|
There are girls. Most of my friends are girls. It's complicated. I'm happy.
Okay now that the life stuff is out of the way and I can type into the void of the internet blogosphere in peace, here's the theory.
It's not a new theory. It's the old BW theory really. Equalize production from your hatches relative to the terran/toss bases. THen adjust drones accordingly.
So far, I've found that 2 hatches and a queen is roughly equal to the same production capacity per terran/toss base. I came to this conclusion by running numbers on the maximum resources a terran or toss player can allocate through basic building structures. Ex. How many resources does toss constantly use for production out of a 4 gate or 3 gate.
Also, a queen produces larva at 4/3 the rate of a hatchery which means you can also adjust the drones necessary for production by 4/3. But wait, how many drones do you need?
Per Hatch of Production
Unit On Minerals Gas Geysers
Drone/Ling 5.55 0
Roach 8 1
Baneling 10 1.5
Hydra 10 1.5
Muta 10 3
Infestor 10 4.33
Corruptor 14 1.5
Ultralisk ~20 ~4
Why I'm doing this:
I think it's a common fallacy that if zerg CAN drone then he SHOULD drone. I think this is bad, because if one doesn't compensate with production along the way, one's income will be erratic and difficult to manage. Building an extra hatchery for every extra 6 drones one makes or a queen for every 8 extra drones leads itself to smoother macro. Further this opens oneself up for timing windows. Also if one CAN add more production earlier, that investment should pay off more than if one adds production all at once later in the form of macro hatches, which will open zerg up to more timings. This is in contrast to other races who slam down a bunch of rax or gateways at once, because zerg production is continuous out of larva. There isn't a sudden burst of resources when one expands, but a continual progression as the expansion is integrally connected to production. (I know you've heard that last part before. But just to reitterate common knowledge.)
Why else is this important? While terran and toss have guidelines that are relatively simple to follow with regard to the amount of production one can have per base, zerg is in much murkier water.
Further let's try applying this to a specific matchup.
Let's say it's ZvP and I"m aiming for a roach/ling composition transitioning into baneling drops.
First off, we know that production wise, we need only have 2 hatches and a queen to keep up with toss production on one base. I scout some form of one base play with 2 gas. Probably a 3 gate expo but potentially a 3 gate dt expo or some form of stargate play.
To keep up with production with my roach/ling composition, I find that if I allocate all of one queen's larva to roaches and the two hatchery's larva to lings, then I'll need
11 drones for lings
11 drones for roaches + 3 drones on 1 gas geyser.
What this means:
I have 22 drones on minerals when I equalize production with toss. I have an extra queen that I can use to gain a slight lead on my toss opponent. However if I'm going to use that queen just for drones/lings , then I need about 7 more drones. That's a total of 29 workers on minerals or 16 in my main and 13 at my natural.
However, I also now know, that in order to equalize my production with my income, in the future I should be looking to take a third base when I make my next 5 or 6 drones. The reason for this is that a hatchery to constantly produce lings/drones needs 5.55 drones. Being able to produce more drones will let you reach your other saturation levels.
So similar to the 5 drone rule in SC1 we have the 6-7 drone rule in SC2. (6-7 because you need 5-6 to constantly produce drones and 1 for the hatchery).
Other interesting things to note, are that if you're opening with ling/bane against protoss, you can switch to ling/hydra with the same income if you plan to maintain the proportion of lings:banes the same as lings:hydras.
Further notes: I'm experimenting now, with taking my third faster and using my second queen primarily for creep. The reason is that once toss takes his expansion, you'll want 2 queens and 4 hatches or some equivalent ratio. 3Q+3H is slightly more than toss production which is also good because you'll maintain a slight lead.
I think this info is useful and will define a much more "explosive" zerg style than the current turtle-macro oriented zerg style that we primarily see these days.
Also, I think this information is useful to note other things about expansion timing. For example if you want to constantly produce roaches out of 2 hatcheries and 2 queens, then you need 38 drones on minerals at optimal saturation. But you can't do that because at 32 drones on minerals, your saturation rate drops off. (most everyone should know this). So we should consider then, expanding to a 3rd or varying our unit composition if we want to produce a primarily roach based army.
More to come and happy starcrafting,
Aletheia
 
|
Hey! This is a reeeally well-thought out and planned write-up, I love the look of this!
I'll definitely do some testing about this myself, good luck to you man! this is really good information - really look forward to this getting developed
|
This is really cool ^^
Might wanna ask a mod to move it to sc2 general/strat, pretty helpful stuff :D
|
Hey,
nice writeup, and although your conclusions aren't really new, but mostly true, I have a few concerns:
First of all:
A hatchery spawns 1 larva every 20 seconds. An inject lasts 40 seconds and gives you 4 larvae. Isn't that 4 / 2 ?o.o
Why do you think people say "You should be ahead one base"? They don't mean "rush for 3base saturation", they mean: get an extra hatch, and not an extra hatch in your base, but make it already an expansion(you get an inbase hatch though if you can't defend that expansion for whatever reason).
As this is really something you should do, it's nothing new and just normal. If you have 2 saturated bases you can't just produce off of 2 hatcheries, you need a third one to be effective enough. This doesn't count in things like teching to have, having 3 evos running and upgrading all tech etc...
I would also not say that it is a rule as this would include that you always have to do it, which is just wrong. It is rather just a "rule of thumb", something you should consider but not always rely on. Sometimes it's just the wrong decision to go for the next hatch, sometimes it's exactly the thing you have to do.
You found an effective way to manage some aspects of your macro, thats good, but don't forget to play reactive, because thats still the key.
I hope you get what I'm trying to say, it's pretty late here. I don't want to go too much into detail
Pretty much
"Scout and then do stuff like this if it is the right choice" is what I'm trying to say, I guess xD
On a side note, what does this have to do with playing turtle-macro or "explosive"? You still do what the situation requires, and with deciding between inbase hatch and extra expansion you can vary pretty much there with achieving the same result production-wise.
But in general I agree with you and I think that there are pretty bad fallacies that many zerg players have. It's good to see that there are still some that try it in a different approach.
|
A hatch spawns 1 larva every 15 seconds. An inject lasts 40 seconds, but actually takes 45 seconds for the queen to have enough energy for the next inject.
Um, with regard to being a base up that's somewhat true. However in terms of production you would need a base and a half up per terran or toss base which is actually a lot more... Although there's a cap to that so it's not such a big deal in terms of income. It's huge ins terms of production. When some casters comment that nestea is playing "risky" with a double expand for his 4th and 5th base against terran, I find that it's actually not risky, but necessary in order to keep up with production.
I understand what you're saying by rule of thumb.
With regard to turtle-macro vs explosive. A lot of zergs I see these days, even at the high masters level, tend to turlte on 2-3 bases macroing up and don't expand. The other half seem to expand too quickly and either die to any timings or overwhelm with their production if economy kicks in. One player I'd like to note here is sheth. Sheth drones A LOT before committing to any units. This is fine of course. However i think these kind of drone:production guidelines allows one to be open to adopting timing attacks while maintaining a constant expansion path instead of simply mass droning. THe issue is, when zerg makes units at the moment, I don't think there are clearly defined metrics that one can use to judge whether they can out produce the terran/toss player, or to see where they stand in relation. I find this a useful metric for comparison as well as a useful tool for planning out my midgame timings.
I guess what I'm saying, is that a lot of players tend to "blindly" expand once they're fully saturated or notice they have too many minerals. In fact though, one can expand faster than that by delaying those extra drones because the resources those drones bring in can't be converted through production. And by setting up the extra production earlier, one can produce more drones later and then set up the next round of production earlier etc. So... I find this more "explosive" as you can actually ramp up a bit faster than just mass droning.
|
On August 24 2011 07:53 Aletheia27 wrote:
I guess what I'm saying, is that a lot of players tend to "blindly" expand once they're fully saturated or notice they have too many minerals. In fact though, one can expand faster than that by delaying those extra drones because the resources those drones bring in can't be converted through production. And by setting up the extra production earlier, one can produce more drones later and then set up the next round of production earlier etc. So... I find this more "explosive" as you can actually ramp up a bit faster than just mass droning.
This makes a lot more sense to me and I pretty much feel the same way 
I kinda go for the same approach but I still base it around scouting information ( consisting of economy, production and army size/army comp) all the time, so I guess you could just think of it as a way of securing an economic adventage while being safe! :D
And oops@Data, kinda mixed that up with something else.
|
I am a little confused about the presentation of the value.
This is saying that people should be expanding or getting a queen every 6-7 drones past the first 6 because they can't support the money with constant larva expenditure?
So you start with 6 drones, make your Ovie, and by the 13 drone mark, you are approaching an income greater than the capacity to produce drones nonstop, so we should expand (re: 15 hatch). Similarly, we can pool at such a point that by the time the queen pops we are at 13 drones, necessary for nonstop drone/ling production off the available larvae.
Does this also suggest that the "optimal" way to take a third would be at the next 7 drone break point? So, by 30 drone mark, since we need 13 drones to nonstop drones off 1H+1Q and 26 for 2H+2Q?
Does this also suggest limits on what you can mass up on X bases? So if you can't produce nonstop Roaches off 2 base, a 2 base timing should, for efficiency's sake, be a mixed Roach+Ling? Similarly with Hydras or any other higher tech unit, I suppose.
I guess I am not exactly sure what conclusions are directly resulting from this vs. what are being inferred from the data at hand.
|
Most of this post is primarily the data.
WIth regard to the queen. It should be 7-8 drones past whatever production point you're currently at. It is not necessarily implying that people should be expanding or getting a queen every 6-8 drones past that production point however because people may choose to switch compositions midway. That switch will require a different amount of drone saturation in relation to the production. If one sticks wit hthe same composition however, then yes, they should be expanding or getting a queen every 6-8 drones assuming it's possible and doesn't put one's base in danger.
I ws implying that before you take an expansion, you should generally produce 7 drones in addition to however many you're using for whatever you're currently producing. This allows you to use your new hatchery immediately without any waste of larva. It's also the earliest you could use that new hatch in terms of your total drone count.
Your statement of why 15 hatch or pool around 14 shows exactly why this type of data works. So yes.
With regard to when to take a third. Un interrupted and purely producing drones, one would take their 3rd around the 33 drone mark. 26 for 2H+2Q and then 6-7 more to sustain the third. This falls in line with the IM fast third response against toss FFE builds.
Yes it also suggests limits on the rates of what one can mass up on X bases. Yes for efficiency's sake it should be a mix of Roach/Ling unless you either cut a queen to hit earlier or are droning behind it instead of lings in order to expand during/after.
|
T.O.P.
Hong Kong4685 Posts
I never heard of this theory when I played broodwar. But your ideas are wrong.
You don't need to be even on production with terran/toss. This isn't how zerg works. You don't produce units out of your hatchery 24/7. All you need is equivalent or greater army value by the time he arrives at your natural.
So until terran/toss moves out to attack you, you drone like mad. Then spam units out of your hatcheries when you see him move out. That's the optimal way to play zerg.
|
http://wiki.teamliquid.net/starcraft/Zerg_Strategy
key ideas section.
With regard to the rest. that's fine. How often are you planning to expand with zerg? How much production do you need so as to be on equivalent or greater army value when terran/toss attacks?
You should be producing something out of your hatch 24/7 I would hope. Otherwise you have wasted larva.
I'm not against droning. I'm against droning more than you're capable of producing out of.
See you at school TOP
|
|
|
|
|
|
EPT
