|
United States4883 Posts
Hey all, I'm sure everyone saw DH Summer and saw Stardust tear through some of the greatest zerg players in the world using a really sick double upgrade gateway-heavy strategy. From what I could tell, the strategy, in a loose form, went something like this:
[UPDATED]
-1-gate FE into 3 gate/forge
-+1 attack, add robo, saturate bases
-Add second forge/Twilight + 3 gates,
-Chronoboost out 2 immortals, take 3rd (~8:00),
-Add 3 gates, start warping in a ton of sentry/zealot
-Add 2 more gates for a total of 9 gateways
-@~12:00 attack with a big sentry/immortal/zealot ball
vs Hyun and Jaedong, Stardust opened with FFE and took his third earlier (around 8:30).
The exact build order is as follows:
+ Show Spoiler +9/10: Pylon
16/18: NEXUS
16/18: Forge
17/18: Pylon
17/18: Gate
17/18: Canon
18/26: Gas
19/26: Gas
23/36: Cybernetic Core
26/36: Zealot
@100 Gas: +1 Attack (5.50)
@100% Cybernetics Core: WG
32/36: Pylon
34/36: Stalker
38/44: Robot (6.15)
Sentry
40/44: 2x Gas
42/44: Pylon
44/52: Forge, then TC (7.00)
Sentry
7.45: 3x Gate
2x Obs
@100% +1 Attack: +2 Attack, +1 Armor
MsC
8.30: NEXUS
Start Immortal production
10.00: 3x Gate
10.30: Charge
12.30: 2x Gate
Of course, Stardust had several different variations depending on the situation: When Stardust scouted Jaedong going swarm hosts, he immediately pushed out before a critical mass; Upon scouting aggression, Stardust would generally defend and whittle away the roach/ling numbers and THEN attack. Stardust handled counterattacks with zealot warpins at the 3rd.
The basis of this build appears to be a heavily upgraded gateway army with A LOT of sentries. The idea of staying equal or ahead of zerg in upgrades is interesting as it keeps gateway units alive for much much longer during engagements and allows Stardust to effectively wear down his opponent. However, the drawback to this build is that it has a severe weakness to mutalisks and solid roach/hydra. It seems that Stardust will make the push in these circumstances anyway and attempt to do as much damage as possible, relying on the fact that the zerg player has to use larva and money to make attacking units and can't 1) expand, 2) drone, or 3) tech.
I like the possibilities of this build. It seems like a very solid playstyle that allows the protoss player to play VERY aggressively and kind of contain zerg the way zerg usually contains protoss. I've played with it some and had some moderate success on the masters level.
The key idea seems to be this: + Show Spoiler +5. Staying on low tech is dangerous vs this build, because FF immortal smashes low tech zergs, yet teching is super dangerous because of the pressure this build applies. It could just be that zergs don't know how to deal with it properly yet but in general it seems like zergs have to delay tech to get units bad against this composition (ling roach, and possibly hydras) or just die before their tech gets off the ground (like JD's swarmhosts in the dreamhack final game). [Courtesy of VikingZombie]
Anyone else have thoughts? Ideas? Opinions? It's such a cool build, I had to bring it to the community to ask what everyone thought.
VODs:
Stardust vs. Jaedong DH Summer 2013 (Not sure how to hyperlink with exact times on videos, but the finals start at 8:01:30)
Stardust vs. Hyun DH Summer 2013
Stardust's stream: http://www.twitch.tv/90stardust/new
DAY9 DAILY: Stardust's PvZ Timing Style
|
these arent that new, I remember seeing them by people like creator (I cant remember the first to do it, maybe seed or someone). on maps like entombed valley in WoL as anti-infestor timings.
|
To be honest I thought it was an awful build, it was basically just seemed like a very late soultrain but with a third and slightly more upgrades. The third especially seemed odd as he got it very late and then basically pushed 1 minute after it's done which makes little sense to me. It hardly payed off at that point and more importantly you hardly have the extra gates to use that bigger income anyway..
The heavy upgrades hardly seem useful for this push either, armor does practically nothing against roach/hydra heavy armies and the attack isn't even that beneficial either with most of the army being sentries.
Basically I think stardust's builds were rough but he micro along with it was fantastic which made up for that. Besides that though his opponents all kind of screwed up the defense and just didn't play that well. Perhaps they were just puzzled by the timings of the attack so close after the third expecting him to go into colossi or something more defensive. Personally I think his builds just hit way too late, soultrain hitting earlier with less upgrades is far more dangerous imo. You don't give such a huge timing window for muta's or other tech.
|
For people learning this, when the OP says lots of sentries he means LOTS of sentries. On Stardust's stream he had upwards of 18-20 sentries some games (and he used their fast attack speed to snipe banelings. It was fascinating), and usually at least 12-16.
This build is really confusing. Looking at it, it shouldn't work, but it does, and very well at that. I watched his stream for a while and he almost never lost with it. You have to have absolutely spot on forcefields to do well though. Anything less and you will lose.
|
So I have some things to add to this discussion that other people havent mentioned.
Usually in a gateway sentry immortal composition, the army will push out with sentries and zealots and a few immortals, and then once the gateways come up pure stalkers will be warped in to reinforce in almost all cases. This makes sense as stalkers benefit much more from FF than zealots do.
However, in this composition, zealots seem to be the army core. I dont think people are giving this as much attention as it deserves because as far as I am aware this is practically unheard of (zealots are generally considered to be pretty bad with the main army in pvz, with a few exceptions being stalker collosus into chargelot archon rain styles, early timings such as nani's man train, and zealot archon vs ling bane kinda stuff). Having zealots over stalkers sure has a few disadvantages, like with stalkers how you're able to cut chunks out of the army and kill it off with virtually 0 losses, however it does have quite a few advatanges I can think of (not relating so much to the timing but purely looking at the unit mix)-
1. Zealots are technically far more efficient as a combat unit per cost based on pure stats alone, and scale far better from upgrades. The double upgrades can easily be afforded with all the gas saved on stalkers with this build.
2. Stardust tends to go sentry zealot immortal and then with his 3rd base gas down he tends to get charge, continue double ups, and eventually add either a dt shrine for harass and archons or a HT shrine for archons, then MUCH later he may get storm. This does 2 things, gives him a huge powerful army mass to pressure with and hit any timings that open up (rather than hanging back and getting fast storm or building up a few collosus or something) which may be only good because zerg isnt used to playing against it, I'm not sure... and also gives him charge, which makes harass in general much stronger with warp prisms or proxy pylons. I love getting charge in pvz but you usually cant afford it this early because getting an upgrade for a non core army unit is expensive and means you can die to roach hydra- this build opens up a lot of harass options alot earlier than other builds would.
3. it gives the army much more mass. 10 sentries and 20 zealots costs the same as 20 stalkers. What seems scarier to you? However, after say a max of 18 sentries or so adding more seems to be redundant, so then with your unlimited FF's you can continue reinforcing your army with PURE minerals which is huge
4. With a sentry immortal stalker composition you can easily cut out chunks of the opponents army and kill them off. However, immortals and stalkers outrange sentries by 1, so unless you either have great micro or zealots to tank (which die really fast without double ups and large numbers) your sentries will move to the front to attack and slowly get picked off by roaches. This can be avoided, but even in pro games I often see the protoss losing a sentry here and there. Stardust basically never loses sentries to roach hydra once he has started reinforcing with zealots, and with his increased sentry count this pretty much means infinite FF's for the entire midgame, something that no other gas inexpensive build can claim.
5. Staying on low tech is dangerous vs this build, because FF immortal smashes low tech zergs, yet teching is super dangerous because of the pressure this build applies. It could just be that zergs don't know how to deal with it properly yet but in general it seems like zergs have to delay tech to get units bad against this composition (ling roach, and possibly hydras) or just die before their tech gets off the ground (like JD's swarmhosts in the dreamhack final game).
6. you can still switch into stalkers with blink later if you want, and arguably they're even stronger at this point because DOUBLE UPGRADES
Anyway these are just my initial impressions on this build and why it is so strong, from watching some of stardusts stream and of course most of the dreamhack games... I would value feedback to see if other people agree with my analysis obviously.
I do see some things that concern me however-
MUTALISKS- yes, just in general. Responding to mutas seems to be stupidly hard with this build, from what I've seen on stardust's stream either you have to scout it out quickly, or lack of roach(+hydra) and push to hit a timing before too many get on the field and kill him, or play a very uphill game with blink stalkers as opposed to the usual stargate response (which fortunately have double upgrades). Something I would probably do is instead of perhaps getting +3 attack, I would begin upgrading shields as armor and shield upgrades are very strong vs muta bounce, and only get +3 attack on the forge that finished +3 armor. However as we all know stalkers are pretty bad against mutas that never engage properly and simply contain or harass, so I'm not sure if the extra stalker upgrades would give you enough strength in your stalkers that you could leave some at home to defend while pushing across the map with enough in your main army to not die. The sentries are acutally pretty helpful too though cause of GS, and having way too many to snipe, and their AA is actually OK (in a husky video maxxed sentry with GS actually beats maxxed mutas) so perhaps this build is actually stronger vs mass mutas than the old standard 4 gate robo sentry expand into 3rd -> TC- gates-> stalker with +2 attack and blink. I'm not sure.
All I've seen from stardusts stream if he actually lets the zerg get mutas up is a battle that looks incredibly uphill, that generally if he wins I feel he just massively outplayed the zerg, which isnt very convincing.
In a time when most P are opening stargate and going skytoss to be safe against mutas (and still often dying to them) this seems like a big flaw in my opinion?
It might also be possible to 100% kill zerg if they dont go through either swarmhosts or hive tech or mass spine before getting mutas, I'm not sure about that either and I'd like other people's opinions. Most of the zergs on his stream that have reached mass muta stage have done so through swarmhosts, as opposed to rushing to it, and he does kill a lot of zergs with the 3 base timing so perhaps some of them are teching mutas?
Oh yeah, mutalisks make it much harder to get warp prisms out on the map which all but deny any chargelot warpins around the map and in the main and stuff, which seems to be one of the biggest perks to this build after the 3 base timing phase. So huge advantage of this build virtually negated.
Swarmhosts- Not too early swarmhosts seem strong against this on stardust's stream too, but he really really likes to delay his colossus and just counter attack everywhere for potentially game ending damage, however if it doesnt get done he has no colossus and seems to be behind.
Maybe the answer here is to get colossus and blink off 3 base, after charge and some warp prisms? usually you wont have charge against swarmhosts before the AA (corruptors) comes out but with this build it seems like you could, so chargelots + WP seems like it has a really stupidly high damage potential against a swarmhost build. However I would prefer to do this behind using my endless FF to delay swarmhosts turtling in my base teching to a standard stalker collosus composition, to force the swarmhosts out of position to attack, whereas stardust likes to keep posturing and pressuring with his army on the map while NOT teching or doing so quite slowly. This is pure theorycraft though so I would like to hear other peoples thoughts, I dont see why you couldnt just go into a standard game at this point- however, well timed swarmhosts seem to be the bane of stardust much more than other toss (who tend to die much later to the ultimate stephano army comp).
Ling/bane (maybe with OL drops)- the sentries are a huge liability but I'm pretty sure you could just go into stalker/archon with a sprinkling of zealots and be fine and play out a standard game, I feel like the 'waste of the extra sentries vs bane drops would be somewhat counteracted by the fact that you had a fast double forge. Again, not sure
anyway that was just some thoughts on what i'd been seeing and tried out myself with very little success, some replies with peoples thoughts would be terrific 
|
On June 21 2013 16:37 VIKINGZOMBIE wrote:So I have some things to add to this discussion that other people havent mentioned. Usually in a gateway sentry immortal composition, the army will push out with sentries and zealots and a few immortals, and then once the gateways come up pure stalkers will be warped in to reinforce in almost all cases. This makes sense as stalkers benefit much more from FF than zealots do. However, in this composition, zealots seem to be the army core. I dont think people are giving this as much attention as it deserves because as far as I am aware this is practically unheard of (zealots are generally considered to be pretty bad with the main army in pvz, with a few exceptions being stalker collosus into chargelot archon rain styles, early timings such as nani's man train, and zealot archon vs ling bane kinda stuff). Having zealots over stalkers sure has a few disadvantages, like with stalkers how you're able to cut chunks out of the army and kill it off with virtually 0 losses, however it does have quite a few advatanges I can think of (not relating so much to the timing but purely looking at the unit mix)- 1. Zealots are technically far more efficient as a combat unit per cost based on pure stats alone, and scale far better from upgrades. The double upgrades can easily be afforded with all the gas saved on stalkers with this build. 2. Stardust tends to go sentry zealot immortal and then with his 3rd base gas down he tends to get charge, continue double ups, and eventually add either a dt shrine for harass and archons or a HT shrine for archons, then MUCH later he may get storm. This does 2 things, gives him a huge powerful army mass to pressure with and hit any timings that open up (rather than hanging back and getting fast storm or building up a few collosus or something) which may be only good because zerg isnt used to playing against it, I'm not sure... and also gives him charge, which makes harass in general much stronger with warp prisms or proxy pylons. I love getting charge in pvz but you usually cant afford it this early because getting an upgrade for a non core army unit is expensive and means you can die to roach hydra- this build opens up a lot of harass options alot earlier than other builds would. 3. it gives the army much more mass. 10 sentries and 20 zealots costs the same as 20 stalkers. What seems scarier to you? However, after say a max of 18 sentries or so adding more seems to be redundant, so then with your unlimited FF's you can continue reinforcing your army with PURE minerals which is huge 4. With a sentry immortal stalker composition you can easily cut out chunks of the opponents army and kill them off. However, immortals and stalkers outrange sentries by 1, so unless you either have great micro or zealots to tank (which die really fast without double ups and large numbers) your sentries will move to the front to attack and slowly get picked off by roaches. This can be avoided, but even in pro games I often see the protoss losing a sentry here and there. Stardust basically never loses sentries to roach hydra once he has started reinforcing with zealots, and with his increased sentry count this pretty much means infinite FF's for the entire midgame, something that no other gas inexpensive build can claim. 5. Staying on low tech is dangerous vs this build, because FF immortal smashes low tech zergs, yet teching is super dangerous because of the pressure this build applies. It could just be that zergs don't know how to deal with it properly yet but in general it seems like zergs have to delay tech to get units bad against this composition (ling roach, and possibly hydras) or just die before their tech gets off the ground (like JD's swarmhosts in the dreamhack final game). 6. you can still switch into stalkers with blink later if you want, and arguably they're even stronger at this point because DOUBLE UPGRADES Anyway these are just my initial impressions on this build and why it is so strong, from watching some of stardusts stream and of course most of the dreamhack games... I would value feedback to see if other people agree with my analysis obviously. I do see some things that concern me however- MUTALISKS- yes, just in general. Responding to mutas seems to be stupidly hard with this build, from what I've seen on stardust's stream either you have to scout it out quickly, or lack of roach(+hydra) and push to hit a timing before too many get on the field and kill him, or play a very uphill game with blink stalkers as opposed to the usual stargate response (which fortunately have double upgrades). Something I would probably do is instead of perhaps getting +3 attack, I would begin upgrading shields as armor and shield upgrades are very strong vs muta bounce, and only get +3 attack on the forge that finished +3 armor. However as we all know stalkers are pretty bad against mutas that never engage properly and simply contain or harass, so I'm not sure if the extra stalker upgrades would give you enough strength in your stalkers that you could leave some at home to defend while pushing across the map with enough in your main army to not die. The sentries are acutally pretty helpful too though cause of GS, and having way too many to snipe, and their AA is actually OK (in a husky video maxxed sentry with GS actually beats maxxed mutas) so perhaps this build is actually stronger vs mass mutas than the old standard 4 gate robo sentry expand into 3rd -> TC- gates-> stalker with +2 attack and blink. I'm not sure. All I've seen from stardusts stream if he actually lets the zerg get mutas up is a battle that looks incredibly uphill, that generally if he wins I feel he just massively outplayed the zerg, which isnt very convincing. In a time when most P are opening stargate and going skytoss to be safe against mutas (and still often dying to them) this seems like a big flaw in my opinion? It might also be possible to 100% kill zerg if they dont go through either swarmhosts or hive tech or mass spine before getting mutas, I'm not sure about that either and I'd like other people's opinions. Most of the zergs on his stream that have reached mass muta stage have done so through swarmhosts, as opposed to rushing to it, and he does kill a lot of zergs with the 3 base timing so perhaps some of them are teching mutas? Oh yeah, mutalisks make it much harder to get warp prisms out on the map which all but deny any chargelot warpins around the map and in the main and stuff, which seems to be one of the biggest perks to this build after the 3 base timing phase. So huge advantage of this build virtually negated. Swarmhosts- Not too early swarmhosts seem strong against this on stardust's stream too, but he really really likes to delay his colossus and just counter attack everywhere for potentially game ending damage, however if it doesnt get done he has no colossus and seems to be behind. Maybe the answer here is to get colossus and blink off 3 base, after charge and some warp prisms? usually you wont have charge against swarmhosts before the AA (corruptors) comes out but with this build it seems like you could, so chargelots + WP seems like it has a really stupidly high damage potential against a swarmhost build. However I would prefer to do this behind using my endless FF to delay swarmhosts turtling in my base teching to a standard stalker collosus composition, to force the swarmhosts out of position to attack, whereas stardust likes to keep posturing and pressuring with his army on the map while NOT teching or doing so quite slowly. This is pure theorycraft though so I would like to hear other peoples thoughts, I dont see why you couldnt just go into a standard game at this point- however, well timed swarmhosts seem to be the bane of stardust much more than other toss (who tend to die much later to the ultimate stephano army comp). Ling/bane (maybe with OL drops)- the sentries are a huge liability but I'm pretty sure you could just go into stalker/archon with a sprinkling of zealots and be fine and play out a standard game, I feel like the 'waste of the extra sentries vs bane drops would be somewhat counteracted by the fact that you had a fast double forge. Again, not sure anyway that was just some thoughts on what i'd been seeing and tried out myself with very little success, some replies with peoples thoughts would be terrific
very nice analysis! i really think P players start to realize that gateway units arent weak at all and much less gimmick than fast hightech compositions were often it is "colossus gets sniped, i lose, colossus doestn get sniped, i win". with mass gateway armies with small support from 2-3 immortals its much less gimmicky and imo more fun to play.
stardust plays a lot of zealot + mass sentries and i think in the future we will see a lot more mass stalker + some sentries play which might be even harder for zerg to deal with now that fungal can be dodged and ling hydra isnt very good vs it at all especially if P mixes in more zealots and sentries once seeing ling hydra. both need double upgrades to be viable but as seen from stardust thats definetly a viable option. and its much more fun to play, play against and watch than mass voids or mass colossus. gateway + immo + archon + HT + phoenix are the most fun units of P
the reason it being more fun is basically: P can move out much earlier and retreat with MsC recall or mass FF + retreat or blink back. so no deathball needed to move out. its more open games were lot of attacks etc. happen. as a Z 16 sentries and infinite FF is both scary and while FF being an anti-micro spell the most hate came from sentry immo in WoL where Z had no answer to mass FF spam. now with SHs, better mutas and maybe even banedrop (which again you couldnt go for in WoL because you would die to sentry immo and still cant go for vs 95% SG openings in meta right now + ovis being very slow) there are options for Z. would be fun if more P started playing like this since vs colossus and/or voids Z has no other option than to fasttech to hive and/or go mass SH turtle which leads to boring long games.
|
On June 21 2013 12:34 Markwerf wrote:
To be honest I thought it was an awful build, it was basically just seemed like a very late soultrain but with a third and slightly more upgrades. The third especially seemed odd as he got it very late and then basically pushed 1 minute after it's done which makes little sense to me. It hardly payed off at that point and more importantly you hardly have the extra gates to use that bigger income anyway..
The heavy upgrades hardly seem useful for this push either, armor does practically nothing against roach/hydra heavy armies and the attack isn't even that beneficial either with most of the army being sentries.
Basically I think stardust's builds were rough but he micro along with it was fantastic which made up for that. Besides that though his opponents all kind of screwed up the defense and just didn't play that well. Perhaps they were just puzzled by the timings of the attack so close after the third expecting him to go into colossi or something more defensive. Personally I think his builds just hit way too late, soultrain hitting earlier with less upgrades is far more dangerous imo. You don't give such a huge timing window for muta's or other tech.
In most of the games I watched, he took his 3rd at 8:30. Imo, it isn't a late 3rd and by the time he push (around 12:00), it already payed off since he doesn't take 5th and 6th gas. The upgrades are an important part of this build and a lot of Protoss (like Tod for example) use a similar double forge build in this matchup. It's true that armor upgrades aren't really important when going Stalker heavy but when going Zealot/Sentry, it's actually a big deal. Hydra only deal 12 damages per shot, so a +1 armor advantage in top of guardian shield mean that hydra only deal 8 damages per shot to Zealots (-50% dps). Zealots and Immortal also scale way better than Stalker with attack upgrades (+2 per upgrad for Zealots and +5 per upgrad for Immortals against armored vs +1 per upgrad for Stalkers).
On June 21 2013 11:31 SC2John wrote:
Hey all, I'm sure everyone saw DH Summer and saw Stardust tear through some of the greatest zerg players in the world using a really sick double upgrade gateway-heavy strategy. From what I could tell, the strategy, in a loose form, went something like this:
-1-gate FE into 3 gate/forge
-+1 attack, add robo, saturate bases
-Chronoboost out 2 immortals, take 3rd (~10:00), +1 armor
-Add second forge/Twilight, warp in a lot of units
-@~12:00 attack with a big sentry/immortal/zealot ball
vs Hyun and Jaedong, Stardust opened with FFE and took his third earlier (around 8:30).
You might want to add this build in the OP:
9/10: Pylon
16/18: NEXUS
16/18: Forge
17/18: Pylon
17/18: Gate
17/18: Canon
18/26: Gas
19/26: Gas
23/36: Cybernetic Core
26/36: Zealot
@100 Gas: +1 Attack (5.50)
@100% Cybernetics Core: WG
32/36: Pylon
34/36: Stalker
38/44: Robot (6.15)
Sentry
40/44: 2x Gas
42/44: Pylon
44/52: Forge, then TC (7.00)
Sentry
7.45: 3x Gate
2x Obs
@100% +1 Attack: +2 Attack, +1 Armor
MsC
8.30: NEXUS
Start Immortal production
10.00: 3x Gate
10.30: Charge
12.30: 2x Gate
|
On June 21 2013 12:34 Markwerf wrote:
To be honest I thought it was an awful build, it was basically just seemed like a very late soultrain but with a third and slightly more upgrades. The third especially seemed odd as he got it very late and then basically pushed 1 minute after it's done which makes little sense to me. It hardly payed off at that point and more importantly you hardly have the extra gates to use that bigger income anyway..
The heavy upgrades hardly seem useful for this push either, armor does practically nothing against roach/hydra heavy armies and the attack isn't even that beneficial either with most of the army being sentries.
Basically I think stardust's builds were rough but he micro along with it was fantastic which made up for that. Besides that though his opponents all kind of screwed up the defense and just didn't play that well. Perhaps they were just puzzled by the timings of the attack so close after the third expecting him to go into colossi or something more defensive. Personally I think his builds just hit way too late, soultrain hitting earlier with less upgrades is far more dangerous imo. You don't give such a huge timing window for muta's or other tech.
From my POV, and note that I don't watch his stream, this build was incredibly effective since it didn't rely on a certain timing. I was wondering in the LR why no one went for straight up mutas and crushed him.
My thought on it is, the zerg is never safe against this style. There is never "ohh he's going colossi, I have 3 minutes until he can push out", he can push with 2,3,4,5 etc immortals and any number of gateway units. That means that the zerg has to play a lot safer and is kind of forced into preemptive units rather than drones, since if Stardust sees zerg with no units, there is nothing stopping him from walking over and killing the zerg.
I think this build is pretty cool. Makes protoss feel a lot less hit and miss, compared to using colossi, vr and hts. With good control it is probably a lot more... stable metagame vice compared to always finding new allins to do like is the design of the sc2 protoss.
|
|
|
they are completely shut down by infestors. they relied on zealot sentries.
|
|
|
United States4883 Posts
On June 21 2013 18:52 Zarahtra wrote:Show nested quote +On June 21 2013 12:34 Markwerf wrote:
To be honest I thought it was an awful build, it was basically just seemed like a very late soultrain but with a third and slightly more upgrades. The third especially seemed odd as he got it very late and then basically pushed 1 minute after it's done which makes little sense to me. It hardly payed off at that point and more importantly you hardly have the extra gates to use that bigger income anyway..
The heavy upgrades hardly seem useful for this push either, armor does practically nothing against roach/hydra heavy armies and the attack isn't even that beneficial either with most of the army being sentries.
Basically I think stardust's builds were rough but he micro along with it was fantastic which made up for that. Besides that though his opponents all kind of screwed up the defense and just didn't play that well. Perhaps they were just puzzled by the timings of the attack so close after the third expecting him to go into colossi or something more defensive. Personally I think his builds just hit way too late, soultrain hitting earlier with less upgrades is far more dangerous imo. You don't give such a huge timing window for muta's or other tech. From my POV, and note that I don't watch his stream, this build was incredibly effective since it didn't rely on a certain timing. I was wondering in the LR why no one went for straight up mutas and crushed him. My thought on it is, the zerg is never safe against this style. There is never "ohh he's going colossi, I have 3 minutes until he can push out", he can push with 2,3,4,5 etc immortals and any number of gateway units. That means that the zerg has to play a lot safer and is kind of forced into preemptive units rather than drones, since if Stardust sees zerg with no units, there is nothing stopping him from walking over and killing the zerg. I think this build is pretty cool. Makes protoss feel a lot less hit and miss, compared to using colossi, vr and hts. With good control it is probably a lot more... stable metagame vice compared to always finding new allins to do like is the design of the sc2 protoss.
I think it's quite interesting to note that this build is actually fairly similar to the soultrain, just a little later. HOWEVER, while they look the same, Stardust's build seems to have a completely different goal in mind. Instead of outright killing the zerg player with a dangerous push, the goal seems to be a soft contain. With so many gateway units outside, zerg has to keep making units and can't afford to take extra gases, can't afford to continue teching, can't expand. Meanwhile, Stardust is VERY SLOWLY adding on tech and getting his upgrades while staying fairly saturated. And unlike the soultrain, where eventually you run out of forcefields and can no longer survive, Stardust continually adds sentries and keeps the MsC around for indefinite pressure.
It seems that the zerg player is slowly mining out while trying to deflect any huge aggression, but they slowly fall behind in upgrades and tech.
On June 21 2013 21:21 Sated wrote:Show nested quote +-1-gate FE into 3 gate/forge
-+1 attack, add robo, saturate bases
-Chronoboost out 2 immortals, take 3rd (~10:00), +1 armor
-Add second forge/Twilight, warp in a lot of units
-@~12:00 attack with a big sentry/immortal/zealot ball I've been doing a fairly similar thing for a long time, only I attack from 2 bases instead of from 3. -1 Gate FE into 3 Gate/Forge -+1, 3 Sentries, Robotics Facility, Natural Gases, Saturate Bases -Twilight Council When Robo Finishes, 3 Immortals From Robo -+2, Charge, Templar Archives As TC Finishes, 4th Immortal -Add Gateways Up To 8 -@~13:00 attack with a big Chargelot/Archon/Immortal ball. I have done a 3 base version as well: -1 Gate FE into 3 Gate/Forge -+1, 3 Sentries, Robotics Facility, Natural Gases, Saturate Bases -Twilight Council When Robo Finishes, Continuous Immortal Production -+2, Charge, Templar Archives As TC Finishes -Nexus -Saturate The Nexus On Minerals Only -Add Gateways Up To 12 (!!!) -@~15:00 attack with a big Chargelot/Archon/Immortal ball. I actually think that this is a lot stronger as it doesn't rely on Forcefields so heavily, although I'm sure that my timings are slower than a pro-gamer could hit them at. People might say that it is weak against Roach/Hydra, but that's not true if you engage in a fairly open space where you can get a good surface area with you're Chargelots. You're essentially using your Chargelots as Forcefields, since they're mostly their to mess with the enemy's AI and to tank for your Immortals/Archons. Immortals and Archons shred Roach/Hydra if they have something buffering for them. It can't be that bad since I have a ~70% win ratio vs. Zerg using this type of style. The only problems with both of these styles is that they're quite weak to Mutalisk builds (you don't have much anti-air) and they're not amazing against SH builds either. Against Roach/Hydra or Ling/Ultralisk builds, though, they're pretty amazing!
I'm not as sure about these builds, they seem quite weak against any very aggressive zerg. The key to Stardust's build is that he gets a 3rd so he can continue to add on units; 2-basing it just makes it an all-in. I also think that your attacks come way too late and allow zerg too much time to breath; a 15:00 zealot/archon/immortal push is no stronger than a 15:00 colossus/stalker/sentry push (in fact, it's weaker). Against a zerg player that goes for something like the old roach/ling max or a hydra/ling player into SH or a fast mutalisk player, you're going to straight up die. Basically, I think that the meta rewards protoss players who step out of their bases earlier; if you're stuck in your base for more than 10:00, you've pretty much tossed the game as protoss.
Note: I say they're bad against hydra/ling and roach/ling because those compositions can contain you inside your base and trade more or less efficiently with your army before you have colossus.
|
I just skimmed over this, and keep in mind when fighting against mutas that stardust doesn't like void rays too much, so that might be one of the reasons why he doesn't transition into airtoss.
|
|
|
Can we get some links to VODs in the OP?
|
United States4883 Posts
On June 22 2013 00:10 Sated wrote:Show nested quote +On June 21 2013 23:03 SC2John wrote:On June 21 2013 18:52 Zarahtra wrote:On June 21 2013 12:34 Markwerf wrote:
To be honest I thought it was an awful build, it was basically just seemed like a very late soultrain but with a third and slightly more upgrades. The third especially seemed odd as he got it very late and then basically pushed 1 minute after it's done which makes little sense to me. It hardly payed off at that point and more importantly you hardly have the extra gates to use that bigger income anyway..
The heavy upgrades hardly seem useful for this push either, armor does practically nothing against roach/hydra heavy armies and the attack isn't even that beneficial either with most of the army being sentries.
Basically I think stardust's builds were rough but he micro along with it was fantastic which made up for that. Besides that though his opponents all kind of screwed up the defense and just didn't play that well. Perhaps they were just puzzled by the timings of the attack so close after the third expecting him to go into colossi or something more defensive. Personally I think his builds just hit way too late, soultrain hitting earlier with less upgrades is far more dangerous imo. You don't give such a huge timing window for muta's or other tech. From my POV, and note that I don't watch his stream, this build was incredibly effective since it didn't rely on a certain timing. I was wondering in the LR why no one went for straight up mutas and crushed him. My thought on it is, the zerg is never safe against this style. There is never "ohh he's going colossi, I have 3 minutes until he can push out", he can push with 2,3,4,5 etc immortals and any number of gateway units. That means that the zerg has to play a lot safer and is kind of forced into preemptive units rather than drones, since if Stardust sees zerg with no units, there is nothing stopping him from walking over and killing the zerg. I think this build is pretty cool. Makes protoss feel a lot less hit and miss, compared to using colossi, vr and hts. With good control it is probably a lot more... stable metagame vice compared to always finding new allins to do like is the design of the sc2 protoss. I think it's quite interesting to note that this build is actually fairly similar to the soultrain, just a little later. HOWEVER, while they look the same, Stardust's build seems to have a completely different goal in mind. Instead of outright killing the zerg player with a dangerous push, the goal seems to be a soft contain. With so many gateway units outside, zerg has to keep making units and can't afford to take extra gases, can't afford to continue teching, can't expand. Meanwhile, Stardust is VERY SLOWLY adding on tech and getting his upgrades while staying fairly saturated. And unlike the soultrain, where eventually you run out of forcefields and can no longer survive, Stardust continually adds sentries and keeps the MsC around for indefinite pressure. It seems that the zerg player is slowly mining out while trying to deflect any huge aggression, but they slowly fall behind in upgrades and tech. On June 21 2013 21:21 Sated wrote:-1-gate FE into 3 gate/forge
-+1 attack, add robo, saturate bases
-Chronoboost out 2 immortals, take 3rd (~10:00), +1 armor
-Add second forge/Twilight, warp in a lot of units
-@~12:00 attack with a big sentry/immortal/zealot ball I've been doing a fairly similar thing for a long time, only I attack from 2 bases instead of from 3. -1 Gate FE into 3 Gate/Forge -+1, 3 Sentries, Robotics Facility, Natural Gases, Saturate Bases -Twilight Council When Robo Finishes, 3 Immortals From Robo -+2, Charge, Templar Archives As TC Finishes, 4th Immortal -Add Gateways Up To 8 -@~13:00 attack with a big Chargelot/Archon/Immortal ball. I have done a 3 base version as well: -1 Gate FE into 3 Gate/Forge -+1, 3 Sentries, Robotics Facility, Natural Gases, Saturate Bases -Twilight Council When Robo Finishes, Continuous Immortal Production -+2, Charge, Templar Archives As TC Finishes -Nexus -Saturate The Nexus On Minerals Only -Add Gateways Up To 12 (!!!) -@~15:00 attack with a big Chargelot/Archon/Immortal ball. I actually think that this is a lot stronger as it doesn't rely on Forcefields so heavily, although I'm sure that my timings are slower than a pro-gamer could hit them at. People might say that it is weak against Roach/Hydra, but that's not true if you engage in a fairly open space where you can get a good surface area with you're Chargelots. You're essentially using your Chargelots as Forcefields, since they're mostly their to mess with the enemy's AI and to tank for your Immortals/Archons. Immortals and Archons shred Roach/Hydra if they have something buffering for them. It can't be that bad since I have a ~70% win ratio vs. Zerg using this type of style. The only problems with both of these styles is that they're quite weak to Mutalisk builds (you don't have much anti-air) and they're not amazing against SH builds either. Against Roach/Hydra or Ling/Ultralisk builds, though, they're pretty amazing! I'm not as sure about these builds, they seem quite weak against any very aggressive zerg. The key to Stardust's build is that he gets a 3rd so he can continue to add on units; 2-basing it just makes it an all-in. I also think that your attacks come way too late and allow zerg too much time to breath; a 15:00 zealot/archon/immortal push is no stronger than a 15:00 colossus/stalker/sentry push (in fact, it's weaker). Against a zerg player that goes for something like the old roach/ling max or a hydra/ling player into SH or a fast mutalisk player, you're going to straight up die. Basically, I think that the meta rewards protoss players who step out of their bases earlier; if you're stuck in your base for more than 10:00, you've pretty much tossed the game as protoss. Note: I say they're bad against hydra/ling and roach/ling because those compositions can contain you inside your base and trade more or less efficiently with your army before you have colossus. The whole point of Stardust's build is to avoid using Colossi as they are easily countered by Corruptors or Vipers, the difference is that he focuses on double upgrades and Sentries whereas I focus on Archons and Immortals. To say that Colossus/Stalker/Sentry works better is to ignore the reason you don't want to be making Colossi in the first place. EDIT: I think Stardust and ToD showed how well Chargelot/Archon/Immortal/Templar compositions can work vs. Zerg. As long as the Zerg doesn't go for Swarm Hosts, I don't see any reason to build Colossi vs. Zerg...
I would say that against any standard zerg army, even before they have viper/corruptor, colossus/stalker/sentry is a stronger army. That's generally why we saw Rain doing big 15-minute pushes with this kind of force and not trying to get a big zealot/archon force. Zealot/archon is great in low numbers and it's a very lean composition, but against a fully loaded zerg army, it's even worse than making a bunch of colossus. IF THIS IS NOT TRUE, then why did no one go for huge zealot/archon pushes at 15:00 in GSL, instead preferring the traditional colossus/stalker/sentry/immortal armies? I'm not trying to disagree, I just see colossus-based armies as a lot more powerful against big zerg armies compared to zealot/archon. I feel Stardust bridges this gap with the double upgrades and forced equal income for much of the game, whereas the build you're talking about allows a zerg to fairly easily take a 4th base, get any tech they want up, and continue upgrades. I don't think this is bad against zergling/ultralisk or some forms of hydra/ling, but I think it has a lot of inherent weaknesses that should probably be looked at. I'll be honest, I used to go balls to the wall zealot/archon against a friend who used a heavy ling/bling mass expanding style of play, and it worked beautifully.
Also, I agree that the ToD games are worth watching. It was super ballsy to just skip stargate when dealing with mutas, but ToD managed to pull it off quite well. On top of that, the ToD vs. Stephano games were incredible examples of how to deal with swarm hosts without colossus. Protoss has a pretty good future with gateway-heavy styles.
On June 22 2013 01:30 Chemist391 wrote:
Can we get some links to VODs in the OP?
Done. Also put up Stardust's stream.
EDIT: Also reformatted the OP a little bit and added a small selection of VIKINGZOMBIE's post to the OP.
|
I think the build is interesting and obviously it has potential from what we've seen StarDust do although it's hard to know the longevity of it. It feels very all-iny although I guess you can transtion. For those needed reaction's I added some notes below for you guys interested. I'm mid-high master on EU with some gm kills under my belt.
For you who are afraid of mutas this build can react back to mutas pretty well since you have 3 base, good upgrades and close by to blink stalkers/ archons.
Vs Swarmhost: If you go and bring an obs you do really well vs a zerg who have swarmhost as long as he haven't built up his numbers. All thanks to ff. The only zerg unit who can hit the protoss army behind perfect ff's is hydras vs sentrys which I feel is why zealots do good in the army more so than vs zerg flanks. With the range upgrade ofc hydras become more of a threat!
If you are vs muta I feel the best way is either to go immediately for the attack and kill, warping in pure stalker and make sure to pop guardian shield.
If you decide to wait it out a lot of blink stalker's and archon of 3 bace I feel does extremely well. Most people underprepare for the counter-attack. Zerg will have such a hard time engaging your main army. Make more cannon's than you ever thought would be needed and leave an archon.
Also something I do which I consider meta-gamy but have worked is leave 2 archon in the prism as a defense mechanism.
On some maps like akilon wastes you can move between your neutral and third this way and when mutas come they will go for the snipe of the prism and you can unload just in time and get of a double shoot with your archon's.
PandaTank busted out the unupgraded version of this vs Stephano during the HSC VII for you who would like to see that. It was on Neo Planet S.
- Mawxie
|
On June 22 2013 02:14 SC2John wrote:
I would say that against any standard zerg army, even before they have viper/corruptor, colossus/stalker/sentry is a stronger army. That's generally why we saw Rain doing big 15-minute pushes with this kind of force and not trying to get a big zealot/archon force. Zealot/archon is great in low numbers and it's a very lean composition, but against a fully loaded zerg army, it's even worse than making a bunch of colossus. IF THIS IS NOT TRUE, then why did no one go for huge zealot/archon pushes at 15:00 in GSL, instead preferring the traditional colossus/stalker/sentry/immortal armies?
I'm assuming Rain's 15 minute pushes from WoL? That was because his pushes were designed to kill the Zerg while his BLs were morphing... And obviously, you need stalkers to shoot up at the morphing BLs, since archon range is uber low.
|
United States4883 Posts
On June 24 2013 11:06 TRAvian wrote:Show nested quote +On June 22 2013 02:14 SC2John wrote:
I would say that against any standard zerg army, even before they have viper/corruptor, colossus/stalker/sentry is a stronger army. That's generally why we saw Rain doing big 15-minute pushes with this kind of force and not trying to get a big zealot/archon force. Zealot/archon is great in low numbers and it's a very lean composition, but against a fully loaded zerg army, it's even worse than making a bunch of colossus. IF THIS IS NOT TRUE, then why did no one go for huge zealot/archon pushes at 15:00 in GSL, instead preferring the traditional colossus/stalker/sentry/immortal armies?
I'm assuming Rain's 15 minute pushes from WoL? That was because his pushes were designed to kill the Zerg while his BLs were morphing... And obviously, you need stalkers to shoot up at the morphing BLs, since archon range is uber low.
There's a reason why players transitioned into broodlords. Roach/hydra fell out of style at some point because after a certain point, a big upgraded, maxed out colossus force could just kill them. The only thing that makes roach/hydra viable in HotS is viper (and hydra speed to some extent). Vipers come out a little faster than Blords, but as a protoss player, you still have an attack timing at around 15:00 (maybe a little earlier), which you can reinforce with zealot/archon.
I'm just not fully convinced that playing a single-upgrade passive zealot/archon style is good or somehow better than a colossus push so late in the game where zerg has at least 2-2 and 4-5 bases. In theory, the only reason why I see it working is because zerg misreads your aggression or is just unprepared for the push. I'd like to see some replays.
BACK ON SUBJECT, though, I feel like Stardust does something fairly similar, he just swaps the order of some things. Instead of getting up a bunch of tech, he gets 1) LOTS of units, then 2) double ups, then 3) chargelot archon. This seems to be a more functional line of thought because it means that you can pressure the zerg, slow down the economy, slow down the tech, so that when you DO make a big zealot/archon push, you're ahead by a full set of upgrades and you have equal economy to the zerg player. The alternate picture is a big zealot/archon/immortal/sentry push at ~18:00 with 3/2 and 4 bases against a zerg player with 2/2 and 4 bases. Again, though, this relies heavily on the protoss being very aggressive and on top of forcefields and reacting to scouting constantly; like all great aggressive builds, it walks a very slim line.
|
United States4883 Posts
Updated the OP to include Day9's recent daily as well as a revised general plan! Hope to see some more feedback on this cool style!
|
|
|
|
|
|
EPT
