EPTI've always done the super standard 1 gas gate expand, and I just can't imagine a single scenario where having just a tiny bit more gas is more beneficial than having a slightly later nexus. Even with only 2 probes in each gas for optimal gas mining efficiency, the nexus is at least later by like 10-15 seconds right?
The HotS Protoss Help Me Thread - Page 284
| Forum Index > StarCraft 2 Strategy |
Make sure you read the OP before asking a question. Asking a question already addressed in the OP will result in moderation action. Also, please put some effort into your questions. Do not just ask for build orders. | ||
|
EngrishTeacher
Canada1109 Posts
I've always done the super standard 1 gas gate expand, and I just can't imagine a single scenario where having just a tiny bit more gas is more beneficial than having a slightly later nexus. Even with only 2 probes in each gas for optimal gas mining efficiency, the nexus is at least later by like 10-15 seconds right? | ||
EsportsJohn
United States4883 Posts
On August 07 2014 11:46 EngrishTeacher wrote: What's the main difference between a 2 gas gate expand a the standard 1 gas gate expand? In what situation would one go fro the 2 gas variant? I've always done the super standard 1 gas gate expand, and I just can't imagine a single scenario where having just a tiny bit more gas is more beneficial than having a slightly later nexus. Even with only 2 probes in each gas for optimal gas mining efficiency, the nexus is at least later by like 10-15 seconds right? I believe I responded to this like 10 pages back, but I'll try to remember exactly what I said. Something along the lines of (we're talking about PvT, I assume): 2-gas expands are going to get a later nexus, but your tech is a little bit more ambiguous (is also good for opening sentry instead of stalker). The main difference usually comes ~6:30-7:30, where you would normally need to throw down your natural geysers with a 1-gate FE. However, since you have so much gas from earlier, you can delay these a little bit in favor of more mineral-heavy things. Otherwise, I think 2-gas openings are superior for oracle openings. Other than that...not much of a difference really. It's just a preference, I think. If you want, check out herO's gas timings, as he has done a lot of 2-gas expands with like 1-2 in each geyser and still gets his nexus down pretty much on time. I'm sure vhapter can give you an idea........ | ||
|
vhapter
Brazil677 Posts
(1) Wait for second geyser to finish to send probes to gas (1 probe on each). This happens slightly before your gateway finishes. (2) Rally next probe to minerals. (16 on minerals without scouting) (3) Rally the next probe to gas and pull one probe from minerals to build your expansion. (4) Rally the next probe to minerals. (16 on minerals again). (5) Rally probe to gas. ^ You can get a slightly higher gas income if you cut scouting at this point. I say this because herO used to scout after starting his nexus (dodging the reaper, ofc). But it's really unnecessary. If you don't cut scouting, you can get your robo at 4:47 which is about as much gas as if you go single gas --> nexus --> msc --> gas --> pylon... but if you cut scouting, you should have a slightly higher gas income, for sure. The interesting thing is that this allows you to switch to a 1 base build more quickly if you get ebay blocked by adding probe to gas earlier. I'm not saying that's a good way to deal with it, but that's one more thing for terrans to consider - one more thing you can do. This build might also allow for a faster nexus while going stargate afterwards (if you send probes to gas right after starting your nexus or something), but I'm not sure how your mineral and gas lines up. All in all, it looks like a very smart way to manage your gas and it makes your build more adaptable. While CJherO doesn't really do it anymore though, it's still a nice thing to consider. And as sc2john said, 2 gas builds are more ambiguous. On a side note, the current double forge colossus build (with fast blink and fast colossus), which is incredibly popular in Korea (CJherO and Zest do it all the time), opens with a single geysers takes gases 3 and 4 at 6:00-6:05 off of 3 stalkers, before any extra gates. If you delay that, the build doesn't line up properly. It doesn't change the fact that 2 gas builds don't need gases 3 and 4 that early though, so sc2john's point still stands. | ||
|
EngrishTeacher
Canada1109 Posts
On maps where it's easy to deny further reaper scouting attempts, I wonder if it's worth the mind game to just mine both gases with 4 or 6 probes, pull probe off gas completely once the reaper is forced out for the first time and play standard from there. The ~20? seconds delay on the nexus is probably worth the response from T (potential turrets in mineral line, early bunker, delayed barracks/tech, etc.). | ||
|
Mojito99
Germany154 Posts
On August 08 2014 23:12 EngrishTeacher wrote: Awesome thanks for clearing up my question. I play T as well, and I can vouch that whenever I see anything other than a standard 1 gas gate expand from the P, I get really nervous and play a lot more conservatively. On maps where it's easy to deny further reaper scouting attempts, I wonder if it's worth the mind game to just mine both gases with 4 or 6 probes, pull probe off gas completely once the reaper is forced out for the first time and play standard from there. The ~20? seconds delay on the nexus is probably worth the response from T (potential turrets in mineral line, early bunker, delayed barracks/tech, etc.). a double gas before expo with 2 in each makes protoss significantly more unpredictable which in my opinion is the biggest strength of the race. Loosing this unpredictability in favour of defensive Colossus styles will be very bad for both players and viewers. | ||
|
Mojito99
Germany154 Posts
any got some written info about the 1gate FE (sentry msc) into SG that zest shows in some of his games. looking for some guides or information how this build works out against the various timings. Also: is there any information on how or perhaps why not you can go IAC into tempest and skipping colossus. | ||
|
Xaeldaren
Ireland588 Posts
- sOs vs Flash for example. Polt has been doing it in his recent games like in the Destiny I tournament. I've been going for 2 additional gates at 5:45 along with 2 stalkers and a sentry, but it's still incredibly difficult to hold with gateway units + overcharge. Best case scenario I lose several probes and gateway units, worst case scenario I lose the natural nexus. Here are some examples of me trying to deal with it and not doing so well: http://ggtracker.com/matches/5316459 http://ggtracker.com/matches/5316461 | ||
|
vhapter
Brazil677 Posts
On August 11 2014 20:15 Xaeldaren wrote: How is everyone dealing with the fast stim attacks that are gaining popularity right now? http://www.youtube.com/watch?v=IqQRe4eqHG8&feature=youtu.be - sOs vs Flash for example. Polt has been doing it in his recent games like in the Destiny I tournament. I've been going for 2 additional gates at 5:45 along with 2 stalkers and a sentry, but it's still incredibly difficult to hold with gateway units + overcharge. Best case scenario I lose several probes and gateway units, worst case scenario I lose the natural nexus. I haven't faced a stim push that fast yet, but I'm well aware of what kind of timing that is (8:10ish stim timing with marine marauder). If you're in the same position as sOs, I believe the best decision may be to pull your probes up your ramp and force field it. It's not economical at all, but if you can't fight without losing a ton of units and if you don't fight you lose almost 20 probes or something... I guess all you can do is let your nexus do its thing and pull back. Perhaps a gateway simcity at the natural + force fields could do the job, but I dunno about that... it would be very map dependent and most likely an annoyance in the long run. But it's still something to think about I guess, and might even come in handy against occasional hellion openers if you can wall one side of your natural with gates, and the other with 2-3 stalkers + a pylon. Adding 2 gates at 5:45 means you're going to have 3 gates up pretty damn early. I highly doubt you need more than 3 stalkers (which is the standard nowadays) and 3 zealots to hold that off with the aid of photon overcharge. I think a 3 stalker opener makes you safer against aggressive builds early on (that is to say, stuff that happens before stim timings), but that's beyond the point. The bottom line is, you can probably tech faster and still be very safe. You can probably delay your gates until 6:45 and be safe enough that you can directly engage the terran army under photon overcharge. A push like the one Flash did shouldn't be enough to take down your nexus, but I can see a follow-up attack taking it down if you can't make enough units in time. What might be killing your nexus is 9:00-9:30 timings. The build I've been doing lately gets very late gates. It's basically msc expand with 3stalkers off of 1 gate --> 7:00ish twilight + robo bay --> double forge --> +2 gates at about 7:50-8:00, although you can easily swap the gates and forges. What I do when I see 3 rax with double techlab + a reactor is get my colossus out asap (9:00-9:15) and build 2 cannons at my natural to buy time for gateway units. I saw a master league protoss talk about this response on reddit and decided to give it try, and it has worked really well for me so far. However, this is not possible to do against that very early stim timing Flash did against sOs, but thankfully it's much less common than a 9:00-9:30 stim/combat shields timings. There's also that double poke build with concussive shells first into a stim push, but I haven't seen that much on the ladder either - they try to force 2 photon overcharges before doing the real push, and the first poke has concussive shells, so if you see marauders during the poke, you probably just want to pull the hell back and don't kite the marines unless the marauder is definitely out of range or you're in range of photon overcharge. I'm not a terran player, but if you want to work on these timings, feel free to hit me up on NA (vhapter #721). I always learn stuff when I do this kind of thing, so I'd be glad to help. Replay analysis + Show Spoiler + Ok, just watched one of your replays. You only have 2 stalkers and a sentry, but the sentry almost dies and serves no purpose at this point because it's one of those 7 min pokes. I think you should have 3 stalkers at this point, and maybe warp in that sentry once wg finishes if you really want it. Since your opponent has a marauder, I wouldn't kite outside photon overcharge range if possible. You let all your stalkers die because you didn't kite your units, which is absolutely essential in order to be cost effective. Keep your sentry away from the fight if possible (up your ramp is good imo). Focus fire a unit with stalkers and move back before marines fire under photon overcharge, then repeat. I don't think you even need to warp in at this point, 3 stalkers with good kiting can do a lot if they let photon overcharge kill their army, and you can even kite them out on the map if there's no risk of losing your stalkers to concussive shells. If they target probes, just pull them away. These losses slow you down quite a bit. While it's not necessary, I think you should consider getting a colossus out as early as 9:00-9:30 with an optimized build, it helps you quite a bit against these timings... some builds get a late colossus, but since you weren't doing anything tech heavy, your colossus was very late imo. Second replay, yet another surprise at the 7min mark. I'd say keep your stalkers on top of your natural's ramp in order to spot these pokes more easily if possible. Sending your observer on the path you're most likely to see their army coming is a good idea imo, although tbh your scouting path was fine on merry go round (just something to think about when you play other maps). Your sentry dies, which makes me cringe. You can't leave your sentry that exposed and you should be watching your army more carefully. It's not like he has stim at this point, so losing using here is unacceptable imho. If anything, you should be killing stuff at this point! There's a reason terrans haven't been doing these pokes without concussive shells lately imo - stalkers can punish marines really well if microed properly. Ok, so there's another poke at 8:30... this time with stim... but just like I said previously, you need more stalkers and zealots to deal with these pokes, and instead of 3 stalkers and 3 zealots, you have 2 sentries, 2 stalkers, and 2 zealots. Instead of waiting for the terran to commit, you keep your army more exposed... when stim is finished, you need to keep your army away from your ramp imo, which is the opposite of early on. So having a few more units as well as punishing the terran for doing an early poke without losing units yourself would have made a considerable difference imo. Don't let them get the first overcharge for free! I mean, if this guy had researched concussive shells first, he wouldn't have gotten stim this fast... you're letting him force the first overcharge too easily for someone who's going straight for stim. | ||
|
TokyoGirl
Japan116 Posts
| ||
|
Xaeldaren
Ireland588 Posts
| ||
|
DBS
515 Posts
| ||
|
Whitewing
United States7483 Posts
On August 12 2014 12:14 DBS wrote: So I play terran and zerg, and would like to also learn some protoss, but the issue is that i'm not sure how to learn a protoss strategy, I don't want to just play all ins, but i'm not sure the proper way of thinking about build orders for protoss, because there always seem to be SO many adjustments to the infrastructure, and i don't know how to manage my economy while also doing those adjustments. So i guess the best way to phrase my question is, What is the best way to learn a protoss macro strategy? Take a standard build as executed by a pro player like Rain or CJ_herO, and copy the build to the letter. Don't make adjustments to the build until you're looking at high masters, just play standard and safe and scout. | ||
|
DBS
515 Posts
On August 12 2014 12:26 Whitewing wrote: Take a standard build as executed by a pro player like Rain or CJ_herO, and copy the build to the letter. Don't make adjustments to the build until you're looking at high masters, just play standard and safe and scout. My issue with this is that if i get attacked all my timings are thrown off. For terran, i just think of the pieces that i need (For example in PvT 5 rax 2 e-bays starport 3 CC) and add in my missing pieces, but that seems to not work at all for me, and i can't even find somewhere to learn what set up i need for different matchups. | ||
|
brickrd
United States4894 Posts
On August 12 2014 13:55 DBS wrote: My issue with this is that if i get attacked all my timings are thrown off. For terran, i just think of the pieces that i need (For example in PvT 5 rax 2 e-bays starport 3 CC) and add in my missing pieces, but that seems to not work at all for me, and i can't even find somewhere to learn what set up i need for different matchups. its just a matter of learning how to think about the timings other races have from the protoss perspective. i know when you start out offracing toss it feels like you just have no units and they can hit you at any time, but once you hammer out a few gmaes you start to figure out how to defend with the minimum amount of units just like you do as zerg while youre trying to drone. just keep scouting expansions, tech and saturation like you do in any matchup. if you play t and z you should be able to look at a terran build and say "whoa, there's a reactor on that barracks already/there's a factory being built, i'm getting attacked" or look at a zerg build and say "that's not enough drones/that third is too late/that gas is too early, i'm getting attacked" and from there it's usually a matter of having enough sentries to buy time for robo tech/more gateways/whatever you need to defend. once you defend the attack then just go back to what you were doing before or counter-allin and you should be golden you don't really need to "set up" anything different from a rain or hero build if you're being hit with a cheesy timing attack, it's a really simple three-step process: 1) identify aggression 2) make sentries 3) continue build if youre talking about defending attacks that come later than when youre trying to establish your natural then your question is getting kind of vague and complex because then youre into the midgame and there are a lot of things to read into depending on whats happening, just like any race... | ||
|
vhapter
Brazil677 Posts
On August 12 2014 13:55 DBS wrote: My issue with this is that if i get attacked all my timings are thrown off. For terran, i just think of the pieces that i need (For example in PvT 5 rax 2 e-bays starport 3 CC) and add in my missing pieces, but that seems to not work at all for me, and i can't even find somewhere to learn what set up i need for different matchups. Your question is too open... too vague... to the point it's actually unclear. It'd be possible to give you a more proper reply if you were more specific. | ||
|
Archiatrus
Germany64 Posts
Up until now I preferred voidrays. I went up to two with one in queue and crossed the map with zealot/sentry support and killed my opponent or at least killed his natural while taking my own. But there is one exception: unfortunately, if he opens phoenix it is a insta loss... and I dont like a opening where I have to pray for the right opening of my opponent. The only solution would be scouting his first phonix (in queue), but I dont think I can sneak into his base between a few seconds before he makes his tech choice and when I arrive at his base with my army. Please correct me if this is wrong. So the alternatives are oracle and phoenix. Imo every P opening should be oracle proof, which means the phoenix is left. But can you be aggressive with a Phoenix opening? And defending seems to be much harder too. So basicly my question is what do you build first in a Stargate opening and why? Or is robo/twilight just better? PvT midgame after 1Gate FE: When what? I do the 1Gate FE from the OP with a robo but then I get a little bit lost what to build when. My basic plan up until now is: get 2 collosi -> get some HT+storm -> get a few extra collosi -> lots of stalker/chargelots/a few archons to get 200 supply. The only problem is that I just build the tech buildings when I feel it is the right moment to do. But this leads sometimes to way too late HTs or way too few units to defend at the beginning. So can somebody write down a basic plan in what order and approximatly when I have to build the following: add x gates (total y) robo bay twighlight council forge(s) third gases at natural The best think would be a something along the lines: "if natural is at full saturation -> start robo bay" or "+1/1 halfway done -> twilight". Thank you | ||
|
vhapter
Brazil677 Posts
On August 12 2014 23:25 Archiatrus wrote: PvP Stargate opening: What do you build (first)? Up until now I preferred voidrays. I went up to two with one in queue and crossed the map with zealot/sentry support and killed my opponent or at least killed his natural while taking my own. But there is one exception: unfortunately, if he opens phoenix it is a insta loss... and I dont like a opening where I have to pray for the right opening of my opponent. The only solution would be scouting his first phonix (in queue), but I dont think I can sneak into his base between a few seconds before he makes his tech choice and when I arrive at his base with my army. Please correct me if this is wrong. So the alternatives are oracle and phoenix. Imo every P opening should be oracle proof, which means the phoenix is left. But can you be aggressive with a Phoenix opening? And defending seems to be much harder too. So basicly my question is what do you build first in a Stargate opening and why? Or is robo/twilight just better? PvT midgame after 1Gate FE: When what? I do the 1Gate FE from the OP with a robo but then I get a little bit lost what to build when. My basic plan up until now is: get 2 collosi -> get some HT+storm -> get a few extra collosi -> lots of stalker/chargelots/a few archons to get 200 supply. The only problem is that I just build the tech buildings when I feel it is the right moment to do. But this leads sometimes to way too late HTs or way too few units to defend at the beginning. So can somebody write down a basic plan in what order and approximatly when I have to build the following: add x gates (total y) robo bay twighlight council forge(s) third gases at natural The best think would be a something along the lines: "if natural is at full saturation -> start robo bay" or "+1/1 halfway done -> twilight". Thank you Single gas (or better yet, read the CJherO 2 gas trick that I wrote a few posts ago) Nexus --> msc Second gas (if you're not doing herO's trick) --> pylon --> stalker --> [robo --> wg] OR [wg --> robo] ^ once your robo is done, make 4 observers, first one for scouting, the other 3 for spotting (or defensive if you see an early factory) +2 stalkers (pylon after third stalker... I won't be listing pylons now, but it's super important not to get supply blocked) 6:00-6:05 double gas at natural 6:50-7:10 twilight (blink asap) + robo bay (colossus asap) [double forge --> +2 gates] OR [+2 gates --> double forge] ^ whichever feels more comfortable, but don't forget to scout with your observer at this point too Once you have about 44 probes (2 bases properly saturated), cut probes in order to get infrastructure up (+3 gates when resources allow, and don't forge thermal lance) and units for defense. Chrono colossi and blink. Most pros chrono colossi and blink with this build until blink finishes and they have at least 2 colossi out, and then they shift their focus to forge upgrades. It's ok to resume worker production temporarily and go up to 19-20 probes on mineral in each base, then you can take a third and continue making probes in order to quickly saturate your third. Going over 20 probes on minerals in each base before starting your third is a bit overkill though. Start thermal lance before building a templar archives (basically, you build your templar archives instead of going up to 3+ colossi, so it's only a matter of when you get enough gas to build). If you're cutting colossi at 2, I suggest getting a warp prism. Most pros have been going up to 10 gates as soon as their third finishes, before taking geysers 5 and 6 (that is to say, with pure colossi builds though, so they can get a lot of units out asap). If you have an edge, it's generally safer to grab your geysers earlier, but you still want those gates up quickly. Feel free to add 2-4 gates at one point later. You can also add a second robo when you have enough spare gas. I don't have a specific time for charge with this build, but I don't delay it for too long. I generally go up to 12 stalkers and decide between making a few high templars and starting charge. In theory, you can start charge as soon as blink finishes though, which is pretty early... anyway, this really depends on how you you manage your gas income. I think it's a good idea to start charge no later than 11:30-12:00 though. | ||
|
DinoMight
United States3725 Posts
I'm finding it problematic against Zerg because they can open something like fast 3 hatch and defend easily in one place. Should one just always FFE on Deadwing? I really don't like that flat ground walk up to the natural choke. It also counters my style of multi-pronged attacks/harass vs all races because they basically just sit their army in one place and you have to fight them there. | ||
|
Mojito99
Germany154 Posts
On August 13 2014 05:03 vhapter wrote: Single gas (or better yet, read the CJherO 2 gas trick that I wrote a few posts ago) Nexus --> msc Second gas (if you're not doing herO's trick) --> pylon --> stalker --> [robo --> wg] OR [wg --> robo] ^ once your robo is done, make 4 observers, first one for scouting, the other 3 for spotting (or defensive if you see an early factory) +2 stalkers (pylon after third stalker... I won't be listing pylons now, but it's super important not to get supply blocked) 6:00-6:05 double gas at natural 6:50-7:10 twilight (blink asap) + robo bay (colossus asap) [double forge --> +2 gates] OR [+2 gates --> double forge] ^ whichever feels more comfortable, but don't forget to scout with your observer at this point too Once you have about 44 probes (2 bases properly saturated), cut probes in order to get infrastructure up (+3 gates when resources allow, and don't forge thermal lance) and units for defense. Chrono colossi and blink. Most pros chrono colossi and blink with this build until blink finishes and they have at least 2 colossi out, and then they shift their focus to forge upgrades. It's ok to resume worker production temporarily and go up to 19-20 probes on mineral in each base, then you can take a third and continue making probes in order to quickly saturate your third. Going over 20 probes on minerals in each base before starting your third is a bit overkill though. Start thermal lance before building a templar archives (basically, you build your templar archives instead of going up to 3+ colossi, so it's only a matter of when you get enough gas to build). If you're cutting colossi at 2, I suggest getting a warp prism. Most pros have been going up to 10 gates as soon as their third finishes, before taking geysers 5 and 6 (that is to say, with pure colossi builds though, so they can get a lot of units out asap). If you have an edge, it's generally safer to grab your geysers earlier, but you still want those gates up quickly. Feel free to add 2-4 gates at one point later. You can also add a second robo when you have enough spare gas. I don't have a specific time for charge with this build, but I don't delay it for too long. I generally go up to 12 stalkers and decide between making a few high templars and starting charge. In theory, you can start charge as soon as blink finishes though, which is pretty early... anyway, this really depends on how you you manage your gas income. I think it's a good idea to start charge no later than 11:30-12:00 though. a quick transition to HT is very very vulnerable to SCV pulls. Anything below 5 Colossus before the transition is very risky. I dont quite understand the gameplan here. You commit heavily to blink which is only good to save your colossus, but how do you expect to keep them alive vs 12+ vikings with only 3 available. You also commit to double forge but then transition into tech which does not benefit of the double upgrades (storm). All in all this seems extremely vulnerable to every sort of pressure Terran put on because you commit to both double upgrades and fast tech. This gameplan will most definitely die to SCV pulls before storm is out - your storm choice is not even a colossus fake or something like that so it may not even be ready for 1/1 pulls or so. something to think about best | ||
|
Whitewing
United States7483 Posts
On August 13 2014 07:40 Mojito99 wrote: a quick transition to HT is very very vulnerable to SCV pulls. Anything below 5 Colossus before the transition is very risky. I dont quite understand the gameplan here. You commit heavily to blink which is only good to save your colossus, but how do you expect to keep them alive vs 12+ vikings with only 3 available. You also commit to double forge but then transition into tech which does not benefit of the double upgrades (storm). All in all this seems extremely vulnerable to every sort of pressure Terran put on because you commit to both double upgrades and fast tech. This gameplan will most definitely die to SCV pulls before storm is out - your storm choice is not even a colossus fake or something like that so it may not even be ready for 1/1 pulls or so. something to think about best That is correct: the idea is that by showing colossus you will discourage an scv pull, or the terran will aim for a higher viking count expecting you to go up to 5 colossus, and then pull their scvs too late. You're playing with the timings with this: it relies on tricking terran. If the all-in comes late from terran, you'll have storm, and colossus will stop bio/mine from slaying you. However, if terran just attacks before storm is finished, you are very likely to die, which was why this fell out of favor in the first place. | ||
| ||
