|
I had posted this in the strategy section, but few people thought it had any merit at all, so I'm posting it here just in case anyone finds it interesting to read.
Ok, this is a theory I came up with while in the shower, so it must be good!
I'm sure you're all aware that when you finish a lair or hive, you gain a larva, right? What about using this to BUY larvae?
Zerg economics is unique and complex in that it's all about larvae management, specifically on whether to make a drone or military unit, but broadly: all your units are made by larvae, and all of your buildings then are made from larvae too. The longer a game goes, the more larvae you have. If you let your larvae go to three, you potentially miss a larvae development cycle, and thus lose a larva. Efficient builds are efficient for zerg when they best use larvae, when to make drones, when to make units, when to expo, when to make a hatch in your base, whether it's better to make a static defense structure and lose a drone or a military unit, etc.
My theory is: A hatchery cost 300 minerals and takes 120 seconds to build, and you lose a drone, so you then have 8 fewer minerals times however many seconds it takes for a drone to bring minerals from a block to your hatchery times the amount of times this could happen over the course of the game. A lair costs 150 minerals and 100 gas, takes 100 seconds to build, and you lose no drone. For half the immediate cost in minerals, along with a third the cost of the hatchery in gas you "buy" a larva. The building also has more hp and allows you to get more tech, keeping it alive longer and if you lose your main lair/hive you have a backup one already at lair-stage, but these are just bonuses, not something the theory strives for. Now, you lose gas by this but gain minerals, and gain a larva which never would have been produced otherwise, on top of still having an extra larva you wouldn't have had if you had made a hatchery (the drone). To put it into the perspective of terran and protoss, it would be like spending 100 minerals and 67 gas to make your factory able to produce two tanks at the same time for one time only, or spending 75 minerals and 50 gas to make one gateway able to produce two dragoons at the same time for one time only. For protoss and terran, this would give you one extra unit, ie you would be buying time, because you'd spend money to gain a unit in the same timeframe. Time, like larvae, is finite, so being able to buy it (or a larva) is allowing you to buy something you normally couldn't get otherwise. It's a bit more complex with zerg though. Imagine if you spend 150 minerals and 100 gas to get a lair with your nat hatch while your main already is a lair. You gain a larva. This larva could be 2 zerglings, or one hydra, which could be one lurker, etc. This means that you've spent money to increase the total possible amount of units you could have at a certain point. More units in the same amount of time, that is by definition, efficient. Imagine though, that you spend that extra larva on a drone instead of another unit. This means, that if you lose a drone to making a hatchery, static defense, or some upgrade or technology building, you have another drone to make up for it. If you need to make a hatchery, by making a lair you gain a drone, so making another hatchery hurts less because you don't lose mining speed, you have the same amount of drones as before you made that hatchery. If you don't have to use that drone immediately, you're now gaining 8 minerals every whatever-amount-of-seconds. After however many cycles of this, you've already not only paid for that missing 150 minerals, but you're gaining minerals. You could even say you're gaining gas, because you could put another drone on gas of a new expo sooner, because you have that extra drone already available for mining, so after a certain amount of time, you've completely paid off the money you spent on that lair to gain a larva, and are now getting more money in the same amount of time, definitively efficient.
In starcraft, all the little things add up over the course of a game. Two more lings makes a big difference over time, even if by itself it seems to make no or little of a difference. It's harder to see as the game goes on longer, and more units are involved, but when you look at low-eco games, you can easily see how the difference of one larva could change the outcome of a game, and these same things work just the same in a longer game, even if it's harder to see. Sometimes you follow a game and you have no idea how one guy won, because he was losing the entire game until he suddenly just wins. Just a matter of who made better decisions and/or was more efficient in the end.
How this could be useful: You're able to use all of your larvae, and are mining at such a rate that the amount of money leftover after using all larvae at all hatcheries is increasing. You feel the need to either make another hatchery or expo, because presumably you won't need a surplus of money in the near future, and want to be able to spend that money. Often, players just go ahead and make another hatchery, often in their base. This accomplishes losing a drone and gaining no minerals, but it gives the player more larvae than they would have by that time. Often I see players add hatches when they don't really need to, or won't be able to support that many hatcheries in the near future. Possibly their money goes high as they're teching, but as soon as they hit hive tech or something they'll have plenty to spend it on, and those extra larvae will just sit and be wasted. If instead of, or possibly to complement getting a new hatchery, players make lairs at all of their current hatcheries, they'd gain larvae to spend their money on without gaining more larvae than they can support. It'll cost money to do so, but if they need to spend money, why not be super-efficient in the process, and give yourself more larvae to spend money on, which according to my theory would pay off in the end anyway?
The theory isn't perfect, and perhaps its application is far less effective than I've made it out to be, but I think it's worth consideration atleast.
I think it probably can be useful in a game, but that you'd still need to learn when to use it and when not to, like all things in sc. Might be awkward to try and teach yourself the correct timing and scenarios when to use it, but imo it'd be worth it once you "mastered" it.
Also you can build hives later, it should still pay off given a long enough game or something.
  
|
In all honesty I never knew that a larva appears when the Lair/Hive completes :p
|
Neither did I. 
But it seems to me that your theory is only useful for a player that NEVER lets any of his hatches get to 3 larva. Although this is technically possible, I doubt it happens very often, especially at the level of your average tl.netter/iccuper. Even so, I think spending one resource (mins/gas) to get another (larva) is only useful some of the time, and may be useless/harmful in other situations.
The part below this may have been addressed in your post, and if so, please ignore it and blame it on my poor reading comprehension:
By your theory, you can spend essentially X to get a larva in some amount of time. (and advance tech if it is your first lair/hive)
Also, by building a hatchery, you can spend about 2X + a drone + the resources that it would have mined + the larva used to replace it + the cost of the drone replacing it to get a larva in a longer amount of time.
Obviously option A looks good. However, Option B has an added bonus in that it creates economically free larva for the rest of the time that it stays alive. This is, to me, a very important point that you may have underrepresented. Idk, I am honestly not comprehending this whole thing entirely, so I may be completely off-base, and if so, I apologize.
|
haha yeah I didn't know that either. Still um... I really don't think it's worth getting a lair JUST for another larva. Do you upgrade like 5 lairs, or what?
|
I don't think 150/100 for 1 extra larvae is as efficient as 350+ mining for unlimited larvae.
It could potentially be useful for some sort of timing push, but missing 100 gas would set the push behind more than 1 extra pair of zerglings (what you would get if you traded 100 gas for 1 larvae to save 150 minerals).
I just don't see this being effective.
Again, I think just building a hatchery and getting 4 drones out of it would be more useful than the 150/100 for 1 extra drone. And it gets exponentially more useful as you get your 5th 6th or 7th drone for only 50 minerals vs 150/50 for that 1 larvae.
A new hatchery also gives you the option of having three larvae if your macro slips up and you end up with extra minerals.
As to your point of having more gas you will save 1 drone + have a drone out 20 seconds earlier for 100 gas. I don't know which is better but you may have a point there. Although as I understand zerg play gas geysers are always kept full regardless (except perhaps early game/timing builds where the build a lair or another hatchery wouldn't come into question.) so you don't get more gas, you would get more minerals.
This actually makes me think of cutting probes or scvs for p or z. At a sacrafice of later economy, you can pump out a few more units a little earlier.
However, for z's their timing attacks usually comes at saving up 3 larvae and getting 3 mutas the second the spire finishes. So I dunno, I'm just having a hard time seeing this as useful although it is intriguing to think about.
|
Yeah, I never noticed that either. But your protoss/terran conversion analogy is a bit flawed; namely, you're buying the ability to make an extra unit about a minute from now (btw your build times are off, unless you're playing at normal speed or something, I timed a bunch of stuff myself and posted it in this blog entry.
I honestly don't think it's worth it based on the expense, even without the fact that you have to wait almost a minute for that extra larva. The only exception is when you're getting lair/hive anyway, you may as well use this free larva. I will keep an eye out for that in the future. One or two extra drones per game could be pretty nice.
|
I can't believe I don't know this but hatcheries/lairs/hives all generate larvae at the same rate correct?
|
I hate to sound noobish, but why is it bad to let there be up to 3 larvae?
Are you just not pumping units consistently enough in that case?
|
not reading that huge essay
|
On July 06 2008 14:27 DoctorHelvetica wrote:
I hate to sound noobish, but why is it bad to let there be up to 3 larvae?
Are you just not pumping units consistently enough in that case?
I think since zerg sometimes find themselves short on larvae early game, they maximize the amount of larvae a hatchery produces. If you have 3 larvae, that means you are losing 1 potential larva every larvae reproduction cycle.
But yeah, on topice: Since zerg often find themselves at a shortage of gas, I'm not sure this would a good idea. IMO, 1 Larva for 150//100 isn't nearly as good as 1 larva for 300 due to the gas that the zerg needs for high tech units like defilers, and ultras.
I could see it being useful if you want to do a extra fast lurk drop or something.
|
Thanks for the replies guys, it helps me think 
Arcology, as far as I know they all make larvae at the same rate.
Dromar I got my times right out of staredit, because I didn't know where else to find them, so I think you're right about that being the times in normal speed.
Randombum, I think your comparison to cutting workers with p/t is fundamentally exactly what this is like, except that ideally my theory pays off over economically over time rather than resulting in a lesser economy at shorter amounts of time (presumably by the time the map is mined out cutting workers will have made no difference, but you'll have mined fewer total minerals before that time, presumably).
The more I think about the theory, the more I think it's more of an end behavior model rather than a normal economic function. It seems to me that the theory only applies as someone said above to someone who plays perfectly, who is always using their larvae, and under the assumption that the game lasts long enough for it to pay off. It also considers minerals to be the limited resource, not vespene gas becaus gas is unlimited theoretically, since you can mine depleted geysers. The theory also considers larvae to be a limited resource, despite actually being a renewable resource. The value which is limited about them though is the maximum amount you can have at any given point. This theoretically optimizes that value, because you kill a larva for a hatchery. As you've pointed out however, a hatchery does give you more larvae, but for the price of one larvae, so making the investment of a hatchery does then pay off better. Losing one larva gives you X amount of larvae every t(larva cycle time) seconds. However, if you still make that hatchery, but invest in a lair solely to get another larva, you then have n (amount of hatcheries) times t (amount of time to get one larva) + 1. If you then consider each living larva is worth more as time goes on (because the existance of one more military unit is definitely worth something, I dunno what it is, but it's worth something, whereas the non-existance of a military unit is hurtful (usually), you find that you have gained whatever that value is, larvae potency maybe? Over time then the larva potency makes up for the lost resources. So like, your opponent possibly can't attack at a certain time because of the existance of a certain amount of military units, and the more units you have, the longer he has to wait, etc)(The longer a drone is alive the more minerals or gas it can mine)(A building's existance allows you to either upgrade your units, gain resources, research for your units, and produce more units, allowing you to effectively use your larvae. A great example of this is in say zvp, lategame, while your money flow is fine, you're able to keep your money low while using all of your larvae, then suddenly he dt drops you and kills your ultralisk den, and you're forced to make lings or hydras in place of ultras, but the larvae appear at the same rate, and the money coming in stays the same, but the money going out increases, and then you might make rash decisions as your money goes up until you've replaced the tech; alternatively if earlier in the game your opponent kills your spawning pool and all you can make is drones, assume you need lings in that situation, making drones will only waste larvae you need to be lings, not making anything won't help you at all, so you're in trouble and larvae for whatever reason are going to waste). The existance of one more unit then is worth the same as any other unit, except that situationally one unit is worth more than another, and that's just basic macro, the understanding of when to make which unit. So I'm saying that, the existance of two more zerglings is equal to the existance of one more drone, so you could say then by the theory that 1 ling egg, or 2 lings, upon hatching is = to T (time alive) times t (time it takes to mine a mineral and bring it to your hatchery) times 8 (minerals in each return). Now, this certainly is a stretch, but if you do consider it that way, then each larva is worth as long as it's alive, assuming you're making the best unit for that situation, be it a drone, zergling, overlord, etc.
I think actually then, that my idea is best used while playing normally, not skipping hatches or anything. You could probably use it for some kind of timing push, or just whenever you have extra money. An investment of 150/100 I'm saying will pay for itself over time, and it doesn't attach additional taxes to your economy (a hatch would, because you would then have to use every larva it produces, or you wouldn't be using the larva which made that hatchery to its full effectiveness, ie being inefficient).
tl;dr Use my theory in conjunction with making new hatcheries, don't replace making new hatcheries with it. It's theoretically worth it I think, if the unit stays alive long enough.
|
On July 06 2008 15:05 FragKrag wrote:
But yeah, on topice: Since zerg often find themselves at a shortage of gas, I'm not sure this would a good idea. IMO, 1 Larva for 150//100 isn't nearly as good as 1 larva for 300 due to the gas that the zerg needs for high tech units like defilers, and ultras.
Yeah, seems like another hatch is a way better deal. It's only marginally more expensive (or less, depending on your need for gas at the time), and keeps producing larva.
|
thats just so absurd and stupid it hurts..
|
|
|
Calgary25954 Posts
No, it's never useful. Even for getting an extra Ultralisk, it's still not useful.
|
your math is wrong because it does not consider anything about timing.
|
I think another problem with ths theory is tha you assume a zerg unit is generally equal in value to a terran or protoss unit, when generally it is not, you compair it to making 2 tank out of a factory when that would be VERY VERY useful for a TvP timing rush, but I can't think of any Zv? situation where the cost justifies the ends. Theres no early game unit that would justifty spending that much more to wait 100 seconds to get a SINGLE extra larva, and late game I'm sure ou can find some spare larva lying around and if you can't then your probably a progamer anyway. It might prove useful to get an extra muta, but it would then also slow your lurker/hive tech or might even cost you the resources to build another muta.
Even if your reworked a build so you would have the resources to afford the extra muta into your initial harrass it would probably slow it down enough where it would be faster and more cost effective to just wait 30 seconds for another larva.
|
On July 06 2008 16:27 Nightmarjoo wrote:
The more I think about the theory, the more I think it's more of an end behavior model rather than a normal economic function. It seems to me that the theory only applies as someone said above to someone who plays perfectly, who is always using their larvae, and under the assumption that the game lasts long enough for it to pay off. It also considers minerals to be the limited resource, not vespene gas becaus gas is unlimited theoretically, since you can mine depleted geysers. The theory also considers larvae to be a limited resource, despite actually being a renewable resource.
I stopped reading after that, huge wall of text is very unfriendly but even that part is dumb. This "theory" would only be applicable early/mid game it seems and your basing it around the idea that gas in unlimited. However gas is also linear and at that point in time, definitely not unlimited. The fact that your considering a never ending renewable resource (larva) as a limited one as well goes to show how backwards you're thinking. The price of an extra lair combined with perfect macro would definitely throw off the timing of your next hatchery and in turn that would fuck the shit out of. It seems to be a tiny play that could push you into an all in situation (by not having proper amount of hatches)
Bathroom theories are fun but some of them need to go down the toilet. Pun intended.
|
|
|
|
|
|
EPT
