EPT[G] Going CC-first Bio in TvP - Page 5
| Forum Index > StarCraft 2 Strategy |
|
WeRRa
378 Posts
| ||
|
mau5mat
Northern Ireland461 Posts
On June 08 2013 03:09 WeRRa wrote: Just wanted to note that Innovation goes 15/17 rax, Flash 14/16 rax and TY 14/15 rax. I guess the 15/16 rax isn't that comon among kespa players and 14/15 or 14/16 rax is pretty save vs early agression. This is really cool actually, thanks for sharing! | ||
|
Jazzman88
Canada2228 Posts
On June 08 2013 02:53 Lock0n wrote: You actually have more marines from 5 min onwards with a cc first build than you would with a 1 rax fe or a 1rax reactor expand build. So this build can crush voidray all ins easy. It's not easy; competent 1-base play of any sort should (rightfully) be very threatening to anyone doing eco-focused builds. It shouldn't be easy to hold an all-in off one base when you go CC-first. Voidray or Stargate all-ins are the strongest weapon you will face, period. 4-gate is not that effective (because as noted you get lots of Marines), DTs should be dealt with by scouting and early Scan, and Blink is actually safer than with 1-rax FE because of the excess minerals. To U_G_L_Y: You are somewhat correct in assessing the difficulty of executing this build below a certain level. Masters and above should be able to utilize this more or less effectively because of their effective scouting. The need to react correctly is huge. I have been playing this build exclusively on larger 4p maps since I wrote this guide, and while my TvP is not a great matchup, it doesn't have anything to do with the CC-first. Correct scouting is and always will be the answer to anything you face with a CC-first. There are so many minerals that there is ALWAYS a correct way to hold the all-in, but it requires a lot of grinding and game sense. I believe it's probably necessary to lose more than you win with this build for a while, during which time you learn what it is that kills you and slowly learn how to tell exactly what's coming without getting that one money Scan or lucky SCV run-by at 6:30. A secondary note: be sure that you are differentiating the opening from the midgame when reviewing 1-base all-ins. Many Protoss like to 1-base versus CC-first, but that doesn't mean the 1-base build is any less lethal versus a 1-rax FE or Reactor FE. In fact, CC-first holds 4-gates MUCH more efficiently that Reactor expand because you get more Marines and Bunkers earlier. When looking at your replays, compare what you had when the attack hit to what you might have with a Reaper expand, or 1-1-1 expand, or Reactor expand. If there's a substantial difference, then you have to start looking at adjustments tailored for CC-first. If there are similar amounts of stuff that you had in both situations, then it's a matter of scouting and reacting entirely, not the build's problem. | ||
|
mihajovics
179 Posts
| ||
|
awakenx
United States341 Posts
Lots of good resources to possibly add to the thread, and ever since OGN lifted the paywall on VODs, ForGG vs HerO is accessible and a good example of CC First working out beautifully. | ||
|
awakenx
United States341 Posts
| ||
|
Lock0n
United Kingdom184 Posts
If you're the OP I imagine you should know better about the build and how it compares to other Terran openings  | ||
|
Jazzman88
Canada2228 Posts
On June 08 2013 16:07 Lock0n wrote: Well in all honesty if you can't hold a void ray all in using a cc first build, then you're not going to hold a void ray all in with any build. Cc first gets you the most marines out of any expand build. And with so many SCVs and potential for bunkers, In my opinion cc first is basically a build order win over void ray all in. If you know of a build that is better against it, then feel free to point it out. And no,1base 3 rax all in does not count as a viable build. If you're the OP I imagine you should know better about the build and how it compares to other Terran openings This is absolutely correct. CC-first in terms of sheer unit amounts will defend better than any other expand build. The reason people think it might be weaker to all-ins is that Protoss see it and instinctively want to punish it, hence they just all in more often versus CC-first. The crucial differences are in scouting, because you don't get the early Reapers and the Mine drops. Remember in WoL, when you had to get the SCV out of the Protoss base by 4:15 before the Stalker pops and then sneak it back in once it showed at your front? You have to do the same thing with this build, but there's still a chance that the MsC or the Stalker get left at home to deny that, so you may have to expend a Scan. In addition, all of the subtle 'reading' moves are something you need to practice with time. Intuiting from the first 3 units he builds that 'oh, no Zealot or Sentry plus early gas? Likely Blink play, cool' or '1 Sentry, no expand, double gas? Check for DTs or Stargate'. All of that comes with time. And yes, expect alterations to the OP to account for ForGG's Hellbat/Marauder focused midgame. | ||
|
A Wild Sosd
Australia421 Posts
| ||
|
Jazzman88
Canada2228 Posts
On June 08 2013 23:16 A Wild Sosd wrote: I really want to play this style but im so nervous that I will just get all inned as soon as they see it. How have peoples experiences been on ladder using this build? If the Protoss pulls an all-in I haven't seen before, I lose the first time. I rarely lose against all-ins I know are coming and I am familiar with. For example, I haven't lost to a 4-gate that I scout coming before 6:00 in a month. Blink all-ins I have now succeeded in figuring out and don't really lose to those anymore. Ditto proxy Stargate plays. The last all-in that caused me consistent trouble was 6-gate off of Nexus-first, but that is a matter of scouting and proper Bunker placement (damn you, Zerus Prime choke point!). Encourage the all-ins! Every one you lose to teaches you a vulnerability that you're not accounting for. Aggression is always easier to execute than defense, so losing to a few all-ins teaches you how to master the build and throw off a wide variety of things and get into that bread-and-butter midgame power mode that is the CC-first strength. | ||
|
Lock0n
United Kingdom184 Posts
| ||
|
Jazzman88
Canada2228 Posts
Discussion of the neat Hellbat/Marauder transition ForGG destroyed Liquid Hero with on Whirlwind the other day in WCS Finals play. Some interesting math that I started looking at while thinking of possible ways to exploit Hero's play (which, let's be honest, was extremely standard aside from the fail Storms and lol control exhibited in that match). I don't think you'll achieve a massive winrate spike because there are options for the Protoss to defend MUCH better than Hero did, but it's a very good composition that deals with a surprising amount of standard deathballs in a way that is much more conducive to standing and fighting that Terrans haven't been able to do very much so far in TvP. | ||
|
TheDwf
France19747 Posts
On June 08 2013 00:49 U_G_L_Y wrote: Maybe I am biased because I have lost 13 games in a row vs Protoss with this build, but I do not believe this is safe or viable below GM. Watching the VODs, there was very little 1 base play. MKP got SMASHED by the voids and I don't think there was anything he could have done to stop it. Innovation built 3 bunkers (which I am still not clear on how he knew it was coming or how he knew when to pull SCVs and lost every SCV at his natural to Huk's 3 gate pressure after expo and was way behind. It was a very close game, despite Huk never having made a single colossus or templar at the 22 minute mark. W T F ? ! Taeja was WAY behind Sase and only won because Sase accidentally split his army up navigating a cliff, I think most diamond players could win that game if they resumed replay right before that mistake. I was hoping that this guide finally would break my TvP funk since HOTS, but honestly watching those VODs after reading your guide through and practicing it 30 + times... I was on the verge of tears last night. Every single Pro Terran win, even top Koreans over foreigners, required the Protoss to fuck up horribly. I don't think you can hold a competent 1 base build reliably with CC first. CC first can defend most if not all 1-base play, but the margin for error is very thin against certain things: one misstep and you instantly lose. As pointed out above, CC first is actually the best build against some 1-base all-ins hitting late (e. g. Blink Stalkers) because of the massive econ advantage you get before the all-in hits. The problem usually comes from the limited information you can collect about which attack Protoss is doing. About the MKP game: bad example since MKP never scouts. Of course you can't defend an all-in if you're out of position on top of not even knowing it's coming. In this case, with +2 bunkers at his natural and SCVs prepared he would have held easily. About the Bogus game: HuK went 5g, not 3g pressure. From memory he was ahead (remember that Protoss has zero tech) after holding the 5g despite losing some of his Marines in front of his natural before the attack hit. About the Taeja game: you mean the one on CK? Taeja was severely behind, yes, but it wasn't specifically because of CC first (though the build does have the horrible drawback of having your two OCs virtually synchronised, thus leaving you with no close scan should DTs appear after you called two MULEs). If it can be of any consolation, losing a lot at first when learning CC first is probably unavoidable. | ||
|
U_G_L_Y
United States516 Posts
My buddy did the exact same build that Rain did except made an oracle instead of a void. I pulled every marine into the main and then the gateway units popped the empty bunkers uncontested. Suppose it had been a robo instead of SG? How do you know where to position your units? Edit: Seed, not Rain | ||
|
TheDwf
France19747 Posts
On June 09 2013 06:10 U_G_L_Y wrote: With my WOL 1 rax fe, stim was started at 5:20 so holding every all in except 4 gate was easier. With CC first, blink is much harder because you have 0 marauders for the first blink in. My buddy did the exact same build that Rain did except made an oracle instead of a void. I pulled every marine into the main and then the gateway units popped the empty bunkers uncontested. Suppose it had been a robo instead of SG? How do you know where to position your units? I go 3 rax add-ons (lab lab reactor with the first 100 gas) against Blink Stalkers and I always have 2-4 Marauders ready for the first wave. What do you mean with "same build that Rain"? | ||
|
mau5mat
Northern Ireland461 Posts
http://www.reddit.com/r/AllThingsTerran/comments/1fw1f5/tvp_cc_first_safest_build_we_have/ | ||
|
Jazzman88
Canada2228 Posts
On June 09 2013 06:10 U_G_L_Y wrote: With my WOL 1 rax fe, stim was started at 5:20 so holding every all in except 4 gate was easier. With CC first, blink is much harder because you have 0 marauders for the first blink in. My buddy did the exact same build that Rain did except made an oracle instead of a void. I pulled every marine into the main and then the gateway units popped the empty bunkers uncontested. Suppose it had been a robo instead of SG? How do you know where to position your units? If Blink is 'much harder' using CC-first, you're either encountering a difficulty with scouting or a difficulty with macro. If you can accurately predict or scout Blink before he starts the extra gateways, you can actually just go straight for Bunkers and Factory tech, pumping Marauders and delaying for a tank, and you hold fairly easily. I have no idea also what you mean by the build that Rain did. The only pro game I have in the VoDs section with Rain was a Robo macro game versus Jjakji that Rain won in the later game. If you're just talking about the LGIM.Seed game that he crushed MKP with Voids using, okay, then your job is to only put as many units as necessary to defend in whatever position. Scouting Stargate, keep 6 Marines at the main mineral line until you have a Turret 50% done there, then shift everything to the natural. Check for an expansion, and if there isn't one, you build another 2 Bunkers and 2 Turrets at the natural expansion, and put a Marine/SCV a ways out patrolling for the move-in. Pull at least 15 SCVs to repair, and you should hold. The technique for holding Robo is similar, but you make more Bunkers instead of Turrets, because all you need to do is break the Immortals, and wait for Stim, and you can win with a counterpush. Stargate needs divided Marines until Turrets are up, then you concentrate at the weak point (front). Robo means you place Depots to watch for Prisms, and concentrate everything at the natural for defense. As to differentiating the two, scouting, scouting, scouting, and more scouting. It's fairly easy to hide at least 1 SCV on the map to check his expansions, and if you're facing a 1-base all-in, wasting a Scan to check his tech is not a bad idea at all. On June 09 2013 06:28 mau5mat wrote: Dwf, what do you think of this variant of CC first proposed on Reddit? http://www.reddit.com/r/AllThingsTerran/comments/1fw1f5/tvp_cc_first_safest_build_we_have/ This is basically what ForGG throws in sometimes, going gasless for a while, and it bears similarity to the two strands of '6-rax' and 'triple CC' described as corner cases in the OP. He gets 4 Barracks to make sure he has tons of Marines, then grabs a third base which ensures the economic advantage, THEN techs rapidly with 2 or 3 gases. The problem I have with that style is more a consistency point than an outright strategic point: going 4 naked Barracks into third Command Center is slightly less greedy than straight triple CC after 2-rax, but yet not as pressuring as 3-rax into quick Medivacs and upgrades. True, you have a large midgame army, but you are guaranteed to be behind on upgrades for the whole game against virtually all standard play by Protoss, and you have a much later Stim/Combat Shields timing with Medivacs, which allows clever play to get ahead in a couple of ways. It can lead to very dynamic games, but I just think going either for a faster pressure in the midgame or straight-up greed is more consistent with the idea and philosophy of the build. It's absolutely viable, but I prefer a more standard follow-up in terms of overall solidity. | ||
|
TheDwf
France19747 Posts
On June 09 2013 06:28 mau5mat wrote: Dwf, what do you think of this variant of CC first proposed on Reddit? http://www.reddit.com/r/AllThingsTerran/comments/1fw1f5/tvp_cc_first_safest_build_we_have/ Doesn't look good at all. Delaying your Medivacs for so long is just giving Protoss a special favour. Going 4 rax is pointless against standard play since you won't be able to pressure (Protoss will just PO and hold with his first units). You will also freelose to some 2-bases all-ins like 6'30 robobay into 2-bases Colossi. I don't understand the reasoning behind his build order. I'll try to answer your post in the Terran Help Me Thread later, by the way. | ||
|
mau5mat
Northern Ireland461 Posts
On June 09 2013 06:52 TheDwf wrote: Doesn't look good at all. Delaying your Medivacs for so long is just giving Protoss a special favour. Going 4 rax is pointless against standard play since you won't be able to pressure (Protoss will just PO and hold with his first units). You will also freelose to some 2-bases all-ins like 6'30 robobay into 2-bases Colossi. I don't understand the reasoning behind his build order. I'll try to answer your post in the Terran Help Me Thread later, by the way. Thanks so much, and to think I was forgotten :') | ||
|
U_G_L_Y
United States516 Posts
On June 09 2013 06:17 TheDwf wrote: I go 3 rax add-ons (lab lab reactor with the first 100 gas) against Blink Stalkers and I always have 2-4 Marauders ready for the first wave. What do you mean with "same build that Rain"? Sorry, Seed vs MKP. Since the rax is later I am afraid to get addons since in WOL I got my first gas before my other 2 rax and bunker for ultra fast stim, but I guess Blink takes longer now so the first blink in doesn't come until 8 stalkers instead of 5? How do you tell if it is a proxy twilight, robo, or SG? I had 2 SCVs out looking for it (never found it), by the time I realized that the third pylon had to be powering a twilight since it wasn't near my base, it was too late to opt for techlabs. Another game I found it but again, it was too late for techlabs, I had bunkers, Ebay, turrets, and he just jumped past it all and kited my naked +1 marines and SCVs to death. | ||
| ||
