[D] Why would you ever hatch first? - Page 5

Adebisi

Canada1637 Posts

October 29 2010 12:32 GMT

#81

On October 29 2010 14:40 LegendLength wrote:
I ran some tests of my usual 16 hatch, 15 pool build against a 15 pool, 16 hatch. The numbers were very similar at time 6:30 . That is taking into account number of drones, amount of minerals and queued production at that time.

I've always believed hatch first is best but this is making me strongly reconsider.

I don't understand the few comments that said, when going hatch first, the queens are free to place a creep tumor. I find there is enough minerals to keep creating drones out of the larvae using constant injection from the start. Although there are some 1 - 2 second pauses occasionally when waiting for 50 minerals.

The build exact build order is:

9 overlord
13 scout
16 hatch
15 pool
18 extractor
17 overlord
25 overlord

All larvae are converted to drones except for a zergling at 19. It also includes getting zergling speed as soon as 100 gas arrives (and not taking the drones off gas afterwards).

The reason you can use up all your larva is because 16 hatch means the hatch is up later than 14 hatch -> less larva, also later pool -> later queen -> less larva. If you go for 14 hatch and pool after the queens will have spare mana from not needing to inject because of more natural larva generation, and I believe slightly weaker econ, basically 14 hatch compared to 16 hatch has more natural larva with a weaker early econ.

I'm pretty sure at least o.o.

Perscienter

957 Posts

October 29 2010 13:01 GMT

#82

I recommend pool first for getting these two zerglings out. For chasing scouts, get some information yourself and shutting down brazen stuff the enemy can throw at you. 15 hatch is also more vulnerable to 2 gate pressure. Maybe hatch first is recommendable on cross positions on large maps like shakuras plateau and metalopolis.

edit: I also remember my 15 hatch getting busted by a 12 p against my 9 z on delta quadrant. The distance was just too short.

ChickenLips

2912 Posts

October 29 2010 13:29 GMT

#83

On October 29 2010 21:09 Palmar wrote:
I think this may be relevant to the thread.

I ran some tests using the build order tester, and it turns out that you lose almost nothing in terms of speed if you're going pure drones, if you 10 pool

link to thread/discussion: http://www.teamliquid.net/forum/viewmessage.php?topic_id=164575&currentpage=2#37

I'm not sure what this implies, being a big fan of hatch first v terran and pool first v toss... but I think I'll start experimenting with the 10pool/16hatch opening. It greatly increases flexibility in dealing with cheese and early attacks. If the economy loss is as little as those very narrow results seem to indicate, then it's worth it on the cesspool of random all-ins and cheese that is ~1600 diamond.

The build order to have 50 drones and 4 lings as fast as possible is:

10 SpawningPool M:204 G:0
10 Overlord M:100 G:0
13 Queen M:153 G:0
16 Hatchery M:305 G:0
17 Overlord M:108 G:0
19 Queen M:202 G:0
22 Overlord M:103 G:0
35 Overlord M:327 G:0
36 Overlord M:489 G:0
36 Queen M:605 G:0
39 Overlord M:405 G:0
40 Queen M:255 G:0
45 Zergling M:286 G:0
49 Zergling M:199 G:0
50 Overlord M:149 G:0

You can shave off a second using 3hatches and some weird timings, but this is the intermediate result that intrigued me... and the final result of the optimizer was only a second better than this (6:17 vs 6:18).

oh, and obviously you want your lings out much earlier in a standard game, but the mineral loss from slightly slower drones is not going to slow you down too much. I also realize this is a simple test, and not an actual build. 50 drones blind is a terrible idea, this is just for discussion.

mind = blown, so excited to test this out when i get home :D

BlackDraft

United States64 Posts

October 29 2010 13:42 GMT

#84

I go hatch first to prevent my opponent from preventing my expo with a scout.

Palmar

Iceland22633 Posts

October 29 2010 15:48 GMT

#85

The current test I'm running is twofold:

First test simply finds the fastest way to 40 drones, 2 hatches, 1 geyser, metabolic boost and at least 1 queen + 4 lings

Second test does the same, except it forces a 10pool and then proceeds with the same final goal. I'll give you the time difference when both tests have had an hour or so to run.

Perscienter

957 Posts

October 29 2010 16:09 GMT

#86

An hour to run? Do you test this out with the real game or just in the simulator? I'd prefer the first one.

Palmar

Iceland22633 Posts

October 29 2010 16:32 GMT

#87

I'm using the zerg build order optimizer.

http://www.teamliquid.net/forum/viewmessage.php?topic_id=160231

It doesn't simulate a predefined build, rather optimizes a build given some predetermined constricting factors.

I have finished the unrestricted test, it managed to achieve the end result in 6:02 using the following build

9 Overlord M:105 G:0
13 Hatchery M:308 G:0
14 SpawningPool M:201 G:0
14 Overlord M:108 G:0
14 Extractor M:55 G:0
17 +1 Drone on gas M:114 G:0
17 +1 Drone on gas M:114 G:0
17 +1 Drone on gas M:152 G:4
18 Queen M:204 G:24
23 Queen M:158 G: 72
25 Overlord M:111 G:89
25 +1 Drone on minerals M:11 G:89
25 MetabolicBoost M:117 G:100
30 +1 Drone on minerals M:10 G:25
31 +1 Drone on gas M:191 G:36
31 Overlord M:191 G:36
31 Queen M:155 G:41
35 Overlord M:181 G:63
36 Queen M:156 G: 73
41 Zergling M:59 G:88
47 Zergling M:61 G:109
50 +1 Drone on gas M:106 G:122

Full details here:

+ Show Spoiler +

@0:00 M:50 G:0 L:3 S:6/10 BuildDrone
@0:13 M:51 G:0 L:2 S:7/10 BuildDrone
@0:15 Spawned: Larva+1
@0:17 Spawned: Drone+1
@0:24 M:52 G:0 L:2 S:8/10 BuildDrone
@0:30 Spawned: Drone+1
@0:30 Spawned: Larva+1
@0:41 Spawned: Drone+1
@0:43 M:105 G:0 L:2 S

/10 BuildOverlord
@0:45 Spawned: Larva+1
@0:51 M:56 G:0 L:2 S

/10 BuildDrone
@1:00 Spawned: Larva+1
@1:08 Spawned: Overlord+1
@1:08 Spawned: Drone+1
@1:08 M:114 G:0 L:2 S:10/18 BuildDrone
@1:08 M:64 G:0 L:1 S:11/18 BuildDrone
@1:15 Spawned: Larva+1
@1:15 M:63 G:0 L:1 S:12/18 BuildDrone
@1:25 Spawned: Drone+1
@1:25 Spawned: Drone+1
@1:30 Spawned: Larva+1
@1:32 Spawned: Drone+1
@1:45 Spawned: Larva+1
@1:50 M:308 G:0 L:2 S:13/18 BuildHatchery
@1:55 M:51 G:0 L:2 S:12/18 BuildDrone
@2:00 Spawned: Larva+1
@2:01 M:53 G:0 L:2 S:13/18 BuildDrone
@2:12 Spawned: Drone+1
@2:15 Spawned: Larva+1
@2:18 Spawned: Drone+1
@2:23 M:201 G:0 L:2 S:14/18 BuildSpawningPool
@2:29 M:57 G:0 L:2 S:13/18 BuildDrone
@2:30 Spawned: Larva+1
@2:40 M:108 G:0 L:2 S:14/18 BuildOverlord
@2:45 Spawned: Larva+1
@2:45 M:55 G:0 L:2 S:14/18 BuildExtractor
@2:46 Spawned: Drone+1
@2:48 M:56 G:0 L:2 S:13/18 BuildDrone
@2:53 M:52 G:0 L:1 S:14/18 BuildDrone
@3:00 Spawned: Larva+1
@3:00 M:67 G:0 L:1 S:15/18 BuildDrone
@3:05 Spawned: Overlord+1
@3:05 Spawned: Drone+1
@3:10 Spawned: Drone+1
@3:15 Spawned: Extractor+1
@3:15 Spawned: Larva+1
@3:15 M:164 G:0 L:1 S:16/26 BuildDrone
@3:15 M:114 G:0 L:0 S:17/26 MineGas
@3:15 M:114 G:0 L:0 S:17/26 MineGas
@3:17 Spawned: Drone+1
@3:17 Mining: +1 on gas
@3:17 Mining: +1 on gas
@3:19 M:152 G:4 L:0 S:17/26 MineGas
@3:21 Mining: +1 on gas
@3:28 Spawned: Spawning Pool+1
@3:30 Spawned: Hatchery+1
@3:30 Spawned: Larva+1
@3:30 M:254 G:24 L:2 S:17/28 BuildDrone
@3:30 M:204 G:24 L:1 S:18/28 BuildQueen
@3:30 M:54 G:24 L:1 S:20/28 BuildDrone
@3:32 Spawned: Drone+1
@3:45 Spawned: Larva+1
@3:45 M:150 G:53 L:2 S:21/28 BuildDrone
@3:45 M:100 G:53 L:1 S:22/28 BuildDrone
@3:47 Spawned: Drone+1
@3:47 Spawned: Drone+1
@3:55 M:158 G:72 L:0 S:23/28 BuildQueen
@4:00 Spawned: Larva+1
@4:02 Spawned: Drone+1
@4:02 Spawned: Drone+1
@4:04 M:111 G:89 L:2 S:25/28 BuildOverlord
@4:04 M:11 G:89 L:1 S:25/28 MineMineral
@4:06 Mining: +1 on mineral
@4:12 M:117 G:100 L:1 S:25/28 Metabolic Boost
@4:15 Spawned: Larva+1
@4:15 M:58 G:4 L:3 S:25/28 BuildDrone
@4:19 M:62 G:10 L:2 S:26/28 BuildDrone
@4:20 Spawned: Queen+1
@4:22 M:52 G:14 L:1 S:27/28 BuildDrone
@4:29 Spawned: Overlord+1
@4:30 Spawned: Larva+1
@4:30 M:110 G:25 L:2 S:28/36 BuildDrone
@4:30 M:60 G:25 L:1 S:29/36 BuildDrone
@4:30 M:10 G:25 L:0 S:30/36 MineMineral
@4:32 Spawned: Drone+1
@4:32 Mining: +1 on mineral
@4:36 Spawned: Drone+1
@4:39 Spawned: Drone+1
@4:45 Spawned: Queen+1
@4:45 Spawned: Larva+1
@4:45 M:241 G:36 L:2 S:30/36 BuildDrone
@4:45 M:191 G:36 L:1 S:31/36 MineGas
@4:45 M:191 G:36 L:1 S:31/36 BuildOverlord
@4:47 Spawned: Drone+1
@4:47 Spawned: Drone+1
@4:47 Mining: +1 on gas
@4:49 M:155 G:41 L:0 S:31/36 BuildQueen
@5:00 Spawned: Larva+1
@5:00 M:194 G:56 L:2 S:33/36 BuildDrone
@5:00 M:144 G:56 L:1 S:34/36 BuildDrone
@5:02 Spawned: Drone+1
@5:05 Spawned: Larva+4
@5:05 M:181 G:63 L:4 S:35/36 BuildOverlord
@5:05 M:81 G:63 L:3 S:35/36 BuildDrone
@5:10 Spawned: Overlord+1
@5:12 M:156 G:73 L:2 S:36/44 BuildQueen
@5:15 Spawned: Larva+1
@5:15 M:60 G:77 L:4 S:38/44 BuildDrone
@5:17 Spawned: Drone+1
@5:17 Spawned: Drone+1
@5:18 M:63 G:81 L:3 S:39/44 BuildDrone
@5:20 M:52 G:84 L:2 S:40/44 BuildDrone
@5:22 Spawned: Drone+1
@5:23 M:59 G:88 L:1 S:41/44 BuildZergling
@5:30 Spawned: Larva+4
@5:30 Spawned: Overlord+1
@5:30 Spawned: Larva+1
@5:30 M:148 G

8 L:6 S:42/52 BuildDrone
@5:30 M

8 G

8 L:5 S:43/52 BuildDrone
@5:31 M:68 G

9 L:4 S:44/52 BuildDrone
@5:32 Spawned: Drone+1
@5:33 M:57 G:102 L:3 S:45/52 BuildDrone
@5:35 Spawned: Drone+1
@5:36 M:68 G:106 L:2 S:46/52 BuildDrone
@5:37 Spawned: Drone+1
@5:38 M:61 G:109 L:1 S:47/52 BuildZergling
@5:39 Spawned: Queen+1
@5:45 Spawned: Larva+1
@5:45 M:163 G:119 L:2 S:48/52 BuildDrone
@5:45 M:113 G:119 L:1 S:49/52 BuildDrone
@5:47 Spawned: Zergling+2
@5:47 Spawned: Drone+1
@5:47 Spawned: Drone+1
@5:47 M:106 G:122 L:0 S:50/52 MineGas
@5:48 Spawned: Drone+1
@5:49 Mining: +1 on gas
@5:50 Spawned: Larva+4
@5:50 Spawned: Drone+1
@5:53 Spawned: Drone+1
@6:00 Spawned: Larva+1
@6:02 Evolved: Metabolic Boost
@6:02 Spawned: Queen+1
@6:02 Spawned: Zergling+2
@6:02 Spawned: Drone+1
@6:02 Spawned: Drone+1
Satisfied.
Number of actions in build order: 76
---Final Output---
At time: 6:02
Minerals: 448 Gas: 150 Supply: 50/52
Drones: 40
Overlords: 6
Queens: 4
Zerglings: 4
Hatcheries: 2
Gas Extractors: 1
Spawning Pools: 1
Metabolic Boost
------------------
Fri Oct 29 16:14:45 GMT 2010: 431818.23382857675

Valkola

Finland128 Posts

October 29 2010 16:40 GMT

#88

If u make spawning pool first then hatcery ( for example 14 pool 15 hatch) u wont have any money to use that spawning pool when it finishes so its basically the same zergling/queen production wise if u go hatch first or pool first but with pool first u have one hatchery more.

BeastofManju

United States79 Posts

October 29 2010 16:40 GMT

#89

1400+ D Zerg.

Everyone is forgetting a HUGE REASON. Only 1 person has mentioned so far... the hatch first will provide 2 Extra supply upon creation.

Markwerf

Netherlands3728 Posts

October 29 2010 16:50 GMT

#90

Why are people even bothered with what build reaches 50 pop the fastest?? It's completely irrelevant...
The real thing that matters is what gets more larva and what collects more minerals. 10 pool is horrible for your economy by delaying drones unneccesary long and just collecting less minerals.
Using the time to reach 50 pop as benchmark is completely retarded.

That said it's a known fact that 14 pool is just a little bit better economy wise then 14 hatch but the difference is nearly negligible. The reason most pro's 14 hatch is that it's just way harder to probe block that way and makes it quite easy to defend as well.

g5000

10 Posts

October 29 2010 17:08 GMT

#91

--- Nuked ---

SeeDLiNg

United States690 Posts

October 29 2010 17:21 GMT

#92

On October 30 2010 01:50 Markwerf wrote:
Why are people even bothered with what build reaches 50 pop the fastest?? It's completely irrelevant...
The real thing that matters is what gets more larva and what collects more minerals. 10 pool is horrible for your economy by delaying drones unneccesary long and just collecting less minerals.
Using the time to reach 50 pop as benchmark is completely retarded.

That said it's a known fact that 14 pool is just a little bit better economy wise then 14 hatch but the difference is nearly negligible. The reason most pro's 14 hatch is that it's just way harder to probe block that way and makes it quite easy to defend as well.

wait... so you're saying that who can mine 50 drones worth of minerals and overlords and get the larvae for it all is "completely retarded"

then follow it up by saying what gets more minerals is important?

can't say I follow what you're getting at.

on topic... I'm hella enticed by this. really want to start playing around with pool first builds. see how our terran and protoss counterparts respond.

LegendLength

19 Posts

October 29 2010 19:08 GMT

#93

On October 29 2010 21:09 Palmar wrote:
[...] it turns out that you lose almost nothing in terms of speed if you're going pure drones, if you 10 pool

[...]

10 SpawningPool M:204 G:0
10 Overlord M:100 G:0
13 Queen M:153 G:0
16 Hatchery M:305 G:0
17 Overlord M:108 G:0
19 Queen M:202 G:0
22 Overlord M:103 G:0
35 Overlord M:327 G:0
36 Overlord M:489 G:0
36 Queen M:605 G:0
39 Overlord M:405 G:0
40 Queen M:255 G:0
45 Zergling M:286 G:0
49 Zergling M:199 G:0
50 Overlord M:149 G:0

I tried testing a rough version of that build, doing 10 pool then hatch as soon as you have too much income. I didn't do it cleanly but it did seem to fall about 500 minerals short at 6:30 compared to the normal builds (16 hatch or 14-16 pool) for the same time reference.

I get the feeling the difference is a lot smaller than 500 minerals though and would like someone more competent to test it. Just to see what the trade off is for the earlier zergling potential (and I guess earlier speedling too?).

ChickenLips

2912 Posts

October 29 2010 19:19 GMT

#94

the problem with testing how to get 50 drones the fastest is that it is just that. 50 drones the fastest. the ZBOO doesn't care how much mineral surplus you wind up with (since droning off of 1 base takes way less than full saturation to do perfectly) and thus readily sacrifices early dronage for faster larvae (since that is what it needs to get the 50 drones ASAP).

With the current form of the ZBOO i think it's pretty good to find the optimal rush builds (f.e. 1 ultralisk rush off 1 base :D) and one shouldn't try to get the best economy with it since those two are generally exclusive with Zerg (you can either go crazy on drones or tech/units)

LWr

60 Posts

October 29 2010 19:42 GMT

#95

The reason to go hatch first is because of larva, but you guys are looking at it the wrong way.
It's not that going hatch first magically makes more larva.

By going hatch first, you have 2 hatcheries by the time the pool finishes. That means you can get TWO queens earlier, and you also have creep at your nat faster, allowing you to put down spine crawlers.

Queens don't use up any larva. 2 spine crawlers = 2 larva, but they do sooo much damage.

Defending early pushes with 2 queens and 2 spine crawlers means you used only 2 larva to defend, while if you used speedlings instead, you would need 5+ larva to do the same amount of damage. Therefore by going hatch first you are saving up at the very least 3 larva, which can be used for drones. And drones are your priority when you FE, because getting a better economy is the whole point of fast expanding!

LegendLength

19 Posts

October 29 2010 21:14 GMT

#96

On October 30 2010 04:42 LWr wrote:
Defending early pushes with 2 queens and 2 spine crawlers means you used only 2 larva to defend, while if you used speedlings instead, you would need 5+ larva to do the same amount of damage. Therefore by going hatch first you are saving up at the very least 3 larva, which can be used for drones. And drones are your priority when you FE, because getting a better economy is the whole point of fast expanding!

After a certain amount of time both builds have 2 queens. I'm not sure which gets both queens out earliest but the pool first obviously gets the first queen out earliest.

So it's still doesn't sway me to go hatch first except for the case where you want to defend very early with that first queen.

The original poster is saying both builds have the same end result once both hatches are running at capacity. But the pool first allows you to get a few zerglings out earlier if necessary.

Gimpb

293 Posts

October 29 2010 22:52 GMT

#97

I think like any expansion it primarily boils down to getting it started and getting it defended.

As for getting it planted, I guess the logic is you might as well do it super fast if you have a chance to. If they build something there or micro a worker around in circles, that costs them time and resources. If you wait around there's more time for them to attack or put something in the way.

On the defense side of things, it's very difficult to kill it with tiny amounts of early units and zerg's reactive production mechanics are a big deal with the army sizes are so small that a zerg can build a larger army than the attacker in less time than it will take to kill the hatch. If you wait a bit you end up with a situation where attackers can really threaten it when it's building and obviously it's easier to defend a completed hatch than one under construction.

Galleon.frigate

Canada721 Posts

October 29 2010 23:12 GMT

#98

This is one of perhaps the most valuable threads for zergs on tl right now and I'm noticing a lot of garbage in here

thank you to the op for bringing this up, and thank you for everyone who is actually running tests, assuming my sc2 was fuctional I'd be doing the same right now.

OP questioning the common assuption that hatch first is better eco, and he is hit with a bunch of people just blindly saying the those same assumptions.

emc

United States3088 Posts

October 29 2010 23:15 GMT

#99

In ZvZ i've found on most maps (steppes doesn't work) that the distances are simply too big for early aggression and the difference between 14 hatch and 14 pool 15 hatch is kind of big. The 14 hatcher can get his defense out before speed kicks in for the 14 pooler with two queens and a couple lings which is more than enough to stop early ling aggression. I've been finding it extremely risky to expand around 21 supply because during hatch building, this is generally when roaches come for a timing attack. I find that an earlier hatch (as long as you aren't getting 6 or 8 pooled) works to your advantage because you can secure a defensive position faster than you normally would. The earlier the better, because you can use all the larve for drones and an army when facing against a 1 basing Zerg. I also like the quick expand because you can use both your queens to spread creep rather than inject larve because your larve count is still high with two hatches. If you don't expand quickly, then your expansion will have to be delayed until you defend their inevitable timing attack.

ZvT and ZvP are situational depending on the map and location. Close posistions on Metalopolis or LT can be scary and I'll generally do 14 extractor 14 pool into safe speedling/baneling or speedling/roaches against Protoss. On 2 player maps like blistering sands, scrap station I'll FE always. On far distance maps, if you go for a speedling build into a later expansion, you'll end up behind and most of the time your harass attacks will get defended quite easily, so it's really a waste of money. What else would you be doing with a speedling build? It's a complete waste of money against competent T or P players, it's almost always better to get less gas for more drones, and spines for defense rather than more lings.

jdseemoreglass

United States3773 Posts

October 30 2010 00:13 GMT

#100

I decided to test this just because the notion that hatch first wasn't more economical sounded absurd to me. Note: I ONLY tested the builds trying to maximize economy (building two queens as soon as possible, spending all energy on larva, all larva on drones and overs, maynarding etc.) so this says nothing about which is better strategy, safer, or creep friendly.
I performed both builds as efficiently as possible multiple times and took the average of each after pausing each game at exactly 6:20. I picked this time because I noted in each build it was the moment with the least units in production, right after drone pops. Here are the results:

14 pool 15 hatch:
Minerals: 590
Drones: 44
Overlords: 7

14 hatch 14 pool:
Minerals: 735
Drones: 48
Overlords: 7

Edit: Added previous posters build
10 pool 16 hatch:
Minerals: 640
Drones: 44
Overlords: 6

I know everyone likes to get caught up on technicalities, but my tests seem to confirm the assumption, that hatch first is more economical. The 10pool build actually did surprisingly well, perhaps we should focus on this fact more. If anyone wants to debate this or suggest a better method of testing, please do.

Prev 1 2 3 4 5 6 Next All

Please or register to reply.

[D] Why would you ever hatch first? - Page 5

Completed

Ongoing

Upcoming