|
So to preface, I have recently taken a break from playing games in SC2, and I have been taking some time to just drill my mechanics and iron out the hiccups in my play.
I belong to the school of thought which feels that macro-focused play is the best way to play zerg. I have been booting up a ton of vs AI games on Scrap Station and just seeing how fast I can get to a max army with maxed upgrades and all the while keeping gas and minerals as low as possible. I'm able to get to my third base fully maxed without floating more then 500 mins or gas, but when I start to produce hive-tech like Ultras or Brood Lords I begin to float a ton of mins. Obviously I need a fourth just for geysers, but fully saturating 3 bases takes about 90 drones, which I have heard other players say is pretty much the optimal amount of drones to have. When I get to 4 bases I generally maynard drones to take advantage of the open mineral fields, and put 6 of those existing drones onto the new gases. So my end-game winds up looking like this: 90 drones, 24 on Gas and 66 on minerals, and 4-5 queens which puts me at around 100 supply dedicated to macro. This leaves only 100 supply for army.
In watching tournament casts and replays of top-level zerg's, I see a lot of games where they get to this point and feel comfortable. They build whatever tech they want, constantly damaging the opponents economy and taking out their expos. Then somehow, with inferior econ, their opponent will somehow manage to get a few more power units, completely trounces the zerg army, and then it only takes one expo falling to evaporate zerg's huge lead, and its GG. I feel like End-game zerg is missing something at all levels of play, and I don't know what exactly it is, but I want to start a discussion on some tactics that could allow us to make our aggressive push more robust and powerful.
Obviously this heavy econ is conducive to a 300 food push style attack, which does work quite often, but in reality this 300-food push is more like a 150 food push because 100 supply was dedicated to econ. I'm wondering what the viability of a 4-base zerg all-in could be.
Here's a basic sketch of how it could look:
1. Max with 100 supply worth of whatever army comp you want and make sure you are fully upgraded.
2. Play extremely defensively until you can bank up a trust fund and max larva.
3. Suicide ALL drones, build a ton of spines, etc etc to free up 90 supply.
4. Produce 200 supply worth of actual units hopefully with money still in the bank, spit larva one last time and send your queens with your army into a good attacking position.
5. Attack and reinforce with all of your remaining larva / resources.
Now I know this is highly situational, and it's going to be rare that you will be able to pool the necessary money and larva, but I do see games where even the highest level players end up with trust funds and banked larva at the later stages of the game, and maybe one out of every 100 games this strategy could be the key to victory.
With an army of this size, you can hold your own at the front of his base with 100 supply worth of units, and then drop 50 supply worth at 2 different mining expos, use multiple nydus worms, etc. The possibilities are truly endless here. Colossus giving you trouble? Against the toss death ball, building even 2 or 3 too many corruptors can lead to a total loss, but when you push with 200 food just build 50 food worth of corruptors and still have a ground army that's 50% larger then what your opponent is used to facing.
Even if your opponent sees you losing drones to gain supply it will be too late for them to do anything but turtle because you will fill up that 100 supply in one or two production cycles. (Depending upon what you build and if you use macro hatches, etc.)
So to summarize, I'm basically I'm posting this to get some feedback from the community, and everything is theorycraft at this point. I'm sure other players have thought of this, but I have never seen this attempted in a high-level game. And I haven't seen anyone suggesting this in the forums. Does anyone think this is viable; why or why not?
|
Not gonna lie I skimmed your OP (I did read the five bullet = points) and I guess it COULD work... although I don't think building spines is realistic. If you have 80 drones you are going to drop 8000 minerals on spine crawlers and you're going to have to have enough money left over to build 80 supply worth of units, which isn't going to happen.
However, what you COULD do is just suicide all of your drones and have enough resources to build 90 more supply. 40 roaches will run you 3k minerals and 1k gas which isn't an unrealistic amount to spend. Since I've never controlled a maxed zerg army that didn't have ANY drones I can't say how strong this would be.
You better be careful with your army composition b/c if you don't make enough corrupters... there's pretty much no turning back. I guess you could try to delay your huge push so you get like 6k/3k or something and you actually could rebuild... but I'd be interested in hearing why somebody who is insanely good says this is a terrible idea.
|
Calgary25990 Posts
That seems like a terrible idea. I'm not sure how to say it besides that. 4 bases is not endgame and definitely not the point where you should be suiciding workers.
|
If the enemy sees you suiciding Drones, they will know all they have to do is hold out a wave or two by massing Cannons/Bunkers/Turrets/PFs. Bad idea.
|
Lol who has 100 drones on 4 bases o.O im at 60 usually, anything over 70ish and i feel like i have over droned.
Its also super all in, if you mess up or the opponent turtles up you have nothing to fall back on, literally.
|
Calgary25990 Posts
On February 26 2011 03:07 uSnAmplified wrote:
Lol who has 100 drones on 4 bases o.O im at 60 usually, anything over 70ish and i feel like i have over droned.
Its also super all in, if you mess up or the opponent turtles up you have nothing to fall back on, literally.
I've played and seen top level players having 90-100 Drones in many games.
|
Full Saturation before major diminishing returns = 2 workers per patch + 6 on gas or = 16+6
Super Saturation or maximum level is achieved at = 3 workers per patch +6 on gas or 24+6
From a theoretical standpoint you would want as many drones as possible before diminishing returns kick-in.
With Zerg this is achievable by expanding rapidly and transferring drones between bases. The reason players make more than 16+6 per base is for when they transfer drones to the next base.
Ideally 22-24 drones per base is what your looking for (after drones become buildings). For 3 bases this is around ~70.
Never sit at more than 90 drones for too long as this will cause your army to be too small to battle properly at 200/200.
Exception in ZvP the rule of ~70 drones is broken because of a plan for mass spine crawlers. An easy solution to not losing mining, establishing a defense for if you lose your army (While remaxing), as well as a few other benefits. Ultimately though any zerg player should reduce his drone count back somewhere between 70-80.
|
This is super all-in, but I really like the idea.
There are times that feel very comfortable, and so do they, where my money is needlessly stockpiling, since I dare not attack, since my second army wouldn't be up before they're in my base, after we fought.
I've not thought of the possibility of getting rid of the drones. I'd probably use my drones to attack their workers lol (Nydus in their bases, then attack with workers), then build as big an army as I can, and show a real steamroll.
Not very practical, but may be the right way to win the rare games where it's applicable.
|
On February 26 2011 03:05 Touch wrote:
If the enemy sees you suiciding Drones, they will know all they have to do is hold out a wave or two by massing Cannons/Bunkers/Turrets/PFs. Bad idea.
You could kill your own drones. Just make sure your opponent doesn't scout it XD
|
This is something that everyone will make fun of, TLO will pull off in some tourney, you'll claim it is, therefore, viable, and the conclusion will still be that it doesn't work.
|
|
|
On February 26 2011 03:15 Chill wrote:Show nested quote +On February 26 2011 03:07 uSnAmplified wrote:
Lol who has 100 drones on 4 bases o.O im at 60 usually, anything over 70ish and i feel like i have over droned.
Its also super all in, if you mess up or the opponent turtles up you have nothing to fall back on, literally. I've played and seen top level players having 90-100 Drones in many games. Thats pretty crazy, i usually practice 60-70 to have good saturation and large enough space for an army.
|
This seems like a really bad idea. I have no clue why you would want to put everything into one big attack.. this is really not have zerg works.
Also, in case you want to know, i find your way to "practise" pretty bad :p
To me it looks like you want to work something like this out because you are scared of playing 1on1 ladder, but at the same time you have noticed that practising against the AI is pretty worthless. So now youre looking for something that you can still practise against the AI of which you hope that it can be applied to a game vs a human...
Just start laddering and forget about everything you wrote there.
|
On February 26 2011 04:00 DarKFoRcE wrote:
This seems like a really bad idea. I have no clue why you would want to put everything into one big attack.. this is really not have zerg works.
Also, in case you want to know, i find your way to "practise" pretty bad :p
To me it looks like you want to work something like this out because you are scared of playing 1on1 ladder, but at the same time you have noticed that practising against the AI is pretty worthless. So now youre looking for something that you can still practise against the AI of which you hope that it can be applied to a game vs a human...
Just start laddering and forget about everything you wrote there.
Well thanks for taking the time to read my post even though you think it's silly 
Is it really a bad idea to test out your build in single-player? I heard LiquidTyler mention that he did this from time to time on his stream, Day9 does this all the time on stream to show a build and suggests doing this to learn builds, (And I may be wrong here) but I am pretty sure I read a post where someone was being coached by Root.Catz and Catz was having him do basically the same thing I was doing here. You're probably right though and my time would be spent better mass-gaming at this point, I just feel really awkward when I get to late-game, and start floating a ton of resources and I felt like vs AI would help me learn my expo and gas timings.
|
On February 26 2011 03:33 JaqMs wrote:Show nested quote +On February 26 2011 03:05 Touch wrote:
If the enemy sees you suiciding Drones, they will know all they have to do is hold out a wave or two by massing Cannons/Bunkers/Turrets/PFs. Bad idea. You could kill your own drones. Just make sure your opponent doesn't scout it XD ......Retarded as that sounds, it may work xD
|
This is dumb for a couple reasons.
1.) As others have pointed out, you're putting all your eggs in one basket.
But more importantly...
2.) Zerg doesn't work that way. There really isn't a Zerg army composition that even comes close to the effectiveness of a Toss (or even Terran) ball. The way we win in the late game is by remaking multiple disposable armies.
You can't do that if you don't have any drones...
|
A max army fight against protoss or terran is a tricky sort of business, and their are many different ways to approach the game in that way. A straight up fight at your front door with a max t or p army generally is bad. Its kind of complicated... I think it is really too simplistic to say that late game zerg is inefficient or terrible, its really a lot about positioning, timing, ect that determines the winner. For the most part, the big battle that determines the game in a zvx usually happens sometime around 12-15 minutes, and usually neither player is maxed. Its at that point, in most games, one player grabs map control and is able to "leverage" a victory, that ultimately comes in the late game. I will agree with you on this, in a late game scenario, against terran especially, where both players are neck and neck the zerg often is at a disadvantage; against terran more so then protoss because marine drops are so hard to deal with on 4-6 bases and so efficient for their cost.
|
You will probably have to throw multiple 200 armys worth of stuff at and endgame P or T to break him.
|
On February 27 2011 01:33 Fateless wrote:Show nested quote +On February 26 2011 04:00 DarKFoRcE wrote:
This seems like a really bad idea. I have no clue why you would want to put everything into one big attack.. this is really not have zerg works.
Also, in case you want to know, i find your way to "practise" pretty bad :p
To me it looks like you want to work something like this out because you are scared of playing 1on1 ladder, but at the same time you have noticed that practising against the AI is pretty worthless. So now youre looking for something that you can still practise against the AI of which you hope that it can be applied to a game vs a human...
Just start laddering and forget about everything you wrote there. Well thanks for taking the time to read my post even though you think it's silly Is it really a bad idea to test out your build in single-player? I heard LiquidTyler mention that he did this from time to time on his stream, Day9 does this all the time on stream to show a build and suggests doing this to learn builds, (And I may be wrong here) but I am pretty sure I read a post where someone was being coached by Root.Catz and Catz was having him do basically the same thing I was doing here. You're probably right though and my time would be spent better mass-gaming at this point, I just feel really awkward when I get to late-game, and start floating a ton of resources and I felt like vs AI would help me learn my expo and gas timings.
I don't think it's a bad idea to practise builds on singleplayer to get them polished up to their optimized version. However, it's no use doing this past the 50 or so food mark, because so many things can happen in a normal game, you'll be practising something you'll probably never encounter on ladder.
It's great to practise the frist 5 min or so of your opening, just to get so used to it you can do it without thinking. Getting each and every drone ASAP, OLs on times etc. all adds up to a big improvement.
I think a better goal for you would be: How much stuff do I have at the 4/5 min mark? Then try the same build again and again (like 15hatch/14pool) and see how you can get 1-2 drones more if you do everything perfectly. You could also compare 15hatch/14pool to 14pool/16hatch or 14gas/14pool/20hatch just to see how much economy to lose by going a safer build.
I think anything past that 4/5 min mark is just reaction to what the opponent is doing, and best practised by actual games, not by single player games.
|
4-base all-in. I hear a lot of inappropriate use of phrases on TL on a daily basis (timing attack comes to mind) but I've never heard of a 4-base all-in before.
edit: just fully read your post. This really is a 4-base all-in. And it's not a good idea.
|
|
|
|
|
|
EPT
