|
Before people start screaming about clones, I've already seen threads about bunker rushes. My question was unanswered in the previous threads, and I'm hoping that people in this thread can provide some useful insight.
I'm not much of a Terran player. Yeah, I've obviously played hundreds of games as Terran in my Starcraft career, but I still don't play Terran very often. In modern times, I'm a Zerg player.
Here's the question that I have: When playing maps on which a zergling-proof wall can be built at the Terran player's ramp (at least when you spawn in the appropriate position), is there any reason not to call bunker rushing a risk-free strategy that should be done every single game?
I've arrived at a logic for this from looking at both the Zerg and Terran players' perspectives.
I will use the following data to support my claim:
(1) Under ideal circumstances on a typical map, the SCV's average mining rate is 68 minerals per minute. (Self-tested in single player, fastest speed, on Python. If my average is a little bit off due to other variables, please tell me.)
(2) Under ideal circumstances on a typical map, the drone's average mining rate is 72 minerals per minute. (Self-tested in single player, fastest speed, on Python. If my average is a little bit off due to other variables, please tell me.)
(3) In the recent game of Jaedong vs. Canata (GAME SPOILERS FROM HERE ON; STOP READING NOW IF YOU DON'T WANT TO KNOW THE RESULT), each player spent the following resources in committing to/preventing the bunker rush:
*Jaedong pulled seven drones off of his line when it first began (4:08 into the game).
*Jaedong lost four drones attempting to defend against it.
*Jaedong lost one zergling attempting to defend against it.
*
*Jaedong's three remaining drones returned to mining one minute and four seconds after the bunker rush began.
*Jaedong lost the following minerals:
-Roughly 504 minerals were lost in the seven drones' potential mining time.
-Exactly 200 minerals were lost from the cost of the dead drones.
-Exactly 25 minerals were lost from the cost of the dead zergling.
-Due to the four drones being lost in the bunker rush, and the fact that the game
continued for another two minutes after the surviving drones continued mining, an
additional 576 minerals were lost in the four drones' potential mining time.
TOTAL MINERALS LOST FROM JAEDONG: 1305
*Canata pulled one SCV off of his line when it first began (4:08 into the game).
*That SCV was later killed by a zergling.
*Canata built one bunker in this rush.
*Canata lost a total of three marines in the process of this bunker rush.
*Canata spent the following minerals on the bunker rush:
-Roughly 204 minerals were lost in the one SCV's potential mining time (until the
end of the game).
-Exactly 50 minerals were lost from the cost of the dead SCV.
-Exactly 150 minerals were lost from the cost of the three dead marines.
-Exactly 100 minerals were lost from the cost of the destroyed bunker.
-TOTAL MINERALS LOST FROM CANATA: 504
We are looking at a defecit of 801 minerals for the Zerg player in a game that was just over seven minutes long.
This is obviously one simple example of a game with a bunker rush, and it is clear that dozens of other variables come into play. Here are some of them.
-In another game, the Zerg player may not have lost four drones.
-In another game, the Zerg player may have lost his expansion.
-In another game, the Zerg player may have reached the same or a better result by pulling more or fewer drones from his mineral line.
-In another game, the Zerg player may have opted to go for something like 9 overpool.
-In another game, the Terran player may have not even gotten the bunker up.
-In another game, the Terran player may have not lost any marines (or lost more) in the process of getting the bunker up.
-In another game, the Terran player may have not lost the SCV.
-In another game, the Terran player may have had to send two SCV's to find the Zerg player's spawning location.
....and the list goes on.
Now, I would like to take into consideration a hypothetical example:
In this game, the Zerg pulls eight drones off of his line to defend against the bunker rush. These drones are off of the line for exactly one minute. The Terran player initially sends two SCV's in order to find the Zerg, and begins building the bunker when he reaches the Zerg's base. He sends two marines to the Zerg, and they are killed by the eight drones. The eight drones then kill the SCV, and the Terran player cancels the bunker.
Basically, the Zerg player completely crushes the bunker rush.
Zerg player's losses: one minute of mining time for eight drones.
Terran player's losses: two marines, one SCV, one minute of mining time for the second scouting SCV, and the fee of canceling the bunker:
...in number form...
Zerg player's losses: 576 minerals
Terran player's losses: 243 minerals, plus the mining time of that SCV
This comes to a defecit of 333 minerals for the Zerg player, and the bunker rush failed miserably. Don't get on me about the cost of that mining SCV because it's hard to find bunker rushes that fail as hard as the one above.
Now, here's the last part that I want to throw in. In the event that the Zerg player fends the bunker rush off and the Terran is left with barely anything, the Zerg player often attempts to send zerglings into his base. What is the Zerg player supposed to do if that base allowed for a zergling-free wall? The Terran player can tech to vultures. He can expand early. He doesn't have to rush into getting turrets because the Zerg is behind by 300 minerals. He can take his time and control the game from that point on.
I'd like to ask my question again now. It is important to note that I am not QQing about bunker rushes. I am not calling Terran overpowered. I'm not a crybaby, so I ask that none of you start sobbing all over my thread either. How can I turn a Terran's bunker rush to my favor? Is there ever an instance where it's unprofitable for the Terran to do it?
The only thing I can think of is the Terran can't 14CC (which is greedy as hell to begin with), along with the fact that the economy from 8 raxing isn't as ideal as an economy can be.
|
I play P, so my thoughts on this are entirely theoretical, but I'm guessing that bunker rushing screws up a T's tech worse than it screws up Z's?
|
On July 25 2009 03:44 CaptainPlatypus wrote:
I play P, so my thoughts on this are entirely theoretical, but I'm guessing that bunker rushing screws up a T's tech worse than it screws up Z's?
Not by any stretch of the imagination.
If you check out Jaedong vs. Canata, Canata got vultures out before Jaedong could even get speedlings.
|
And that's the problem with going 12 hatch when you scouted 8rax with your drone.....
+ Show Spoiler +A.k.a. boxer vs yellow semis EVER OSL (for 12 hatch vs 8 rax..)
As a non-zerg player I would assume that 12 pool, overpool, and 9 pool have much less of a problem defending against 8rax... but not totally sure about that. 12 pool might still have some problems, but probably not overpool/9pool.
Basically, if you're greedy you can get owned easily. There's really not much you can do except hope for worse micro from your opponent + micro really well yourself so you don't lose drones. AFAIK if the terran is smart in their follow up with walling/vultures you're basically screwed if you don't defend well off the start.
|
I think this is a very interesting thread.
After watching the vod though, Canata went 8rax, so we need to account for the potential economy lost and minerals forgone compared to a 9 depot, 10-11 rax strategy.
|
Eshlow: The issue is not defending the rush. If you read the 'hypothetical example' section you will see that the OP addresses the situation in which the Bunker Rush is defended, to all extents and purposes, perfectly.
The point he is raising is that even after this perfect defense the Zerg is economically behind, and any possible Zergling advantage is negated by the Terran player's wall. In this situation going for a Pool first build would in fact exacerbate the problems, not reduce them (Even worse economy, even more 'useless' Zerglings.)
|
After watching the vod of the game you mentioned (not disclosing players again for spoiler purposes) a few times, I do not think the Z could have micro'd much better than he did.
But let's be realistic--the Z got unlucky on his scouting and did not see the barracks until almost TOO late to uncommit to the 12-hatch. As eshlow mentioned, 12-pool is already a very shaky defense. By the time Z scouted the barracks, there was not a whole lot the Z could do to prevent the economic damage inevitably done by the bunker rush.
One problem I do have with the OP's analysis:
You only started counting at the moment the T decided to bunker rush from an 8 rax. What about the decision to go 8 rax in the first place? How much economic damage is the T doing to himself relative to the Z by going this BO?
Does that economic sacrifice justify the 333 worst case? The 801 mineral actual-game case? I think you will find that the economic sacrifice of 8 rax vs. 12 hatch is somewhere in between those 2 figures.
Btw: I question the 333 worst case. The 1 minute figure is the result of 2 unsuccessful attempts to crush the bunker rush by the Z and the subsequent interception of reinforcing marines. If, indeed, the bunker rush was successfully "crushed" as you say with no loss the the Z, you will not have the 1:04 mining delay.
|
This is very interesting and i do believe you have a point, but are you saying that the bunker rush was a proxy rax or walk from base? If it is a proxy you will most likely have to build another rax to make that ling-free block. Other than that i guess you have a point, it kind of seems too simple when you put it that way lol. The only problem i see is that you (as the T) would have to macro harder which would allow the bunker rush to be countered easily otherwise you would not take full advantage of the situation.
|
i disagree, despite the points as to minerals lost, it delays the terran players expansion, slowing him down, also, a loss of the rines gives the zerg player temporary map control and more strength unitwise, and, if defended right, I'm pretty sure that the zerg player will come out ahead due to the t will have less when his lair tech comes out as opposed to the t doing a normal fast exp build. a terran stuck at a disadvantage without map control when the zerg has either mutas or lurkers is hard to come back from.
refering to canata vs jaedong game
+ Show Spoiler +I'm pretty sure that Jaedong lost not due to the bunk rush, more to his reaction to the vulture, when he sent his lings past it towards canata's base, allowing the vulture to get up his ramp - if he had delayed with lings, they prolly would have died, but the sunk at his nat could have finished allowing him to repel it from his main, and leaving him in the game.
|
@Newguy:
+ Show Spoiler +I believe he gg'd at the wraith. Zergling counter was only natural, he made a split second decision when he saw the vult to keep going, in hopes that his popping lings could block the ramp in time for the sunken to pop. And anyway that point, he really had no choice but to go for that counter. If he couldn't do any meaningful damage back to Canata, that tech/econ deficit would have been insurmountable.
|
besides not factoring in the minerals lost due to 8-raxing (which are probably pretty signifigant later depot=later scvs=later everything)
There is also a production disadvantage the terran faces from forfeiting map control.
If we take the example where the Zerg crushes the bunker rush and assume they built 8 lings. After those lings the Zerg can just pump drones safely. If the terran fast expands it is more time before they put any pressure on the Zerg, and if they tech to vultures a sunken will easily defend.
Either way I think the Zerg ends up ahead. The instantaneous mineral losses might favor the terran, but the Zerg can easily regain any losses and come out ahead by pumping drones.
EDIT: If that was confusing
what I'm trying to say is that your looking at it the outcome of the rush backwards. Instead of looking at how many minerals each player lost you should look at what they have after the rush.
In your example it might look like this
Zerg: 13 drones, 8 lings, 2 hatcheries with extractor or third hatch coming soon (map control alloing drone production)
Terran(i don't play terran so im not really sure about these): 11 SCVs, 2-3 marines, one rax, CC, gas/second rax building/factory or saved minerals (depending on follow-up).
Looking at it like this I'd say Zerg is ahead.
|
On July 25 2009 04:27 Newguy wrote:i disagree, despite the points as to minerals lost, it delays the terran players expansion, slowing him down, also, a loss of the rines gives the zerg player temporary map control and more strength unitwise, and, if defended right, I'm pretty sure that the zerg player will come out ahead due to the t will have less when his lair tech comes out as opposed to the t doing a normal fast exp build. a terran stuck at a disadvantage without map control when the zerg has either mutas or lurkers is hard to come back from. refering to canata vs jaedong game + Show Spoiler +I'm pretty sure that Jaedong lost not due to the bunk rush, more to his reaction to the vulture, when he sent his lings past it towards canata's base, allowing the vulture to get up his ramp - if he had delayed with lings, they prolly would have died, but the sunk at his nat could have finished allowing him to repel it from his main, and leaving him in the game.
Bunker rush also potentially KILLS zerg expo, forces him to make lings, takes a grip of drones off of mining for a very long time, and potentially kills quite a few drones (and we all know that a zerg losing one drone hurts more than a T or P losing even two or three...especially in the early game).
A T should NOT expo after bunker rushing (unless he kills the hatchery...but even then). That's...not a great idea. It depends on how much damage you do, but the T should continue playing safe or aggressively...an immediate expo doesn't fit in either of those categories.
|
On July 25 2009 04:53 lt.dunbar wrote:
besides not factoring in the minerals lost due to 8-raxing (which are probably pretty signifigant later depot=later scvs=later everything)
There is also a production disadvantage the terran faces from forfeiting map control.
If we take the example where the Zerg crushes the bunker rush and assume they built 8 lings. After those lings the Zerg can just pump drones safely. If the terran fast expands it is more time before they put any pressure on the Zerg, and if they tech to vultures a sunken will easily defend.
Either way I think the Zerg ends up ahead. The instantaneous mineral losses might favor the terran, but the Zerg can easily regain any losses and come out ahead by pumping drones.
EDIT: If that was confusing
what I'm trying to say is that your looking at it the outcome of the rush backwards. Instead of looking at how many minerals each player lost you should look at what they have after the rush.
In your example it might look like this
Zerg: 13 drones, 8 lings, 2 hatcheries with extractor or third hatch coming soon (map control alloing drone production)
Terran(i don't play terran so im not really sure about these): 11 SCVs, 2-3 marines, one rax, CC, gas/second rax building/factory or saved minerals (depending on follow-up).
Looking at it like this I'd say Zerg is ahead.
I think you vastly overestimate the usefulness of those 8 leftover lings. Zerg doesn't need to have zerglings out in earlygame zvt to pump drones. Making zerglings in early zvt sucks, especially with ZvT these days. Bunker rushes almost always get followed up by a vulture, which makes your leftover lings useless.
The fact that zerg can pump drones after the bunker rush is an odd point to make. Terran is going to be making scvs the entire time except for a few exceptions. And idk where you got those numbers, but no zerg is going to come out of a bunker rush with 13 drones and 8 lings unless the rush was a complete failure. It's more realistically going to be something like 10 drones 8 lings. Then you are forced to screw your economy even more by making a sunk to stop the vulture since 8 lings do nothing but delay a vulture for 35 seconds.
|
Our argument is that yes, it forces the zerg to cut economy to defend, but the terran has already cut about an equal amount of economy to get an 8 rax bunker rush in the first place. Bunker rushing with 8 rax is effective versus 12 hatch on close positions, but it's not unstoppable and if it fails to do damage (remember, the example you're basing this on is progamers), you're pretty far behind.
|
zerg will be behind for maybe 2 minutes if he 12 hatched and kept his expo. 2 haches (or 1) can pump drones A LOT faster than 1 cc can, and zerg is free to do it (because T pretty much lost his early force and inevitably zerg will have more lings). If the zerg puts up 1 sunken to block vults, then he is still ahead for quite a while.
edit: turned mins to minutes for clarity 
|
In my original post I noted that there is a certain amount of economic sacrifice made when going 8 rax. I believe that, compared to 14 CC, yes, 8 rax results in a lame economy.
However, while the arguments against my point are good, and while if they're true then I'm absolutely wrong about bunker rushing seeming to be a risk-free strategy, I wouldn't mind seeing some mathematical evidence here. Exactly how much of an effect does going 8 rax have on your economy in the first place? We can't only compare it to 14 CC, but how about other builds? Did you take into consideration the fact that Canata did in fact expand after Jaedong fended the rush off?
Also, I think that you're greatly overestimating the bad effect that this has on the Terran's economy. Everybody seems to believe that the Terran's economy is in shambles after the rush fails, and somebody even suggested that he has fewer SCV's than the Zerg has drones. This is absolutely untrue. The Terran player can easily keep constant production up on the SCV's, and even manage to throw down an expansion before 25 supply.
This continues looking to be like something that very well should be standard. Hell, I'm even thinking that fake bunker rushing, just to pull 7-8 drones off of the line for 30 seconds, would be a great idea.
|
I think your post should have a spoiler alert in the title....once someone is already reading a post on bunker rushing.....and then you mention a game, you already spoiled it for then. Even though you didn't tell them the results, you told them how the game is gonna play out.
|
I actually did some testing on this awhile ago
At the 3 minute mark:
8rax, 9depot, 15depot, with 2 SCVs sent out as soon as rax finishes
12SCVs mining, 2 SCVs rushing, 1538 minerals mined
9depot, 11rax, 16depot, 18CC, with SCV scout sent at 12 supply
16 SCVs mining, 1 SCV scouting, 1656 minerals mined
So the terran actually cuts 3 scvs just to do 8rax/9depot
|
8 Lings could really delay the vulture from attacking your base or even manage to kill it. If you have 8 Lings, he just can't leave his base with a wall-in and no units to defend. You have the threat to destroy or for him to pull scv's just to repair.
|
You're not really counting it right imo
Instead of doing what you did you should see what each player has after the bunker rush.
the zerg : 2 hatcherys, pool, X drones, gas?, X overlords, X zerglings, etc
the terran: 1 CC, 1barracks, gas(if meching), X SCVs, Xmarines
Then, you compare it to if the bunker rush had not happened and the terran went 10rax -> FE and you can see the effects of the bunker rush
If you have good micro/get lucky/your opponent sucks you can hold off bunker rushes without suffering any casualties
Bunker rushes are a part of the game, sometimes starcraft is like rock paper scissors it can't be helped. You should be enjoying the game instead of whining about minor imbalances.
|
|
|
|
|
|
EPT
