EPTDo you have replays against bulldog/immortal busts with a warp prism? I do that vs tech/expand openings a lot and it works really well, i'm curious to see how it does at GM level.
[G] StimmedProbe's 5 Factory Mech TvP - Page 2
| Forum Index > StarCraft 2 Strategy |
Teoita
Italy12246 Posts
Do you have replays against bulldog/immortal busts with a warp prism? I do that vs tech/expand openings a lot and it works really well, i'm curious to see how it does at GM level. | ||
|
Akamu
United States309 Posts
| ||
|
StimmedProbe
Hong Kong98 Posts
On October 26 2012 16:12 Moonling wrote: After watching the Antiga game my only question is why do you need 4 tech lab facts? At no point during that game did you produce 4 tanks at a time, generally you stuck with 3 at a time and just produced a hellion out of the 4th tech lab fact. Could you not get the same production with 3 tech facts, and 1 reactor? And maybe get an amory for +1 instead of that other factory?? My reasoning is because I feel like you are going all-in because if you don't do damage the Toss will be up in tech (He was about 3 min from Storm, and if the push fails he will have his 3rd which means multiple Robo's. Where has you have 0 upgrades, you do have a 3rd, but not in the positon to get out vikings if he pumps out massive amounts of coli after your failed push I get 4 factories with tech lab because that spends your 3 base gas income perfectly. With 6 gas, you earn 672 gas per minute, and 4 factories with tech lab consume 667 gas per minute. You will also spend 100 gas a minute on banshees. Again, you gotta produce hellions from time to time if your macro is perfect, but most of the time I find that 4 tl facts + 1 reactor fact + 1tl port macros perfectly off 3 bases for another sharp timing. Upgrades do not matter very much, although they are nice. Your positioning and whether or not you are sieged up properly determine the outcome of the game much more than a +1 upgrade. On October 26 2012 17:03 blublub wrote: This build seems to have potential. But is it possible to win againts equall oponent that goes immortal/archon ? Tanks bearly do any damage to both the units. Do you incorporate some ghost or more banshes when you see alot of immortal and templar ? Do you kite them with helions to lower their shields ? I always struggle with mech againts immortals. You earn 448 gas per minute off 4 gas, you think you can get 4 obs, 6 stalkers, at least 3 sentires, and immortals and archons out? It ain't happening. The banshees already force at least 3 obs from protoss and 6 stalkers. Check my calculations below. On October 26 2012 17:56 brofestor wrote: prefer a 111 with reactor hellion plus medivac drop into expand before going mech in tvp. with a dual prong hellion harass you can get a lot of probes taht allows you to play mech nicely. in these cases it seems ur opponents arent used to dealing with mech (all going herp derp colossus as if expecting mmm all the time), with correct scouting chargelots/immortals/archons/templar shld stop this push. I crush chargelot/immortal/archon/templar harder than colo based play lol... That's why there are no replays of it. It is a 1a stomp.. You have roughly 3000 gas to spend on stuff to beat the stuff in my screenshot. Go ahead and try to find a mix that beats it lol.. Warpgate + Robo + 6 stalkers + 3 sentries + 4 obs = 1050 gas. You got 2000ish left. Twlight/Templar archives, -350. So you wanna make 6 archons? Oh you want charge as well? Then make that 5 archons.. oh and no immortals then.. On October 26 2012 18:06 Teoita wrote: Cool stuff! Do you have replays against bulldog/immortal busts with a warp prism? I do that vs tech/expand openings a lot and it works really well, i'm curious to see how it does at GM level. I'll ask my practice partners to try it. | ||
|
y0su
Finland7871 Posts
| ||
|
DKR
United Kingdom622 Posts
Would be great to see some replay's with early harrass | ||
|
Salient
United States876 Posts
| ||
|
Methy
United Kingdom74 Posts
Mech will always be the 'catch em with their pants down' in WoL, because if a protoss knows mech is coming, he should never lose. The standard 1 gate expo => 3 gate => colossus + double forge is about the worst possible thing the protoss could be doing. Colossi are not all that useful against pure mech, and upgrades are flat out a waste of time vs mech. As such, any time your opponent gets colossus + double forge, you should treat it as a horrible misread. This makes mech a wonderful laddering strategy, the overwhelming majority of Protoss simply have no idea how to deal with mech, and will play standard and get crushed. On the other hand, those that do respond correctly will similarly crush you with little difficulty. That being said, the strongest thing about this build is the fact that it *looks* so much like a 1/1/1 in the early stages - which will often cause a protoss to cut probes at ~30 at which point they fall too far behind to catch up. I just feel it's always worth mentioning that builds like these can never become the standard because of how bad they are when your opponent is expecting them. (If you don't believe that, consider that Mech TvP is similar to broodwar TvP, but the terran has a few disadvantages - tanks are weaker, hellions are way worse than vultures - while protoss has a few advantages - dragoons can blink, zealots can charge, immortals stomp everything. The only 'new' thing for the terran that could threaten that balance is the thor - which is vulnerable to templar anyway) | ||
|
Methy
United Kingdom74 Posts
| ||
|
`dunedain
655 Posts
Thanks yet again for showing that there are still people out there working hard on furthering the strategies that we Terran rely on to survive throughout these dark times. Will try to refine this out on my practice account, thanks again! | ||
|
brofestor
Singapore101 Posts
On October 26 2012 19:22 StimmedProbe wrote: I crush chargelot/immortal/archon/templar harder than colo based play lol... That's why there are no replays of it. It is a 1a stomp.. You have roughly 3000 gas to spend on stuff to beat the stuff in my screenshot. Go ahead and try to find a mix that beats it lol.. Warpgate + Robo + 6 stalkers + 3 sentries + 4 obs = 1050 gas. You got 2000ish left. Twlight/Templar archives, -350. So you wanna make 6 archons? Oh you want charge as well? Then make that 5 archons.. oh and no immortals then.. i really doubt it...pls show replays of that happening outside of some early immortal/prsim bust, colossus always FARE WORST against mech builds since they are practically shite against mech and the gas will be better use for immortals... when i mean about the compo is that it would have prob at max, a few archons with a mostly chargelot/immortal army, sprinkling of stalker and templar (feedback/storm) for those banshees. All your replays so far just show that ur opponent know how to do colo builds very well but fail to adapt to mech since they probably meet ur build only like once every hundred PvT they have. Upon scouting with obs, the toss should have stop colo production, get super fast 3rd for the gas (since you cannot pressure with mech early) to build chargelot/immortal. and pls they dont need so many obs, 2 plus a cannon or 2 will be enough. If there are a few games where toss PLAYED OPTIMALLY TO MECH and yet ur mech can still crush them, MVP wld be using mech to TvP his way to the bank alr. | ||
|
DKR
United Kingdom622 Posts
On October 26 2012 21:38 Methy wrote: I have a big problem with this sort of guide - something that all 'Mech in TvP' posts fail to recognise - your opponents are essentially misreading your build, and reacting terribly, which is what allows you to win. Mech will always be the 'catch em with their pants down' in WoL, because if a protoss knows mech is coming, he should never lose. The standard 1 gate expo => 3 gate => colossus + double forge is about the worst possible thing the protoss could be doing. Colossi are not all that useful against pure mech, and upgrades are flat out a waste of time vs mech. As such, any time your opponent gets colossus + double forge, you should treat it as a horrible misread. This makes mech a wonderful laddering strategy, the overwhelming majority of Protoss simply have no idea how to deal with mech, and will play standard and get crushed. On the other hand, those that do respond correctly will similarly crush you with little difficulty. That being said, the strongest thing about this build is the fact that it *looks* so much like a 1/1/1 in the early stages - which will often cause a protoss to cut probes at ~30 at which point they fall too far behind to catch up. I just feel it's always worth mentioning that builds like these can never become the standard because of how bad they are when your opponent is expecting them. (If you don't believe that, consider that Mech TvP is similar to broodwar TvP, but the terran has a few disadvantages - tanks are weaker, hellions are way worse than vultures - while protoss has a few advantages - dragoons can blink, zealots can charge, immortals stomp everything. The only 'new' thing for the terran that could threaten that balance is the thor - which is vulnerable to templar anyway) People were saying very similar things about Mech TvZ until a few months ago. For the timing the build hits it's clearly very hard for a Protoss to deflect no matter their composition. If they go for a 2 base play then the Terran can take a defensive advantage and get his/her third up. Until enough high level players truly experiment with Mech TvP it's incredibly hard to say that it definitively won't work. Also, whoever is right or wrong in the mech debate doesn't particuarly matter. The majority of people are not Masters+ and are only really looking for ladder builds anyway. | ||
|
Methy
United Kingdom74 Posts
On October 26 2012 22:08 DKR wrote: People were saying very similar things about Mech TvZ until a few months ago. For the timing the build hits it's clearly very hard for a Protoss to deflect no matter their composition. If they go for a 2 base play then the Terran can take a defensive advantage and get his/her third up. Until enough high level players truly experiment with Mech TvP it's incredibly hard to say that it definitively won't work. Also, whoever is right or wrong in the mech debate doesn't particuarly matter. The majority of people are not Masters+ and are only really looking for ladder builds anyway. It's not about composition - it's about playstyle. Fundamentally, Mech cannot put on substantial pressure until *at least* 10 minutes, and even then usually 13 or 14, they can harass, sure, but that's it. Standard Protoss play involves making sure you're safe from attacks in the early parts of the game. Think about how long it took protoss to figure out straight up 50 food stim timings? A long time, and builds that they do now incorporate being safe against them (early 3 gates + sentries) Against mech this isn't an issue, you play it out like broodwar, a third base off 1 or 2 gates and maybe a robo (though you don't need the robo because of the lack of mines). Taking a third at 6 minutes, walling it in tight to prevent hellion run byes, chronoboosting right up to 70 probes at 10 minutes, ignoring upgrades all together, skipping sentries (except for maybe having 1 to deflect hellions), and then flooding with gateway units/immortals. That is how you deal with mech, and it is incredibly effective. On the other hand, that playstyle will cripple you against bio. If you're trying to match 2 base vs 2 base PvT mech, the Protoss player should always lose, no one is disputing that. This build is immensely strong in those situations because of how efficient mech armies are. The issues with TvZ mech *WERE* simply 'composition' issues - roach busts before siege mode were crippling, but getting early tanks against mutalisks was similarly crippling, the solution was to add in banshees to keep you safe against roaches, and effectively transition straight to thors for protection against mutas. | ||
|
Methy
United Kingdom74 Posts
Any protoss with expertise in broodwar PvT should be absolutely fine against any mech build in SC2, with the one possible exception of mass thor + banshee - but that gets stomped so stupidly hard by feedback that it's hardly an issue. edit - I'd like to actually point out that Thor Banshee was a fairly popular build for a long time in the early stages of the game - scans + thors long range AA taking out observers first, and then cloaked banshees dominating because of the lack of detection. This was extremely strong after they made strike cannons a cooldown - taking away the thor energy. So strong in fact, that they had to revert the change and give thors energy, making the composition extremely vulnerable to templar play (feed back on thors and banshees, air units general succeptability to storm because of a high degree of stacking, etc) | ||
|
Methy
United Kingdom74 Posts
While this build will bring players huge success on the ladder, I feel it should be explicitly stated in the OP that you will be walled against anybody that *does* understand how to respond correctly, but that those players are the extreme minority, even in Master/Grandmaster league. If you're happy to accept that in a small percentage of cases you will lose the game where there was quite literally nothing you could reasonably have done to win it - a fundamental problem with just about any laddering build - then you can ignore this. It is, however, a very prominant weakness with the build. | ||
|
mizU
United States12125 Posts
| ||
|
Nyast
Belgium554 Posts
Mech scares me. Not because I think it's unbeatable, but it's exactly as Methy said: we face this build so rarely on ladder that we don't know anything about timings and good reactions. With more experience I have no doubts it wouldn't be terribly hard to hold it. | ||
|
Methy
United Kingdom74 Posts
On October 26 2012 22:41 mizU wrote: How would I beat this as a Protoss if I face it on ladder? What have you lost to? As I said above, play it like Broodwar. If you're unfamiliar with Broodwar then play it out like this: (Ideal strategy - you know mech is coming before the game starts) + Show Spoiler + - Any 1 gate Expo - 3rd Nexus at ~6:00 (if you're uncomfortable with deflecting hellions off just one gate, do add a second) - Robo asap after this since cloaked banshees will be coming at ~7:30 (should be able to get your first observer out comfortably) - get a single forge at your third, and upgrade attack (not armour) without chronoboosting (upgrades are largely irrelevant against mech - to see this, work out the number of attacks required for stalkers and zealots to kill tanks and hellions with and without upgrades - compare this to the terran side of the picture) - Wall in both your natural and your third to prevent hellions getting into the mineral line (again broodwar style) -Go up to ~ 10 gates (should all be done before 10 minutes) -Constantly produce chronoboosted immortals out of your robo after 2-3 obs You then have two paths to follow: - Double robo with constant immortal production before twighlight OR - Twighlight researching blink then charge - Early 4th base with templar You should be trying to deny him expos as much as possible. Don't ever try to push in to a heavily fortified position. Always try to set up flanks (flanks are your BIGGEST friend against mech) Since he doesn't have mines, he can't really prevent this, as such you can make it a general rule to always attack from *at least* 2 different angles, and pick where you want to engage accordingly. If the game ever ends up split map, get a mothership and go for mass recall into the main (again similar to broodwar style) Sacrificing your entire army to kill off production facilities if needs be, and then flooding with reinforcements (similar to zerg) Your general mentality is that he has the cost effective army, so you cannot win a fight cost for cost. His army is immobile however, so cannot shut down expansions. As such you're playing like a zerg - getting a big economic lead and flooding with low cost units (going for backstabs/base trades are also quite effective) Now ideally you will not know mech is coming, and so will open up 1 gate FE => 3 gate robo. This is less than optimal against mech builds but still winnable *if* you throw down a third as soon as your obs gets to his base to reveal mech, and play as detailed above. | ||
|
Methy
United Kingdom74 Posts
On October 26 2012 22:50 Nyast wrote: I recently faced this as toss on ladder, and despite fast 3 bases, upgrades and a heavy chargelots/archons army, I got demolished despite higher army pop. Mech scares me. Not because I think it's unbeatable, but it's exactly as Methy said: we face this build so rarely on ladder that we don't know anything about timings and good reactions. With more experience I have no doubts it wouldn't be terribly hard to hold it. Committing to upgrades is, unfortunately, a bad thing against mech as they do so little for you. The idea is that the same square of 'dragoons/zealot vs tank/vulture' exists - Tanks want to be firing at stalkers not zealots, hellions want to soak up stalker hits to protect the tanks, while firing at zealots, zealots want to draw tank fire allowing stalkers to close with the tanks. This is why blink is so helpful - zealots run in initially to draw the tank fire, but unlike broodwar, the 'dragoons' can simply ignore the 'vultures' - they can blink directly onto the tanks while the tanks are firing at zealots. Add in the power of immortals against BOTH tanks and vultures, and... well... *ouch* As such, archon/chargelot is not actually an optimal composition *anyway.* That being said, it is still far more about playstyle than about composition. You need to play fairly similarly to zerg, taking expansions early and getting a huge economic lead. | ||
|
alpenrahm
Germany628 Posts
first off, i deeply appreciate your work done with this guide and im already a fan of your idea. I like how you basically skip the reactor rax and the armory in favor of earlier gas and more factories to get out that monstrous amount of tanks. however i am concerned about this build in a couple of ways. 1. you stay on 1 rax production up until 7 minutes ( which is when your first banshee pops) and completely rely on that for defence. meaning that, once you put down the factory you are commited to go through with it all the way and have no way to hold any 3 gate blink or 4 gate all in because you will eventually just get outproduced. Even just holding some well executed 3 gate pressure will prove to be quite a difficult task. 2. what would you actually do if protoss would, however wierd it may be, proxy stargate you. build vikings and micro like a boss? 3. this build is completely all in. once you lose your tanks you are done for. you have no upgrades. your 3rd is super late. Also, in the ohana game vs this gm dude you won because he played greedy and didnt respond to your tech at all. while you made as many units as you possibly could, this guy went for thermal lance colossi, 2-2 uprages, charge and a 3rd with 7 gates. so its no wonder that you were ahead in supply. i suggest that you should think about doing this build with a cc first 2 rax opener (just to be a little safer against early pressure not for constant marine production) and skip one factory in favor of an armory and armor upgrades. I know this would cut into your tank count but you should be able to get a decent lategame setup up with this. edit: your execution was flawed too. you could have had 10 more scvs by the time you moved out. | ||
|
Teoita
Italy12246 Posts
On October 26 2012 22:41 mizU wrote: How would I beat this as a Protoss if I face it on ladder? What have you lost to? Other than bulldogging it (not sure if this works vs gm's but i've beaten monk with it so eh), i think a 1gate fe -> some sort of robo/3gate into forge/twilight council opening would do very well vs this kind of stuff. I never was a fan of colossi vs mech, and the ability to get charge, blink, archons and immortals vs the push should do extremely well. That said yeah mech stuff is so unexplored, it's kinda hard to say. More than anything, it would come down to positioning. Getting a reactionary warp prism to flank with 2 immortals and a round of warped in zealots would also be really strong for instance. | ||
| ||
