|
Ok, so I've been having a bit of trouble in my ZvP. A lot of my problems come from the strength of 2-gate openings, for which I have discovered that just getting a crap-ton of slings and +1 melee works quite effectively, and sometimes wins me the game immediately after obliterating their attack.
The problem arises, however, in games where I can hold off the 2-gate, and they transition into 3-gate/robo with mostly a stalker/colossus mix. I know the general answer here is Ultras, which is fine and dandy, but the problem is getting Ultras out fast enough when you're already a bit behind from having to prepare for a 2-gate opening that threatens to become 4-gate. So I've been toying around with ideas of how to best hold off the stalker colossus army until I can get sufficient income to support a sling/ultra army or even a sling/corruptor--->broodlords gameplan.
The answer: Banelings!
This may seem counter-intuitive because banelings have such low health and colossi already incinerate ling armies in seconds. Also, nothing in the stalker/colossus army is light armored, so banelings only do 20 damage vs 35 (prior to upgrades). Those are true sentiments, but here's what I discovered in various unit testing sessions:
A Sling/Bling army effectively destroys an equal food Stalker/Colossus army for half to cost in gas and nearly equivalent minerals!
Why?
1) Banelings do all their damage up front.
2) Banelings do considerable splash damage.
3) The remaining stalkers/colossi that the banelings do NOT kill are sufficiently softened up for Slings to clean up afterwards.
This is, of course, assuming you have the speed upgrade for BOTH zerglings and banelings. Non-upgraded banelings will likely never reach their targets. Obviously, you would also want to try to engage on creep in a real match.
I also have only tested this in a unit test map, and with no upgrades for either party except zergling/baneling speed and colossus range. I plan to do more testing and post the results in this thread, varying the stalker:colossus ratio and the overall food count on both ends, as well as with and without upgrades on both ends.
Anyway, I wanted to share this discovery, as I previously thought the banelings would never work against this, but have since changed my mind. The nice thing about ling/bling is that they share upgrades with Ultras, which is likely what you are trying to go for later against this composition.
Thoughts?
About tests:
All tests conducted on creep, with a flank on the stalker/colossus ball, and timed such that the zerglings and banelings engage at roughly the same time.
Five tests at each scenario were conducted.
Speed upgrades were researched for zerglings and banelings, and range researched for colossi.
=============================================================
Tests/Results:
48 Food (9 Stalker + 5 Colossi) vs (29 Banelings + 67 Zerglings)
11.2 Zerglings survive
~16.7% Zergling survival rate
48 Food (12 Stalker + 4 Colossi) vs (28 Banelings + 68 Zerglings)
19.8 Zerglings + 0.2 Banelings survive
~29% Zergling survival rate
W/ +1 Melee upgrade:
24 Zerglings + 0.4 Banelings survive
~35.3% Zergling survival rate
W/ +2 Melee upgrade:
34 Zerglings + 1.6 Banelings survive
~50% Zergling survival rate
48 Food (15 Stalker + 3 Colossi) vs (27 Banelings + 69 Zerglings)
25.4 Zerglings survive
~36.8% Zergling survival rate
I'll post more tests when I have a bit more time to run them later. Thanks!
|
if it works 5 times i believe that it wasnt just coincidence 
|
A good Protoss who has 3-4 four sentries with his army will just forcefield and render your banelings useless
|
yeah, the surge damage "upfront damage" of banelings i was thinking might have something.
a lot of prolbems zerg have been having, i thought maybe the answer is baneling or infestors., with their splash. cz u just cannot go straight up single targeting units like ling roach hydra.
and if stalkers have blink, then that means u still kill the colossus, and u can back away.
|
whatever happened to hydra roach?
|
Seems interesting, but I'm worried about what happens if your opponent goes for +2 attack really fast while getting his army up. +2 colossus will one shot anything ending in "-ling" regardless of your upgrades. That could be really bad for your confrontation.
This also feels like it would be less and less effective the smaller the encounters become, meaning if he opts to harass you, or try a pronged attack, you could be in some serious trouble.
|
if you lay down a trap with burrowed ones it might work if there is no observer nearby
|
On August 30 2010 07:34 STS17 wrote:
Seems interesting, but I'm worried about what happens if your opponent goes for +2 attack really fast while getting his army up. +2 colossus will one shot anything ending in "-ling" regardless of your upgrades. That could be really bad for your confrontation.
This also feels like it would be less and less effective the smaller the encounters become, meaning if he opts to harass you, or try a pronged attack, you could be in some serious trouble.
it just means u sandwich, not just one side. which is something everyone should always get in the practice of doing as zerg. it's pretty much superior 99.9% of the time than one siding.
But yeah, you would need a lot of baneling; numbers below critical mass would produce bad results.
Even if stalkers have upgrades and let's throw in blink too... you will still melt any zlots and sentries as well as colossi. stalkers blink away, but if everything else dies, it's not a problem. Colossi are you only true problem. After reducing his colossi numbers, you can then back away your blings, macro, tech, counterattack, whatever.
In fact, it would actually be helpful if stalkers blinked away, kinda like how if marauders moved away from your banelings and left marines there. Stalkers are armored, and they are larger sized, reducing the AOE effectiveness of banelings
|
You were probably really really ahead without knowing, had really super ups vs no ups, or the protoss were really really bad. I mean... just think about the numbers at the most basic level and you'll see what you say is nonsense.
|
On August 30 2010 07:40 SuperJongMan wrote:
You were probably really really ahead without knowing, had really super ups vs no ups, or the protoss were really really bad. I mean... just think about the numbers at the most basic level and you'll see what you say is nonsense.
so you're saying "do the math on the economic viability"
|
Banelings work against any bunched up army if you can get them into the bunch. They work well against tanks as well as an example, but you have to do a drop to hit tanks most of the time.
|
I really can't see this working well, it only takes one cliff or blink and you have no response. Considering how littered maps are with cliffs, I'm going to remain very skeptical until I see some replays. Even grabbing everything behind your first few stalkers would delay the rest while you get a second or third volley off. I guess no one has done it to me, so maybe it works well, but I just don't see it. Replays?
|
On August 30 2010 07:39 Alphasquad wrote:
if you lay down a trap with burrowed ones it might work if there is no observer nearby
this is true but if they are going colossus there will prolly be an observer chilling with the army (although whenever I use banelings v P they are like "do banelings even work v P?" so it might work lol)
|
I updated the OP with some limited test results.
Here they are for those too lazy to scroll up:
+ Show Spoiler +All tests conducted on creep, with a flank on the stalker/colossus ball, and timed such that the zerglings and banelings engage at roughly the same time.
Five tests at each scenario were conducted.
Speed upgrades were researched for zerglings and banelings, and range researched for colossi.
=============================================================
Tests/Results:
48 Food (12 Stalker + 4 Colossi) vs (28 Banelings + 68 Zerglings)
19.8 Zerglings + 0.2 Banelings survive
~29% Zergling survival rate
48 Food (15 Stalker + 3 Colossi) vs (27 Banelings + 69 Zerglings)
25.4 Zerglings survive
~36.8% Zergling survival rate
I'll post more tests when I have a bit more time to run them later. Thanks!
I'll post more results later.
EDIT: I should note that in my testing, I make just enough banelings to equal 50% the gas spent by the opposing player.
|
...you're not factoring in forcefield
|
On August 30 2010 08:16 setanta wrote:
...you're not factoring in forcefield
That's not a stalker/colossus army ,is it? 
But yes, obviously if you throw different units in for either army, things change. We're talking about someone who is going strictly stalker/colossus or stalker/zealot/colossus.
|
any decent player is going to assume 'stalker/colossus' involves some sentries, which it should, and as such banelings are very ineffective at dealing with this kind of army
|
You're looking at it in a vacuum. No protoss player will not have sentries in the army. While yes, stalker colosuss would die to this, that'll never realistically happen. Even if they didn't originally plan for sentries as soon as they see a pure melee army they'll trap all your lings and kill half of them.
|
Go Ultra/ling. Its eats threw stalker/colossus like nothing. You should be able to get them out in time off 3 bases. Dont go take your forth and fast tech to hive and keep scouting to make sure he is going colossus. Cause my ultras dont so well verus voidrays : (
|
You're looking at it in a vacuum. No protoss player will not have sentries in the army
I watch replays of high level Protoss who don't use sentries all the time so you can't just assume they are there.
|
|
|
|
|
|
EPT
