|
Ok, this is my first thread and I hope this hasn't been suggested yet. I searched and found 1 thread that hinted at this, but he wanted to change the spawn larva mechanic and didn't connect the dots on what could be done about this as is.
Whenever I watch a game being casted, the caster predictably clicks on a queen to view its energy. "With perfect macro a queens energy should never be above 25. He needs to spawn larva as soon as they pop." This is seen as just common sense and nobody seems to argue.
But then I thought about the numbers and it seemed to me like there was something wrong. A hatchery spawns a larva every 15 seconds as long as there are less than 3, and spawn larva generates 4 larva in 40 seconds. The problem is that those numbers dont play well with each other at all.
This means that spawn larva actually reduces a hatcherys natural production by waiting 10 seconds on the larva timer then resetting it. This is as far as the other thread got.
So what if you waited for that third larva to pop before spawn larva reset it? Would that actually increase production or hurt it?
I wrote out a list going through the cycle to see what effect it would have. Assuming you could spend your larva as soon as they pop (best case scenario as this would only really benefit already great macro players) I ran 5 cycles of spawn larva, one the normal way, and one waiting 5 seconds after sl pops to cast it again.
Normally, 5 cycles of sl takes 200 seconds and yields 30 larva. Thats 1 larva every 6.66 seconds. Waiting 5 seconds in between, those 5 cycles take 225 seconds and yeild 35 larva. That is 1 larva every 6.42 seconds.
The moral of my unnecessarily long post? The next time you are criticized for high queen energy, laugh at them as you now have more larva and energy to spare for creep tumors.
Btw if this has been discussed before, feel free to close this thread and laugh at me, as I have wasted alot of time.
|
cool I love stuff like this ^^
|
But keep in mind that you lose larva spawning time the whole time you have three larva anyway, so the difference is negligible at best, non-existent at worst.
|
in theory it makes some sense but it real practical use, like me and all the other day9 viewers during the no queen funday monday, the hatch without a queen is so slow on larvae production, with good macro you're larvae should never go to 3 on a hatch if your less than 3 base, as well as the fact queen injects stack, without a queen on a hatch you need another inbase hatch to support production or else you go insanely high in resources, this will never be imlemented at high level play
|
in the endeffect nobody except the computer has perfect queen timing anyway and I believe queens get slightly more energy then 25 per 40 sec anyway.
|
On November 08 2010 11:43 Blobskillz wrote:
in the endeffect nobody except the computer has perfect queen timing anyway and I believe queens get slightly more energy then 25 per 40 sec anyway.
They actually get slightly less, like 23-24.
|
This would only apply to people that CAN and WANT to spend all their larva immediately. This applies to absolutely no one when a game lasts longer then 15 minutes. I have yet to see any professional Z be able to spend every single larva that came from natural production the second it came up let alone keep their queens energy below 25 on 3 bases +
|
A slight miscalculation: it takes longer than 40 seconds for a queen to generate 25 energy. (44.4444 seconds, according to the Liquipedia page on Energy)
So...it might be self-optimizing?
|
well actually you dont have to spend every larva as it pops, just get below 3 very quickly after sl pops, then keep it below 3 before the next round pops.
Edit:
@Ichabad: ah that is interesting. I wonder if that was planned out? This could still be applicable though on a queen that already has a bit of excess energy. I often go to a hatch a bit too soon and wait a sec for the larva to pop then sl again immediately, which is still worse than waiting the 5 or so seconds. Perhaps I should change the moral to "if your larva just popped, go do something else and come back." But thats not as rewarding as my fist moral.
|
i dont think this works because you cant actually spend every larva u have.
u want to be not missing injects so u can stockpile larva, and use like 10 or more at once.
so if you have 4 larva at a hatch, its not producing any more. and if ur not going to use that larva, its better to inject asap.
|
Unless you are at 200 food, there is no benefit to stockpiling larva, just as it is bad to stockpile mins. And I am the first to tell you that I suck, but on 1 or 2 hatches I dont have a problem keeping my larva < 3 most of the time. I'm not saying that this is gamechanging or anything, but waiting for that 3rd larva is always beneficial if you are spending them.
|
I thought about that alot too. Actually I think there're 2 ways to play Zerg.
One is to play with normal mechanic production like Terran or Protoss (which is maximizing your production capacity using multiple production buildings, aka sc1:bw). With this type of play, you have to constantly check your production like Terran. If you think about it, since 1 larva is spawned every 15 seconds, no units have building time under 17 sec, so no matter what you build(assuming you build 1 unit at a time like Terran), you will always have larva available.
pro: Produce units overtime, have multiple hatcheries to work on, not depend on particular buildings,
con: production capacity is not forgiving (if you forget to build your units like Terran players, you're screwed), hard to defend air in the beginning (if you absolutely don't go Queen, which is stupid anyway), lots of money went into production buildings like Terran
The other is to play like current Zerg players, which is using queens on hatchery,
Pro: burst produce units at once, production is very forgiving (if you forgot to build 5 drone in 50 sec, you still be able to build 5 drones faster than a Terran (50+ 17sec compared to 17x5=85 sec)), save money from production buildings, easy to spread creep if you have multiple queens, easy to defend air since you have queens.
Con: production capacity depends on hatchery, if you have 3 and lose 1, you lose 1/3 of your production, if you lose your queen you get a pretty hard hit in production capacity too, overlord/supply management which is pretty crucial, larva timing.
Since the game is in its early stage, I hope zerg players will expand the first build too.
|
On November 08 2010 12:47 Tyler918273 wrote:
Unless you are at 200 food, there is no benefit to stockpiling larva, just as it is bad to stockpile mins. And I am the first to tell you that I suck, but on 1 or 2 hatches I dont have a problem keeping my larva < 3 most of the time. I'm not saying that this is gamechanging or anything, but waiting for that 3rd larva is always beneficial if you are spending them.
Unless of course you have a new tech structure, such as say a spire, and you want to stock pile some resources and larva to make a bunch when it finishes.
|
Early game (like 5-7mins), it can be better to use starting energy on creep tumors instead of larva inject, because you don't have the money to spend all of your larva anyways. 2 Hatches without larva inject would be able to support one saturated base for a very brief amount of time.
However, inject larva allows a Z player to quickly mass an army, quickly switch unit composition, etc. Also, I find that you are always going to have a need to have lings around, whether it's to stop drops or to support your other units, so in that way, larva is always needed.
|
On November 08 2010 12:58 mrblue182 wrote:Show nested quote +On November 08 2010 12:47 Tyler918273 wrote:
Unless you are at 200 food, there is no benefit to stockpiling larva, just as it is bad to stockpile mins. And I am the first to tell you that I suck, but on 1 or 2 hatches I dont have a problem keeping my larva < 3 most of the time. I'm not saying that this is gamechanging or anything, but waiting for that 3rd larva is always beneficial if you are spending them. Unless of course you have a new tech structure, such as say a spire, and you want to stock pile some resources and larva to make a bunch when it finishes.
Yes and this is why I love zerg. The mechanic is so dynamic that the strategic ceiling is ridiculously high. The most efficient way to produce is actually going to depend on what you are trying to do. So while waiting for mutas, injecting immediately is best but then to make more lings it is better to wait the 5 seconds and spend larva immediately. But if you started saving larva too early you will have less for lings. I just think its cool how this macro mechanic could factor in to high level play in a few years.
|
don't forget transfuse. i think ppl don't realize the queen is like a chronoboost/mule/scan all in a unit that can attack land and air. zergs really gotta use 'em more.
|
As long as you spend the larvae immediately, there should be no delay in the natural spawn. Or at least it would be negligible.
However, often times a zerg player will spawn multiple queens. Particularly, on a pool-first build, they may send a queen made at their main to their natural. In this case, the queen may accumulate more than 25 energy before it can spawn larvae at the expansion hatch. So if, say, your queen had 30 energy when the new hatch formed, you would never be able to have less than 5 energy on the queen for the duration of the game, even if you hit all of your spawn timings perfectly. Some times I think casters and the like make too much of a deal out of a queen having 5-10 more energy than they "should" early on, since this can often just be the result of long walk times or queens being made a bit earlier than necessary.
|
On November 08 2010 13:19 GeorgeForeman wrote:
As long as you spend the larvae immediately, there should be no delay in the natural spawn. Or at least it would be negligible.
However, often times a zerg player will spawn multiple queens. Particularly, on a pool-first build, they may send a queen made at their main to their natural. In this case, the queen may accumulate more than 25 energy before it can spawn larvae at the expansion hatch. So if, say, your queen had 30 energy when the new hatch formed, you would never be able to have less than 5 energy on the queen for the duration of the game, even if you hit all of your spawn timings perfectly. Some times I think casters and the like make too much of a deal out of a queen having 5-10 more energy than they "should" early on, since this can often just be the result of long walk times or queens being made a bit earlier than necessary.
You clearly don't play Zerg. If you hit your spawn perfectly, you will lose 2-3 energy per round. If you're spot on perfect, you'll actually get "Not Enough Energy!" error messages.
|
On November 08 2010 14:13 MangoTango wrote:Show nested quote +On November 08 2010 13:19 GeorgeForeman wrote:
As long as you spend the larvae immediately, there should be no delay in the natural spawn. Or at least it would be negligible.
However, often times a zerg player will spawn multiple queens. Particularly, on a pool-first build, they may send a queen made at their main to their natural. In this case, the queen may accumulate more than 25 energy before it can spawn larvae at the expansion hatch. So if, say, your queen had 30 energy when the new hatch formed, you would never be able to have less than 5 energy on the queen for the duration of the game, even if you hit all of your spawn timings perfectly. Some times I think casters and the like make too much of a deal out of a queen having 5-10 more energy than they "should" early on, since this can often just be the result of long walk times or queens being made a bit earlier than necessary. You clearly don't play Zerg. If you hit your spawn perfectly, you will lose 2-3 energy per round. If you're spot on perfect, you'll actually get "Not Enough Energy!" error messages.
Haha, yeah that happens to me all the time. And then i get annoyed and decide to not make such a big deal out of Spawn Larva and my macro slips a bit.
Stupid mental monologues.
|
I think this can be summarized as: if you use the larva that has just spawned and are about to inject, it's better to do so 5-6 seconds later.
This is actually quite handy for things like 7RR when you want all the units to spawn at the same time.
|
|
|
Until you are maxed, you want to get to a point where you have this dynamic between your money and your larva. It is the basic concept of "Do I have enough larva resource to spend this money on?". There are multiple avenues of going about this. A player like FruitDealer will skimp a little bit on the larva-based spending in his mid game, and instead, pump massive amounts of upgrades with their money. Then suddenly in the late mid game, he starts expanding like crazy through the use of his technological advantage, thus resulting in more larva to spend the money on instead of upgrades. Some people get more queens with excess money to aid in creep spread and early defense. The resource of the larva is the linchpin of the zerg race. Maximizing the efficiency of it's availability is the key to any good plan. If you are not capitalizing on full larva production potential, your plan WILL have holes in it, and your money WILL start to stockpile, thus causing you to make misleading decisions, like upgrading or teching when you can't actually afford it (this is due to stagnant larva, which is a RESOURCE). Queens are a very important part of the equation. Having your money get a little bit high as a zerg is perfectly okay, so long as you either a) have a larva production cycle occurring with the potential to spend what you currently possess or b) the appropriate means to tech with excess resources from your larva cycle.
There is lots of stylistic potential for this unique resource that zerg has. I don't think that this theory of the queen's larva inject taking .2 seconds off of a larva spawn could not be applicable due to the very delicate and precise timing window it would require over a VERY extended period of time on multiple hatcheries with results that would be compromised with an inject that occured one energy too soon, or was delayed too long, requiring you to perform the task flawlessly 5 times in a row to make up for one single mis-timing. It is more efficient to just inject asap.
|
|
|
|
|
|
EPT
