my goal is not necessarily to defeat my opponent out right with the first pop. Once my nydus pops and my roaches are in his base I obviously go for his mineral line which automatically forces him to pull his workers. I then scout the whole base with my roaches. depending on the army he may or may not be able to defeat my roaches out-right. If he does then I obviously asses the situation and decide what to do next (try to build another nydus with my second OL to try to finish this game, expand and tech switch etc...) If I see that they are going air (a banshee or a VR rush) I will immediately put down a hydra den and evo chamber and build another queen. My initial roaches will obviously die but not before doing enough damgae to hold him off while I build hydras and another worm if he destroyed my first one. Basically if this doesn't win outright it's usually at least a good first harass that should put you at least on even economic footing with your opponent and allow you to safely expand. Obviously the worst possible outcome is for them to spot your nydus and destroy it before it finishes. In this case your chances of winning are slim (which is why as I mentioned before this is a sort of an all-in build). If they do spot your nydus worm then try to build one with your second OL. Quite honestly this works more often then it should since people will usually be happy that they spotted the first worm and not go searching for another one.
A quick note on preventing scouting. Obviously it is important that your opponent does not know exactly what you are doing. Therefore it is extremely important to prevent your opponent from scouting you while your nydus network is building. If you're playing against a terran and he hasn't sent in an SCV yet then I would recommed building your worm away from your main base with an OL spewing creep that way you shouldn't worry too much about being scanned. Other then that it's not too bad if your opponent sees what you are doing since in my experience I don't think a lot of people will think I'm going nydus worm when they see me going with the early lair and actually often if I can let a terran see an early lair but not let them scout the roach warren they think I'm going early mutas and I come in to their base with roaches against marines and towers in their mineral line. Obviously it's an easy win in that case.
Replays
http://www.sc2replayed.com/replays/137740-1v1-protoss-zerg-lost-temple
http://www.sc2replayed.com/replays/127204-1v1-protoss-zerg-blistering-sands
http://www.sc2replayed.com/replays/137751-1v1-terran-zerg-lost-temple
http://www.sc2replayed.com/replays/137754-1v1-protoss-zerg-lost-temple
A couple of different replays just to show the different variations this build can take. One is against a 2700 diamond Toss so it can still work against diamond players however I must admit it's getting harder and harder to win with this build once I get higher and higher on the ladders.
Another replay is of a Terran going banshee just to show that it can even work against air.
Conclusion that's basically it. Again I know it's nothing revolutionary but I haven't seen the build order for this yet (either on the forums or on Liquipedia) so I decided to write one up since I've gotten a lot out of this community and wanted to contribute something of my own. Please let me know what you think. Comments, questions, and concerns are welcome.
|
What is the point of building 3 overlords at 26 supply, Is it best for the zerg to make the exact number of overlords he needs then to make the units when hes doing a 1 base attack?
|
I haven't had a chance to see reps yet - but looking at your BO it seems a bit sus. Surely there's massive window your leaving open to early mid game attacks from P or T. By 26 food I will have had to fend off 2-3 pushes depending on what tactics the enemy is doing.
Investing such a huge amount of gas for what is essentially a mineral line harras is questionable, and I doubt this will work more than once in a game either so the returns aren't going to be massive.
If your thinking of doing this kind of harras you're much best off going for overlord ling or bling drops as these can be used throught the game - ovie carpet bombing an army etc. Roaches aren't ideal either, though enough can one shot a worker I think it's way too much of an investment to sucide them in this way. Afterall, wouldn't 7 or 8 zealot counter attack be seriously dangerous - you wouldn't really have much of a home army? A marine ball would also threaten you badly straight after.
The returns are no way near high enough for this to be a game winning cheese or all in I'm afriad.
Nydus worms are awesome don't get me wrong but such an early nydus push will killl your economy. I'm in silver and few of my recent opponents would fall to this.
I'll watch your replys and may eat my words but I don't thinks so. Thanks for the suggestion though Fry Bender!!!
|
The build is what it is, I don't really see the relevance of a build that can't work at a decent level though.
That said I don't know if I'm convinced that this strategy can't work even though the OP admits it. I think it would all be in the execution, usually the case with everything, almost any strategy will work even in mid masters from my understanding if it is well executed.
|
May I suggest an improvement?
Don't use roaches, use speedlings. Fast expand for the extra larvae, and because it better disguises that you are doing a nydus build. Add a 2nd geyser and start speedling upgrade while lair is morphing. With 2 geysers, you'll have 200 gas by the time lair finishes, and another 100 gas by the time the nydus finishes.
With 2 hatch speedling you can nydus in like 35 - 50 speedlings, with your original build you are probably nydusing in about 8 - 12 roaches. Speedling DPS is much better, so you can do more damage faster, and they perform well in an open main, their only drawback being it takes a while to nydus in mass speedling from one worm.
|
Looks really good.
I have to point out though that it might be a bit all-in...
|
On February 11 2011 17:16 Cabracan wrote:
What is the point of building 3 overlords at 26 supply, Is it best for the zerg to make the exact number of overlords he needs then to make the units when hes doing a 1 base attack?
The reason why I do 3 overlords at once is because I actually supply lock myself at 26/26 drones so there isn't anything else I can make with the extra larvae. Now the reason why I supply lock myself is becuase I want to get my econ up as fast as possible and there is no need for any roaches until the nydus worm so the timings actually work out that way.
Deleuze wrote
haven't had a chance to see reps yet - but looking at your BO it seems a bit sus. Surely there's massive window your leaving open to early mid game attacks from P or T. By 26 food I will have had to fend off 2-3 pushes depending on what tactics the enemy is doing.
Actually that's very rarely the case. In all the games that I have played using this build (and it's been a lot) I very rarely get harassed before the nydus worm is out. I'm pretty sure that it's because I'm not fast expanding and my opponent is weary of sending anything in early to a one-basing Zerg. Actually the best thing for this build is for that initial push to come at around 6:20-6:30 right as my nydus netwrok is finishing up. I have plenty of roaches to easily stop the push with the support of the queen and even drones if needed and even though he spots my nydus worm he has no forces in his base to stop it and if he pulls all his workers it usualy is still not enough time to stop the worm from finishing and then I can instantly send the remaining of my forces to his base.
Deleuze wrote
If your thinking of doing this kind of harras you're much best off going for overlord ling or bling drops as these can be used throught the game - ovie carpet bombing an army etc. Roaches aren't ideal either, though enough can one shot a worker I think it's way too much of an investment to sucide them in this way. Afterall, wouldn't 7 or 8 zealot counter attack be seriously dangerous - you wouldn't really have much of a home army? A marine ball would also threaten you badly straight after
The reason why this is better then an OL drop is obvious I think. As soon as the nydus worm pops you will have forces in his base including the 2 roaches that just finished building 2 seconds ago. With an OL drop you also don't get the instant reinforcements streaming into his base. Finally if he does counter attack you and tries to go for a base trade then just put all your roaches back into the nydus after you did enough damage but before his counter arrives and stop his counter attack. Again the beauty of this build is being able to basically have your whole entire army in two places at once.
telfire wrote
The build is what it is, I don't really see the relevance of a build that can't work at a decent level though.
That said I don't know if I'm convinced that this strategy can't work even though the OP admits it. I think it would all be in the execution, usually the case with everything, almost any strategy will work even in mid masters from my understanding if it is well executed.
While I never said that this build won't work at all, it does becomes much harder to get the nydus worm in their base due to better pylon/depot placement as well as better scouting and just paying attention better to the minimap in general. But that being said I absolutely agree that if you can actually get the nydus worm into his base then yeah this will probably work even at the master level since 10-12 roaches at 6:40 minutes is not easy to stop once they're actually in your base.
|
It's important that your build have a transition from a completely standard opener, and that you're only doing this if you scout something strange like forge fast expand, or wall/tech terran.
Further, you need to show that you can beat these tech timings at a critical window.
Also, you may want to look at this months old post regarding a nydus wyrm 7RR:
http://www.reddit.com/r/starcraft/comments/e0qcr/new_evo_chamber_discovery_7_roach_nydus_worm_rush/
as well as this one, which is the speedling variant mentioned in an earlier post:
http://www.teamliquid.net/forum/viewmessage.php?topic_id=184904
Personally, I recommend never doing an offensive, unprompted nydus on a turtling player. Nydus should be reactionary, and one of the golden rules of sc2 is that you should never drop a turtling player.
That is exactly what your build and the two above aim to due, and as such, if he spots it, since his units aren't far away, your rush may fail.
Another golden rule of SC2 is that if the hard counter to your build is simply scouting it, your build is probably not good. If any player scouts your double gas and/or fast lair, he will simply attack and kill you or inflict irrecoverable damage.
|
A Nydus network in general is a good building to have but I don’t believe you can base an entire strat around it. I like to use the Nydus for Islands, furthest bases from rally point, and potential sneak attacks. Also it is great for defending your mineral line if you get late game against terran, when you got like 4+ bases. When the terran drops on you load up the nydus and tell them to go mine somewhere else and just bounce them around until units can kill off the drop.
|
On February 12 2011 01:27 michaelhasanalias wrote:It's important that your build have a transition from a completely standard opener, and that you're only doing this if you scout something strange like forge fast expand, or wall/tech terran. Further, you need to show that you can beat these tech timings at a critical window. Also, you may want to look at this months old post regarding a nydus wyrm 7RR: http://www.reddit.com/r/starcraft/comments/e0qcr/new_evo_chamber_discovery_7_roach_nydus_worm_rush/as well as this one, which is the speedling variant mentioned in an earlier post: http://www.teamliquid.net/forum/viewmessage.php?topic_id=184904Personally, I recommend never doing an offensive, unprompted nydus on a turtling player. Nydus should be reactionary, and one of the golden rules of sc2 is that you should never drop a turtling player. That is exactly what your build and the two above aim to due, and as such, if he spots it, since his units aren't far away, your rush may fail. Another golden rule of SC2 is that if the hard counter to your build is simply scouting it, your build is probably not good. If any player scouts your double gas and/or fast lair, he will simply attack and kill you or inflict irrecoverable damage.
Thank you for the Reddit link. I think my build is better then that since my nydus worm pops at 6:40 (30 seconds later) but I have 10 more drones 2 more overlords and 3 more roaches so it's not as all-in as that build.
As far as the speedling nydus worm rush goes in my experience that just doesn't work. and the problem is that it takes the speedlings way too long to get out. By the time your opponent's army finds your worm (approximately 5 sec after it pops) I can only get out 10 speedlings after that the army literaly kills a speedling every .5 seconds as they come out of the nydus. It's literaly like the speedlings line up to get killed one at a time. Getting ten roaches to come out in 5 seconds is a whole lot better as 10 roaches in a line is a nice formidable force.
Unfortunately I don't have any replays but I think this build is a hard counter to Toss's Forge FE. I have not once lost to a Toss FE using this build. I'm not sure on the timings of a Forge FE but the nydus worm gets 10 roaches in their base at 6:40. With another 2 roaches coming in at 6:50 and then as many reinforcements as you can make. In y experience an FE toss will not have anywhere near an army enough to stop that kind of a push. Usually the way that game goes is he transfers all his drones to the expansion and builds as many cannons as he can to stop me from goign down his ramp. I destroy his main base and keep building roaches after which if he has built too many cannons to push into his natural with my roach army then I just contain him since I now have a huge army of roaches and full map control and he has only an expo with nothing but too many drones and too many cannons.
I'm not sure why you wouldn't want to do a nydus worm on a turtling player? Usually at 6:40 a terran will have a bunker or two and a siege tank at the entrance. I'm nydusing into the back of his base which means that the bunker is useless and the siege tank has to unsiege. For a turtling toss he will usually have a couple of cannons at his ramp, again useless. My biggest problem with this build (basically the only time when I automatically lose even after poping my nydus into his base succesfully) is against a terran who has a big bio ball with an emphasis on marauders vs marines. And that's because marauders kill roaches so well that I can't even do enough damage with my initial ten roaches to make a difference. After this the terran just counters me and it's gg for me. (which is why I try super hard to prevent a terran from scouting my roach warren for as long as possible. But against turtling players I have had really good success. So I'm not sure I see the logic of not nydusing a turtling opponent?
|
On February 12 2011 01:28 P0ckets wrote:
A Nydus network in general is a good building to have but I don’t believe you can base an entire strat around it. I like to use the Nydus for Islands, furthest bases from rally point, and potential sneak attacks. Also it is great for defending your mineral line if you get late game against terran, when you got like 4+ bases. When the terran drops on you load up the nydus and tell them to go mine somewhere else and just bounce them around until units can kill off the drop.
I think that in bronze-plat you could base your whole strat on it. It got me into diamond, for whatever that's worth. And I agree with you on taking island expos. On Scrap station my first expo is always a nydus on the island. If the game goes to that point my fourth is a nydus into the expo at the very bottom that's protected by rocks (again nydus). Same goes for Lost Temple
|
On February 12 2011 02:08 FryBender wrote:Show nested quote +On February 12 2011 01:28 P0ckets wrote:
A Nydus network in general is a good building to have but I don’t believe you can base an entire strat around it. I like to use the Nydus for Islands, furthest bases from rally point, and potential sneak attacks. Also it is great for defending your mineral line if you get late game against terran, when you got like 4+ bases. When the terran drops on you load up the nydus and tell them to go mine somewhere else and just bounce them around until units can kill off the drop. I think that in bronze-plat you could base your whole strat on it. It got me into diamond, for whatever that's worth. And I agree with you on taking island expos. On Scrap station my first expo is always a nydus on the island. If the game goes to that point my fourth is a nydus into the expo at the very bottom that's protected by rocks (again nydus). Same goes for Lost Temple
Nydus Worms are good to use when you start bringing home the real bacon and are having trouble keeping up with the macro.I mean having basically a teleport is great especially for hydras/queens or any other slow unit.
|
Don't forget to use queen's creep tumor in enemy base once the nydus pop, even if your initial force is destroyed, the creep will prevent the enemy to build or worse if you focused the command center and forced him to lift. Creep tumors give you vision for another nydus.
|
As far as the speedling nydus worm rush goes in my experience that just doesn't work. and the problem is that it takes the speedlings way too long to get out. By the time your opponent's army finds your worm (approximately 5 sec after it pops) I can only get out 10 speedlings after that the army literaly kills a speedling every .5 seconds as they come out of the nydus. It's literaly like the speedlings line up to get killed one at a time. Getting ten roaches to come out in 5 seconds is a whole lot better as 10 roaches in a line is a nice formidable force.
Yeah that's a good point. Like I mentioned, the speedlings take longer to exit the nydus than the roaches, so its definitely a consideration when deciding between the two. If you want to be really annoying send drones through the nydus to make spine crawlers (or spore crawlers if you know he's got air), but don't destroy your economy doing this kind of stuff. The aforementioned Queen to lay a creep tumor is a great technique, versus protoss especially. I had a game back in beta where I laid two creep tumors in his base during a nydus raid, and within a few minutes I quietly had creeped 75% of his main on Desert Oasis (no forge or observers, haha).
In general I would only nydus a player who was teching and/or fast expanding. If that is what you mean by turtling, then sure. But many players turtle while building up a huge tier 1 - 1.5 force, so be wary of nydusing into a death trap like that, as it won't end well with you being down 300 gas (400 if u count lair) from your nydus play.
|
On February 12 2011 02:04 FryBender wrote:Show nested quote +On February 12 2011 01:27 michaelhasanalias wrote:It's important that your build have a transition from a completely standard opener, and that you're only doing this if you scout something strange like forge fast expand, or wall/tech terran. Further, you need to show that you can beat these tech timings at a critical window. Also, you may want to look at this months old post regarding a nydus wyrm 7RR: http://www.reddit.com/r/starcraft/comments/e0qcr/new_evo_chamber_discovery_7_roach_nydus_worm_rush/as well as this one, which is the speedling variant mentioned in an earlier post: http://www.teamliquid.net/forum/viewmessage.php?topic_id=184904Personally, I recommend never doing an offensive, unprompted nydus on a turtling player. Nydus should be reactionary, and one of the golden rules of sc2 is that you should never drop a turtling player. That is exactly what your build and the two above aim to due, and as such, if he spots it, since his units aren't far away, your rush may fail. Another golden rule of SC2 is that if the hard counter to your build is simply scouting it, your build is probably not good. If any player scouts your double gas and/or fast lair, he will simply attack and kill you or inflict irrecoverable damage. Thank you for the Reddit link. I think my build is better then that since my nydus worm pops at 6:40 (30 seconds later) but I have 10 more drones 2 more overlords and 3 more roaches so it's not as all-in as that build.As far as the speedling nydus worm rush goes in my experience that just doesn't work. and the problem is that it takes the speedlings way too long to get out. By the time your opponent's army finds your worm (approximately 5 sec after it pops) I can only get out 10 speedlings after that the army literaly kills a speedling every .5 seconds as they come out of the nydus. It's literaly like the speedlings line up to get killed one at a time. Getting ten roaches to come out in 5 seconds is a whole lot better as 10 roaches in a line is a nice formidable force. Unfortunately I don't have any replays but I think this build is a hard counter to Toss's Forge FE. I have not once lost to a Toss FE using this build. I'm not sure on the timings of a Forge FE but the nydus worm gets 10 roaches in their base at 6:40. With another 2 roaches coming in at 6:50 and then as many reinforcements as you can make. In y experience an FE toss will not have anywhere near an army enough to stop that kind of a push. Usually the way that game goes is he transfers all his drones to the expansion and builds as many cannons as he can to stop me from goign down his ramp. I destroy his main base and keep building roaches after which if he has built too many cannons to push into his natural with my roach army then I just contain him since I now have a huge army of roaches and full map control and he has only an expo with nothing but too many drones and too many cannons. I'm not sure why you wouldn't want to do a nydus worm on a turtling player? Usually at 6:40 a terran will have a bunker or two and a siege tank at the entrance. I'm nydusing into the back of his base which means that the bunker is useless and the siege tank has to unsiege. For a turtling toss he will usually have a couple of cannons at his ramp, again useless. My biggest problem with this build (basically the only time when I automatically lose even after poping my nydus into his base succesfully) is against a terran who has a big bio ball with an emphasis on marauders vs marines. And that's because marauders kill roaches so well that I can't even do enough damage with my initial ten roaches to make a difference. After this the terran just counters me and it's gg for me. (which is why I try super hard to prevent a terran from scouting my roach warren for as long as possible. But against turtling players I have had really good success. So I'm not sure I see the logic of not nydusing a turtling opponent?
Regarding the bolded text:
You could attack at 7:10 and maybe have 20 more drones and 8 more roaches.... or attack at 14 minutes and have 3 bases and 60 roaches. The point is that your nydus needs to hit a critical timing window where the terran (or whomever you do it to) can't beat it if he does X. For example, if he rushes banshee, that 6:10 rush will beat him because he doesn't have anything but a few marines (killed by roach). If you wait until 6:40, the banshee will be out and kill all your roaches while he pulls his workers to his ramp.
Also, 6 workers have the dps to kill a nydus worm, so if he spots it and ONLY has workers, he can still defend.
I'm not claiming that your build is bad and/or not useful (although it does appear to not hit any relevant timing windows), but I would like to see some evidence of two things:
1) How do you transition from your build if you scout something that would hard counter it? (i.e. 2-rax push, 6rax marine/marauder, any 4wg variant)
2) What builds does this one offer a hard (or soft) counter to? I believe any early unscouted nydus play beats forge FE, but this is reactionary and not a build.
Here's a game where I did a reactionary nydus vs forge FE. It hit much later but remained as effective because it was unscouted, and was able to hit the timing window (which is LARGE for forge FE if unscouted):
 vs  Michael vs FECKIT ![[image loading]](http://www.topreplays.com/img/layout/download.png) reactionary nydus vs. forge FE
|
|
|
|
|
|
EPT
