Building a new hatchery requires a spawning pool

[GoogleCodeExporter]

What steps will reproduce the problem?
1.  Regardless of setting waypoints or a final build, the algorithm does not 
allow for a Hatchery then Spawning Pool build.  If waypoints are used, it never 
hits them.  If a final version is used, then the spawning pool is always built 
first.
2.
3.

What is the expected output? What do you see instead?

What version of the product are you using? On what operating system?
Latest one downloadable.

Please provide any additional information below.

Original issue reported on code.google.com by rpanu...@gmail.com on 3 Nov 2010 at 2:20

[GoogleCodeExporter]

I'm pretty sure I've seen a hatch first build. I've noticed that it almost 
always goes pool first. I think pool first may just be better in most cases. If 
you want quick lair tech units or upgrades, you need a quick pool to get a 
quick lair. 1 hatch + 1 queen is more larva production than 2 hatch no queen. 
And finally, I think the only reasons to get hatch first are reasons the 
program doesn't consider, such as 2 fast queens for creep tumor or so the creep 
from the 2nd hatch spreads in time to place early crawlers.

Original comment by Sketh...@hotmail.com on 4 Nov 2010 at 6:31

[GoogleCodeExporter]

But that isn't accurate, larva are produced at 1 every 10s, a queen spawn is 4 
every 45s.  That means larva is produced faster at 2 hatch.  Also, if you go 
hatch then pool you have the capability of getting 2 queens simultaneously as 
soon as your pool is done if you time it right.  So that would increase your 
production capacity well beyond 1 queen 1 hatch.  14 / 15 Hatch, Pool, 1-2 D up 
to 16, either get OL's or wait for second hatch thats 4 food for 2 queens + 
minerals or train OL's for OL's + queens + more drone production.

Original comment by rpanu...@gmail.com on 4 Nov 2010 at 12:26

[GoogleCodeExporter]

This issue appears to be related to way points failing to clear.  This is 
likely to occur if a previous build specified zerglins and/or spawning pools in 
a way point.  The effect can sometimes be negated by changing the prior, 
now-unwanted objectives to higher numbers than were previously used, running 
the optimizer for a few seconds, changing the unwanted objectives back to zero, 
and using the desired objective (2 bases).

Original comment by qwerty10...@gmail.com on 5 Nov 2010 at 3:35

[GoogleCodeExporter]

Your numbers are wrong. A hatch produces 1 larva every 15 seconds and a queen 
spawns 4 every 40 seconds. 2 hatch generates .1333 larva per second, while 1 
hatch 1 queen generates .1666 larva per second (assuming that you are 
constantly using the larva).

The only reason to get two queens that early is for creep tumors, which the 
program doesn't consider. 2 hatch 2 queen that early wouldn't help drone 
production, it would actually hurt it. You don't have the economy to use all of 
the larva produced by 2 hatch 2 queen and therefore the 150 minerals spent on 
the 2nd queen would be better spent on 3 drones.

Original comment by Sketh...@hotmail.com on 5 Nov 2010 at 9:48

[GoogleCodeExporter]

Spawn larva takes 40 seconds to finish, and then pops one larva per half second 
or so, and takes 45 seconds to generate the energy for the next spit.

I think a common misconception is that two bases immediately gives you more 
economy. It only gives you more economy if you have more than 16 drones on 
minerals.

uh... So. Is there a consensus on this yet?

Original comment by Frit...@gmail.com on 6 Nov 2010 at 1:36

• Changed state: Discuss

[GoogleCodeExporter]

Just to add my comment to the mix: I believe it doesn't build hatch then 
spawning pool as there is no economical incentive to do so. It will rarely go 2 
hatch at all for anything unless it is a very larvae heavy build (like mass 
ling) or a long term build where it will saturate the 1st base.

Original comment by kel.bolden@gmail.com on 10 Nov 2010 at 2:55

[GoogleCodeExporter]

I've watched games before where players will build a second hatchery before 
getting a spawning pool, but it's rare.

Original comment by mike.angstadt on 10 Nov 2010 at 3:28

[GoogleCodeExporter]

While it may be natural for a player to go Spawining pool before Hatchery it 
may not be the fastest build for their requirements. For example if a player 
wished to see the fastest route to 40 drones and the algorithm determines 
queens are not worth the cost, the spawning pool does not only seem out of 
place but will slow down the build, this would probably be appropriate as 
another option in the settings page.

Original comment by nafets...@gmail.com on 16 Nov 2010 at 4:44

[GoogleCodeExporter]

I think that it counts the Hatchery that you start with as one base.  So, if 
you enter "1" in the "Bases" box, it will consider that waypoint met because 
you always start with one base.  If you enter "2", then it correctly builds a 
second Hatchery before a Spawning Pool.

Original comment by mike.angstadt on 19 Nov 2010 at 9:43

[GoogleCodeExporter]

But if you only specify 40 drones it seems to build a spawning pool regardless. 
Or at least it did when I tried it. I'll retest sometime this weekend.

Original comment by nafets...@gmail.com on 19 Nov 2010 at 9:52

[GoogleCodeExporter]

Yeah that's happening for me too.  I guess it determined that it's faster to 
create the necessary tech to build Queens to get the extra larvae?  It's giving 
me a time of 5:34 with 40 Drones, 5 Overlords, 2 Queens, and 1 Spawning Pool

Original comment by mike.angstadt on 19 Nov 2010 at 10:03

[GoogleCodeExporter]

Greater than around 17 drones, you need queens or more hatches to keep up with 
the larva count.

Original comment by Frit...@gmail.com on 21 Nov 2010 at 9:41

• Changed state: Done
• Added labels: Milestone-0021