Extractor 'trick'

[Gifted]

Zabi has it right, this is one of the stables of a few builds.

[Hav0x]

I wouldn't say its "intended" but it isn't condemned by blizzard.

[Hammy]

In BW you lost 13 mins by using an extractor trick, not including the mining time lost.
Basically it was only used in 9pool builds because the timing was just perfect: when your pool was done, your OL poped up, and you could make your 6 lings.

In SC2, you'll only lose about 6mins, which isn't too big a deal, and with improved pathing you would waste even less mining time. I can see this as a decent addition to certain builds.

You could potentially double-extractor if we see any very greedy builds (builds that involve making anything before the first OL), though I haven't seen any yet, since such a raw rush would fail to worker pathing.

[RamiZ]

You could potentially double-extractor if we see any very greedy builds (builds that involve making anything before the first OL), though I haven't seen any yet, since such a raw rush would fail to worker pathing.

This. There are 2 extractors, so you can do it with double the trick and get 2 more Drones or w/e for 10 minerals loss... I think it is stronger than it was in SC1 cause of this but it still isn't game breaking...

[Alex06]

The only way to fix it is to make constructing buildings take up a point of supply.

Which would be SO ridiculous.

No, actually, they could simply make it so that the supply for the drone is only removed once the building is actually finished.

And you don't have to build an extractor. You can also build, say, a hatchery with it or something else. Which means you aren't limited to doing this with 2 drones max at the beginning of the game.

In the original Starcraft it was actually a net mineral LOSS because of the way building cancellation was refunded. Go look at any current pro reps and you won't see zergs doing any extractor tricks.

Now that extractors are only 25 minerals, it might be a gain, but then again overlords build insanely fast anyway. If there is a benefit from this mechanic it is very very slight and nothing worth complaining about.

They cost 25 gas now? O_O

[Santrega]

No, actually, they could simply make it so that the supply for the drone is only removed once the building is actually finished.

Alex, I believe thats exactly what they were saying... It's ridiculous to have a unit thats morphing into a building to take up supply.

What they should do is give a bigger mineral penalty for extractors. So if you start one and cancel it, it will cost 50% of it's cost. I don't remember if the extractor is 75 or 100, but it would effectively cost 70-100% of a drone's cost. Thats a much more difficult penalty to bear for this, effectively solving any problem blizzard might see.

I personally don't think this is a problem, but if they need to fix it, the penalty of canceling a building is where it should be fixed.

[RamiZ]

They cost 25 gas now? O_O

You mean Minerals, yes, since there are less gas in it, and you also need more Workers for it.

[Aldrius]

No, actually, they could simply make it so that the supply for the drone is only removed once the building is actually finished.

That's exactly what I just said... it'd be ridiculous.

Protoss and Terran start with 11 supply anyway. Zerg start with 10. (That's no different than SC1, though, is it? Didn't Zerg have 9 and P/T have 10?) So the extractor trick is just evening the odds as it is. :P

[GnaReffotsirk]

I've seen this happen just now.

on 10th you build pool
spawn drone
extractor trick 11th drone,
extractor trick 12th drone,
cancel both extractors,
spawn overlord
queen
lings
... so on.

it doesn't factor in. It just delays your overlord. However, I have yet to see the difference personally.

With chrono boost, protoss can 15 gate while this is happening.

terrans no idea.

Edit:

Tbh, if someone would lend me an account, I will document starting builds, and try everything possible. I'd record them and upload on youtube, also providing replay packs for review. All this vs. the AI, so no worries with me messing with anything else.

I want to see what comes up for everything, in terms of resources, potentials, etc. These tests will be at 5 min, 7 min, and up to 10 minutes lengths.

The idea is to exhaust what we can do and how that would set one up towards the next minute or two.

If not, I hope beta participants would do the above.