Yeah, my bad for misunderstanding the scenario you were talking about. I was thinking about a much later point in a game.
1 Like
The thing that offsets this is a cannon rush is almost guaranteed to fail if itâs caught early enough.
See your nat being cannon rushed? Wait for the cannons to complete, cancel the hatch and remake it at ur 3rd/4th, etc. Then go back and taje care of the cannons/pylon later before it becomes a warp gate beach head.
Where cannon rushes do the best is when the opponent does the wrong things to beat it. Now if weâre talking about PvZ proxy 3 gate cannon rush then it gets blatantly unfair for the zerg, you canât really shut that down easily because thereâs no set in stone counter.
1 Like
It was scouted though. He sent over a probe scout, saw no expansion, saw the gap in the minerals, checked for a hidden natural behind the mineral wall, then went up and scouted it directly.
The scout directly transitioned into Stats holding the all-in, and even with the scout Shin nearly ended the game. It was purely on Statâs micro - and itâs unquestionable that he did micro - that caused him to hold even with a setup in the main base to prevent his death on the all in.
Shin also had multiple opportunities to take the consolation prize of killing the natural nexus and at least put himself on better footing, but he didnât. He also got supply blocked multiple times when he lost a few overlords and the proxy hatch to the single void ray after it got out.
Scouted after making critical investments. You donât need a nexus in that scenario. 1 chronoboost hard counters it because you canât mine out the minerals. 1 adept shuts that rush down.
No, he had nothing out on time. He was fighting banelings with probes. His scout was too late to be useful. The take away is that Probes and a shield battery are enough to beat a proxy hatchery.
1 Like
Nope, doesnât work. Protoss cannon the other locations. You lose 300 minerals and he loses 250. He gets an expo and you are stuck on 1 base.
1 Like
Yeah I know for a fact youâre wrong lol. No discussion to be had, move along.
1 Like
https://i.imgur.com/tnonGVA.png
They eventually nerfed this build. I was one of the few people on planet Earth to be able to beat partingâs version. 3 hatch before spawning pool beats this rush. 14 hatch into 20 hatch is the exact timings, and it works because the creep is so fast, at the natural, that it blocks buildings, and the drone transfer looks like a drone-pull allin. So, you have to use drone pull allins on occasion or the protoss wonât buy into the lie. You actually transfer them to your third. You can verify all this by looking at the minimap.
The only pro player that I saw beat proxy robo, during its prime, was Serral. It was unbeatable if you played standard openers, and nobody sane would try to beat it with 3 hatch because itâs such a counter-intuitive way to solve the problem, but itâs the correct way to beat it. This build is a strategic hard counter to proxy robo. Serral beat it with a standard opener, which is very impressive in its own right.
Serral has only recently started to do hatch-before-overlord builds, which I started doing back in 2018 because I was so fed up with cannon rushes & this hard countered them. Here is Serral using hatch-before-overlord to beat Heroâs cannon rush (hero looks like a clown for even trying):
https://www.youtube.com/watch?v=6NxRSMULsUs&t=4525s
This is why I was easily beating parting on a regular basis. Hatch before overlord is an extreme hard counter to cannon rushes. The problem is that itâs vulnerable to a protoss who recognizes the hatch timing and chronoboosts probes instead of cannon rushing. The fact that hero didnât know the hatch timing tells you how uncommon it is to counter cannon rushes like this. They can also take a faster third because you have a later zergling speed. Itâs a losing build order if the Protoss has his head screwed on right, but vs a cannon rush abuser itâs a free win.
It has a few advantages. An obvious one is that itâs easy for a weak protoss to misread the build (they think itâs a normal hatch first, which is vulnerable to cannon rushing) so it baits weak protoss into a free win for the zerg. The other is that if he sends his probe to block the third, you get a free kill on his natural pylon and a few probe kills (you sneak out 8 lings). You can drone pull allin half the time if youâd like, and you can even do it with the third base.
I beat parting so many times with the drone pull allin variant which is why in the game above he thought the drones leaving my main were going to his natural, when in reality they were off to the third. This is the power of mixing in drone pulls and proxy hatcheries. Protoss have a very large early game vulnerability in that they canât tell what build you are doing if you mix in these few variants.
TLDR, Iâve beaten the best cannon rushers on Earth, and yet Bnet forum clowns will tell you âYeah I know for a fact youâre wrong lol. No discussion to be had, move alongâ like youâre some noob who has no idea what heâs talking about. Welcome to bnet. Land of the delusionally enlarged egos.
2 Likes
OK I really donât care, write another essay. Iâve seen it work, just because you canât and think youâre God doesnât mean youâre right. GL bro.
1 Like
If the cannon rusher sees youâre making no effort to save your expansion they wonât commit as many cannons and the ones that know how to do it properly always send out a second probe to scout and continue the rush if you kill the first probe for example.
Hereâs some tips: https://www.youtube.com/watch?v=GStkjoQhVjY
are we talking about vision or scouting?
Because creep is not really scouting. first oracle ~4min, I canât scout much with creepâŚ
Oh yeah, not impressed. Ah huh. Parting was the best cannon rusher on earth. Beating him is the definition of having beaten cannon rushing, lmfao. Code s players were knocked out of the gsl with those builds. But yes, you are right about cannon rushing because you say so. No supporting evidence or game theory. Danktemplar says all you have to do to beat it is expand again. Boom. Problem solved.
I really, really donât care about what you say. Bad reaction to cannon rushes lose games at all levels of skill. Code S doesnât mean infallible. Especially when it comes to hardcore cheese builds.
P.S. Gaulzi is one of the best cannon rushers out there, Parting is just mid at it. Good execution but zero inspiration. None. The best cannon rushes happen with eco cheese behind it or with some form of tech for a 1-2 punch to capitalize on traditional responses to cannon rushes.
Code S doesnât see that crap all too often, they have their own meta and their cheese is TAME by ladder standards.
1 Like
Code s meta doesnât include cannon rush because top zergs/terrans are ridiculously good at shutting it down. Parting took it 1 step higher and made it work. Some 4.7k cannon rusher from eu doesnât compare. Part of ehat makes cannon rushes good is that you pair it with strong play in general. Parting is a world champion for crying out loud, so he had outstanding play in general. Vs a 4.7k cannon rusher, you can make loads of mistakes and still recover because the rest of their play isnât refined. Vs parting, 1 drone kill to many and itâs gg because his followups are so fierce.
Serral narrowly beat this proxy rush and bnet forum users will legit say a 4.7k player wouldâve done better than parting did. Itâs so detached from reality that youâre practically talking about bigfoot and unicorns at this point.
If I had to rank famous cannon rushers, Iâd say sos then parting, then grimmy. Grimmy is an eu gm with about 5.7k mmr.
The difference is thereâs a dance when itâs pros matching other pros. Harstem will wait to see a drone pull before slapping down extra cannons which leaves him more vulnerable to stuffing it up unlike like your average pleb. The amount of times Iâve stopped a cannon rush in its tracks only for a second probe to turn up and dump all their resources into cannons is astounding. I think team game maps have much better cannon rushing spots and thereâs an expectation due to matchmaking to hold it off efficiently.
The thing that a lot of toss donât understand is that what separates a successful cannon rusher from a clown is the ability to control the ramp. Denying the natural is not sufficient to win. You have to have control over the ramp. Without control over the ramp, the zerg can cancel and double expand at the third and fourth. He can get units out and threaten a counter attack. He can transfer workers. He can get units onto the low ground and bypass the cannons to attack the pylons directly. This is what Zergs did back in WoL when they were just figuring this stuff out. Youâd see games where top Zergs were taking their natural as their 4th base. The cannon rusher is committed to that position and he needs to keep the zerg committed to that position too. Ramp control was identified as a crux of successful cannon rushing literally 10 years ago, itâs just that SC2 knowledge has regressed. Itâs actually rare knowledge nowadays.
Thatâs one of the reasons why my styles were absolutely busted vs cannons. Proxy hatcheries bypass the âramp controlâ issue very well. My winrate was so high vs cannon rushers that I practically considered it a free win. This was true even during the proxy robo and proxy stargate era, when it was virtually impossible to win as zerg at a high level. Things became more complicated when they buffed stargate openers, because void ray mobility is a significant threat to a zerg who bypasses the cannon issue by just expanding elsewhere. You are more spread out that normal, and void rays at one point were so busted you couldnât keep your main and third even if you kept them tightly packed. So good luck keeping them when your third is your nat and your fourth is your third.
To be frank, there are no cannon rushers with the same level of skill as what existed during 2018. They are all gone. The knowledge has been lost. Cannon rushers today are total clowns who have no clue what they are doing. Itâs sad to be frank. The only cannon rusher that Iâve run into in recent days who came even remotely close to the difficulty of 2018 toss was a 5.7k protoss called ChaeWon who would stream snipe me. Itâs amazing to see the difference between the current protoss and the old protoss styles. Huge difference in skill level. Cannon rushes used to be impossible to beat without refined responses, now itâs just keeping your second overlord at home and pulling drones the moment you see a probe go towards the minerals. There was some serious mind-reading voodoo required to beat protoss because theyâd know how to fake you out and how to commit after faking you out about faking you out. It was legit poker inception embedded in SC2. Youâd have to practically be a mind reader to know, based on probe movements, whether it was legit or a fake.
Thatâs why I started doing the hatch before overlord builds. I realized the faster creep impeded cannon placement & faster lings killed any cannons that would normally complete. In addition to that, it would outright kill most gateway players. There used to be a 6k mmr protoss called âPlayaâ who would die literally every time I faced him to the first 8 lings made off of a 14 hatch 16 pool. Youâd get a pylon kill at their natural and like 8 probe kills. He started blind chronoboosting a zealot against every barcode he faced because of me. Oh, he also used to camp a probe behind his zealot because I bypassed his zealot with a mineral walked drone in conjunction with the 8 lings. Toss were so clueless on how to deal with hatch before overlord builds that theyâd blind make a zealot and pull a probe off mining to camp behind him.
It used to be that maps were very tightly packed at the natural, and thatâs what made cannon rushes good. But thatâs also what made hatch before overlord builds such a hard counter. Tightly packed natural means creep spread blocks the cannons more. You can see that in the screenshot vs parting. He canât build anything around his proxy because itâs covered in creep.
i think cannons are good in pvp. i think they are bad in pvz. protoss doesnât get creep, they canât produce military out of their nexus, so they have more things to defend (gateways, pylons, hatcheries). ive been playing about two months and i am 3.7k ish with both random and protoss. i played a ton of broodwar. the first thing i thought in this game was wow⌠cannon rushes are way better. and they are, but they still suck against zerg
Zergs only have to defend their natural with workers, which is conveniently located right next to the mineral patches they mine from. If you send 4 workers for each building toss makes , youll always come out ahead econ wise. the nice part is, unlike terran and protoss who have to defend with 4 workers as well, zerg workers can immediately go back to mining after destroying the pylon or toss canceling (remember to shift queue)
zergs can move their overlords, pylons cannot move
zergs can also 14 hatch like others have mentioned
zergs can also hold the ramp and mine gas during a cannon rush, something a protoss cannot. you cannot cannon the zerg main
zerg also has access to ravagers on hatchery, so they can immediately threaten a counter attack
zerg can also just proxy hatchery in your base while this happens and force you to cannon two locations
if you have issues defeating a (scouted) cannon rush with zerg in sc2 you are straight up bad. if you hear our hive cluster is under attack and expect to magically hold it at that point, youâre not going to.
heres a little guide for you adventurer:
pull 4 drones and click every new building he makes
in this video, serral fights with drones against cannons for no reason, he was never in any danger of dying here. lol. the creep was already too far.
those two cannons left at 2:10 cannot kill the hatchery with the additional lings
also, he scouted it after the pylon finished. the first two cannons are 75% finished before a drone even hits it. he still has enough money for 8 lings, and a queen after the drone pull
Serral loses 5 drones, is 5 workers down, but has an expansion. he holds the rush at 2:10. by 3:15 he is already even in workers. hero at that point has not even started warpgate.stargate is done 1.5 minutes late, serral has creep out to his 3rd before the adept even gets across the map.
so yeah its not the exact game state zergs are used to playing in, but i think most zergs would take ^ this game state over the default any day of the week. hero was also mining with 1 probe out of a gas from midway through the rush until the end, and still his tech is that late.
zerg holds the rush after scouting canons at 75%. you know which other race in sc2 can do that? none bro. that is auto lose in both other matchups
You are right but for the wrong reasons. 1 cannon was in range of the hatchery and the other 2 are there to protect the 1. 3 is the magical number needed to bring down a hatchery no matter what (as long as they can complete). No number of lings will be able to bring it down if the buildings are properly walled.
The reason serral shouldnâtâve pulled the drones is because 99 times in 100 the protoss will properly wall the cannons and the drone pull will do nothing. That protoss is just absolute dumpster tier to the point that walling in some cannons is apparently too hard.
Yes, having double the bases & double the worker production turns out to be pretty good. Being down in workers was more of a choice of build order than it was anything else. Thatâs why I pointed out previously the way to beat hatch before overlord, with a cannon rush, is to not cannon rush.
Properly executed, the protoss wins 99% of the time there. You canât defend a cannon rush with that late of a scout. When done properly, the protoss will invest into 3 cannons and 1 pylon, but after seeing the zergâs response he will cancel the third cannon. Some pylons are built for walling, and those are cancelled at the right time too. He will kill the hatch & he will have control over the zergâs ramp, making it impossible to expand elsewhere.
This protoss did the whole rush wrong. It was in the wrong place, wasnât walled in, he overmade cannons, didnât time the second and third cannons , and didnât cancel the excess cannons. Basically he screwed it up, panicked and tried to compensate by spamming cannons all over the ground. Easily the worst cannon rush I have personally ever seen. He had invested so much that even if he got the hatch kill heâd still be behind because the zerg can double expand.
A cannon rush properly walled wonât let more than 1 ling attack a single cannon. These cannons had a full surround. It was honestly one of the best games of SC in recent years, because it made some people realize how bad the top protoss are (compared to Serral / the top protoss from history). The reason protoss donât win premiers is because they suck, and they suck really bad.
i think he realized he had no chance of beating serral unless he cheesed and got in his head maybe cannon hm a few times and win out of the frustration that follows. i agree that this is a big mismatch
you claim that properly executed, the protoss wins 99% of the time. if this is true, why are protoss not spending hours practicing cannon rushes?
and yes, i got your edit, this was a very bad cannon rush. iâm just saying that often the circumstances that follow this rush, because it is typically not game ending, often have diferent advntages based on how much both players committed to that moment and the moments that lead up to them. serral didnt blindly make a gas for example he scouted with his second ovie and wasnt planning aggression. if he had been a gas pool zgamer well he probably gauges how much protoss is committing to the hatchery, and goes for ravagers.
ravagers are incredibly strong in this kind of situation, very underrated. and lair nydus is also very underrated in this situation as well. when you are behind, you are not dead. its definitely very easy to press an advantage in this matchup but if its playable i think the better player drags it out or presses and wins in the following moments depending on the calculus of commitment.
i get it, the calculus on this rush is incredibly hard to do in the other matchups, and you are truly blind in those. in zvp you can at least scout him with overlords during and after the rush and that can giv eyou some kind of indication of his commitment, and you have less buildings to defend, but also have a ramp to defend. so its a very different thing rush entirely
1 Like
Hero misplaced his first pylon by 1 hex, and that is why his rush failed. Cannon rushes are strategic. Strategy emphasizes quality over quantity. Pro players arenât strategic, they are APM spammers. Modern pro level SC2 is about spamming slightly more apm than the other apm spammer, and getting small advantages from it which compound into a win over a 20 minute game. This emphasizes quantity of actions as more important than quality. Because of this, in this version of SC2, strategy is basically irrelevant. Uthermalâs posting videos of beating GMs with half-HP units. That just wouldnât be possible in a strategic game.
Strategy has no significant impact on the outcome, so players put their focus into what does affect the outcome, and thatâs multitasking. Thatâs why great multitasking players can play a fierce macro game but look like total idiots if they try a cannon rush. Cannon rushes are strategic, the quality of actions matters much more than the quantity, and their brains are wired to spam massive quantities of low quality actions. Macro games are about APM spam where the quantity matters much more than the quality. Making a slight mistake in a macro game is basically irrelevant, making a slight mistake in a cannon rush is catastrophic.
Yes, the outcome is very complex & strategic. Itâs easy for either player to come out behind depending on how things go. But, based on the timing of Serralâs scout, a competent cannon rusher would win that game 99 times in 100. Properly walled, the prootss kills the hatch for what amounts to 1 pylon and 2 cannons invested. He ends up with a faster natural, and control over the zergâs natural (ideally control over the ramp as well).
They are great except that they are useless vs a void ray followup. What zerg needs to deal with a void ray is queens and a tightly packed defense. The cannons force the zerg to expand more spread out, making creep difficult, and adding more attack vectors for the void rays. The dead hatchery also cuts zergâs queen production in half, which is equivalent to half the creep spread. You simply donât have enough money to make drones, re expand, make ravagers, and spread creep. There isnât a viable combination there. The best you can do is to double expand, ignore the cannons, and mass queens. You can keep up in eco, positioning is hard, but you will be very behind in upgrades & tech. Thatâs why the cannon rush needs to be placed so that it has control over the ramps. If you have that, then the zerg canât double expand, and ravagers are the only option. Toss follows up with 1 base stargate, and wins.
It depends on the variation. If itâs a gateway before nexus, he will make a stalker and you have no idea if he took a natural or not. There could be a robo or a stargate hiding behind the cannons, or a 4 gate, or a hidden nexus (effectively a double expand from the protoss). Itâs not a good position to be in.
Forge cannons nexus is less robust, because it only has an economical followup. But, the economy shoots off to the moon in about 0.1 seconds flat, so thatâs the advantage. The zerg is vastly behind, has to jailbreak himself before the protossâs economy has left orbit.
There are loads of other variations. For example, you can do a 1 gate proxy cannon rush, or a 2 gate proxy cannon rush. You can do loads of different followups. Grimmy had a feirce 2 gate followup which was designed to hard counter the ravager response. Heâd make 1 adept to shade up the ramp for vision, and then heâd chrono stalkers off 2 gates. 2 gate stalkers hard counters 1 base ravager.
Then there are the robo and stargate followups, with a variety of timings.
1 Like