|
(Note: This thread is only for discussing non-expanding builds. FE is always more economical if it can be executed without safety or hatch-blocking concerns.)
A Second Look at Zerg Openings
I'm sure many of you remember my previous threads testing different Zerg opening builds and analyzing the 11overpool 18hatch in particular. Those threads I believe did a lot of good work, but at the time we were all pretty much groping in the dark. It was difficult for people to see the forest for the trees and actually apply the data meaningfully. Eventually I accepted the criticisms of a vocal minority and abandoned my efforts.
However, after recieving quite a bit of positive and curious feedback from people on battlenet, streams, and forums, I have decided to once again turn my attention to an analysis of Zerg openings. With each attempt, I have learned a little more and refined the process and the data to be more meaningful. I have been asked several times given this data why the pros choose the openings they do. I always assumed that the pros remember what worked in broodwar. Back then, delaying spawning pool production would always result in increased economy. However, with the introduction of queens and the spawn larvae mechanic, is this still the best decision? At what timing and ordering of production can we maximize both mineral and larvae production in a way that is viable strategically? This is the question I have been attempting to answer.
In order to prevent any confusion regarding gas timings, I decided to incorporate gas into every build order. I used the fairly standard ZvZ opening, 14gas/14pool in order to have a set time for gas. Therefore, every build gets an extractor at 1:40 and continues to mine gas throughout the test. I recognize that in most situations, getting gas this quickly is not necessary, but I did this primarily to decrease some of the bickering that was previously present regarding safety and viability, and to negate the argument that somehow gas timings would affect one build to a significant degree relative to another. After incorporating gas into each build, I came up with identical ordering of results as before, hopefully negating such arguments.
Data: (Note: DET stands for double extractor trick.)
(Excel Spreadsheet Stats)
(Typical 1Base Larve Production Rate)
What does it all mean?
Note that the build which mined the most total resources, 14pool, also resulted in the fewest final drones. The build with the worst mining rate, 10pool, resulted in the highest total larvae count. There is a clear inverse relationship between minerals mined and larvae produced in the short run. (Please do not confuse this with minerals OWNED... We are NOT talking about spending or keeping resources when we say minerals mined.) This is common sense when you consider that any initial investment will sacrifice in the short run for gains in the long run. It is clear that as Zerg players, we are faced with a constant decision of trade-off between resource production and larvae production.
From watching the pro-Zergs play, it appears they tend to favor more early resources instead of more early larvae, whether knowingly or unknowingly. Are they correct in this assessment? Won't greater larvae production eventually result in the greater economy in the long run? These are very complicated questions with many variables and unknown factors, but if we ever hope to answer them, we have to start somewhere.
For some, this data will be a bit confusing, therefore I have offered a straightforward interpretation of the data and what it potentially means.
1) Which pool-first build gives the most larvae?
(Graph of Differences in Drones Produced with 14Pool as a baseline)
This is difficult to determine using specific moments in time, because the spawn larvae mechanic is constantly fluctuating larvae rates. In order to reach an estimate of which build yields the most total larvae, we must analyze a simple moving average (SMA) of the differences in drones produced for each build. That data can be seen here:
(Simple Moving Average of difference in drones produced with 14Pool as a baseline)
Therefore, we can see that the 11overpool build will yield the highest average larvae counts of any other pool-first build tested. Eventually larvae production will stabilize for every build, and additional larvae production will only be possible with the addition of hatcheries. But unlike resource count, the faster pool builds will have a permanent larvae advantage for the duration of the game without taking into account expansion timings.
2) Which pool-first build gives the most resources?
(Graph of Differences in Minerals Mined with 10Pool as a baseline)
Typically, the longer you delay your production of a spawning pool, the faster you will be able to gather minerals, but only in the short run! Note from the graphs that the mineral advantage of the 14 pool begins to decline at the 4:20 mark. If you look at the other graph, you will see this is the exact moment when the faster pool builds catch up and surpass in drone count. Therefore, once you pass the 4:20 mark, the mineral advantage begins to slowly decline and become overtaken by the faster pool builds due to their faster rate of drone production. Which build will yield greatest long-run economy is complicated by expansion timings and perhaps can be covered again later.
3) Which builds should we not use?
The following builds should be avoided: 10pool, 12pool-DET, and 12overpool-DET.
The 10pool and 12pool-DET should generally NOT be used because they do not produce sufficient additional larvae to justify their reduced resource count compared with the 11overpool. Notice that the 11overpool, 10pool, and 12pool-DET builds all end up with very similar larvae counts. However, the 11overpool build can give more than 100 extra minerals in comparison, and is therefore the preferred option. Likewise, the 12overpool-DET build should generally NOT be used. It gives you up to 35 minerals more than the 11overpool, but results in up to 2 fewer larvae, which is not a good trade-off.
4) Which builds should we use?
The build you should use is of course dependent upon the map, matchup, and your goal or play style. Here are some factors to consider:
In ZvZ, assuming you aren't quickly expanding, I think 11overpool is clearly best. An increased larvae count is a significant advantage in such an aggressive matchup. The extra minerals of the later pool builds generally aren't needed in this matchup if you are playing pool first, since you will typically not be fast-expanding, and you will be gathering more than enough in order to constantly produce lings and banelings. Getting an earlier pool will also give you the flexibility of going all-in if you scout a player going hatch-first against you, and also give you greater safety in defending against 6/7pool cheese. Due to all these factors, I would recommend 11overpool with gas at 1:40 instead of the more common 14gas/14pool opening in ZvZ.
In ZvP I would lean towards the 14pool build, because it is fairly common to expand at some point in the 20's supply, and also to get a roach warren and/or spines, all of which require a higher mineral count advantage as opposed to larvae.
In ZvT I almost always favor opening hatch first, but if you are going for an aggressive baneling bust or roach rush, 11overpool should always be best.
5)
Real-world application and comparison
In order to test my theory of the relative advantages of 11overpool vs 14pool in speedling/baneling builds, I set up an experiment. I played repeated tests according to the following general guidelines:
@1:40 - Geyser @16 drones - pump 100% zerglings/overlords @100 gas - Metabolic Boost @50-100 gas - Baneling Nest Continue pumping speedlings and spending 100% of remaining gas on banelings
In attempting to maximize the use of the 11overpool's larvae count while putting all available resources to their best use, I believe I have developed perhaps the most efficient baneling build possible for Zerg. I am fairly confident that no other build can produce a greater amount of banelings and speedlings in a similar timeframe, but I am open to any challengers.
In adapting this build to a fast baneling bust, it also surpassed the current Liquipedia entry for baneling rush, which is very similar but has some small obvious errors in the ordering and timing.
Best Results @ 6:00:
14Pool 15 drones, 27.04 Speedlings, 7 banelings, 5 overlords
11Overpool - encRoach ZvZ build 15 drones, 33.42 Speedlings, 7 banelings, 5 overlords
Build Orders for encRoach Baneling Openings: + Show Spoiler + ZvZ, Speedling heavy or most aggressive: 10 Extractor Trick 11 Overlord 11 Spawning Pool 11 Geyser (immediately after drone) 15 Queen 17 Overlord @100 gas - Speed, remove 1 worker from gas 18+ Mass speedlings 21 Overlord @75 gas - Baneling nest 26 Overlord -Continue pumping speedlings and spend 100% of gas on banelings.
ZvT, baneling heavy, or fastest bust, 5:20 target: Same as above, EXCEPT: @100 gas, do not remove worker from gas @50 gas, baneling nest Cut overlord at 26
Replays:
This test confirmed that in ZvZ or a ZvT baneling bust, there is a slight advantage to going 11overpool when compared to 14pool due to the larvae advantage. This of course is a subjective assessment depending upon a players style and average expansion timing. As was expected, there were more minerals left unspent by the 14pool build. Therefore, with a differing build, it would be possible to divert minerals into gas for more banelings. I will not attempt to compare the relative worth of a baneling to speedlings here since that is a highly subjective and situational assessment. However, it must be noted that this added flexibility of additional banelings must be contrasted with the 11overpool's flexibility of creating a higher number of either drones or speedlings if your opponent goes for play which negates banelings such as roach openings.
A look at the in-base hatch
The in-base macro hatch is an interesting concept. It is difficult to say how many situations exist where you would prefer a second hatchery in your main as opposed to at a natural expansion, but let's not suppose they do not exist. How efficient would such a build be?
I conducted two tests. The first test includes gas at the 1:40 mark to provide a fair comparison with the other builds. In the second test, I eliminated gas altogether in order to test maximum drone count. What I found was very interesting, but it must be noted they they underscore the production capacity of such builds. 6:00 was too short a time to compare the power of double hatch/double queen to single.
InBase w/gas- 3160 Minerals Mined, 37.82 Drones
InBase wo/gas- 3600 Minerals Mined, 41.71 Drones
11Overpool w/gas- 3154 Minerals Mined, 35.24 Drones
Both of the in-base hatch builds surpassed the pool-first builds in workers, and were either better or comparable in minerals. At first glance it would appear that even an in-base hatch is always better than pool first, but we must consider other factors before jumping to such a conclusion. The in-base hatch builds were only so efficient because they continually produced drones and thus were able to actually take advantage of nearly all the additional production capacity. In a real-game, this would typically not be possible due to the necessity of acquiring an army and upgrades, therefore leaving many larvae incapable of being spent at all. I decided to dig a little deeper to determine when, if ever, an in-base hatch was the best option.
Obviously the sole advantage to an in-base hatch as opposed to an expansion is the greater safety afforded by the building location/terrain. Clearly this safety is an asset, since saturating quickly is both what this build is capable of and requires in order to put its high production capacity to actual use. I envision such a build being most useful on maps such as Backwater Gulch, where your natural + main are practically impossible to defend against a 4gate. So here is the general theme such a build would take:
1) Take advantage of terrain to safely and quickly saturate your main. This can most easily be accomplished by utilizing the two fast queens to block your ramp and the proxy hatch creep near the ramp for spine crawlers. 2) Take advantage of this quickly saturated main and very high production capacity to pump tons of speedlings for map control or aggression. 3) Take advantage of this map control in order to expand and once again quickly saturate you natural. Repeat.
Here is a test of maximum production capacity of zerglings beyond the six minute mark to illustrate the points at which each build has a comparable production advantage. It could be suggested that the in-base hatch is to vulnerable to a more standard build before the 6:00 mark, but let's not forget that we can very quickly get two queens out to block the ramp, and also remember the defender's advantage as far as distance travel is concerned.
I took this general strategy to the master ladder for a few games for fun and testing and had surprisingly good success with it. Of course there are always weaknesses to any build, and much of the execution has to be ironed out, but it is clearly something for us to consider seriously.
Replays:
EDIT: ZvZ replay removed at player request.
I'm not sure how much we can draw from the replays. Obviously the wins don't say too much, considering my opponents played imperfectly, but I find the production capacity at least to be pretty amazing, and the possibilities in ZvZ given the fast double-queen ramp block to be potentially very effective in trying to increase saturation in a matchup that has become so hyper-aggressive. I will continue experimenting with the in-base hatch in the few situations where it could perhaps be warranted.
Summary, tl;dr
Note, these are very simplistic generalizations of the conclusions I reached for people who don't want to read through all the data:
Never 10 Pool, and never double-extractor trick. Start going 11Overpool for speedling/baneling play in ZvZ or cheese ZvT. Keep going 14 Pool in ZvP. Consider an in-base 14 hatch on maps with a difficult to defend natural in ZvZ or ZvP, using queens to block the ramp and spines to defend while you quickly saturate and use high production capacity for mass speedling map control.
|
But the reason you do something like 10 pool is for the earlier first round of Zerglings for "cheese" type openers, and you should never double extractor trick because it would require 100 saved up minerals, which we already knew.
The main reason to go early pool is for the TIME at which the Zerglings pop, and as a result we adjust gas timings ourselves to get speed most optimally while continuing ling or drone production.
All in all, no new information other than that 11overpool is still a good middle of the road opening BO.
edit: I would say that the hatchery timing (in-base) is most applicable to team games for timing pushes, but in 1v1 there are so many variables that can delay the very delicate ZvZ ling timing that it becomes less than desirable. In ZvP on just about any map I would suggest that early ling aggression is stronger than an early hatch for raw production, although as the game drags on the extra hatcheries DO really help.
|
lots of info to go over...very nice read and who would've thought the differences to be so large.
good job OP
|
Well done research as your previous investigation. Congrats. BTW: I am playing in base hatch for some time and it combines economic advantage with safety. Also the larva richness allows to cover drone losses extremely quickly in early .. midgame. You can pump 3 queens at once (after taking nat) in case of VR or phoenix. You are not screwed when loosing nat to early 1 base all-in, because you can keep up ling production. I did not do the numbers, however my winrate is substantly higher with in-base hatch first builds (fewer losses to cheese and early harrass).
15 hatch in base, 15 gas (remove 1..2 drones after getting 100 gas) 14 pool 16 ovie (except when doing 18 hatch) 18..24 hatch at nat (opt to cancel it in case of very early pushes)
|
I think I understood most of this. Im going to try this out a bit in some practice games and see how it works out, thanks for the extremley in depth post!
|
Thanks for this, it was quite an interesting read.
|
I think you should be a bit more clear on what you are measuring and how you are measuring it. I'm assuming these are simulated measurements assuming you just pump pure drones until some time?
If so, I'd be curious how these builds fare against various early pressure builds. It stands to reason that having more drones earlier is beneficial for facing early pressure when you must spend larvae and resources on lings and not drones.
Take a 2 rax all-in as an example, is an 11 overpool going to be able to drone in the way you simulated and be safe? You must recognize that the earlier pool + queen means that the time afterwords must be spend droning to "catch up" with the later pools.
|
Great info JD, good to see it in graphs and tables.
I'd been experimenting with a 14 Hatch in-base against Protoss, but it's hard to quantify its benefits except that it's a whole lot easier to defend than an expansion. The problem is it makes most P's think you are massing Lings to all-in and they go extremely heavy on Zealot/Sentry.
Haven't tried it in ZvZ though, mostly because I just hate Ling/Bane wars.
|
Great info with lots of objective thought. Will start 11 overpooling and see how it goes.
|
much better than your first attempt at this research. obviously well planned and detailed, and clearly a lot more effort went into making this information not only accurate but readable and easy to understand. well done OP.
Take a 2 rax all-in as an example, is an 11 overpool going to be able to drone in the way you simulated and be safe? You must recognize that the earlier pool + queen means that the time afterwords must be spend droning to "catch up" with the later pools.
in the tl;dr section he suggests the 11overpool for ZvT cheese only, such as baneling or roach allin. under normal circumstances he recommends hatch first.
|
You need to keep in mind that just because a build may favor minerals over larva, doesn't make it a bad build, as in the mineral favoring builds you can do things like get a gas faster, or a roach warren faster, since you have extra minerals. All that really matters is that you have just enough money to spend all your larva, whatever your build may be.
|
well done op. lots of great information throughtout the post. i am really curious about the in base hatch. i remember back in the day seeing players like sen experiment with this with little luck but i like the idea of getting quick saturation early and have the capability to masd produce units at critical moments. I will play some practice games doing this and hopefully i have some success because im getting pretty bored of the standards
|
On March 19 2011 10:29 uobradbury wrote: well done op. lots of great information throughtout the post. i am really curious about the in base hatch. i remember back in the day seeing players like sen experiment with this with little luck but i like the idea of getting quick saturation early and have the capability to masd produce units at critical moments. I will play some practice games doing this and hopefully i have some success because im getting pretty bored of the standards
Thanks, I appreciate it. I've also become fairly intrigued by this possibility of in-base hatch. I keep going back to that second replay where the toss goes a standard 3gate-expand and just gets so completely swamped by speedlings it wasn't even close. On maps where a natural is very difficult to defend, I've had a ton of success with this so far. For the first time I feel comfortable having no maps vetoed at all so I can keep testing this strategy.
|
Good job mate, you've clearly put a lot of effort into this. Ill be adding it to my thread soon mate.
|
Nice work, I've been a little harsh before ('the vocal minority') because I found your previous thread lacking in accuracy but this one is well represented.
The conclusions are basically the same as before that 14 pool is economically better but 11 overpool has more larvae (which might catch up the economy later but is unlikely to). It would be nice to link or put the build orders in a spoiler again as I'm a bit unsure how many lings you produce when the pool pops. When I tried 11 overpool with gas (around 17 pop) and 2 sets of lings upon completion of the pool i never had enough minerals to get a 2nd queen directly AND a hatchery. That mitigated the entire advantage of the fast pool and means the build wasn't good in my opinion (for ZvP that is).
If its possible to do 11 overpool with 4 (or 2) lings when the pool completes and no delay on your 2nd queen with reasonably fast gas it could be a great build for ZvP on some maps. The reason would be that earlier lings mean you can chase away their probe earlier and thus get your hatch down earlier and they might be forced to get their zealot earlier then they want (zealot before 2nd gas or possibly even being forced to chrono it).
So overall great post, good to see your stance about the quality of the build overall has changed. For me the only thing missing is the exact build orders for 11 overpool specifically as that isn't mainstream really.
|
Wow, great great great great work. You have so many good contributions to this community. Good job and thank you! The quality of your post is outstanding and impressive.
Thanks for the summary, though I read all of it too ^_^
I'm glad that the in-base hatch has proved to be somewhat viable. It's always been a fun idea to me.
|
I've been doing in-base hatch in all my ZvZs and you can mass speedlings before they start to saturate their expo if they decide to in which case you can simply mass more speedlings, ling peel banes and eventually win. At least that's how most of my games turned out. Even when I lost 6-8 lings due to slow reaction, I had plenty more to run in.
|
big up, for your investigations! Very interessting and nice to read, your thread should be sticky, one of the best threads on tl
Cheers
|
I tested the 11 overpool build a bit more with ZvP in mind because I think the build has some potential there.
Hatch blocking is so extremely common in ZvP, especially on 2 player maps like xel naga, that the earlier queen and lings an earlier pool provides could also allow for a faster expansion because you can unblock with natural with some lings. In ZvP getting gas on time is also a must though as getting ling speed around the time or just after 4 gates hit is crucial. So I tried to find if the 11 overpool build could be adjusted to fit these criteria: - build at least 1 pair of lings immediately as the pool finishes - you can only expand after your lings have reached your natural - start collecting gas on time, so geyser no later then 17 pop. - make 2 queens ASAP after pool, thus first right away and second right after the first. - no larvae waste except perhaps to get the pool earlier.
So I tried to see if 11 overpool could fit this criteria and it just barely can't in my testing. If you get gas and 2 lings and the expo you will just be short on money to make your 2nd queen on time. Then your 2nd queen can't be made immediately after the first which is bad because it means your first queen has to inject twice and you can't start a tumor as early, also your injections won't be synchronized..
However if you go pool a little bit later, 12 pool after extractor trick, you can manage all this and actually have no larvae waste. That means you'll be able to expand faster against a probe block AND have more larvae at the cost of a few minerals (and not scouting). Overall that seems to be worth it on some maps where scouting isn't needed that much.
Build: 11 overlord, after extractor trick 12 pool 16 gas (just before pool finishes) @ 100% pool, queen + 1 pair of lings, will go to 18/18 18 overlord @ 100% lings, expo. 2nd queen after first @ 100 gas, lingspeed and take drones of gas.
12 pool after extractor trick basically seems better then a normal 12 pool (slightly more minerals). Definately seems worth trying for ZvP.
|
Markwerf: What is your timing on your second overlord? Also, what is the timing of your ling speed in relation to the more commonly used 14 gas / 14 pool build?
|
Good post JD.
A logical conclusion that for pool first builds, there are delayed advantages to having the pool earlier (more larvae). In other words, later pool is not necessarily better than earlier pool.
|
To the OP: So, your graphs show that eventually an 11 Overpool build will get you more drone production than a 14 pool. My question is this: what if you can't make drones because of pressure from your opponent? You are then forced to make lings/blings, which will beat back the pressure, but you will have a pretty terrible economy relative to what a 14 pool would have.
In other words, if your opponent pressures you with any sort of early rush, you won't have the chance to benefit from that drone production advantage that your graph shows.
I played a ZvZ yesterday where I tried the 11 overpool build, and my opponent did a later pool (probably 14). I lost the game almost instantly when I tried to get the drone advantage because my opponent just pumped lings from his hatch and ran over me...to survive, I would've had to pump lings just like he did, only my economy would be worse because I wouldn't get the chance to reach that drone production boost. A 14 pool gives you the 14 drones RIGHT AWAY, and if you choose to pump lings after that point, you will still have those 14 drones mining the whole time, which will give you a better economy versus someone that gets earlier pool/lings/queen because they will have less drones mining while they're defending your attack (unless they somehow use way less units to beat back your attack so they can make drones with the remaining larva; however this is risky at best, especially in ZvZ where your opponent can scout exactly what you're doing with his early scout drone).
Any clarification is appreciated - thanks.
Also thanks for putting so much work into analyzing the 11 Overpool strategy - I will definitely use it if you can clarify how it can be effective versus an opponent who knows its limitations.
|
So if you 11 overpool against 14g14p, when do you start making zerglings? still around 15 food or do you keep your larvae since 11 food ?
|
On March 20 2011 02:26 DaeWang790 wrote: Markwerf: What is your timing on your second overlord? Also, what is the timing of your ling speed in relation to the more commonly used 14 gas / 14 pool build?
Second overlord is made at 11 instead of 9 so would be about 20 secs later. Gas collection starts about a minute later then a normal 14 gas 14 pool build so speedling tech is also about a minute later.
I've never really understood why the 14 gas 14 pool build became the norm lately in ZvP, I don't see what the point is of rushing lingspeed so much. Sure you need it reasonably quickly so you don't lose to stalker kiting or 4 gates but starting it ASAP when the pool finishes seems excessive. All i ever see zergs use it that early for is a bit of easier map control.. From what I see more zergs are already getting gas a bit later though, going for 14 pool 16 gas for example just to get those extra few minerals (40 or so i estimate).
|
To skypig: The differences in larvae can be seen from the graphs to equalize at the 4:20 mark. I'm not sure how quickly a rush can come, but I would be surprised if it was before 4:20. I'm not sure on the timing of a 6pool so I'm not sure if it's an issue in that situation, but in every other, both builds will be able to equalize in drone count before pressure can arrive.
You can note from the testing I did 11overpool vs 14pool that both builds had the same number of drones once I started producing zerglings, so pressure interfering with economy shouldn't be an issue in that regard.
For people asking about specific build orders for 11overpool: I did provide a general build order, but it was in a spoiler so maybe people missed it. I found it was usually best to build around 15 or 16 drones in order to have enough minerals to spend constant larvae injects on lings and banelings.
If anyone else has any questions, feel free to ask. And thanks everyone for the positive feedback!
|
On March 20 2011 04:01 jdseemoreglass wrote: For people asking about specific build orders for 11overpool: I did provide a general build order, but it was in a spoiler so maybe people missed it. I found it was usually best to build around 15 or 16 drones in order to have enough minerals to spend constant larvae injects on lings and banelings.
If anyone else has any questions, feel free to ask. And thanks everyone for the positive feedback!
I saw the build orders for your TvT and TvZ aggression openings but I didn't see a 11 overpool build including expoing, which I assume you used for testing the income and drone numbers.
|
On March 20 2011 04:19 Markwerf wrote:Show nested quote +On March 20 2011 04:01 jdseemoreglass wrote: For people asking about specific build orders for 11overpool: I did provide a general build order, but it was in a spoiler so maybe people missed it. I found it was usually best to build around 15 or 16 drones in order to have enough minerals to spend constant larvae injects on lings and banelings.
If anyone else has any questions, feel free to ask. And thanks everyone for the positive feedback! I saw the build orders for your TvT and TvZ aggression openings but I didn't see a 11 overpool build including expoing, which I assume you used for testing the income and drone numbers.
None of my testing for this thread included an expansion. I wanted to start off as simple as possible. If you are planning to expand quickly, I would suggest going with 14pool instead of 11overpool because the extra minerals will allow you to get the hatch faster and defend it easier imo. As I explained in the OP, builds that require an expansion are much more resource intensive than larvae, and so pool builds with more early resources should be favored in such cases.
|
On March 20 2011 03:57 Markwerf wrote:Show nested quote +On March 20 2011 02:26 DaeWang790 wrote: Markwerf: What is your timing on your second overlord? Also, what is the timing of your ling speed in relation to the more commonly used 14 gas / 14 pool build? Second overlord is made at 11 instead of 9 so would be about 20 secs later. Gas collection starts about a minute later then a normal 14 gas 14 pool build so speedling tech is also about a minute later. I've never really understood why the 14 gas 14 pool build became the norm lately in ZvP, I don't see what the point is of rushing lingspeed so much. Sure you need it reasonably quickly so you don't lose to stalker kiting or 4 gates but starting it ASAP when the pool finishes seems excessive. All i ever see zergs use it that early for is a bit of easier map control.. From what I see more zergs are already getting gas a bit later though, going for 14 pool 16 gas for example just to get those extra few minerals (40 or so i estimate).
I guess I really meant to ask when you get your 3rd overlord. You didn't specify it in the build order you had presented.
I personally go with the 14gas/pool simply because I use it in most matchups. Also, it allows for cheese if the situation calls for it.
|
So this thread finally caps out at two pages and officially dies. I go to check out the threads that are burying this one. First I glance at a thread proposing basically a carrier rush. I decide to skip ahead and open up the next one. In the OP, it is literally suggested that you build a hatchery in your opponent's base, cancel it, build an evo chamber on the left-over creep, and wait for it to die, so that you can scout with broodlings. I was trying to determine whether there was some elaborate troll going on before I proceeded to read through 3 pages of debate regarding whether an overlord scout costs 100 or 200 minerals since you need to build another one to replace it. Finally, I open one more thread. It is literally someone typing ebonics in order to use the word "pimp" as an acronym for macro.
I want to officially apologize to TL for any times that I defended the SC2 strategy forums. I used to be insulted by the elitism of people bashing a whole section on this site. I'm sorry guys, I literally had no idea. I promise I will never commit the sin again of defending the indefensible.
|
On March 20 2011 15:25 jdseemoreglass wrote: So this thread finally caps out at two pages and officially dies. I go to check out the threads that are burying this one. First I glance at a thread proposing basically a carrier rush. I decide to skip ahead and open up the next one. In the OP, it is literally suggested that you build a hatchery in your opponent's base, cancel it, build an evo chamber on the left-over creep, and wait for it to die, so that you can scout with broodlings. I was trying to determine whether there was some elaborate troll going on before I proceeded to read through 3 pages of debate regarding whether an overlord scout costs 100 or 200 minerals since you need to build another one to replace it. Finally, I open one more thread. It is literally someone typing ebonics in order to use the word "pimp" as an acronym for macro.
I think what's worst is the thread that is discussing how 10pool is a viable (and good) build for ALL match-ups, based upon personal opinion/experience, whereas you are here doing a solid experiment showing that 10pool is actually the worst opener to do (unless you're cheesing, but he's recommending doing an economic 10pool against T and P).
I'd just like to say thanks for posting this thread, I find it very helpful (even though I'm a 15 hatch kinda guy myself), and it's definitely interesting to see the comparison between the different openers. I always thought that the earlier pool --> earlier queen could have some advantages over the seemingly more greedy later pools, but it's great to see the numbers presented this way and so clearly to confirm this.
I also liked the analysis of the in-base hatch, although could you add numbers comparing the in-base hatch w/ no gas to a 11 overpool w/ no gas?
Also, do you have any future plans for doing a similar comparison with standard hatch first openings (15 hatch 15 pool is my favorite so I'd be interested in comparing that with 15h/14p, 14h/13p, and 14h/14p)?
|
I just want to say thank you for doing this, as I read your first research as well and I'm equally impressed. Needless to say, I like your findings here.
I suppose the takeaway from this is:
-ZvP: Opening 11 overpool 18 hatchery is probably the way to go. Thought I'd have to do some testing comparing it to say a 14 pool 15 hatchery (I know you said no comparing it to FE for a variety of reasons, but I feel those reasons are nullified if you go pool first)
-ZvT: Like you said, you should probably just learn how to safeguard a 15 hatch.
-ZvZ: Do the 11 overpool build if you're looking for the early victory.
Is there anything you currently are experimenting with now? A new build order perhaps? Maybe it's built around banes or something? I'd love to hear what's going on with all of that; since obviously the 11 overpool is such a nifty start.
|
On March 20 2011 15:25 jdseemoreglass wrote: So this thread finally caps out at two pages and officially dies. I go to check out the threads that are burying this one. First I glance at a thread proposing basically a carrier rush. I decide to skip ahead and open up the next one. In the OP, it is literally suggested that you build a hatchery in your opponent's base, cancel it, build an evo chamber on the left-over creep, and wait for it to die, so that you can scout with broodlings. I was trying to determine whether there was some elaborate troll going on before I proceeded to read through 3 pages of debate regarding whether an overlord scout costs 100 or 200 minerals since you need to build another one to replace it. Finally, I open one more thread. It is literally someone typing ebonics in order to use the word "pimp" as an acronym for macro.
I want to officially apologize to TL for any times that I defended the SC2 strategy forums. I used to be insulted by the elitism of people bashing a whole section on this site. I'm sorry guys, I literally had no idea. I promise I will never commit the sin again of defending the indefensible.
It is an unfortunate characteristic of forums that the amount of replies in a thread are often dependant on how many people disagree with you and not how many agree with you. Your previous thread had a shitload of comments because many people disagreed, flamed, bashed and what not. This thread however is quite accurate in it's data but not really new either because it's a recap of what has been said before. That basically means noone argues and you get almost strictly positive replies but very few of them at the same time. People are more vocal about what they hate then about what they like unfortunately...
Number of replies shouldn't be the measuring stick for the quality of a thread but unfortunately replies are the only thing that keeps a thread floating at the top.
The strategy forum can only be improved by implementing a rating system imo where good threads instead of threads with lots of replies stay at the top. Or mods would have to get alot more active in removing crappy threads....
|
On March 20 2011 15:25 jdseemoreglass wrote: So this thread finally caps out at two pages and officially dies. I go to check out the threads that are burying this one. First I glance at a thread proposing basically a carrier rush. I decide to skip ahead and open up the next one. In the OP, it is literally suggested that you build a hatchery in your opponent's base, cancel it, build an evo chamber on the left-over creep, and wait for it to die, so that you can scout with broodlings. I was trying to determine whether there was some elaborate troll going on before I proceeded to read through 3 pages of debate regarding whether an overlord scout costs 100 or 200 minerals since you need to build another one to replace it. Finally, I open one more thread. It is literally someone typing ebonics in order to use the word "pimp" as an acronym for macro.
I want to officially apologize to TL for any times that I defended the SC2 strategy forums. I used to be insulted by the elitism of people bashing a whole section on this site. I'm sorry guys, I literally had no idea. I promise I will never commit the sin again of defending the indefensible.
=) well replacing overlords is expensive, 100 for the initial and then 100 for the replacement overlord and then 100 for renewing the sac'ed one .. that's 300 minerals oO.
I'd be very interested in analysis of in base hatch play .. my experience tells me that this is a very safe and economic build (with major benefits beyond the 6'00 minute mark), however i am a pretty bad player, so it would be interesting to do the numbers (e.g. "1 queen early 3rd hatch at nat" vs "2 queen late 3rd at exe" vs "delayed queens + instant 3rd at nat"). Also it seems to me it is possible to delay the pool even further because your main is easier to defend ..
Can we agree on a simple standardized testing scheme (incl. 1 scout and speed timing) ?
|
very useful information. i'll have to take this into consideration with my 10 pool build i posted for defensive fast queen production.
maybe an 11 pool would be slightly better over having more creep and queens on the map.
Hmmm. Ponders.
|
Nice read, always nice to get some more information regarding this. Thanks
|
On March 20 2011 15:25 jdseemoreglass wrote: So this thread finally caps out at two pages and officially dies. I go to check out the threads that are burying this one. First I glance at a thread proposing basically a carrier rush. I decide to skip ahead and open up the next one. In the OP, it is literally suggested that you build a hatchery in your opponent's base, cancel it, build an evo chamber on the left-over creep, and wait for it to die, so that you can scout with broodlings. I was trying to determine whether there was some elaborate troll going on before I proceeded to read through 3 pages of debate regarding whether an overlord scout costs 100 or 200 minerals since you need to build another one to replace it. Finally, I open one more thread. It is literally someone typing ebonics in order to use the word "pimp" as an acronym for macro.
I want to officially apologize to TL for any times that I defended the SC2 strategy forums. I used to be insulted by the elitism of people bashing a whole section on this site. I'm sorry guys, I literally had no idea. I promise I will never commit the sin again of defending the indefensible.
There's...nothing to discuss. Don't take everything so personnally. Your last thread was a good idea but I was personnally put off by how you reacted to any criticism. Now that no one is arguing with you you should be happy.
I don't understand what you want, 10 pages of people congratulating you?
|
good read, thanks for posting.
I don't really have any questions, but I am slightly surprised at the findings regarding in-base 14hatch. It's definitely something to consider.
Actually, at least insofar as I can tell, wouldn't the tradeoff between 12 pool and 14pool be worth taking in almost every situation?
|
How exactly do you get such precise data? Are you using a program to analyze replays? Or do you just make sure to be precise with the ingame timer?
|
So this thread finally caps out at two pages and officially dies. I go to check out the threads that are burying this one. First I glance at a thread proposing basically a carrier rush. I decide to skip ahead and open up the next one. In the OP, it is literally suggested that you build a hatchery in your opponent's base, cancel it, build an evo chamber on the left-over creep, and wait for it to die, so that you can scout with broodlings. I was trying to determine whether there was some elaborate troll going on before I proceeded to read through 3 pages of debate regarding whether an overlord scout costs 100 or 200 minerals since you need to build another one to replace it. Finally, I open one more thread. It is literally someone typing ebonics in order to use the word "pimp" as an acronym for macro.
Sadly that works. Well against terran too. Block my wall in with a proxy hatch, cancel and put an evo. Rush with a 9 pool or earlier pool build that was delayed JUST a bit. Cheese and win.
|
iAmJeffrey: The thread he is referring to isn't talking about some sort of wall-delay so you can ling all-in. He thinks he is playing a legitimate macro based game and that it's the more cost effective way of scouting something like a 4 gate. He also believes sacrificing an overlord costs 200 minerals. Not exactly a TeamLiquid material thread.
|
On March 20 2011 22:34 MilesTeg wrote:Show nested quote +On March 20 2011 15:25 jdseemoreglass wrote: So this thread finally caps out at two pages and officially dies. I go to check out the threads that are burying this one. First I glance at a thread proposing basically a carrier rush. I decide to skip ahead and open up the next one. In the OP, it is literally suggested that you build a hatchery in your opponent's base, cancel it, build an evo chamber on the left-over creep, and wait for it to die, so that you can scout with broodlings. I was trying to determine whether there was some elaborate troll going on before I proceeded to read through 3 pages of debate regarding whether an overlord scout costs 100 or 200 minerals since you need to build another one to replace it. Finally, I open one more thread. It is literally someone typing ebonics in order to use the word "pimp" as an acronym for macro.
I want to officially apologize to TL for any times that I defended the SC2 strategy forums. I used to be insulted by the elitism of people bashing a whole section on this site. I'm sorry guys, I literally had no idea. I promise I will never commit the sin again of defending the indefensible.
There's...nothing to discuss. Don't take everything so personnally. Your last thread was a good idea but I was personnally put off by how you reacted to any criticism. Now that no one is arguing with you you should be happy. I don't understand what you want, 10 pages of people congratulating you?
No, in the past he was defending the strategy forum as a viable place to come and discuss builds and strats in a mature manner. He's simply pointing out that he's taken a look around, and he's apologizing for his defense.
|
On March 20 2011 15:25 jdseemoreglass wrote: So this thread finally caps out at two pages and officially dies. I go to check out the threads that are burying this one. First I glance at a thread proposing basically a carrier rush. I decide to skip ahead and open up the next one. In the OP, it is literally suggested that you build a hatchery in your opponent's base, cancel it, build an evo chamber on the left-over creep, and wait for it to die, so that you can scout with broodlings. I was trying to determine whether there was some elaborate troll going on before I proceeded to read through 3 pages of debate regarding whether an overlord scout costs 100 or 200 minerals since you need to build another one to replace it. Finally, I open one more thread. It is literally someone typing ebonics in order to use the word "pimp" as an acronym for macro.
I want to officially apologize to TL for any times that I defended the SC2 strategy forums. I used to be insulted by the elitism of people bashing a whole section on this site. I'm sorry guys, I literally had no idea. I promise I will never commit the sin again of defending the indefensible.
I'd like you to answer my question a bit after this post, but also bumping for the reaasons you mention.
I really hope the mods just start teamp- or perma-banning these idiots instead of warning them.
It's one thing to use the [L] tag, and make a shitty post but at least try to self-segregate yourself from the higher level discussions. It's quite another to be like "hey, I'm a gold player, but I'm high level, and I know this will work in the GSL against MC because he has never seen an offensive evolution chamber before."
|
Would you be willing to have a look at earlier pools as well?
I'd be interested to know the economic gain/loss from 7pool vs 6 pool, etc etc as compared to traditional openings. Perhaps there is a nugget in there?
I just watched a Junwi v Huk game yesterday where he 7pooled huk and ended up with a 90 food advantage. certainly it wasn't only due to the 7 pool (which did no physical damage due to forge), but the idea that it's possible to build up a 90 food advantage (200 to 110) by 15-18 minutes off a 7 pool opening seems pretty interesting.
|
What is this? a PhD dissertation?
User was temp banned for this post.
|
How many times did you repeat each build order to obtain those stats?
|
Amazing analyses, thanks so much for the hard work encRoach, I've followed several of your threads and they are very helpful, thanks for taking your time to share this.
|
Informative and an really interesting read. Good job!
|
JD,
I read your first thread on the 11 overpool and enjoyed it a lot. This one is a great followup. It's a huge challenge to get calculations like these up to the point where they reflect everything a player needs a build to do. It seems like you're steadily working in that direction.
If you plan on continuing this work I have a request: I think it would be great to look at how gas timings affect the production of the builds. This would probably be easiest for the 11 overpool build, since gas always comes after the pool. At what point does delaying gas not affect your production significantly anymore? Can you actually get it earlier than 1:40 with minimal penalty? Or if not, can delaying to, say, 1:55 push your production even higher?
Doing this for the 14 pool build is harder because moving the gas to after the pool changes everything and brings in a big additional variable. But it looks like you're putting a lot of time into these, so comparing the various possibilities for gas timings on the "good" builds (i.e. not 10 pool, DET, etc) to see how much production you sacrifice for getting gas at a given time would be a great next step. Spanishiwa's no gas build is quite viable, at least at low masters, so pushing gas back to recover production isn't necessarily a bad idea.
Anyway, I look forward to the next post, whatever it may be.
|
Very very interesting post and a good read!
You discussed pool first builds indeep in this topic, but do you have any plans to compare different hatch first (timings) with each other?
|
This is a very deep analysis indeed. Thank you for this!
|
11 pool seems perfectly valid in cheese builds. Good players will probably scout it and assume some sort of ling based cheese though. Especially if you do it multiple times to the same guy.
Not a fan of the inbase hatch, judging from the replays you float gas in one of them and cant even afford to spend all your larvae. Maybe just make one queen rather than two? and thats an awfull lot of spinecrawlers.
It has some merit in the fact that you can produce alot of zerglings and i will have to applaud you for defending that half assed 4gate. In the one replay you pulled guys off gass it seemed much more potent, though i will have to question the amount of drones produced. If you intended to win the game with those lings, wouldnt halting production at a spesific drone count be better than preparing for an eventual expansion?
|
I'm really intrigued by the in base hatch at 14 supply(just watched all of your replays), I can think of so many games where we end up needing a macro hatch anyways and to top it off you can setup such a nice wall in at the top of your ramp( would be a god send against these Terran clowns who open with 6+ hellions). Also I would be willing to bet I can saturate my natural faster using the inbase hatch strat than I would with my normal 14gas 14 pool ( expanding around 22 supply ). Has anyone else been working with this any? Something about it seems so solid to me in the early game. I mean you get creep to your natural right off, nice ramp block that can be removed when desired ( spinecrawler uproot ), can pump 3 queens at once if you suspect air, can quickly snap enough drones to saturate your natural once you gain map control, and lastly you can punish people so hard with mass lings early.
At any rate I think I'm going to play around with this some, very interesting stuff thanks for the writeup.
|
This is the type of analysis that really helps me feel confident with different BOs. Don't worry about sub-par threads, just keep doing solid work and it'll absolutely benefit the community.
On April 09 2011 19:57 ray4ever wrote: You discussed pool first builds indeep in this topic, but do you have any plans to compare different hatch first (timings) with each other?
It'd be great to see different hatch first BO's as well.
|
Thanks alot man I really appreciate this. Ive been trying your inbase hatch build alot with good success recently. Thanks
|
Wanted to add I tried the in base hatch in a few matches today ( I'm mid masters ) and it worked pretty well actually. Comparing the replays to my normal build it seems like I'm able to hit a nice timing where I can fully saturate my main with almost no risk so when I go to take my natural I'm just pumping units. I find it very hard to change openings though after playing 14gas 14pool ( 22 expand ) for so long so I end up playing kind of sloppy in the early stages of the mid game. At any rate I went 3-3 using in base hatchery, won a zvp zvt and zvz doing this .. could have some potential and it feels really solid against early game aggression.
|
Thanks for all the research you put into this thread!
|
OMFG wall of information brain on meltdown trying to read it... ahh! But no seriously good work on the data collection and graphs... well done!
|
Thehitman Build: 12 pool, 17 queen, 16 drones into overlord, 18 nat hatchery, 18 extractor. build 12 zerglings, lair, infestor pit. Mass zerglings and infestors, start ground armor as soon as you have half the infestor pit completed.
This is so far only viable against protoss 3 gate into base, 1 gate into base or 4 gate rush. For 6 gate rush you need roaches and spine crawlers.
|
I love threads like these -- especially your last one and this one. Please keep them coming. I think it's well thought out threads like these that constantly question all the "perfect" timings that makes us all better as players and observers. Thanks!
|
This may seem like a basic question... but for those of us Noobs out there...
11 Overpool means.... ?
11 Overlord Drone back to 11 11 Spawning Pool?
Am I right? Thanks for this info man. Really great stuff. =)
|
its just 11 overlord 11 pool you dont have to drone back up from making an overlord but yeah
|
On April 10 2011 10:12 Marzocchi wrote: This may seem like a basic question... but for those of us Noobs out there...
11 Overpool means.... ?
11 Overlord Drone back to 11 11 Spawning Pool?
Am I right? Thanks for this info man. Really great stuff. =)
11 Overpool means using the extractor trick to get to 11 supply, making an overlord, and then making a spawning pool. It works out fairly well, because by the time your overlord finishes, you will have 3 larvae that can quickly produce 3 drones to catch up to 9 overlord builds fairly well. It is much better than pool followed by overlord, because there is a significant time when you are larvae unspendable due to supply cap.
I really want to thank everyone for all the positive feedback on this. I was really disheartened that I put all this work into it and then it practically died, but luckily someone bumped it for more people to see and get some much needed analysis.
I'm not sure yet about doing an analysis of hatch first builds. There are so many complicating factors that make comparisons difficult, such as maynard timing, spine timing, queen timing, etc. Also in hatch first builds there is a deficit of minerals, making gas timings very complicated to hash out efficiently. There is a lot of disagreement as far as the best queen timing, speed timing, etc. For example, spanishiwa seems to be in favor of very late gas in his builds. These things cannot be scientifically analyzed imo, they have to simply be tested in game first to make determinations.
|
Thanks for the awesome extensive information about the different Zerg openings. It kind of underlines what i was suspecting: 11 Overpool and 14 Hatch are the most solid openings from a resource point of view.
|
Hey jdseemoreglass, don't feel bad. I use the 11 overpool in most of my matchups and I believe it is superior all around against protoss at least, and at least beneficial against many types of terran openings. If nothing else, you've changed my style.
Frankly, I watch tourneys and watch people like idra get bunkered and quit, and destiny get cyber cores in his nat into 4 gate, morrow getting pylon blocked from their nat by an FE toss on taldarim, and fruit dealer losing 4+ drones early to marines and it seems so pointless.
According to my tests with a drone scout, a 14 pool has a measly 50 mineral bonus after 7 minutes. All because they refuse to have a slightly earlier pool and end all the useless aggression and hatchery delay tactics. Compared with potential for zerglings 25 seconds earlier, 25 second earlier queen and inject or tumor, no hatchery block possible, period, and certain early expands by toss become unviable on certain map positions, it should be a no brainer. Every pro should be using it.
So thank you, man. Cheers.
|
First of all. Great great post. Really informative and should probably be spotlighted for the effort put in.
I used to 10 overpool in the beta (in all matchups), and I want to explain my thought process behind abandoning the build.
http://wiki.teamliquid.net/starcraft2/Lalush_build (long time ago... =) )
The reason I didn't do the gas trick to go up to 11 drones, was because i wanted to make a queen and 6 lings instantly upon completion of the spawning pool at 13/18 supply. Then I'd drone from 16/18 to 18/18, after which I took my gas to go down to 17/18 while making another drone to minimize the downtime. After that larva inject and an overlord, and then like 7-8 drones.
The build lined up real nice. Back then people had bad multitasking because the game was so new, so the lings would almost always pressure them into making mistakes.
Additionally. Roaches were totally awesome with 2 armor and 1 supply. ZvZ was all about roaches. Lings were a total waste. And the 10 overpool build let me put on pressure while building 7-8 drones from the early larva inject, and then use the next round of larva for roaches. I always came out ahead because no one would open speedlings back then.
What made me abandon 10 overpool in ZvZ was the 14gas 13/14 pool build after roaches were nerfed. You have a huge problem in that you cannot use your first batch of larva on purely drones. You have to use them on slow lings to stay alive. And those slow lings are USELESS. They only serve as a placeholder to keep you alive until you get roaches (which your economy is way too bad for if you have to make lings instead of drones with the first batch from larva inject).
ZvZ is all about who can get away with making the most drones without dying. A non speedling build simply cannot put pressure on a speedling build, and it cannot keep tabs on if the opponent is making drones or continuing to make units for an all-in. Thus, it became an indesirable build order. Whereas I used to be able to keep tabs on my opponents (when they didn't open speedlings), now they are the ones keeping tabs on me. From me keeping them in the dark, they are now keeping me guessing and in the dark.
That's why 10/11 overpool sucks in ZvZ (unless your opponent goes for a non instant speedling build).
ZvT was the first matchup I gave up 10 overpool in. Making 6 lings and pressuring had 0 effect vs a wall-in. Furthermore, you could not expand with slow ling vs reapers. You would have to put down a roach warren before you'd expand. Back then, at least in the beginning of the beta, hydras were awesomely good (+10hp and higher fire rate). So I'd often get away with roach warren into expo, or just 1basing into hydras and expo. But the nerf of hydras and the emergence of the reactor hellion build made me give up on hydras as well as the 10 overpool build in ZvT. You simply cannot make drones at the timings you need to make drones in order to make the 10 overpool economical.
ZvP was the matchup where I held on to the 10 overpool build the longest. I still believe it might be a viable and functional build (although I wouldn't transition into hydras today). What made me TOTALLY give up on 10 overpool in ZvP was the invention of the 4gate rush. I simply could not hold a 4gate with an 10 overpool without being absolutely economically crippled. Sometimes I'd hold it for 5-10 minutes but eventually die because his crappy 1base economy pretty much matched mine. Once the sentry numbers start to go over 10 (even if he is on 1base and has no intention whatsoever in expanding), shit becomes reaaaal hard. You simply cannot find any space/time for building up drone numbers while defending a 4gate with an early pool opening.
Early in the beta hydras were much better as stated before. Also, the common protoss 1base "all-in" build was 3gate robo immortal. Which 10 overpool into hydras absolutely destroyed, since that build struck so late. Fast expansion builds were also pleasant to play against, as this build allowed you to drone up heavily while teching fast and doing extremely fast 2 base hydra "all-ins" (which always worked). At least until people started making shitloads of sentries while rushing to colossi.
The protoss 4gate build is the absolute most extreme and earliest rush though. And it cannot be effectively held with a 10 or an 11 overpool. It alone is pretty much the reason not to use 10/11 overpool. You can pull it out in certain situations in boX series. It's excellent in providing easy wins vs forge FE protosses (either through the early lings or through quick roach follow ups).
And it might be effective against a sentry expand style as well. But the thing is that an experienced Protoss player will know to 4gate vs it. Problem for Protoss players is that this build is so out of style that they likely would not know how to properly react. And thus it could be used a couple of times with great effect.
Anyway. I just thought I'd provide an outlook on 11 overpools practicality in high level play, since I consider myself very experienced in using the opening. It was the only opening I used for months into the beta.
|
It doesn't matter how small a mineral bonus a 14pool theoretically has after 7 minutes. If your opponent does an aggressive opening, you'll see why top players opt for the mineral bonus as opposed to the larva boost. You are stuck with a tech disadvantage and way less strategical flexibility using a 10/11 overpool.
Additional larva are of no use unless they can consistently (vs a wide range of strategies) be used to make drones. Sadly, SC2 is just a bit too fast paced of a game, the rushes are just a bit too extreme for the 11 overpool to be viable (at least in my personal experience).
The reason top players opt for the mineral bonus is that they maintain that bonus throughout defending a cheese or an all-in strategy. The timings of the 11 overpool simply don't line up nicely enough to be able to make drones at the times you need to while defending your (delayed) expansion with a tech disadvantage compared to the 14pool.
11overpools main disadvantage compared to 14/15hatch is the extreme delay in expansion timing. And the fact that the expansion, the bonus larva from inject, and the situational need for a roach warren or a speedling upgrade all compete for your resources at the same crucial timing in the game.
11overpools main disadvantage compared to 14pool, is its inability to match 14 pool's tech while maintaining a decent economy (the mineral bonus issue). Also: what was written two paragraphs above, the timings don't line up as nicely in juxtaposition with the opponent's possible build orders.
14/15 hatch and 14pool don't put you in the position to make impossible choices. That's what ultimately makes them the best.
|
do you have a thread on hatch first somewhere? what's the scenario when you would do 14 gas/14 pool? i think it greatly depends on scenario, but it is intuitive that a later pool means more minerals. is there an analysis on 15 to 20 pool? just for completeness? keep up the good work! P.S. I wish there are some build testers for protoss that can crunch out viable amounts of sentries to protect an FE. Also, i've always had an itching question to know when it is worth it to cut probes for an expansion. on another subject...it kinda bugs me that there is no auto-probe production, auto-inject, and the like... also it would be kinda nice to have some kind of build tester that speed plays a given build (to account for unit movement and drone transfer).
|
@Lalush
I like most of your analysis, however, note that in my testing the 11overpool IS a speedling build. It gets gas at the same time as the 14gas/14pool, and has a larvae advantage in comparison.
I agree that 11overpool in ZvT is not economical and ZvP is not safe. However in ZvZ I think it still has real potential, both in it's larvae advantage and in it's flexibility.
|
On April 10 2011 12:42 LaLuSh wrote: It doesn't matter how small a mineral bonus a 14pool theoretically has after 7 minutes. If your opponent does an aggressive opening, you'll see why top players opt for the mineral bonus as opposed to the larva boost. Furthermore, you are stuck with a tech disadvantage and less flexibility using an 10/11 overpool.
Additional larva are of no use unless they can consistently (vs a wide range of strategies) be used to make drones. Sadly, SC2 is just a bit too fast paced of a game, the rushes are just a bit too extreme for the 11 overpool to be viable (at least in my personal experience).
I think you're missing the versatility of 11 overpool openings. I don't think one should open 11 overpool to get 6 lings and pressure. For the reasons you mentioned, you can't do damage to justify it and it sets you behind economically to the point you can't defend things like 4 gate. Instead, the 11 overpool (into 14-15 gas) is an opening that's comparable to 14 gas 14 pool except that one trades a handful of minerals for some extra larva. if you open with only a pair of zerglings to deny scouting and poke at the front, 11 overpool holds 4 gate just about as well as 14 gas/14 pool. Expanding at 21 (a few seconds after gas pool), you can afford pure zergling production while still building up enough for spines or a baneling nest against a 1 gas 4 gate, and you have more larva with which to produce those zerglings. When it comes to 4 gate defense, you effectively trade a short time on spines/tech for more speedlings. The speed on 11 overpool 14 gas is only a couple seconds behind 14 gas 14 pool. The timing on the injects and speed for the build is actually perfect for making speedlings with your second inject and using the lings to delay a 4 gate or expansion.
In TvZ, 11 overpool is again, I think, best used as a speedling opening. It has an easier time with aggressive 2 rax play than the alternatives due to the extra larva, and if no such aggression is forthcoming, then the larva give you plenty of drones to power econ with.
Similarly, in ZvZ, 11 overpool 14(15) gas is a variation on speedling/(baneling) that trades seconds on speed for more larva. It's a perfectly viable alternative to gas/pool, and the relative advantages and disadvantages of each are fairly subtle. I would use this opening anytime I wasn't hatching first.
Even gasless 11 overpool openings don't lend themselves best to lots of early slow lings. For instance, 11 overpool into extractor trick to get an expansion at 19/18 with a pair of lings to scout and deny scouting is effectively on par economically with any other pool/hatch opening.
|
On April 10 2011 13:05 IzieBoy wrote: do you have a thread on hatch first somewhere? what's the scenario when you would do 14 gas/14 pool? i think it greatly depends on scenario, but it is intuitive that a later pool means more minerals. is there an analysis on 15 to 20 pool? just for completeness? keep up the good work!
From my testing I concluded that getting a pool beyond 15 supply was kind of a waste because you will have 200 minerals in reserve anyways, so there is really no reason not to get it.
I did do some analysis on hatch first, and concluded that it is always the most economical option. Hatch first should be your preferred option, however it runs into problems in regard to early pressure, particularly in ZvZ, and also hatch blocking against protoss.
|
Sweet info! Bookmarked for later reading.
|
Also, it's worth noting that 11 overpool 14 gas is practically optimal for various one base all-ins with either roaches or zergling/baneling and denies the initial scout extremely early. Also, if you're willing to give up seeing the inside of the P or T base (mostly, their number of gases) early, you can forego a drone scout, since the fact that you get an early pool and lings makes you safe against cheese and lets you check whether they're 2 raxing or 2 gating or whatever, which helps make up for any minor economic inferiority.
|
What good is larvae if you don't have the income to support it? Yes, zerg players want larvae to produce units, but you want to arrive at a state where your income matches your expenditure on larvae. This is why you see pros favor econ over larva builds. If your larvae production exceeds your income, it's the equivalent of a terran building 5 barracks when he can only produce out of 3 simultaneously. A zerg needs to determine what units he wants to produce (ie. cost per larva) and then add drones to increase his income such that he can then increase his larva output and continue producing more and more of the same unit. Alternatively, he can change the unit he is producing (ie. from a roach to a mutalisk, effectively doubling the cost of each larva) and not add any further larva production, or even enter into a state of excess larva.
|
On April 10 2011 13:44 LilClinkin wrote: What good is larvae if you don't have the income to support it? Yes, zerg players want larvae to produce units, but you want to arrive at a state where your income matches your expenditure on larvae. This is why you see pros favor econ over larva builds. If your larvae production exceeds your income, it's the equivalent of a terran building 5 barracks when he can only produce out of 3 simultaneously. A zerg needs to determine what units he wants to produce (ie. cost per larva) and then add drones to increase his income such that he can then increase his larva output and continue producing more and more of the same unit. Alternatively, he can change the unit he is producing (ie. from a roach to a mutalisk, effectively doubling the cost of each larva) and not add any further larva production, or even enter into a state of excess larva.
With 11 overpool, if you play it economically (i.e. produce only the early pair of lings for chasing off the scout), you have more than enough income to support your extra larva.
|
Nice, gonna try to implement the 11overpool.
Thanks for your work! I've always been curious as to what the differences between these builds were. I always blindly follow the pros. lol
Cheers,
|
A question to everyone who is willing to answer:
Let's say I go for a early +1 Ling/Bane aggression opening, the matchup being ZvZ.
The opener I use at this moment is:
14 Extractor 14 Spawning Pool 15 Overlord Take no workers off-gas.
Make as much Lings as possible, a Baneling Nest when you have 50gas after Ling speed and when the opposing scout is away.
An Evolution Chamber at ~20 supply slightly after the first Queen is out because I start having excess money due to larva shortage. +1 when it's done, the Chamber is done near 100gas. Take a worker off-gas if you don't need that many Banes.
Attack and expand when you get the chance.
The analysis says that I should Overpool in ZvZ if I'm not going for an FE. Is that the case here?
I need minerals for Zerglings, Banelings(25 isn't much but still...), +1 and the buildings required for the three. I also want to expand when I have the chance. I usually make one Drone after my pool is done, usually after the Evo, to cover the loss from the Baneling Nest and the Evolution Chamber. The Overpool probably requires me to be way more sharp with my Droning.
Can an Overpool support all this? Considering that I probably need to drone up before I can start the aggression, does it make any difference compared to 14/14?
|
On April 10 2011 17:49 Airact wrote: A question to everyone who is willing to answer:
Let's say I go for a early +1 Ling/Bane aggression opening, the matchup being ZvZ.
The opener I use at this moment is:
14 Extractor 14 Spawning Pool 15 Overlord Take no workers off-gas.
Make as much Lings as possible, a Baneling Nest when you have 50gas after Ling speed and when the opposing scout is away.
An Evolution Chamber at ~20 supply slightly after the first Queen is out because I start having excess money due to larva shortage. +1 when it's done, the Chamber is done near 100gas. Take a worker off-gas if you don't need that many Banes.
Attack and expand when you get the chance.
The analysis says that I should Overpool in ZvZ if I'm not going for an FE. Is that the case here?
I need minerals for Zerglings, Banelings(25 isn't much but still...), +1 and the buildings required for the three. I also want to expand when I have the chance. I usually make one Drone after my pool is done, usually after the Evo, to cover the loss from the Baneling Nest and the Evolution Chamber. The Overpool probably requires me to be way more sharp with my Droning.
Can an Overpool support all this? Considering that I probably need to drone up before I can start the aggression, does it make any difference compared to 14/14?
I don't really like the build you're doing, but I tried 11 overpool 14 gas, droning to 15 (no scout), making one pair of zerglings, queen, overlord, then two drones (making 17 drones) and then pure zerglings and the rest of the tech you wanted, and you can just barely constantly produce lings if you're perfect with spawn larva. You can't afford an expo unless you skip an inject or make more drones early. However, overpool is nice in that your injects come so early that you can afford the larva to make extra drones and still have the potential to be aggressive. The timings will be a little different - try it yourself (maybe drone to ~20, if you want an expo?) and compare them.
|
Hey lalush, you're definitely better than I am, and I'm going to take your word that you've explored this build and found it lacking but I don't feel your main criticism (the delayed hatch) is valid.
The hatchery is not delayed by anymore than the average 14 pool. Sure a 16 hatch is earlier, if you get it up on time, but about 50% of the time it doesn't go down on time on a 14 pool. I've been watching tourneys and some zergs give up putting down a hatch until zerglings spawn at 22+ supply, often times they have to knock down a pylon for even longer. With an 11 overpool your hatch goes down at 18, before the overlord every time, unless there's an engineering bay in the way. That is early enough in my experience to have the hatch up completely by the time 4 marines arrive at your base, and if you build the crawler when the hatch is at 20% in your main, you can run it down to your lower hatch and have it burrow the moment the hatch pops, rather having to delay marines with drones while you wait for spines to build. Besides clearing out pylons, those lings can perform scouting for bunkers and pylons around the hatch so you can send your overlord someplace more interesting.
I would really like to know which builds it is not viable against so I could practice it with a friend and find out myself if there is indeed a build that I cannot hold against. The only one I've had trouble with is 4 marines, 6 scvs hits at just the right timing, but it is harder to hit that timing with the larger maps these days.
Edit: Also as for the, can't spend minerals on drones on first inject thing, I take up a tact where I get gas early on toss, and just use the first inject on a tumor. The tumor will have time to extend into my nat and that gives me a lot more leeway in defending a 4 gate, and connects my bases quickly so that I can deal with air a little better early and use my queens for defense. Against zvt, if I'm being two raxed, I can use the larvae for extra lings. Since the queen came out 25 seconds earlier, and that's 60% or so of an injection timing, it's almost like free stuff. I'm not totally sure whether this is optimal, but it feels good.
|
On April 11 2011 03:19 onmach wrote: Hey lalush, you're definitely better than I am, and I'm going to take your word that you've explored this build and found it lacking but I don't feel your main criticism (the delayed hatch) is valid.
The hatchery is not delayed by anymore than the average 14 pool. Sure a 16 hatch is earlier, if you get it up on time, but about 50% of the time it doesn't go down on time on a 14 pool. I've been watching tourneys and some zergs give up putting down a hatch until zerglings spawn at 22+ supply, often times they have to knock down a pylon for even longer. With an 11 overpool your hatch goes down at 18, before the overlord every time, unless there's an engineering bay in the way. That is early enough in my experience to have the hatch up completely by the time 4 marines arrive at your base, and if you build the crawler when the hatch is at 20% in your main, you can run it down to your lower hatch and have it burrow the moment the hatch pops, rather having to delay marines with drones while you wait for spines to build. Besides clearing out pylons, those lings can perform scouting for bunkers and pylons around the hatch so you can send your overlord someplace more interesting.
I would really like to know which builds it is not viable against so I could practice it with a friend and find out myself if there is indeed a build that I cannot hold against. The only one I've had trouble with is 4 marines, 6 scvs hits at just the right timing, but it is harder to hit that timing with the larger maps these days.
Edit: Also as for the, can't spend minerals on drones on first inject thing, I take up a tact where I get gas early on toss, and just use the first inject on a tumor. The tumor will have time to extend into my nat and that gives me a lot more leeway in defending a 4 gate, and connects my bases quickly so that I can deal with air a little better early and use my queens for defense. Against zvt, if I'm being two raxed, I can use the larvae for extra lings. Since the queen came out 25 seconds earlier, and that's 60% or so of an injection timing, it's almost like free stuff. I'm not totally sure whether this is optimal, but it feels good.
I used delayed hatch as an argument for hatch first and not 14pool. Hatch first builds pretty much always go down on time nowadays, I don't know what you're talking about. No top zerg player goes hatch first vs Protoss anymore, and that's the only matchup where hatch first builds wouldn't go down on time.
Also, excuse me, but you just don't seem to know what you're talking about or you're just talking from the experience of a gold or platinum league player defending 2rax or reactor hellion. The way you describe defending a fast expansion with an inbase spine crawler will never work against a properly executed 2rax. You seen good terran players kite slow zerglings with marines? He will lose 2 marines while you lose 30 zerglings. 11overpool vs Terran is a highly uneconomical build order. You need more tech/buildings/units/drones than you can afford to make. And the extra larva that's the entire point of the 11 overpool build more often than not have to be wasted in order to afford spine crawlers, queens, roach warren, ...every unit/building in general that require more minerals than you can spare and ends up leaving your larva idle.
Against reactor hellion you pretty much need to go roach warren before expand. The fact that you even have to go roach warren nullifies the entire point of the build order. The point of extra larva is that they need to be used for drones in order for the 11overpool to be able to catch up to the 14pool in economy. Now you used up 1 early drone for a 150 mineral roach warren. And then you had to build at least 3 roaches instead of urgently needed drones.
You will kill yourself using the 11overpool against any decent Terran. You are stuck with 15-16 drones, 2hatches, 1roach warren, 3 roaches, 1 queen. At the same timing a hatch first would have ~22 drones, 2 queens, 2hatches, and 2 spine crawlers defending the same build order.
The inefficiency of the 11overpool's opening timings will just keep compounding throughout the game.
Against a 2rax 1 spine crawler will be nowhere near enough. Even if you make a bunch of lings and the Terran loses their entire marine force -- if the terran isn't completely incompetent -- a 11overpool will have killed itself from not making drones while the terran was building a CC inbase and constantly producing SCV:s. 1base 4rax all-in builds weren't popular back when I played this build, but I'd say they would be close to impossible to consistently defend with an 11overpool opening (even moreso because you are completely blind with slow lings and marines chilling with 0 risk outside your base).
The entire point of this thread is whether greater larva production will result in greater economy in the long run. If you so much as have to waste your first larva inject on lings to defend a 2rax: 11overpool will already start to lag behind. The test in OP tests optimal conditions where you can build drones freely. But having to stay at 13-14 drones for a minute in early game as opposed to being at 17-18 with a resource greedy build will have a great effect on the effectiveness of the build order.
Pros have simply realized that with 11 overpool, you will frequently have larva that you cannot use. And excess larva is just wasted larva and a wasted potentially better economy. That potentially better economy in turn enables you to support more larva faster in the future, at a timing where the larva won't just be wasted larva -- but backed up with an economy that can actually support both it and defending a rush/cheese.
|
Thanks for the response. I don't really lose to 2 rax very often, but I'll experiment with a practice partner to see for myself.
|
On April 10 2011 13:12 jdseemoreglass wrote: @Lalush
I like most of your analysis, however, note that in my testing the 11overpool IS a speedling build. It gets gas at the same time as the 14gas/14pool, and has a larvae advantage in comparison.
I agree that 11overpool in ZvT is not economical and ZvP is not safe. However in ZvZ I think it still has real potential, both in it's larvae advantage and in it's flexibility.
Yes but in your testing you don't build lings and you just keep building drones, am I right?
You have to realize that in a real game the income rate will be skewed in the 14gas/14pool build's favor for several minutes due to the fact that the other build has to produce lings and not drones.
I know it has a larva advantage. But that larva advantage won't be used on drones by anyone doing an abnormally quick pool/speed build. The point of those builds is to use the larva advantage to overrun your opponent with speedlings. Although, only an amateur doing a 14gas 13/14pool will get overrun by such a build.
If the game is to pan out according to your testing, the larva advantage zerg has to make drones with his larva inject. And then the build will end up exactly where I described it. Pushed back into its own base, with 0 clue if the opponent is using his next 6-7 larva on drones or lings, and with 0 clue if the opponent has a baneling's nest on the way or a roach warren -- or neither.
And that's basically where theory and practice diverge. Information is also a resource in SC2 =)
|
On April 11 2011 08:04 onmach wrote: Thanks for the response. I don't really lose to 2 rax very often, but I'll experiment with a practice partner to see for myself.
Excuse me for being overly harsh in my post. I haven't tried 11overpool speedling opening so maybe I'm talking out of my ass. But I somehow can't imagine the timings lining up nicely with such an early gas into speedling research, and with the need of taking an expansion, and then the larva from the inject all popping early and craving your resources simultaneously.
I will try it and see for myself before completely dismissing it.
|
I have been using the 11 overpool in ZvZ for some time and it has gone from being my worst matchup (that used to make me rage) to being my best matchup.
I drone to 14 and right before my pool finishes I make my gas; that puts me on 13 supply so that I can immediately make a queen and six lings when the pool is done. This supply caps me on 18 and unfortunately I have to wait a larva before I can make the overlord. This means my minerals build up quite a bit, which I normally use to throw down a hatchery (sometimes I have tried skipping speed and going straight for a roach warren - depends what my 6 lings find in my opponent's base). The gas timing is 'normal' so I can opt for speed, banelings (with or without speed first), or roaches or whatever (the extra hatch also allows for mass ling all in). Lately I've just been trying Nestea's routine of getting 24 drones and then just pumping roaches and trying to kill my opponent with them, possibly including a fast +1.
The other thing I like about this build is that I don't have to worry about 6 pools. I don't drone scout at all in ZvZ anymore. I mean, I didn't always lose to them or anything, but now it's not even close :D
|
|
On April 11 2011 08:11 LaLuSh wrote:Show nested quote +On April 10 2011 13:12 jdseemoreglass wrote: @Lalush
I like most of your analysis, however, note that in my testing the 11overpool IS a speedling build. It gets gas at the same time as the 14gas/14pool, and has a larvae advantage in comparison.
I agree that 11overpool in ZvT is not economical and ZvP is not safe. However in ZvZ I think it still has real potential, both in it's larvae advantage and in it's flexibility. Yes but in your testing you don't build lings and you just keep building drones, am I right? You have to realize that in a real game the income rate will be skewed in the 14gas/14pool build's favor for several minutes due to the fact that the other build has to produce lings and not drones. I know it has a larva advantage. But that larva advantage won't be used on drones by anyone doing an abnormally quick pool/speed build. The point of those builds is to use the larva advantage to overrun your opponent with speedlings. Although, only an amateur doing a 14gas 13/14pool will get overrun by such a build. If the game is to pan out according to your testing, the larva advantage zerg has to make drones with his larva inject. And then the build will end up exactly where I described it. Pushed back into its own base, with 0 clue if the opponent is using his next 6-7 larva on drones or lings, and with 0 clue if the opponent has a baneling's nest on the way or a roach warren -- or neither. And that's basically where theory and practice diverge. Information is also a resource in SC2 =)
Since this is the second time you misstated my actual testing methods, I'm going to have to assume you didn't thoroughly read the OP. It is incorrect that in the testing I build only drones, in fact I included a whole section entitled "Real world application." In this testing, both the 14 pool and 11 overpool had reached identical worker counts before producing zerglings, negating the argument that the build has to produce drones instead of lings to "catch-up," unless you are referring solely to the 100 mineral disadvantage, which is actually the trade off for additional larvae. The net result was an increase in the maximum number of speedlings for the 11overpool, disproving the claim that you "can't spend your larvae anyway."
I can accept disagreement as to the viability of different builds or whether you prefer minerals to larvae, but I would hope a blue poster wouldn't comment without thoroughly reading the OP, because it might unfairly influence future reader's opinions. Here is the section I am referring to:
Real-world application and comparison
In order to test my theory of the relative advantages of 11overpool vs 14pool in speedling/baneling builds, I set up an experiment. I played repeated tests according to the following general guidelines:
@1:40 - Geyser @16 drones - pump 100% zerglings/overlords @100 gas - Metabolic Boost @50-100 gas - Baneling Nest Continue pumping speedlings and spending 100% of remaining gas on banelings
In attempting to maximize the use of the 11overpool's larvae count while putting all available resources to their best use, I believe I have developed perhaps the most efficient baneling build possible for Zerg. I am fairly confident that no other build can produce a greater amount of banelings and speedlings in a similar timeframe, but I am open to any challengers.
In adapting this build to a fast baneling bust, it also surpassed the current Liquipedia entry for baneling rush, which is very similar but has some small obvious errors in the ordering and timing.
Best Results @ 6:00:
14Pool 15 drones, 27.04 Speedlings, 7 banelings, 5 overlords
11Overpool - encRoach ZvZ build 15 drones, 33.42 Speedlings, 7 banelings, 5 overlords
Build Orders for encRoach Baneling Openings: + Show Spoiler + ZvZ, Speedling heavy or most aggressive: 10 Extractor Trick 11 Overlord 11 Spawning Pool 11 Geyser (immediately after drone) 15 Queen 17 Overlord @100 gas - Speed, remove 1 worker from gas 18+ Mass speedlings 21 Overlord @75 gas - Baneling nest 26 Overlord -Continue pumping speedlings and spend 100% of gas on banelings.
ZvT, baneling heavy, or fastest bust, 5:20 target: Same as above, EXCEPT: @100 gas, do not remove worker from gas @50 gas, baneling nest Cut overlord at 26
Replays:
This test confirmed that in ZvZ or a ZvT baneling bust, there is a slight advantage to going 11overpool when compared to 14pool due to the larvae advantage. This of course is a subjective assessment depending upon a players style and average expansion timing. As was expected, there were more minerals left unspent by the 14pool build. Therefore, with a differing build, it would be possible to divert minerals into gas for more banelings. I will not attempt to compare the relative worth of a baneling to speedlings here since that is a highly subjective and situational assessment. However, it must be noted that this added flexibility of additional banelings must be contrasted with the 11overpool's flexibility of creating a higher number of either drones or speedlings if your opponent goes for play which negates banelings such as roach openings.
|
Have you tested this in team games?
While I understand your focus on the economical impact, the reason 10 pool is used almost on a constant manner in team games is that is the rush is one of the fastest early pressure + sling push before 5 minutes. Suppose I use 11 OP instead of 10P, have you tested how much time will my initial 6 lings for an early push be delayed? I know that 12DE, then pool delays them by like 20 seconds which makes your initial 6 ling rush useless.
|
I got so excited when I saw lalush had posted in this thread, but what a letdown in him not fully reading through the ZvZ applications of overpool.
At any rate, good work OP, I will glady try this stuff out on the ladder today and edit this post with some of my personal experiences.
Regardless of what I think when I come back, I love the analysis here and the conclusions drawn from it. Overpool sounds like the gateway to a super flexible early game, easily transitioning in to ling/bling or even some 1-base roach play due to the mineral > larva conversion.
Thanks OP.
|
Could someone be so kind and explain to me why this whole discussion is based on table that clearly is not accurate?
I have opened that table multiple times, I have went throught those charts and I simply cant figure how those numbers could be accurate, no matter how much I would like to trust them. Basic idea of this thread may be right, but so far whole discussion about it is based on not accurate and not checked table which simply doesnt make sense. Its like deciding the outcome of this "study" and the making the study to look like what maker wants it to show.
So wheres the evidence? Lets see...
All builds starts with same 7th drone, then 8th drone... yet 3 first builds have 5 more minerals. This should have already alerted people and maker of that table to realize hes evidence and test is not going where it should be, table that measures 6 minutes in 20 second intervals having 5 minerals mistake at the first 40 seconds of it. That means the lack of runs of each build is way too small for it to be any kind of evidence and as such faster start/movement/execution may cause huge differences at the 6 minute side of the table, also human error is way too big if even first levels are not scaling, how can later levels be when time and errors multiplies?.
Also minerals gathered by drones are not even close to be equal within same time periods. One good example is 12 pool DET and period 1:20 to 2:20 during that time that build goes from 11.65 drones to 11.24. Minimum drone count being 10. Ok so theres at minimum 10 drones mining for whole time. How come during that minute and those time intervals resource income is as follows: 1:20 =118 (11.65 drones), 1:40=135(11 drones) 2:00=170(10 drones)!!! 2:20=125(11.24 drones). Topping 170 minerals in 20 seconds when dronecount is at 10!! Doesnt alarm anyone? Guy goes from 11.24 (12th drone coming) and 10 mining down to 10 by pool and extractor, yet making more money than was making while having more drones, yet after that when he gets more hes income drops. Umm to me this doesnt make sense, hes not building lords, hes not using drones to make buildings, hes saving for overlord, but that doesnt reduce drone count, so why when he spends 100 minerals to make 2 drones according to table hes income drops by 55? Thats 1 drone more... where did it go?
Same happens with pretty much all builds, which only means theres problem at calculation of these mineral amounts. Same happens all over the board with other mistakes, rendering these kind of tables without large enought pool for each and every build, pretty much useless.
I am sorry that I signed up to give this kind of negative response, much apreciate all the hard work you have done but this is simply one of the things where mistakes costs us Zergs a lot in focusing to wrong stuff. I hope someone puts their time to study these things one build at a time to remove the affect of human error and can give us accurate data, specially with how much resources each drone has produced, and thus eliminating these kind of big errors in stuff that should be clearly visible to everyone.
Just look and actually study the numbers and see yourself.
|
Depends how he measured Blackwolf. If he clicked the mineral patches every 20 seconds in the replay to see how much was left then the error won't be "huge" as you say. It'll stay roughly at +-5 (depending on when he pauses the replay).
I did re-test 11overpool into 16 drones though using his build. Didn't get the same results as the table.
11overpool after 4:00 -- 1540 minerals.
14gas/13pool (forgot 14gas/14pool, i always do 13pool) after 4:00 -- 1735
OH, I just remembered i played my normal build and didn't make 16 drones.
14gas/14pool after 4:00 -- 1800
Test conducted on Scrap station.
On March 19 2011 04:14 jdseemoreglass wrote:
Build Orders for encRoach Baneling Openings: [spoiler] ZvZ, Speedling heavy or most aggressive: 10 Extractor Trick 11 Overlord 11 Spawning Pool 11 Geyser (immediately after drone) 15 Queen 17 Overlord @100 gas - Speed, remove 1 worker from gas
Comments: In a real game, after putting down 11 spawning pool, you don't have enough minerals to make a drone and put down your gas at 1:40. When I tested the build I built a gas at 10, although at 1:43 because i mistimed it slightly and then a drone.
If you want to reverse the order, gas would have to be put down around 1:45-1:46, or later.
There is a bit of larva downtime after putting down the 11pool. You will have 3 larva for 5-10 ingame seconds depending on whether you drone before gas or drone after gas.
15 queen comments: In a real game, if you want to start your queen immediately after pool finishes, you have to 14 queen. Making a drone first, will give you a downtime of 5-6 ingame seconds before the queen can be started. Larva downtime of the 15th drone compared to 15 queen build (~10-11 ingame seconds).
I redid the 11overpool test again to improve execution and ended up at 1550 minerals @ 4:00.
Perhaps with your method (of not putting down the buildings when they have be put down) I'd stand to gain 15-20 minerals more (being generous). That does not explain the 100+ mineral discrepancy between your results and mine.
*Edit: I didn't remoe 1 worker from gas in my testing. Since I assumed he didn't either when making the tables. I kept drones mining on gas while testing both builds.
*Edit2: Removing 1drone from gas would give approx 35-40 more minerals. Still a discrepancy.
What I suspect: OP removed all drones from gas @100 gas when he tested 11overpool, while removing none while testing 14gas/14pool.
*edit 3: Continuing to build drones after 16 drones would not have explained the discrepancy either. Since the 17th drone finishes at around ~3:40.
*edit 4: Using your build order and replicating it as best as I could (making gas at 1:45...), making drones as fast as I could (17th drone finished at 3:45 and 18th drone finished at 3:47). I managed to squeeze it up to 1590 minerals @ 4:00.
I doubt it could be made much better. Without the 17th and 18th drone, it'd be 1570 minerals.
|
United Arab Emirates492 Posts
@Lalush, could you please check how the (idra's zvp opener) 15gas/14pool compares to the ones mentioned by you. Larva/mineral advantage over 14gas/14pool?
|
Idra most commonly opens 14pool and ~15gas.
And that's probably the most common opener at the moment in ZvP (basically the only I use). The gas first into pool openers have gone slightly out of style, except with a few players like morrow (who i saw deviate from his 14gas 14pool for the first time vs MC at dreamhack invitational, in favor of a 14pool in one of the games).
|
United Arab Emirates492 Posts
On April 14 2011 11:49 LaLuSh wrote: Idra most commonly opens 14pool and ~15gas.
And that's probably the most common opener at the moment in ZvP (basically the only I use). The gas first into pool openers have gone slightly out of style, except with a few players like morrow (who i saw deviate from his 14gas 14pool for the first time vs MC at dreamhack invitational, in favor of a 14pool in one of the games).
@Lalush, sorry to ask you here but what about on long maps where the protoss is most likely going to forge expand, do you recommend going 14pool/16hatch (delayed hatch if pylon blocked) or would you still suggest 14pool/15gas is the most optimal build?
Thank you for the wealth of knowledge you provide on these forums .
|
On April 14 2011 12:00 Gunman_csz wrote:Show nested quote +On April 14 2011 11:49 LaLuSh wrote: Idra most commonly opens 14pool and ~15gas.
And that's probably the most common opener at the moment in ZvP (basically the only I use). The gas first into pool openers have gone slightly out of style, except with a few players like morrow (who i saw deviate from his 14gas 14pool for the first time vs MC at dreamhack invitational, in favor of a 14pool in one of the games). @Lalush, sorry to ask you here but what about on long maps where the protoss is most likely going to forge expand, do you recommend going 14pool/16hatch (delayed hatch if pylon blocked) or would you still suggest 14pool/15gas is the most optimal build? Thank you for the wealth of knowledge you provide on these forums .
If you scout forge expand from protoss and intend to play straight up macro you don't need to mine gas with 3 drones. I usually use 1-2 drones and get delayed speed or no speed upgrade at all.
Just mine the gas slowly for an eventual lair.
But playing macro vs FE toss is stupid in itself, so what does it matter what build you use ^^ I don't think the small variations matter much in the end.
Idra seems to prefer the fast hatchery whenever he can get it. I prefer getting the queens on time.
A matter of preference. I just haven't got the 14pool/16hatch timings mapped out as well in my mind, so I stay away from it.
|
Just to clarify due to some confusion in this thread. I have been assuming people are following from my previous threads where I detail all of my methods and provide replays of each test.
To come up with the mineral counts in the table, I used the replay bar on pause to reach exact 30 second intervals. Then to calculate minerals mined, I used the following formula:
Minerals spent tab + Current minerals owned - 700 initial starting minerals = total resources mined.
Drones in production were prorated depending upon their percentage of completion.
All of the data in the initial table came from maximizing economy and producing no army or buildings besides a spawning pool, queen, and gas. After repeated tests, I used the data from the replay which had the highest efficiency and discarded the rest.
The data at the second half of the OP follows the build orders I provided for speedling/baneling play, and is therefore distinct in it's production and gas mining, and therefore won't align perfectly with the data in the table.
I have replays available of every test by request, though they are from an older patch.
|
thumbs up Jd, in Comparison for the crowd i go ~ 1775 ressources collected 150minerals 100 gas unspendt @4.00, havin a spinecrawler against 6poolern cause i try it in 4v4s
but, im fixing my problems with this build by getting 2 extractors harvesting for banelings and 7minutes lair tech by getting 2nd extractor at the 5.20; because of the fact when ur attacking with the 11pool banes, the fights will be in his eco, no matter of his single speedlings try forcing banes to explode, u just need enough gas to constantly morph 2 banes while u stress with the later speedlings, after bane-chess i Defend my expo with a pack of roaches and lings scouting his roachcount and putting down a spire for the pre-Infestor timing kill ;> if he will go all in - fine save up some banes
|
Awesome writeup, thanks for all the hard work. I'm definitely going to try and incorporate this into my gameplay, and maybe 2v2s as well.
One note tho, perhaps a second hatch would be more viable if it's positioned correctly. My idea is to place it at the top of your ramp, and use it to partially wall it. Then, correctly placed spines (using creep from the spawned hatchery) would then give zerg a total wall off, something no zerg is able to do normally.
Maybe this could even be placed at chokes such as shattered temple, where instead of early expanding you early hatch at the choke, giving you more larvae and the opportunity for a safer expand against early toss/terran agression (i dont see this helping as much with zerg).
Actually I think I'm going to try this out then post a discussion thread... Either way great writeup, thanks a ton! :D
|
Awesome thread, well thought out and well presented!
|
you're a fucking god. I hope you can write more stuff like this, love the amount of research done. still going over data. Also do you plan on doing more zerg opening research? mind going over morrow's gas first build?
|
Might not be related but what about the hatch first builds? Which one is the most efficient (in larvae and/or resources)? 14hatch/pool or 15 or 16 or something else (11 -.-)?
|
this post is so fucking awesome! new home away from home
ty op for all the work and effort, really appreciated
|
|
On April 10 2011 13:16 jdseemoreglass wrote:Show nested quote +On April 10 2011 13:05 IzieBoy wrote: do you have a thread on hatch first somewhere? what's the scenario when you would do 14 gas/14 pool? i think it greatly depends on scenario, but it is intuitive that a later pool means more minerals. is there an analysis on 15 to 20 pool? just for completeness? keep up the good work! From my testing I concluded that getting a pool beyond 15 supply was kind of a waste because you will have 200 minerals in reserve anyways, so there is really no reason not to get it. I did do some analysis on hatch first, and concluded that it is always the most economical option. Hatch first should be your preferred option, however it runs into problems in regard to early pressure, particularly in ZvZ, and also hatch blocking against protoss.
Might just message you on bnet since I've seen you on more recently, but in case you read this first...
In your testing, what food hatch did you test?
I've heard silly things like 6hatch could be economically superior if you could get away with it (for example, maybe in team games??
I understand the reasoning behind 14hatch (earliest you can hatch first without larva capping)
And also 15hatch (the earliest you can hatch without excess larva)
What I'm interested in is, is there an earlier hatch-first build that nets some advantage that maybe doesn't cut nearly as much econ as it seems. (something akin to 11 pool 18hatch, but for hatch-first).
I've been messing lately with 11 hatch 12 pool and having some interesting results.
I'm curious if you have any data on any early hatch-first builds?
|
Very interesting thread. Interesting to see that 12 pool does very well compared to 14 pool, with only slightly less resources but slightly ahead on larva.
I was kinda expecting it to do even better on larva, since the 12 pool is down only very slightly later than an 11 overpool; some of the difference might just be a quirk of periodic larva timings and smoothing/sampling. But unlike 11 overpool, 12 pool avoids the mining hit caused by going for an extractor trick instead of building an overlord on 9.
Anyway, it confirms my anecdotal feeling that 12 and 14 pools seem to have no noticeable economic difference, and I like the slightly earlier lings. Seems to me like I should use 12 pool as my standard pool first option.
|
On March 19 2011 04:14 jdseemoreglass wrote:
In ZvZ, assuming you aren't quickly expanding, I think 11overpool is clearly best. An increased larvae count is a significant advantage in such an aggressive matchup. The extra minerals of the later pool builds generally aren't needed in this matchup if you are playing pool first, since you will typically not be fast-expanding, and you will be gathering more than enough in order to constantly produce lings and banelings. Getting an earlier pool will also give you the flexibility of going all-in if you scout a player going hatch-first against you, and also give you greater safety in defending against 6/7pool cheese. Due to all these factors, I would recommend 11overpool with gas at 1:40 instead of the more common 14gas/14pool opening in ZvZ.
In ZvP I would lean towards the 14pool build, because it is fairly common to expand at some point in the 20's supply, and also to get a roach warren and/or spines, all of which require a higher mineral count advantage as opposed to larvae.
In ZvT I almost always favor opening hatch first, but if you are going for an aggressive baneling bust or roach rush, 11overpool should always be best.
I just wanted to say that since your posts long ago on the subject, I've adopted 11overpool as my standard in ZvZ and ZvP (I go for a ling heavy ZvP with delayed roaches though), and I hatch first in ZvT, though I'm open to changes in ZvT right now as I'm pretty lost in that matchup currently.
I think that 11overpool in ZvZ is really great for the reason you mentioned. There is a time where ling speed finishes, and your opponent literally cannot have more lings than you. They must defend with something else, be it sunks+lings, banelings, roaches, queens blocking the ramp can help, etc.
But I do want to mention that I've tried being very aggressive early with 11overpool against hatch first, and it just has not worked for me. I've found it best to just harass drones/stop them from mining, or maybe get a kill if you're lucky, keep the lings running around to scout a roach warren or baneling nest, but keep them alive to be part of your big attack when speed kicks in.
ZvP I think 11overpool is great because (1) I don't want to deal with cannon rushes when the Protoss opens forge FE and I open hatch first, and (2) I can drone to 15, then make a pair of lings and queen to get to 18/18, and the lings come out at a good timing to chase the probe and deal with expansion blocking antics. If they don't cancel their pylon, they can delay the hatch a bit, but it's not very long considering the Protoss lost 100 minerals doing so.
|
Im really confused - if this thread puts an end to the econ debate, why is there a 2nd look?
It's also a bit confusing since some people like speedlings, some don't.
I'm reading ZvZ = 11pool/14gas is better than 14gas/14pool for the same thing For Zvt 'spanishiwa no gas' = 14 hatch 15 pool
I like speedling openings in ZvP, so I might as well go 11 pool 14 gas in that match up too?
in ZvZ 11pool/14gas is better than 14/14 because although you 14 has more money, 11 has the larva and you can't spend the money of 14 anyways while 11 has better macro in that the larva and mineral income is more sustainable.
|
On May 17 2011 08:51 michaelhasanalias wrote:Show nested quote +On April 10 2011 13:16 jdseemoreglass wrote:On April 10 2011 13:05 IzieBoy wrote: do you have a thread on hatch first somewhere? what's the scenario when you would do 14 gas/14 pool? i think it greatly depends on scenario, but it is intuitive that a later pool means more minerals. is there an analysis on 15 to 20 pool? just for completeness? keep up the good work! From my testing I concluded that getting a pool beyond 15 supply was kind of a waste because you will have 200 minerals in reserve anyways, so there is really no reason not to get it. I did do some analysis on hatch first, and concluded that it is always the most economical option. Hatch first should be your preferred option, however it runs into problems in regard to early pressure, particularly in ZvZ, and also hatch blocking against protoss. Might just message you on bnet since I've seen you on more recently, but in case you read this first... In your testing, what food hatch did you test? I've heard silly things like 6hatch could be economically superior if you could get away with it (for example, maybe in team games?? I understand the reasoning behind 14hatch (earliest you can hatch first without larva capping) And also 15hatch (the earliest you can hatch without excess larva) What I'm interested in is, is there an earlier hatch-first build that nets some advantage that maybe doesn't cut nearly as much econ as it seems. (something akin to 11 pool 18hatch, but for hatch-first). I've been messing lately with 11 hatch 12 pool and having some interesting results. I'm curious if you have any data on any early hatch-first builds?
I experimented with very early hatch (to deny blocking). The best IMHO is 13 hatch with triple extractor trick. You put the hatch before the first overlord, but get up to 13 drones by doing a triple extractor. The follow up is then ovie and 13/14/15 pool. You use the drone from the nat extractor trick to scout, else a double extractor => hatch => 13..15 pool will be roughly the same. You get a very early hatch down (~1'50) without too much larvae loss, however it will cost you some economy. The problem is, that you are still extremely vulnerable to cannon/bunker rush, and because of the early hatch you often do not have the option to cancel in case of all in's.
Abandonned it, but i played it for a while so it is not completely unplayable especially on small maps where your nat is blocked easily.
|
On May 18 2011 14:20 Belial88 wrote: Im really confused - if this thread puts an end to the econ debate, why is there a 2nd look?
In the original thread where that line was used, I was referring specifically to the debate of whether hatch first or pool first is more economical. It was shown after repeated tests that hatch first is always ahead in minerals mined.
I intended this thread to explain the data on pool first builds and show that there is a trade-off relationship between minerals and larvae.
It's also a bit confusing since some people like speedlings, some don't.
I'm reading ZvZ = 11pool/14gas is better than 14gas/14pool for the same thing
In ZvZ, speedlings openings are typically standard. Some people prefer hatch first on certain maps, but in general speedlings is the proper way to open. The difference between 11pool and 14pool is that 11pool gives you more total speedlings than the 14pool, and also gets them out faster.
I like speedling openings in ZvP, so I might as well go 11 pool 14 gas in that match up too?
in ZvZ 11pool/14gas is better than 14/14 because although you 14 has more money, 11 has the larva and you can't spend the money of 14 anyways while 11 has better macro in that the larva and mineral income is more sustainable.
If you like a certain opening, by all means do it. But I will let you know that it is standard to expand against protoss once you clear your natural of any probes and pylons. Going 14pool instead of 11overpool will give you a slight resource advantage, allowing you to drop your hatch faster or get a faster roach warren or spine.
---------------------------------------------------------------------------------------------------------------------------------------------------
I've been wanting to do some testing of 11overpool vs 14pool timings against protoss expansion blocks. If anyone is willing, I would like someone to test 1v1 me to block my expo with a pylon each game. Or, if someone has experience with the map editor and can show me how to make a map that can reset with a pylon in place, I would really appreciate it. If anyone else has questions regarding my testing, I'll be happy to answer them in this thread.
Also looking for any master Zerg's who are good with baneling play and would be willing to try some practice games against 11overpool.
|
|
I just found this thread today and I'm surprised how zvx is pretty much following what is in the op. I'm not sure if the players consciously know the math behind it or if it's just a ton of trial and error.
|
For all intents and purposes, spanishiwa style is identical to 14-15 hatch in econ. The difference being that you don't mine gas so you will obviously get more minerals.
|
Your economy explodes with the spanbuild. I usually pool 15 and almost didn't know what to do with all the minerals I had ;-)
|
Been a while since someone has done a post.
SO, if we were to compare the 11 overpool to the 14 pool. We would find that the 11 overpool would trade some mins for extra larva via an earlier queen WHICH leads to better defence against early cheese.
Although, excuse me if someone has already posted this question but if the 11 overpool was to make 6 lings, would the affect on the econ be any worse than that if it was an 14 pool building 6 lings as soon as the pool pops? And how badly it affects the econ. All has to be relative to the 14 pool.
Also if someone could run down the BADS of the 11 overpool hence we might find out what counters it. So i think i read, in ZvP early stalker pressure is a problem cause speed would be researched to late due to a late gas ect.
A reply would be great.
|
the numbers that impressed me the most actually were the 12 pool standard, it seems to result in an extremely favorable position, and I could definitely see its potential in ZvZ, seeing how it could punish fast expands better than 14 pool standard, and still maintain a decent economic status.
|
played around with 10 pool, it has a major timing advantage: because of the fast queen, the first 4 injected larvae spawn round 4'00 they come in right to build 8 lings in case a bunker/cannon rush is incoming.
Over pool is ~20 seconds later, which defintely makes a difference. In case you are not rushed, just pump drones.
also note that a lot of effort is put on how to save 50 minerals delaying pool, however delaying gas for 30 in-game seconds yields ~100 minerals. Additionally scouting with early lings is cheaper and more effective (2 scouts for 50 mins). I 'd rather have later speedlings than a later pool :-)
So 10 pool to me looks like a very flexible and economic build, I trade early game safety and option for early agression against a later speed. On large maps actually an overpool is more efficient as the timigs shift some seconds, so overpool inject is in time to fight early harrass.
|
Super awesome, thanks a ton! This really helps me, especially in ZvZ. I'm glad someone did this because it's what I really want to know about my builds but don't have time to figure out for myself. Probably one of the best posts in the strategy forum this year.
|
Im going to state a few things. one, I got a lot of success in lower leagues out of double macro hatch in base and NO natural. There's obvious strengths to it: your army all comes out together instead of split apart, leading them to be stronger on a defensive front, and you can saturate super quick if you aren't attacked. Queens can be massed very quickly in case of banshee/vr rush or other air play. most people I used this against assumed I was using an ice fisher build. never heard of it, and was simply applying the lessons from SC1 to SC2.
my first 10 games in SC2 I saw the obvious advantage in an early queen, but this went by the wayside as I naturally attuned to more standard builds through trial and error. However, dont underestimate the 10 pool.
you get out 4 extra larvae a full cycle sooner than ZvZ standard openers, and against terran or protoss, this means a slightly earlier saturation with another cycle of larvae popping sooner, meaning 2 rax or other very early aggression can be held off with 8 lings jumping out just about the time they get to your base. On the other hand, if they have tried and FE or, a protoss specifically, has been slow about getting their wall up, you can force them to spend early resources on defense and mess with their build timing, possibly even kill them outright with good micro, and certainly destroy parts of their wall. This does not work on giant maps like tardalrim, but on maps like metalopolis or shakuras, can work ~50% of the time if you do it blindly. Its like an eco heavy ling cheese rush. However, it has the advantage that you can either morph drones for higher saturation sooner, or morph lings to punish some risky play. the 11 overpool is much less capable of punishing risky play or very early aggression. In doing both, you want a decisive advantage that affords you minimal losses while getting rid of their strategy, which will cost them in the long run.
As a rule, if I know nothing about the other player, and no scouting info can be accurately had early on, I go 10 pool v Z, 13 V T, and either 15 or 15 FE/14pool v P on 4 location maps. I go for more eco earlier on larger maps, where any rushing from either side will be almost useless, and for earlier pool on shorter maps, where I can either defend a rush, rush myself, or get early high saturation. Seems to work.
I should've clarified that I get pool before OL at 10. thats what makes it extremely fast eco or rush defense.
|
Thank you, I believe this can be used by other races to get optimal information, such as the more generic early pressure builds for other races will be most affective before/during/after 4:20. The safest would be at 4:20 where a generic build could work if you are blind. I will look into this and see if I can exploit it in the meta-game (tldr comments, only OP). Sorry if someone ninja'd me
|
Likewise, the 12overpool-DET build should generally NOT be used. It gives you up to 35 minerals more than the 11overpool, but results in up to 2 fewer larvae, which is not a good trade-off.
Ok, I haven't done a spreadsheet here or anything, but something's bugging me about this claim.
11 Overpool wastes some larva regen; I want to say it sits at the max 3 larva for about 10 more seconds than 12overpool-DET, which is around 60% of one larva generated. In return it gets a pool (and therefore Queen) about 6 seconds earlier. Spawn larva produces 4 larva every 40 seconds, so 1 larva every 10 seconds, so by delaying the pool we're looking at around 60% less of one larva generated.
Which is to say I'm calculating +0.6 -0.6 = exactly the same larva generated between 11 Overpool and 12Overpool-DET.
Where do you get "2 fewer larvae"?
|
you probably dont have the minerals to spend the larva afterwards, or you sit longer before making pool because you have 3 drones while waiting for the money to double extractor trick 2 drones, thats 150 minerals before the overlord.
|
On August 18 2011 04:54 metroid composite wrote:Show nested quote +Likewise, the 12overpool-DET build should generally NOT be used. It gives you up to 35 minerals more than the 11overpool, but results in up to 2 fewer larvae, which is not a good trade-off.
Ok, I haven't done a spreadsheet here or anything, but something's bugging me about this claim. 11 Overpool wastes some larva regen; I want to say it sits at the max 3 larva for about 10 more seconds than 12overpool-DET, which is around 60% of one larva generated. In return it gets a pool (and therefore Queen) about 6 seconds earlier. Spawn larva produces 4 larva every 40 seconds, so 1 larva every 10 seconds, so by delaying the pool we're looking at around 60% less of one larva generated. Which is to say I'm calculating +0.6 -0.6 = exactly the same larva generated between 11 Overpool and 12Overpool-DET. Where do you get "2 fewer larvae"?
11 overpool idles larvae creation for ~5 to 7 seconds. if you drone micro for better income, you can reduce this even more. Regarding larvae: larvae spawns 10 seconds earlier throughout the early game. So each 40 seconds thee 11 overpool is actually 4 larvae ahead for 10 seconds. if you do then average this over 6 minute, it turns to be out to be more "larvae rich" (i don't agree). Actually 12 det overpool gets the same amount of larvae, but 10 (are you sure about this timing?) seconds later.
In my opinion in real games timings are more important than +- 50 minerals income. So i am going for 10-overpool in order to get my expansion and lings out earlier. Getting the expansion down ~15 seconds earlier also results in one more larvae generated. Additionally i like to kill the scout and scout with initial lings as soon as possible, so the ~10 seconds pool delay between 11 overpool and 10 overpool do matter to me more than losing an early larvae. Earlier lings = earlier safe expansion = more larvae+income = earlier queen = more options (e.g. all in against ffe).
|
On August 18 2011 05:08 BinxyBrown wrote: you probably dont have the minerals to spend the larva afterwards, or you sit longer before making pool because you have 3 drones while waiting for the money to double extractor trick 2 drones, thats 150 minerals before the overlord.
Correct, you have to do the first Extractor Trick at about 100 minerals rather than waiting for 150, or you'll hit 3 larva (which you want to avoid). Leave that drone as an extractor until you do the second extractor trick, and then cancel both extractors.
|
On August 18 2011 05:16 Schnullerbacke13 wrote: Actually 12 det overpool gets the same amount of larvae, but 10 (are you sure about this timing?) seconds later.
Here's what I know on timings.
If I am as flawless as I am able to get, then 12Overpool-DET almost--but not quite--doesn't waste larva (maybe one second of three larva after the pool goes down and I don't quite have 50 minerals for a drone).
The extra drone costs essentially 61 minerals (50 + 6 for cancelled extractor + 5 for missed mining cycle). 61 minerals is...roughly one drone mining cycle for 12 drones.
One drone mining cycle is...(researches) about 7 seconds.
This means for 12Overpool-DET the pool goes down 7 seconds later, but one extra larva is used.
In the mean time, 11 Overpool would get the Pool 7 seconds earlier, but use one less larva in this time. A hatchery produces larva once every *doublechecks* 15 seconds. So...15-7 = 8 wasted larva regen time.
(My mistake--I was mixing the larva regen time (15) with the Drone build time (17)--which is how I got to 17-7 = 10 in that post).
So...ok, this is more like +0.7 - 0.55 = +0.15. (11 Overpool gets more larvae, but it's still definitely not a larva gap of 2).
|
On August 18 2011 05:49 metroid composite wrote:Show nested quote +On August 18 2011 05:16 Schnullerbacke13 wrote: Actually 12 det overpool gets the same amount of larvae, but 10 (are you sure about this timing?) seconds later.
Here's what I know on timings. If I am as flawless as I am able to get, then 12Overpool-DET almost--but not quite--doesn't waste larva (maybe one second of three larva after the pool goes down and I don't quite have 50 minerals for a drone). The extra drone costs essentially 61 minerals (50 + 6 for cancelled extractor + 5 for missed mining cycle). 61 minerals is...roughly one drone mining cycle for 12 drones. One drone mining cycle is...(researches) about 7 seconds. This means for 12Overpool-DET the pool goes down 7 seconds later, but one extra larva is used. In the mean time, 11 Overpool would get the Pool 7 seconds earlier, but use one less larva in this time. A hatchery produces larva once every *doublechecks* 15 seconds. So...15-7 = 8 wasted larva regen time. (My mistake--I was mixing the larva regen time (15) with the Drone build time (17)--which is how I got to 17-7 = 10 in that post). So...ok, this is more like +0.7 - 0.55 = +0.15. (11 Overpool gets more larvae, but it's still definitely not a larva gap of 2).
as pointed out, the larvae advantage seems "computed" by calculating-in the slightly earlier inject over time. In real games the minor income difference will be less significant compared to pool timing .
|
I always thought 12 pool DET was a bit off in timings, oh well
|
in my own personal testing of 11 overpool and 12 overpool DET
the build order optimizer shows them being very equal
but in my testing of replays, i do both builds and see which one has more larva/minerals/stuff at the 4 minute mark. the 12overpool DET has about 100 more minerals and 1 less larva
i dont know what is causing this. i dont think im making mistakes in either build, i try to do them perfectly each time
|
Hm, did a couple practice runs comparing...
12overpool DET and... 9 overlord, 12 pool
The latter was consistently ahead by a small margin in-practice (like...pool 2 seconds earlier).
This could just be me not being very good, though, what with 9 OL, 12 pool being a lot easier to execute (I'm like...diamond on the NA server, I guess? So yes, there are probably people who can cancel extractors faster than I)
|
Hi,
And, what for 13pool and 15pool reach mineral/larvea gain?
|
I'm glad you cleared that up, I'm sick of hearing that 11p18h is the most economical pool-first when that simply isn't the case.
|
I have played you many times before on NA, very good anzalysis.
|
@OP 11 pool in ZvZ.. Sure you can't spend your resources if you're doing 1 base no tech but that is plain bronze.. If you're baneling all in, speedling expansion or ~16 drone roach warren off 14g14p you can easily have zero resources. Both the roach build and speedling expand struggle to get enough resources: resources are scarce, larvae is abundant!!! It's only with a no-expand baneling build that larvae is scarce and resources are abundant.
You basically have to invent 11pool vs 1414 and 11pool vs FE, play 200+ games vs GMs and supply replays of how it holds up against all the different playstyles before your recommendation has any legitimacy whatsoever. Not saying it's bad. Just saying it's completely undemonstrated and thus the recommendation has 0 grounding until it IS demonstrated. I'm also claiming that your claim that you can't spend resources is dead wrong (i.e. 17-19 supply roach warren, speedling expand, 1414 baneling all in a suspected hatchgaspool).
But I deeply appreciate your research. This is really the only criticism I can field for all that you've said.
Also 13h15p and 14h15p are the best if you don't get gas. In a realistic game you're putting 3 on gas and spending at least 100-100 on speed, all before 28/28. You're also -1 drone for a long time if you're scouting. With all this you can't even spend your larvae off 13h15p and the hatchery ceases to generate larvae. This is why I choose things like 15/15/17 15/16/16 etc etc even though it has less larvae. On the face of it, this seems to be more efficient because you can keep your money at 0-0 and have 0 larvae, but with 13h15p you can keep your money at 0-0 but have tonnes of larvae without the natural hatch generation being in use [because you've had to make at least 2 lings perhaps, -1 drone for scouting, and 3 drones on gas, and 100 mineral expenditure on speed, and possible need for spine of you're vsing reactor hellion or in ZvZ FE vs FE etc.]
|
Hey OP, you are a serious nerdboss
That said, I Think it would be awesome if you could provide us some additional data.
You said that all builds have gas timings of 1:40, does this mean that all builds can research metabolic boost at the same time? I sense that trying to do this would perhaps conflict with the timing for when some of the early pool builds would prefer to build a queen - but maybe not. Perhaps you could rejig some of the gas timings to give us some practical ideas of how games may play out..
So, it would be cool to have some timestamp data for when a pool can go down, and when metabolic boost can be started (i.e. pool is up and we have 100 mineral/gas).
It would also be cool to see when a queen would pop with each build.
|
|
|
|