Loading...
  OR  Zero-K Name:    Password:   

Starting Build Sequence

33 posts, 4478 views
Post comment
Filter:    Player:  
Page of 2 (33 records)
sort
12 years ago
So I was going to make a tutorial, but I realised that I'm not 100% on what a good way to start a game is, and I'd like to ask you guys how you prefer to do it. Assumptions: You have the default Strike Commander, three mex spots and you're going Cloaky Bots. And it's 1v1.

Personally, I would build:
1- Cloaky Bot Factory (queue up 5 Glaives, 1 Rector on repeat)
2- Mex
3- 2-3 windgens
4- Mex
5- Radar
6- Lotus
7- 2-3 windgens
8- Mex
+0 / -0
Skasi
12 years ago
Really depends on the map. I'd never build windgens on CCR or Red Comet. Anyway, giving it a try:

1) Areamex 3 metal spots
2) SPACE-plop Cloakfac
2a) Queue 3 Glaives, Recon, Warrior, Zeus
3) Queue 4 solargens
4) 1-2 LLTs/MTs
5) Move to the closest hill (not backwards!), plop radar
+0 / -0
1- mex
2- factory
3- 1 con, then raiders on repeat
4- 2x mex
5- 1-2x solars
6- 3-5x winds
7- radar

Con sometimes does the last 2.

If map is small, your start is spread out and you don't have a good anti-raider raider, make up to 1 llt + 1 defender before expanding. If map is large, your start is close together and you have a good anti-raider such as scorcher or bandit, defend your base with radar + raiders.

Comm expands forward, towards the enemy, relying on his own weapon to defend the mexes behind him until metal is back into the positive.

Con expands sideways, defended by raiders, trying to stay in defended territory or making LLT's as he goes if he can't. Sometimes con loops back around to assist factory/make energy depending on my income and whether there is any more expansion to take.

Once I'm in positive metal income, I make a solar next to each extractor and ensure all routes have at least 1 LLT on them, that the base has at least 2 turrets (LLT/MT) in it if i didn't start with them.

There are a lot of good strategies that skip the early con or don't comm expand but i favor getting my income up quickly so I can sustain the -20 BP you start with.
+0 / -0
12 years ago
i am not sure if it is useful to write the tutorial for a baseline comm since in most 1v1 games you'll have an energy cell.

Apart from that, like Skasi said, it really depends on the map. Things to take into account are
-the rush distance (determines radar timing) you can even delay the radar for a long time on some maps!
-altitude of your start location or the presence of hills
-early intel from your first scout, a dart can be scout the other base very quickly on say red comet
-your strategy! You can decide to play all economy for the first 4 minutes and not to raid. This gives a completely different buildorder then a strategy where you stay on 2-3 mexes for a long time.

It is better to state some general rules of how to start a 1v1 game.
-plop lab
-grab your 3 starting mexes asap
-make a radar
-do not stall on energy, add E-structures as you take the mexes one by one.
-make sure you starting area is covered with an LLT/MT and/or your commander or your own units

Emblis already did some sort of tutorial on 1v1 and how to start. It was pretty good iirc
http://emblis.blog.com/2012/02/22/how-to-survive-the-early-game/

+0 / -0
12 years ago
Skasi,
"5) Move to the closest hill (not backwards!), plop radar"

You want to put radar specifically on a hill. Does it work better in high places?
+0 / -0

12 years ago
Yes. Radar is blocked by terrain, so putting it on hills always gives better coverage.
+0 / -0
12 years ago
radar line of sight works the same as "real" LOS It get blocked by mountains and can see more behind one if it is on a higher position
+0 / -0
Skasi
12 years ago
Oh, I forgot one thing: Try to place the radar so you cover map borders. On Red Comet, for example, it's possible to cover the whole front (north to south) with a single radar. In 1v1s I always try to place my 2nd radar where it does that.
+0 / -0
12 years ago
Ooh, thanks for this info, gentlemen!

I always lived under the impression the radar has a fixed sensor radius and that's that.

I've read all the guides and unit information and cannot confirm it was mentioned there. This information is not readily available to everyone. Maybe the radar mechanics should be explained briefly in the unit description?
+0 / -0

12 years ago
>1) Areamex 3 metal spots
>2) SPACE-plop Cloakfac

What does space do?
+0 / -0
Skasi
It puts the command at the start of the queue. Usually it takes some time before I can decide on a factory, so I want my commander to build mexes meanwhile.

Shift+space usually puts the command between the two closest (so you could areamex a couple of mexes, then shift+space windgens or solars between and your commander would build stuff with as little walking as possible). But that doesn't work with areamex since the latest mex-placement change. :(

Done, Sotha. It's not long, but clear. :)
+0 / -0

12 years ago
Eh if you just turn on radar + LoS mode it shows this.
+0 / -0
I'm not a good player so this may not be good advice, but when I was starting out I put some thought and experimentation into an initial build order. My goal was to maximize early econ production under the very circumstances you describe (which is what a new player would likely be doing). Here's what I came up with.

Pregame: Start immediately adjacent to a mex spot in a cluster of three. Time spent walking is time you're not making metal or units.

Com: Facplop Cloaky Bots
Com: Area mex the three mex spots

While the Com is walking and building, get your factory making units:

Fac: Turn on repeat and turn off auto-factory-assist (should set these by default using "Initial States" in the Game Menu)
Fac: Set a waypoint a short ways ahead of the fac
Fac: Alt-queue a glaive
Fac: Alt-queue three rectors
Fac: Alt-queue four more glaives
Fac: Queue your early-game unit mix. Example: two glaives, zeus, two glaives, warrior, two glaives, zeus, two glaives, rector.

Alt-queue says "add this unit to the queue after the currently building unit, and don't add it to the repeat loop". This sequence will produce five glaives, followed by three rectors, followed by your mix on repeat.

Once you've got your factory busy, add to your Commander's to-do list:

Com: Space-queue two solars next to each mex
Com: Shift-queue assisting the factory

Space-queue inserts the solars into the build queue, so that the Com will build them in the right place and time as he's working on the mexes already in his queue. [ [i]Second Edit: See below for clarification.[/i] ] Shift-queue adds the factory assist order to the end of the queue, so that once he's done building his three mexes and six solars he'll contribute his buildpower to the factory.

If you work fast, you can get all these build orders queued up while the glaives are coming out.

Spread the glaives out in a line somewhat ahead of your base to get some forward visibility and defense, but not too far - they aren't scouting, they're protecting.

As the rectors come out, give them build orders:

Rector #1: Cap mexes. Queue up a long series of nothing but mexes, starting with the closest ones and working in a single general direction outward (i.e. don't have him going back and forth).

Rector #2: Follow behind #1 and add two solars per mex.

Rector #3: Go directly to the best close spot for a radar (a nearby hill if there is one, or the edge of a clear space if there isn't one) and put up a radar and a defensive tower. Then start following in the general direction of Rectors #1 and #2 and add defensive towers and radars wherever it seems they'd be useful.

Give each Rector a long series of build orders all at once, so that they stay busy for a long time without further intervention.

As the Rectors set off on their tasks, use the Glaives to protect them. Don't just set them to guard - use them to spot and clear the territory ahead of where they're going, and keep them close enough to cover the builders but far enough ahead that you can intercept approaching threats.




With this build order I'm trying to do the following:

a) Minimize idle buildpower
b) Minimize m-stalling
c) Minimize e-stalling
d) Maximize output of useful units
e) Protect against early scouting and rushes
f) Maximize territorial control and economic expansion (i.e. GET MOAR MEXES!!)
g) Minimize the micromanagement - use automation and long queues of identical actions to reduce the amount of decision-making and mouse-clicking you have to do

What I'm not doing here is scouting. But it would be easy to take one or two of the first Glaives and send them off towards the enemy instead of guarding the base. Maybe add another initial Glaive to compensate so that you've got one or two scouting and four or five defending.

Also, if you can keep up with the micro, you could add another two Rectors to the initial set and send them off to build mexes and solars (one just doing mexes, the other following behind just doing solars) in the opposite direction from the first pair. The problem with that is now you have two fronts to protect, and not enough glaives to cover both fronts.

The nice thing about this build order is that it generates a ton of productive activity in your base while giving you just a few things to concentrate on - in the early stages, mostly you'll be driving your Glaives around exploring the territory near your base and protecting your builders, while your builders just go on doing their thing without your attention.
+0 / -0
Skasi
12 years ago
I'd probably never get more than 1700 elo that way, Eddie. :)
Don't you stall metal like mad? Three rectors right at start.. seriously?

Give your factory a fight command. That way most units wont suicide when moving to front. Builders will assist. (I don't know why there's a "make builders assist" button, I'd never set a factory to move and it doesn't even support the fight command)

Long build orders are a waste of apm, usually. Builders will need to flee, assist, repair, reclaim or anything else long before they even get half way through their command list. Only exception I ever experience is a windgenfarm builder.
+0 / -0
>Don't you stall metal like mad? Three rectors right at start.. seriously?

No, because their buildpower is idle a lot due to walking time. So they don't drain your metal like they would if they were just standing around assisting your factory. By the time the third one comes out, you've got enough income to support the rate at which they can build more income.

>Long build orders are a waste of apm, usually.

At that point in my build order there's nothing else to spend APM on anyway. And while they might not make it all they way, they could, and will be more likely to if you keep them protected. But yeah, the first casualty of battle is the plan and all that...
+0 / -0
Skasi,
"Done, Sotha. It's not long, but clear. :)"

Thank you kind sir! That was a nice thing to do, many newbies might have had no clue like me. I never have time to play with the LOS buttons during the game: everything is so hectic and often losing focus for few seconds means all the front-line troops died.

The fight order for a factory is also cool idea, I never figured out that one. I've just chosen the factory and right click-dragged a line where I want my new units. For some reason this works poorly, they do not obey the order to get into the drawn line. Instead the units go into a single clump and often never reach the destination. When they don't get to the rally point, they do not get assigned the hotkey I've given to the unit types and I'm fighting in the front without realising my units are stupidly stuck somewhere near the factory.

Any suggestion how to fix that, or will the fight order help? Gotta test that...

One more question: why don't the suicide bots have damage data in the unit listings? For example tick doesn't tell how much emp damage it does, so a newbie like me can never know how exactly many ticks they need to paralyze a goliath and so forth.

Sorry, looks like this thread is OTing into "Sotha's stupid questions."

To the main topic:
I usually start with
1) cloaky bot (instantly cue: rector, 5 x glaives 5 x rockos. Sometimes I feel inclined to make a few scythes and terrorize the enemy economy. That's fun. Even got a few factories down thus far.)
2) build mexes and one solar panel near each one of them so that everything goes into the network.
3) basic defense, one radar near the factory, one lotus per mex/solar panel cluster
4) once the rector is ready, move it forward and start capping frontline mexes using glaives and rockos as support.

But don't listen to me, cause I still suck in this game.
+0 / -0
12 years ago
i tried starts with 2 cons, and even then i stall metal like crazy.
3 rector

3 * 5

15
comm 10
factory 10

thats a whopping 35 buildpower! on 3 mexes you are hardly getting 10
there has to be a lot of walking to do on the map you have in mind

i'd stick with one maybe two constructors at start
+0 / -0


12 years ago
All buildpower is not the same. Facs can only make units; cons can make income.

Yes, with that much buildpower I'll stall. That's okay. I want as much metal as possible to be spent on making income; whatever's left over can go towards units. So I set my Rectors' build priority to high, and make three Rectors. What's left over is more than enough to make enough units to protect against early attacks.

"Okay, fine," you say, "but the Commander is a con too. Why not use one Rector and the Com for capping mexes instead of three Rectors?" Good question! The answer is that I want to keep the Com in my base instead of sending him into the field, for two reasons. First, he's a good base defense. With him in the base and the glaives as forward protection, I don't need to spend time/metal/BP on an LLT/MT. Second, it doesn't take long before I've got enough income to exceed the factory's BP, and by then I really do want to be making units. So the Com stays behind to start assisting the fac.

If your metal and BP were split neatly between making more income and making more units, the rate at which you could grow your income would be limited by a) your current income and b) the portion of your income that you spent on making more income. But it's not a neat split. It's a lumpy split, and that makes a difference.

With your con priority set to high, the rate at which you can increase your income is limited by your cons' travel rate, not your current income or their buildpower. Sending cons in pairs - one for mexes, the second for energy - maximizes the rate at which you can add income. Splitting into multiple pairs - one pair in each expansion direction - would increase your income even faster by working in parallel, but as I mentioned that means having a larger front to protect.
+0 / -0
>I've just chosen the factory and right click-dragged a line where I want my new units. For some reason this works poorly

Right-click-drag works differently for a single unit than it does for a group of units. A group will move to the line in formation and spread out across the line. A single unit will travel along the line and stop at the end of the line.

Selecting the factory counts as selecting a single unit. So the factory is giving each of its newly-built units an order to travel to the start of your line, then travel along the line, then stop.

Not at all what you wanted, I'm sure.

>When they don't get to the rally point, they do not get assigned the hotkey I've given to the unit types

There's an option to assign your units to an autogroup as soon as they leave the factory instead of when they reach the rally point. I highly recommend it, and I'm not sure why it's not the default.
+0 / -0
12 years ago
I can get behind a many-con start as I've tried that before. It requires a nice open map with lots of metal (at least 8 defensible spots on the way to enemy) and not too many teammates.

Most team pub games start at 6v6 and go upwards. I find it better to build early game trap-style units to bait enemies into feeding me metal.

"standard" start for me:
Cloaky fac -> mex -> solar (skip if ecell comm) -> mex -> solar (skip if etc.) -> mex -> solar -> radar -> walk to nearest mex and cap (repeat until seeing enemy)

Fac goes on repeat 2x glaive, rector, glaive. After 3rd glaive, alt-add 1 tick to queue and add one to back of repeat cycle.
1st rector builds an llt then caps if available, runs to front and plonks an llt if I met enemies already, or reclaims if available.
2nd rector builds an mt/llt and solars as needed, reclaim nearby trees before building solars though. Build wind:solar ratio of 3:1 if height is favourable.
Additional rectors expand sideways if can, forward to repair/reclaim if expansion directions are assigned already.

After that, depending on what I face, I'll alt-queue a zeus, warrior or rockos as I need. After first caretaker, snipers if I'm not facing spiders, hammers if I face defences, get a few zeus as well.
If opponent is tanks, I'll drop ticks and start conventional armies.

If I start to pile too many glaives and ticks (aka. enemy not attacking), I mine the ground in front of my forward def, leave a small garrison of glaives and take the excess to help my ally. Hopefully, the "weakening" of my defences will bait enemies into attacking, which with tick mines = juicy reclaim.



Obviously, this is a "typical" start. Which has happened about <10% of the time.
But when I go cloakies, it's usually tick-based.
+0 / -0
Page of 2 (33 records)