|
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. |
You should always replace drones, why would you be starved on resources? Spines are literally free. Then when BL tech comes out, I go from 100 drones to around 70-80, but I will still mass spines and replace drones as I make spines.
Check out the replays in my ZvP guide. The "Extreme Lategame" ones. You'll see how I drone up, and how ridiculous I am with spines. I don't just build 20 spines in the middle of the map, I'll literally put 10-20 spines around my expansions.
Just be careful that you replace drones. If you don't replace them, you can get into really shitty spots where you have no minerals and can't afford any broodlords and just die to a counterpush after a big battle or something.
Also, you should re-root that 20 spine wall. The reason you make that spine wall is because you are worried about a frontal attack before broodlords are out. Once broodlords are out, you don't really care about a frontal attack anymore, it's warp prism harass that's a problem, so uproot all those spines and put them at expansions and around key tech. Check out that "Zerg sim city" thread recently about sim city around your greater spire. I don't hesitate to make like 6+ evo chambers to wall-in my greater spire in the extreme lategame to make sure it doesn't get sniped. Nowadays I'm pretty conscoius though, that I'll position my pools, evos, and roach warren in early game so that they set up a nice windmill sim city for lategame.
- 1500 masters zerg. really though check out my guide, there's 2 reps i just added on extreme lategame, i think itd help you.
|
On August 30 2012 10:49 Belial88 wrote:
Do you guys drone scout in ZvP?
I've been toying around with not drone scouting in ZvP. I got proxy gated twice, out of like 4 times this entire season, in a single session (damn playing at noon is so damn cheesy...).
I'm thinking of not drone scouting on 2 player maps, and drone scouting only on 4 player maps, including maps like Shakuras. Basically, any map where the overlord can get to their natural by like, 16 supply.
I mean, I could have held that proxy double gate inside my base on shakuras if I had drone scouted -_-
I threw up the 2 spines when I saw his empty ramp/natural but I would have definitely had them in time if I drone scouted.
Always, always, always, always drone scout in ZvP.
Honestly, I'd gladly give up a bit of mining time to scout his opener (at my level it's ~30% gateway expands), gas steal if he's on 1 base (don't have to worry about standard DT timing), delay his nexus, sometimes even hatch-blocking if he's going nexus first, and then if he's cheesing me I can proxy hatch in his base. Proxy hatches are always good -- nobody knows the right reaction. If they pull probes they lose more mining time than you spent, if he doesn't get up the right number of cannons in time I can start creep spread + spines in his base....just in general, drone scouts always pay off and give extra safety. That one time in a thousand games he's 4 gating, you'll see it coming easily. It just eliminates a lot of silly losses.
In the example in your thread about the pylon/forge wall-off: you can use that drone to take an expansion at your third! Then you could go 2 base macro hatch lair like it's ZvP, but he's invested tons of money and tech and econ and army into trying to kill you. Granted, if he scouts your third, you'll probably lose it, but it's worth it imo. Also good against cannon rushes, especially if he doesn't block your ramp (say he does it behind the mineral line)
Idk, just what I do. I'd rather play at a slight, slight disadvantage to have a much higher chance of defending cheese. Also, with a drone scout, you know where to send your 9 and higher overlords 
|
On August 30 2012 10:49 Belial88 wrote:
Do you guys drone scout in ZvP?
I've been toying around with not drone scouting in ZvP. I got proxy gated twice, out of like 4 times this entire season, in a single session (damn playing at noon is so damn cheesy...).
I'm thinking of not drone scouting on 2 player maps, and drone scouting only on 4 player maps, including maps like Shakuras. Basically, any map where the overlord can get to their natural by like, 16 supply.
I mean, I could have held that proxy double gate inside my base on shakuras if I had drone scouted -_-
I threw up the 2 spines when I saw his empty ramp/natural but I would have definitely had them in time if I drone scouted.
I do dronescout at 12 in ZvP, but only because I like to sneak in hatch first ( and some proxies just turns hatch first into an auto-lose situation. When I've decided to go for 15 pool straight from the get-go, I either send my first overlord to the vision of the ramp before sending it over to my opponents base. The build won't really change unless you get proxied anyway if you open 15 pool.
The only thing you really need to see that early is the probe and note where he is coming from ( is it the optimal path for a regular scouting probe ect ) so sending out a dronescout on 2 player maps can be a little bit of a loss if you don't plan on taking advantage of that information.
|
On August 31 2012 03:37 Mavvie wrote:Show nested quote +On August 30 2012 10:49 Belial88 wrote:
Do you guys drone scout in ZvP?
I've been toying around with not drone scouting in ZvP. I got proxy gated twice, out of like 4 times this entire season, in a single session (damn playing at noon is so damn cheesy...).
I'm thinking of not drone scouting on 2 player maps, and drone scouting only on 4 player maps, including maps like Shakuras. Basically, any map where the overlord can get to their natural by like, 16 supply.
I mean, I could have held that proxy double gate inside my base on shakuras if I had drone scouted -_-
I threw up the 2 spines when I saw his empty ramp/natural but I would have definitely had them in time if I drone scouted. Always, always, always, always drone scout in ZvP. Honestly, I'd gladly give up a bit of mining time to scout his opener (at my level it's ~30% gateway expands), gas steal if he's on 1 base (don't have to worry about standard DT timing), delay his nexus, sometimes even hatch-blocking if he's going nexus first, and then if he's cheesing me I can proxy hatch in his base. Proxy hatches are always good -- nobody knows the right reaction. If they pull probes they lose more mining time than you spent, if he doesn't get up the right number of cannons in time I can start creep spread + spines in his base....just in general, drone scouts always pay off and give extra safety. That one time in a thousand games he's 4 gating, you'll see it coming easily. It just eliminates a lot of silly losses. In the example in your thread about the pylon/forge wall-off: you can use that drone to take an expansion at your third! Then you could go 2 base macro hatch lair like it's ZvP, but he's invested tons of money and tech and econ and army into trying to kill you. Granted, if he scouts your third, you'll probably lose it, but it's worth it imo. Also good against cannon rushes, especially if he doesn't block your ramp (say he does it behind the mineral line) Idk, just what I do. I'd rather play at a slight, slight disadvantage to have a much higher chance of defending cheese. Also, with a drone scout, you know where to send your 9 and higher overlords
All of the situations that you named just kind of seem like things that you will never see. Especially on 2 player maps their is almost nothing you can gain by drone scouting. Ya if he is gate expanding (which isn't very likely) you can see if he takes a 2nd gas or not, but you should be able to figure out what he is doing anyways by what units come out of his gateway and have your overlord spot for a nexus.
|
On August 31 2012 05:12 Moosegills wrote:Show nested quote +On August 31 2012 03:37 Mavvie wrote:On August 30 2012 10:49 Belial88 wrote:
Do you guys drone scout in ZvP?
I've been toying around with not drone scouting in ZvP. I got proxy gated twice, out of like 4 times this entire season, in a single session (damn playing at noon is so damn cheesy...).
I'm thinking of not drone scouting on 2 player maps, and drone scouting only on 4 player maps, including maps like Shakuras. Basically, any map where the overlord can get to their natural by like, 16 supply.
I mean, I could have held that proxy double gate inside my base on shakuras if I had drone scouted -_-
I threw up the 2 spines when I saw his empty ramp/natural but I would have definitely had them in time if I drone scouted. Always, always, always, always drone scout in ZvP. Honestly, I'd gladly give up a bit of mining time to scout his opener (at my level it's ~30% gateway expands), gas steal if he's on 1 base (don't have to worry about standard DT timing), delay his nexus, sometimes even hatch-blocking if he's going nexus first, and then if he's cheesing me I can proxy hatch in his base. Proxy hatches are always good -- nobody knows the right reaction. If they pull probes they lose more mining time than you spent, if he doesn't get up the right number of cannons in time I can start creep spread + spines in his base....just in general, drone scouts always pay off and give extra safety. That one time in a thousand games he's 4 gating, you'll see it coming easily. It just eliminates a lot of silly losses. In the example in your thread about the pylon/forge wall-off: you can use that drone to take an expansion at your third! Then you could go 2 base macro hatch lair like it's ZvP, but he's invested tons of money and tech and econ and army into trying to kill you. Granted, if he scouts your third, you'll probably lose it, but it's worth it imo. Also good against cannon rushes, especially if he doesn't block your ramp (say he does it behind the mineral line) Idk, just what I do. I'd rather play at a slight, slight disadvantage to have a much higher chance of defending cheese. Also, with a drone scout, you know where to send your 9 and higher overlords All of the situations that you named just kind of seem like things that you will never see. Especially on 2 player maps their is almost nothing you can gain by drone scouting. Ya if he is gate expanding (which isn't very likely) you can see if he takes a 2nd gas or not, but you should be able to figure out what he is doing anyways by what units come out of his gateway and have your overlord spot for a nexus.
Not only that, but if your overlord sees a gateway you can send a super late drone scout, float through the zealot, and see his second gas, if you want to know whether or not he is 4-gating. I agree no reason to drone scout on 2 player map unless you want to do 15 hatch or hatch block.
|
On August 31 2012 03:58 Chaosvuistje wrote:Show nested quote +On August 30 2012 10:49 Belial88 wrote:
Do you guys drone scout in ZvP?
I've been toying around with not drone scouting in ZvP. I got proxy gated twice, out of like 4 times this entire season, in a single session (damn playing at noon is so damn cheesy...).
I'm thinking of not drone scouting on 2 player maps, and drone scouting only on 4 player maps, including maps like Shakuras. Basically, any map where the overlord can get to their natural by like, 16 supply.
I mean, I could have held that proxy double gate inside my base on shakuras if I had drone scouted -_-
I threw up the 2 spines when I saw his empty ramp/natural but I would have definitely had them in time if I drone scouted. I do dronescout at 12 in ZvP, but only because I like to sneak in hatch first ( and some proxies just turns hatch first into an auto-lose situation. When I've decided to go for 15 pool straight from the get-go, I either send my first overlord to the vision of the ramp before sending it over to my opponents base. The build won't really change unless you get proxied anyway if you open 15 pool. The only thing you really need to see that early is the probe and note where he is coming from ( is it the optimal path for a regular scouting probe ect ) so sending out a dronescout on 2 player maps can be a little bit of a loss if you don't plan on taking advantage of that information.
That seems like its your fault for trying to sneak a hatch first. Even if you DO get it down, tosses that goes FFE can just cannon and kill you. If you go 15 pool there is no reason u should die to ANY type of proxies so there is no need to scout. By 2 30~~ IF you have no see the probe scout around your base with a drone (you want to minimize mining loss time). There is nothing to see if you drone scout at 12 in zvp. Grats you saw that he doesnt have anything in his base.... oh look the zealots are already here! So what did that scouting info do? Nothing. Send your first ov to scout and second ovie to your ramp because you are losing ovie scouting time keeping an ovie over a ramp that is not doing anything.
|
On August 31 2012 05:23 phodacbiet wrote:Show nested quote +On August 31 2012 03:58 Chaosvuistje wrote:On August 30 2012 10:49 Belial88 wrote:
Do you guys drone scout in ZvP?
I've been toying around with not drone scouting in ZvP. I got proxy gated twice, out of like 4 times this entire season, in a single session (damn playing at noon is so damn cheesy...).
I'm thinking of not drone scouting on 2 player maps, and drone scouting only on 4 player maps, including maps like Shakuras. Basically, any map where the overlord can get to their natural by like, 16 supply.
I mean, I could have held that proxy double gate inside my base on shakuras if I had drone scouted -_-
I threw up the 2 spines when I saw his empty ramp/natural but I would have definitely had them in time if I drone scouted. I do dronescout at 12 in ZvP, but only because I like to sneak in hatch first ( and some proxies just turns hatch first into an auto-lose situation. When I've decided to go for 15 pool straight from the get-go, I either send my first overlord to the vision of the ramp before sending it over to my opponents base. The build won't really change unless you get proxied anyway if you open 15 pool. The only thing you really need to see that early is the probe and note where he is coming from ( is it the optimal path for a regular scouting probe ect ) so sending out a dronescout on 2 player maps can be a little bit of a loss if you don't plan on taking advantage of that information. That seems like its your fault for trying to sneak a hatch first. Even if you DO get it down, tosses that goes FFE can just cannon and kill you. If you go 15 pool there is no reason u should die to ANY type of proxies so there is no need to scout. By 2 30~~ IF you have no see the probe scout around your base with a drone (you want to minimize mining loss time). There is nothing to see if you drone scout at 12 in zvp. Grats you saw that he doesnt have anything in his base.... oh look the zealots are already here! So what did that scouting info do? Nothing. Send your first ov to scout and second ovie to your ramp because you are losing ovie scouting time keeping an ovie over a ramp that is not doing anything.
First off, hatch first isn't an insta-loss to anything involving cannons. Sure if you allow cannons to go up behind your mineral line, you will be at a great disadvantage, similar to 15 pool. But stop acting like hatch first is an instant ticket to losing to cannon rushes, because it only is if your drone micro is abysmal.
Second off, not sending your first overlord over the ramp or not dronescouting at 12 ( atleast to the ramp ) will give the opponent the opertunity to do just what that huge zvp ramp block is about, 10 pylon and forge block off.
Thirdly, that 12 drone scout will see all that it needs to see ( gateway first? hatch first will have a huge advantage. Attempt at nexus first? A little opening for hatch block and a much easier time to go hatch first as well ). You won't get completely gutted by going for a dronescout and it allows for much more adaptability.
Lastly, I never suggested keeping that overlord on top of the ramp. I suggested moving it to the ramp before sending it over to the opponent to make sure you won't get 10 pylon forge blocked. Please do read my post fully before fuming at me for being wrong.
|
I can normally hold off anything terran or protoss can through at me, but ZvZ is my weak point, I NEVER win against them, all of the builds I try fail. Any little trick? anything at all could help! pretty please with sugar on top
|
On August 31 2012 05:12 Moosegills wrote:Show nested quote +On August 31 2012 03:37 Mavvie wrote:On August 30 2012 10:49 Belial88 wrote:
Do you guys drone scout in ZvP?
I've been toying around with not drone scouting in ZvP. I got proxy gated twice, out of like 4 times this entire season, in a single session (damn playing at noon is so damn cheesy...).
I'm thinking of not drone scouting on 2 player maps, and drone scouting only on 4 player maps, including maps like Shakuras. Basically, any map where the overlord can get to their natural by like, 16 supply.
I mean, I could have held that proxy double gate inside my base on shakuras if I had drone scouted -_-
I threw up the 2 spines when I saw his empty ramp/natural but I would have definitely had them in time if I drone scouted. Always, always, always, always drone scout in ZvP. Honestly, I'd gladly give up a bit of mining time to scout his opener (at my level it's ~30% gateway expands), gas steal if he's on 1 base (don't have to worry about standard DT timing), delay his nexus, sometimes even hatch-blocking if he's going nexus first, and then if he's cheesing me I can proxy hatch in his base. Proxy hatches are always good -- nobody knows the right reaction. If they pull probes they lose more mining time than you spent, if he doesn't get up the right number of cannons in time I can start creep spread + spines in his base....just in general, drone scouts always pay off and give extra safety. That one time in a thousand games he's 4 gating, you'll see it coming easily. It just eliminates a lot of silly losses. In the example in your thread about the pylon/forge wall-off: you can use that drone to take an expansion at your third! Then you could go 2 base macro hatch lair like it's ZvP, but he's invested tons of money and tech and econ and army into trying to kill you. Granted, if he scouts your third, you'll probably lose it, but it's worth it imo. Also good against cannon rushes, especially if he doesn't block your ramp (say he does it behind the mineral line) Idk, just what I do. I'd rather play at a slight, slight disadvantage to have a much higher chance of defending cheese. Also, with a drone scout, you know where to send your 9 and higher overlords All of the situations that you named just kind of seem like things that you will never see. Especially on 2 player maps their is almost nothing you can gain by drone scouting. Ya if he is gate expanding (which isn't very likely) you can see if he takes a 2nd gas or not, but you should be able to figure out what he is doing anyways by what units come out of his gateway and have your overlord spot for a nexus.
I don't know, I guess things differ? I actually experience gateway openings a lot of the time, and if I drone scout I can get the gas right after my pool, which is awesome. I also like to go hatch first, and against gateway openers I'll open hatch-pool-gas, to get a fast ling speed + 2 queens. While you're right, I just like the extra, faster information I get from drone scouting. And again I can delay their expands if they delay mine.
|
On August 31 2012 03:37 Mavvie wrote:Show nested quote +On August 30 2012 10:49 Belial88 wrote:
Do you guys drone scout in ZvP?
I've been toying around with not drone scouting in ZvP. I got proxy gated twice, out of like 4 times this entire season, in a single session (damn playing at noon is so damn cheesy...).
I'm thinking of not drone scouting on 2 player maps, and drone scouting only on 4 player maps, including maps like Shakuras. Basically, any map where the overlord can get to their natural by like, 16 supply.
I mean, I could have held that proxy double gate inside my base on shakuras if I had drone scouted -_-
I threw up the 2 spines when I saw his empty ramp/natural but I would have definitely had them in time if I drone scouted. Always, always, always, always drone scout in ZvP. Honestly, I'd gladly give up a bit of mining time to scout his opener (at my level it's ~30% gateway expands), gas steal if he's on 1 base (don't have to worry about standard DT timing), delay his nexus, sometimes even hatch-blocking if he's going nexus first, and then if he's cheesing me I can proxy hatch in his base. Proxy hatches are always good -- nobody knows the right reaction. If they pull probes they lose more mining time than you spent, if he doesn't get up the right number of cannons in time I can start creep spread + spines in his base....just in general, drone scouts always pay off and give extra safety. That one time in a thousand games he's 4 gating, you'll see it coming easily. It just eliminates a lot of silly losses. In the example in your thread about the pylon/forge wall-off: you can use that drone to take an expansion at your third! Then you could go 2 base macro hatch lair like it's ZvP, but he's invested tons of money and tech and econ and army into trying to kill you. Granted, if he scouts your third, you'll probably lose it, but it's worth it imo. Also good against cannon rushes, especially if he doesn't block your ramp (say he does it behind the mineral line) Idk, just what I do. I'd rather play at a slight, slight disadvantage to have a much higher chance of defending cheese. Also, with a drone scout, you know where to send your 9 and higher overlords
Eh. I can tell exactly what Toss is doing nowadays without having to have a drone in his base, if he's either 1 basing or 2 basing. The only recent example i can recall of a game where I didn't know how to respond (read: you dont need to know what he's doing, you just need to know how to respond, 2 gas quickly and not many gates = get spores and queens, no gas, get roach earlier, etc) was because the guy was a moron and went robo all-in at like 13 minute and hid the robo but realyl I had lost the game with really, really bad macro by the 8:00 mark more than anything else, i would have lost no matter what toss did that game.
That's why I started just pulling my drone home immediately after scouting Toss' opening for the last couple months, but recently I'm starting to think there's no point in even scouting Toss' opening on 2 player maps, as long as my overlord arrives to his main ramp around when my pool finishes.
I don't think using an outside drone vs a ramp block to make an expo is a good idea. Any good toss will just warp a single pylon by the hatch, and then start 1 cannons when it's near done, and a 2nd if it finishes, and cancel the 1st cannon if the hatch is cancelled and not make the 2nd if it's cancelled, so that Zerg ends up losing way more than Toss. A single drone won't do anything against a ramp block or cannon rush, and I have no problem dealing with cannon rushes (ezpz).
I agree with your sentiment about why not drone scout at our level of play... but I'm starting to think recently that I'm really not getting any valuable information with my drone. I think I'm just only going to drone scout on maps that an overlord can't 100% be by toss' natural ramp by the time pool finishes.
|
First off, hatch first isn't an insta-loss to anything involving cannons. Sure if you allow cannons to go up behind your mineral line, you will be at a great disadvantage, similar to 15 pool. But stop acting like hatch first is an instant ticket to losing to cannon rushes, because it only is if your drone micro is abysmal.
No... hatch first is an autoloss to a high level cannon rush. I don't think at diamond or even lower masters you'll really see strong cannon rushers. At higher masters you'll run into people who just only cannon rush and god at they not only amazing at it, but abusive as fuck. I've started to respect cannon rushers recently just because the ones I've faced have been really damn good at it (the few that win that is, I rarely lose to it).
I think maybe it's possible to survive a cannon rush attempt with hatch first, but you are going to be so far behind economically because you pulled at least 4-5 drones to just hold/patrol around the hatch to deny pylon/cannon placement that you would have been way ahead economically if you just went pool first.
Also, I've tested this with GMs, this isn't my own experience I'm talking about here. I used to think the same thing, then some GMs showed me that it's impossible to hold a proper cannon rush that will block you out. 15 drones will not stop 5 cannons going down when they are placed behind minerals or put in a proper position where you can't hit them, on top of 3-5 other cannons warping in around it. With pool first, you will have lings out before this conundrum occurs.
When does your Overlord arrive to the Toss' ramp though, with the route you put it on? I think I may do that, if the overlord gets to their natural ramp in time to throw down 2 blind spines if I see an empty ramp/natural.
|
I just had this game:
http://drop.sc/244989
Air toss. He goes double stargate, and since I responded with infestors there's not really much pressure I can do in response. Whatever, fine, so I can get hive and stuff out super easy right?
Well i see the quick 2 base fleet beacon and know he's going air toss. I go double spire before hive, since I know that in these sorts of games upgrades are just so crucial.
But I'm literally on 7 bases, and I am just soooo starved for gas trying to get 3/3/3 (maybe in hindsight I could have ditched melee, whatever, attack more important with infested terran being so important) and 3/3 corruptors.
I kill his entire air fleet multiple times but I don't know, it's like I can't really do anything to push him back in a timely manner, and then he masses lots of chargelots/archons to support a small air fleet made again instantly. Ugh.
It was a tough game. Maybe I lost too many infestors? The micro was pretty goddamn hard, corrupt everything, focus fire everything (so if you queue up a focus fire shift on multiple like carriers, when you cast corruption on them, it resets them to a-move or do they continue with their focus fire queue?), IT spam plus FG. I was just so energy starved the whole time.
I've faced air transitions, and those imo are usually really easy to deal with. Maybe I was racing upgrades too hard, and had too little army early on? I just felt so damn starved on gas the entire game.
|
On August 31 2012 09:56 Belial88 wrote:Show nested quote +First off, hatch first isn't an insta-loss to anything involving cannons. Sure if you allow cannons to go up behind your mineral line, you will be at a great disadvantage, similar to 15 pool. But stop acting like hatch first is an instant ticket to losing to cannon rushes, because it only is if your drone micro is abysmal. No... hatch first is an autoloss to a high level cannon rush. I don't think at diamond or even lower masters you'll really see strong cannon rushers. At higher masters you'll run into people who just only cannon rush and god at they not only amazing at it, but abusive as fuck. I've started to respect cannon rushers recently just because the ones I've faced have been really damn good at it (the few that win that is, I rarely lose to it). I think maybe it's possible to survive a cannon rush attempt with hatch first, but you are going to be so far behind economically because you pulled at least 4-5 drones to just hold/patrol around the hatch to deny pylon/cannon placement that you would have been way ahead economically if you just went pool first. Also, I've tested this with GMs, this isn't my own experience I'm talking about here. I used to think the same thing, then some GMs showed me that it's impossible to hold a proper cannon rush that will block you out. 15 drones will not stop 5 cannons going down when they are placed behind minerals or put in a proper position where you can't hit them, on top of 3-5 other cannons warping in around it. With pool first, you will have lings out before this conundrum occurs. When does your Overlord arrive to the Toss' ramp though, with the route you put it on? I think I may do that, if the overlord gets to their natural ramp in time to throw down 2 blind spines if I see an empty ramp/natural.
I've faced plenty of two-probes-pulled straight up cheese cannon rushes before that knew just how to place the cannons down behind the mineral lines and abuse highground and whatnot. I'll grant you that it's much much easier to lose to than any other early cheese. But to claim that there are only cannon-rush-protosses in high masters would be a fallacy.
I will grant you that in certain cannon rushes it is much more economical to go for a pool first than a hatch first ( mostly because you have to pull the drones for longer ) and in those cases any economic advantage you will have gotten from hatch first over pool first is nullified. But an autoloss granted that you don't let any blocked-behind-mineral-line cannons get started seems a bit of an overstatement.
I tend to send my overlord straight to the main when I dronescout, and that usually arrives at give or take 3:00. The slight detour will delay that by about 15 to 20 seconds ( depending on the map ofcourse ).
I'll go and calculate just what you are losing when you are going for a dronescout ( 9, 10 and 12 supply dronescouts ) just to chart in the exact amount of delays in a build. Because I'm fairly certain that timewise the delay is minimal.
|
On August 31 2012 14:26 Belial88 wrote:I just had this game: http://drop.sc/244989Air toss. He goes double stargate, and since I responded with infestors there's not really much pressure I can do in response. Whatever, fine, so I can get hive and stuff out super easy right? Well i see the quick 2 base fleet beacon and know he's going air toss. I go double spire before hive, since I know that in these sorts of games upgrades are just so crucial. But I'm literally on 7 bases, and I am just soooo starved for gas trying to get 3/3/3 (maybe in hindsight I could have ditched melee, whatever, attack more important with infested terran being so important) and 3/3 corruptors. I kill his entire air fleet multiple times but I don't know, it's like I can't really do anything to push him back in a timely manner, and then he masses lots of chargelots/archons to support a small air fleet made again instantly. Ugh. It was a tough game. Maybe I lost too many infestors? The micro was pretty goddamn hard, corrupt everything, focus fire everything (so if you queue up a focus fire shift on multiple like carriers, when you cast corruption on them, it resets them to a-move or do they continue with their focus fire queue?), IT spam plus FG. I was just so energy starved the whole time. I've faced air transitions, and those imo are usually really easy to deal with. Maybe I was racing upgrades too hard, and had too little army early on? I just felt so damn starved on gas the entire game.
haha belial I played that guy yesterday too.
http://drop.sc/245001
I'd say the most important difference is that I did a more thorough job delaying his bases, and taking them out via counterattack. the map was easier for me, I suppose.
I just watched the replay again and realized that I never finished the drop... I loaded up a big ling drop then got distracted. That was intended for taking out his main if he ever expanded again or pushed. it definitely would have cleaned out his main.
I kind of half base raced, because my infestors were out of position and my corruptors were on the late side (had to get rid of lings for supply :p, normally I get drops earlier so it is easier to get rid of the zerglings). Keeping infestors alive is a pretty big deal in terms of not getting gas starved.
My generally rule against skytoss is to focus on expansion management, using zerglings (drops and runbys) and infestors (when they have a ton of cannons, its not worth throwing away zerglings, but you can still use mass infested terran to get rid of the cannons. think about curious vs squirtle from TSL4) and get a huge economy lead so I can flood mass double upgrade corruptors til he dies. Don't let him get his 3rd so easily after opening double stargate. Don't let him get a 4th ever. (or if he takes his 4th, make sure you kill either his main or his natural, or his 3rd or something).
|
Alright, so here's the data I've gathered on drone scouting. I've done 4 games where I sent my drone across Daybreak LE, moved all the way into the main to check gasses, and sent it back again to mine. In all the examples the dronescout took roughly 2 minutes and 11 seconds, which amounts to 90 minerals lost on a close mineral patch, which takes about 7 seconds to make a trip on that, (2x60 + 11) / 7 = 18 , 18 * 5 = 90 ).
I had three games with a dronescout and one without, the baseline. In all the games I went for 15 pool, 16 overlord and waited for the 300 minerals of the hatch to appear. The stats of the baseline game are the following:
- 200 minerals for pool @ 2:00
- 100 minerals for overlord @ 2:23
- 300 minerals for hatch @ 2:53
In the baseline game nothing was delayed, all the drones lined up perfectly so there is no mining time lost there.
Game with a dronescout with the 9th drone ( rally point to opponents main while an egg )
This game had quite a few hiccups in the build. For example, the 10th drone had a slight delay of about a second. The double-drone at 11 and 12 respectively weren't able to be made at the same time ( 12 drone was delayed by a second as well ). The 13th drone had another 1 second delay. These delays amount to a whopping 2.5 minerals lost in mining time compared to if they alligned properly.
Onto the real meat of the game, the spawning pool, 3rd overlord and hatch timings
- 200 mins for pool @ 2:10
- 100 mins for overlrod @ 2:33
- 300 mins for hatch @ 3:05
As you can see, it is a pretty universal delay of about 10 seconds across the board. The 9 dronescout also arrives at the base at 1:41, which is way earlier than any 15 hatch or 15 pool would go down. So even if you want to dronescout, this seems like an obvious no-go.
Game with 10 supply dronescout ( once 10th drone is in production, send a drone to opponent immedately )
Unlike the dronescout at 9, this opening had no hiccups in drone production. The 10th drone obviously wasn't delayed and the 12th drone aligned nicely with its 11th neighbor. The 13th drone had no delay, right when the larva fell off it could be built. Compared to the 9 dronescout it also arrived at the ramp 16 seconds later at 1:57, which is still more than enough time to decide between pool first or hatch first.
- 200 minerals for pool @ 2:05
- 100 minerals for overlord @ 2:29
- 300 minerals for hatch @ 3:00
Game with 12 dronescout ( send drone afther 12th drone is in production )
The difference between a 10 dronescout and a 12 dronescout is roughly 10 minerals mined extra, which is traded for arriving at the base 14 seconds later ( or 30 seconds compared to a 9 dronescout ) at 2:11. As with the 10 dronescout, there are no hiccups in the production. What is interresting though is the building timings:
- 200 minerals for pool @ 2:02
- 110 minerals for overlord @ 2:28
- 300 minerals for hatch @ 3:00
As you can see, the difference in timings is abysmal, most of which can be accounted for by drones in the mineral line moving around too much. The drone arrives at the ramp late enough to not being able to send a drone to the natural should you want to go for a hatch first. The later-supply dronescout may make it seem like it is more economical ( and it is, by about 10 minerals compared to 10 dronescout ) but infact it just doesn't allow you to exploit the data that you scout because you make the decision before you see something critical, or lack thereof, in his base. You can basically see if he cheesed or not, but after that its use comes down to slightly later nexus delay harassment or mineral line harass ( which would further lengthen your drone staying there, making it less economical anyway ).
My conclusion is as followed:
While it is absolutely clear that a 9 dronescout is detrimental to all of your builds and delays everything by a good 10 seconds, the difference between the other two is small enough that I'm even going to give the advantage to the 10 dronescout compared to the 12 dronescout simply for the reactability.
Both of the later dronescouts are behind the control by about 5 to 7 seconds in overall building delay in the most optimal conditions or a pylon block. As it gets the hatch down later than the no-drone-scout baseline game by about 5 seconds, there is no shortage of larvae production ( you need more than 15 seconds delay to cause a single loss in larvae production on a hatch ). That 5 seconds would be enough for the slightly quicker larva to be returning on its first mineral trip ( +5 minerals-ish compared to dronescouting ). A probe would have to delay a dronescouter by about 8 to 10 seconds to make an actual larva lost compared to fully optimal no-probe-block no-drone-scout game.
TL;DR - If you want to go dronescout, go 10 drone scout. The difference in early game timings compared to not dronescouting is small enough for anybody that isn't near GM league to even feel a bump in anything whatsoever, and even then it is mostly aestetical. If you are below mid masters and you say you won't dronescout because it isn't as economical, I will laugh generously as you try to justify being good enough to make use of that extra 5 seconds.
I will remind everyone that all of this was in ingame seconds, if you convert the difference to real time, the delay would be between 3.5 and 5 seconds.
|
I've faced plenty of two-probes-pulled straight up cheese cannon rushes before that knew just how to place the cannons down behind the mineral lines and abuse highground and whatnot. I'll grant you that it's much much easier to lose to than any other early cheese. But to claim that there are only cannon-rush-protosses in high masters would be a fallacy.
I will grant you that in certain cannon rushes it is much more economical to go for a pool first than a hatch first ( mostly because you have to pull the drones for longer ) and in those cases any economic advantage you will have gotten from hatch first over pool first is nullified. But an autoloss granted that you don't let any blocked-behind-mineral-line cannons get started seems a bit of an overstatement.
I tend to send my overlord straight to the main when I dronescout, and that usually arrives at give or take 3:00. The slight detour will delay that by about 15 to 20 seconds ( depending on the map ofcourse ).
I'll go and calculate just what you are losing when you are going for a dronescout ( 9, 10 and 12 supply dronescouts ) just to chart in the exact amount of delays in a build. Because I'm fairly certain that timewise the delay is minimal.
Just because you've beat it in Diamond, does not make it true. I think you are too low level to really know better here. I'm not even saying "Hey, I'm 1500 masters, this is MY experience", because for a long time I thought you could hold cannon rushes with hatch first too. This is something that a couple of GM's brought up to me, and proved to me. You just can't beat a high level cannon rush with Hatch First. It's impossible - the cannon rusher can afford enough building right before a pool finishes from a hatch first, that literally, pulling 4 drones to attack each cannon, would not be enough. You'll have 16 drones, against 4-5 warping cannons, many of them blocked off. This critical mass stage is prevented when going pool first because you'll have lings out by then, so ling+drone can take them down, but when going hatch first, you won't have the lings to deal with it.
I'm not saying I'm better than you here at all, as I'm too low level to really know better. This is going off of what multiple GM's proved to me.
http://www.teamliquid.net/forum/viewmessage.php?topic_id=280674¤tpage=3
You can check out this thread, i did a quick search and here's some discussion on the topic where ROOTrsvp, a pro player, states you can't hold hatch first against a proper cannon rush.
I hate to say "You are diamond and don't know what you are talking about" but it's kind of true here. At one point, it was "You are a mid-masters and don't know what you are talking about" when I was the one saying that you could go hatch first against cannon rush. Then some GMs and pro players made me look like an idiot.
|
Thanks for your analysis of your drone scout stuff though Chaos. But some points I'd like to say, i guess:
- why would you 9 drone scout (rally egg)? Even on daybreak, I find that sending a drone at 9, as in pulling a drone returning cargo, after the overlord is made, will get to the opponent's base in time to block a hatch first as well as reactively hatch/pool first. It's really pulling it at 10, because at 10 is when it's returning it's cargo and you send it out, if that makes sense. I know you are talking ZvP, but my point is that you can reactively hatch first or not with a 10 drone scout even on daybreak.
I'd recommend you bring the drone home immediately after scouting too, so you can get back to mining with it. The drone, no matter when you send it, will be gone for 2 minutes, meaning ~80 minerals lost. It is rather small of a loss, imo.
I do agree with your conclusion though. Drone scouting should be encouraged. in ZvZ I think 10 drone scouting is worth it to block hatch first, win against early pools (which happen enough to justify it, imo, if they never happened ever, drone scouting wouldnt be worth it), and against hatch first you block their hatch first and go 17g/17p reactively which gives a net gain over the other player (assuming they dont go 15h/17g/17p blind, which wouldnt make much sense....).
In ZvP, im not sure about drone scouting. I guess it's a matter of how much more economic it is than 14p/16h blind, although you have to drone scout on 4 player maps (but on those maps you couldn't 10 drone scout either). The problem with hatch first is that it autoloses if they go forge first, which happens more than nexus first (or hell, lets say it/s 50/50) so I don't see why I would 10 drone scout when the majority of the time I will just see that Toss went forge first and oh, I can't go pool first anymore. Maybe if like 90% of the time Toss went nexus first, then I'd go 10 drone scout 15 hatch first.
But another problem is that toss who go nexus first, which you'd scout to determine if you should go hatch first or not against, 9 probe scout, meaning they will always block your hatch first and if you pull 2 drones they will just put down a pylon. That's the problem with your theory here. Toss will just block your natural and nothing you can do. It's not like T/Z drone scouts which you just pull 2-4 workers against so you can throw down the hatch (2 vs Z, 4 vs Ebay block sort of crap).
So... if you are going to drone scout in ZvP, I'd say drone scout at 13, not 10. I don't think 10 drone scouting is a good idea in ZvP since Toss will just block your hatchery and prevent you from going hatch first. Also, from what I understand, gateway openers that reactively go forge, then zealot+cannon rush will autowin against hatch first too, so even if Toss isn't going FFE and is going gateway first, you lose. Not to mention hatch first will lose to proxy 2 gate, which you can't tell until after you throw down the hatch first (a proxy 2 gate can have a pylon on the low ground). And that most toss go forge first.
But i guess you assume you can hold hatch first against cannon rushes, which is the problem of your analysis.
|
Hey, I was wondering if someone could help me figure out what my problem is. I know it has something to do with macro but I don't know specifically what the problem is. I end up floating a TON of money into the late game and can't find the larva to spend it on. Here is a ZvP and ZvT, the ZvP I lose very stupidly and the ZvT I win but not in a satisfying way.
ZvP
ZvT
Thanks in advance!
|
On September 01 2012 07:40 Belial88 wrote:
Thanks for your analysis of your drone scout stuff though Chaos. But some points I'd like to say, i guess:
- why would you 9 drone scout (rally egg)? Even on daybreak, I find that sending a drone at 9, as in pulling a drone returning cargo, after the overlord is made, will get to the opponent's base in time to block a hatch first as well as reactively hatch/pool first. It's really pulling it at 10, because at 10 is when it's returning it's cargo and you send it out, if that makes sense. I know you are talking ZvP, but my point is that you can reactively hatch first or not with a 10 drone scout even on daybreak.
I'd recommend you bring the drone home immediately after scouting too, so you can get back to mining with it. The drone, no matter when you send it, will be gone for 2 minutes, meaning ~80 minerals lost. It is rather small of a loss, imo.
It was just for comparisons sake. I used to do 9 dronescout like a year ago and just wanted to know how much it'd hurt me is all.
On September 01 2012 07:40 Belial88 wrote:
I do agree with your conclusion though. Drone scouting should be encouraged. in ZvZ I think 10 drone scouting is worth it to block hatch first, win against early pools (which happen enough to justify it, imo, if they never happened ever, drone scouting wouldnt be worth it), and against hatch first you block their hatch first and go 17g/17p reactively which gives a net gain over the other player (assuming they dont go 15h/17g/17p blind, which wouldnt make much sense....).
In ZvP, im not sure about drone scouting. I guess it's a matter of how much more economic it is than 14p/16h blind, although you have to drone scout on 4 player maps (but on those maps you couldn't 10 drone scout either). The problem with hatch first is that it autoloses if they go forge first, which happens more than nexus first (or hell, lets say it/s 50/50) so I don't see why I would 10 drone scout when the majority of the time I will just see that Toss went forge first and oh, I can't go pool first anymore. Maybe if like 90% of the time Toss went nexus first, then I'd go 10 drone scout 15 hatch first.
I get where you are coming from. But even if I concede for the sake of furthering the discussion that forge first is indeed an auto-loss against an early cannon rush, going Nexus-Forge delays the forge for long enough that the delay of the pool to cancel out. 15 hatch 15 pool gets the pool down about a minute later than a regular 15 pool, the trade off is roughly the same for protoss but you'll have 3 more larvae than a 15 pool opening would have in the long run.
On September 01 2012 07:40 Belial88 wrote:
So... if you are going to drone scout in ZvP, I'd say drone scout at 13, not 10. I don't think 10 drone scouting is a good idea in ZvP since Toss will just block your hatchery and prevent you from going hatch first. Also, from what I understand, gateway openers that reactively go forge, then zealot+cannon rush will autowin against hatch first too, so even if Toss isn't going FFE and is going gateway first, you lose. Not to mention hatch first will lose to proxy 2 gate, which you can't tell until after you throw down the hatch first (a proxy 2 gate can have a pylon on the low ground). And that most toss go forge first.
Fair enough, I see your point since the thing you are scouting the most is the probe, and that'll be at your base around the 13-14ish supply mark anyway. In which case, against an intelligent protoss that controls his probe well, you won't get that hatch first up anyway ( or atleast, so incredibly late because of the later pool that you're actually behind a 15 pool opening ). Compared to just straight up cannon rushes, gateway -> forge cannon rushes are much easier to deal with and much more all in. The delay is long enough that if we consider a decent map size the zealot won't be there when the cannons go up anyway, you won't really have to deal with him until later when your pool is about to finish.
I don't see how he could get that rediculous amount of cannons up that you describe in the forge first example while also sending a zealot across the map. Sure if you let him get those sweet spots cannon wise, you roll over and die, so you don't let them get those unsurroundable spots by parking drones infront of the probe. I have never seen anybody that didn't straight up kill me with gateway-forge come back from being behind that far economically. If he isn't going for that gateway-forge style, because you know, he wants to get the other advantages of getting an earlier gateway out like teching, you will have a nice advantage.
I'll be happy to concede on these points, but I'm still convinced that on 4 player maps when the probe doesn't scout you first, going hatch first in the cases of a gateway opening or nexus first is a viable option. Whether it is on 2 player maps where the probe always gets to your natural in time, I don't tend to sneak a hatch in if the probe is moving around at the right places anyway. So I'll conclude that on 2 player maps it is much more scouting and probe related to get a hatch first off than it is for doing it blindly, which frankly purely relies on luck if forge first is an inevitable loss of the hatchery.
|
On September 01 2012 07:48 Marth753 wrote:Hey, I was wondering if someone could help me figure out what my problem is. I know it has something to do with macro but I don't know specifically what the problem is. I end up floating a TON of money into the late game and can't find the larva to spend it on. Here is a ZvP and ZvT, the ZvP I lose very stupidly and the ZvT I win but not in a satisfying way. ZvPZvTThanks in advance!
Typing from phone, please excuse typos.
zvp: super early gas, why? that's your 17th drone gone. you end up being down workers forever (you only catch up at 7min)
where is the third queen? your third is done, no queen for awhile, and late drones.
some supply blocks, e.g. 54 supply.
no scouting, colossi on the way and spire not started. overseer in and still no spire after seeing support bay. spire starts later, guess you're a little lucky there
need a macro hatch once three bases are saturated.
you have double his supply when you retreat from one colossus. this guy is all kinds of behind. he went 6gate colo blink with double ups. weird, he may just turtle and deathball you.
13 min: you are super maxed, banking 2k minerals, and need to start dropping more hatches.
you really really need to scoot in with roaches when engaging. 80% not shooting in some cases.
you then don't remax. well, you are low on larvae because you stopped injecting (before 8 min, all queens low energy; 16 min, 3 queens near 100 energy).
remax means rebuild the instant you lose roaches. you constantly have idle larvae, 10+, after the first losing engagement at his natural. that's lost larvae from natural spawning, in addition to potential army.
be pro, say gg.
summary for zvp: macro hatch after third is running, stay on top of injects so your bank has meaning, actually use larvae to replace army.
zvt: you don't maynard to natural or rally all new drones to natural, main at 19 workers to 2-3 at nat for awhile.
early bane nest. meh. just mentioning it because you seem not to consider the importance of each drone early on.
you went to 33/60 supply at one point (26 drones). bad. don't do that. see, 7 idle larvae for a bit.
prebuild a queen for your third, please.
again, need macro hatch, maybe 2 for ling heavy zvt.
you always seem to grab gas a bit late, watch that.
you have 15+ idle larvae for nearly one full ingame minute (11:30).
again, 20+ idle larvae at 13:00ish. all this time your fourth is done and empty.
you haven't scouted for his third at all.
37 idle larvae. giant supply block alongside. kill me now.
also you forgot 2-2 for lings.
no gas for banes, tons of infestors, not enough lings. could end badly.
jeez you have 20+ idle larvae almost constantly from 13 min on until you max out at like 18 min.
morph greater spire or start ultra cavern on hive completion.
no creep spread, at 20min.
8k minerals. 5 hatcheries.
terribly slow remax after losing two fights. injects also stopped again after max.
why 70 banes? no broodlords? theres a ton of tanks to stop your ultras, don't dump everything into banes.
zvt summary: macro hatches. use your godawful stockpile of larvae, it's shameful. stop losing infestors, don't just hold down the F key.
overall summary: use your larvae. add macro hatches, one after third base for sure, prob another after fourth base. use your larvae to saturate your new bases. use your larvae (yes, repeated). keep injecting even when maxed.
EDIT: Here's a weak (high diamond i think) zvp replay from yesterday of me fighting a similar 2-base toss, going blink+colossi and unable to take his third (unlike your toss opponent, he doesn't go double ups).
http://drop.sc/246180
some funky stuff aside (i take far third base because of bad positions), it works out similarly. notice that while I have idle larvae, it's because i've spent the majority of my resources, and thus can't actually morph them into anything. you had problems using larvae in the 120-130 supply range. i also overtake toss in worker count by 34ish, even after making a decent amount of lings; your game had you overtaking at 40ish.
because of the terribad positions, I decided that I needed a stronger standing army, so I made a ton of spines and remade the supply as army, so I tried to pick off what I could and just keep a strong army. notice how even in the second-to-last big fight, my supply never dips below 170, I pretty much spam rebuild things as I'm fighting. critical in these bad positions, where he can warp at my doorstep, but always important. as you can see from the last fight, if you rebuild as the fight still rages, the remax is so fast that he only has one warp-in cycle before i hit him again with a new army.
|
|
|
|
|
|
EPT
