|"MAN! I leave for a minute or two to comment on Cashew's story and there are FOUR comments already! I love this site. Did you guys actually read the thing?"|
|FFA tactic: hyper expand mass BCs|
|Date: ||01/06/00 12:01|
|Game Type: ||Starcraft|
|Report Rating: , # of Ratings: 5, Max: 10, Min: 9|
Lifetime Rating for shockwave: 8.9231
A FFA on Lost Temple,|
early expand tactics)
Map: Lost Temple
Races: T, P, Z
Starts: T:3, P:9, Z:12
Some of my most enjoyable games
are on FFA's, but most are just too involved to BR sensibly. I'm
reporting on a game I played last night because I had an opportunity
to try various interesting strats. There are no screenshots, I don't
expect this to be a stellar battlereport, but I'm writing to share
some new strategies that I think may give a beneficial read.
Last night, I was moping around
for a quick 2on2 and ended up in a pub game on Lost Temple. My
opponents had fairly good records, and I got an incredibly lousy
partner (7-14-0 or something similar), so I figured this would be a
challenging game. Unfortunately, my partner left the game
immediately after it started, a common occurence and I assume
because he randomed and didn't get his race of choice, so one of my
opponents suggested a FFA and we decided to duke it out. Normally I
assume that when one person suggests FFA immediately and the other
agrees wholeheartedly immediately, this means I'm in for a 2on1
(although that was not the case this time), but I figured that would
be even more challenging.
Unfortunately, I had randomed to Terran. I was at 3, randomed
ZERG was at 12, randomed TOSS was at 9, my partner who left the game was
TERRAN at 6.
I always choose Protoss on FFA's. This is
personal preference, but my inclination with Protoss is to
just ride it out into late game and then to use my BGH-like
accumulated resources to create a "Finger Of God" type
offensive. For those of you unaware, FOG simply refers to the
almost unstoppable corsair+dweb/carrier/arbiter/shuttle+temps
combo. Toss also of course have the ability to exceed supply
using dark archons, and supply is always a factor in late game
FFAs. Finally, Protoss tend to have the most cost effective
attacks and defenses, which is important in FFAs.
The Master Plan I adopted in this Terran FFA
was quite simple: Expand immediately, stay alive long enough to mass
BCs, clean house. There is a twist to it, though...
In a previous 2on2 game, I had witnessed a truly bizarre strategy
on LT by Whoop/Ronin's team where Ronin built 2 early CC's to float
to the island expansions. In fact, I have witnessed even stranger
tactics in team games (usually larger ones such as 3on3's) where the
Terran player will build up to 10 SCVs, execute an SCV rush on an
opponent's base (usually picking on Zerg), and meanwhile float his
CC to a nearby island and start his econ over. So all credits go to
Ronin for his strat, which I'm copying here with some mods.
Obviously, islands afford the best defensees. In
FFAs or even team games, especially on maps like Lost Temple,
games are often decided by which team is able to control
islands effectively. Islands on LT in particular have an
insane amount of resources. If you can secure an island
expansion without your opponent's knowledge and use that as
your primary resource intake despite even conceding your more
lightly defended main, you can still win the game.
My build went 8th->depot,
10th->rax, then bunker, depot, CC, CC, gas. All structures were
all placed at my choke in a classic Terran wallin, and I added a
bunker behind for extra security. Here is a pic stolen from
Heartcutter's report at http://www.battlereports.com/viewreports.php?reportnum=1677
that illustrates the wall-in that you can use at Temple 12 or 6:
In case anyone is still unaware, the front-gate
Terran wallin prevents all graphically medium sized units such
as zealots, hydras, SCVs, lurkers, DTs, etc. from being able
to get past your choke without first destroying one of the
structures comprising the wallin. This is particularly
effective against zealot rushes, and you can e.g. hold off an
indefinite number of computer zealots because they will just
run back and forth trying to get your marines behind the
wallin without realizing that they need to destroy the depot
first to get to them.
There are a couple of drawbacks to
front-gating. One is that if you are early ling rushed, you
will be at a disadvantage because your marines will be at your
gate, whereas your SCVs will be back at your mineral line
where they cannot immediately help defend. The other is that
it leaves fewer buldings for your wagon wheel at your base and
splits your defenses.
Since this was a FFA and I didn't fear any early ling attacks, I
knew that my wallin would be a sufficiently safe idea. Meanwhile, I
sent two SCVs to distant spots on the map: one to the northwest side
outside of 12's base, another to a spot between my base and 6, and
started CC construction at both places. These were to immediately
float to island expansions.
It's important that you construct your CCs at
spots that are close to the islands so that there is minimal
float time. However, you should also construct them out of the
way so that casual forces don't venture upon your CC. It can
be disastrous, and there will be no way to recover, if some
unit chances upon your SCV and destroys it before it can
complete the CC. If you are close enough to the expansion, it
may be worth it to construct your CC in base or nearby and
then float it a longer distance. On some maps such as
Aftershock, island expansions are fairly close to most starts
so you can just make your CC at your base.
No matter what
you do, though, it is almost always best to make a CC at a
safe location if you can help it and then float it to your
expansion. This is particularly useful on a map like Lost
Temple when you are expanding to your natural: Make your CC on
your plateau, then float it down. This way, your CC cannot be
destroyed during construction, and your opponents will also
not be able to recon to see that you're building a CC there
until it's completely done.
Up to this point, there was really no action, as expected. Nobody
was attacking, and I assumed that my opponents would be pursuing
tech builds and expanding to their naturals soon. It's rare to see a
rush BO in a FFA or early low tech units, although I'd certainly
keep this an option if I can execute small cost-effective attacks.
For example, Mark4's one-marine-one-SCV early game rush can punish a
Zerg player going late pool with minimal loss to yourself.
Should I attack early in a
Attacking early can be effective in taking out one
opponent, but I find that they generally leave you at a
disadvantage as a result. Obviously, most people will not
4-pool, since you may kill one opponent, but it leaves you in
a 1on1 where your remaining opponent has a far better resource
advantage than you. Furthermore, early game attacks that pit
unit against unit are rarely extremely cost effective even if
you win the battle, and thus can set you both at a
disadvantage in relation to your other opponents.
philosophy in FFAs is "maintain the status quo, unless you
are the one violating it". Always try to make sure there
is a balance of power and that one player is not getting too
strong compared to the rest. Don't focus on attacking one
player too much, and don't waste your units. E.g. If A is
getting too powerful, I tell B and we both concentrate on A
and avoid conflict with each other. I focus on extremely cost
effective attacks such that I can weaken my opponents without
hurting myself. After all, if A attacks B, both are weakened
and C ends up being more powerful in comparison.
Although I have no guarantees
that my opponents will adopt the same FFA philosophy -- e.g. I have
been victim of very counterintuitive strats where in a 3-way FFA,
one person wastes a lot of money offensive cannoning me aggressively
such that I die but he is severely handicapped in the resulting 1on1
-- I bank that they are at least semi competent and will not throw
forces away at my light defenses.
I guess wrong. ZERG, who is at 12 next
to my 3, scouts me with an overlord and attacks with a modest force
of lings and hydras. I have a filled bunker and a tank, but siege
mode is still researching out of my single fact. Everything at my
choke is destroyed and I am left to fallback to my main and start
rallying my tanks there. I have no marines since I have gone 1 rax,
but I have ample SCVs. Taking marginal losses, I group some SCVs
around one of my tanks that has just finished constructing, tell the
rest to attack, and try to delay until my other tank finishes
building. I end up taking marginal losses and ZERG eventually loses his whole attack force
spitting at my tanks being repaired by 5 SCVs each. This was my only
setback: if siege can finished in time, I could have held my main
without losses and gotten my economy rolling much much earlier.
When setting up siege defenses, some people like
to stick their tanks flush against their ledge so that the
sieges have maximum coverage. This can often be a bad idea
unless you're playing against another Terran who is using
tanks against you. The closer your tank is to the ledge, the
more exposed it is to counter attack from below the ledge. If
you turbo newbie, for example, it is often a good idea to
recess your tank a bit from the ledge so that it is completely
safe from counter attacks from dragoons or hydras. Your tanks
still have tremendous range and should be able to cover your
choke or your turbo newbie sufficiently: what you sacrifice in
range, you get back in added safety.
At this point, I comsat TOSS and see
what I dread most: 2 starports and 2 cyber cores. I notify ZERG of this, complement him on his attack, and
announce to him that I'm crippled and that he should think about
defending himself against TOSS if he
doesn't want to get overrun as a result.
I find communication to be imperative in FFAs.
Some may disagree, but I have no qualms about announcing to
the whole world that my recon shows one opponent going DTs
early, or that another has secured a strategic expansion, etc.
I don't adhere to any rules in FFAs other than forbidding
permanent alliances, and I will say anything as long as it
ultimately benefits me. After all, "There can only be one,"
and everyone in a FFA should understand that temporary cease
fires or agreements are still just a result of each person
selfishly trying to win. Note though that some people do not
believe in talking during FFAs. If you're unsure, communicate
rules beforehand so that someone doesn't get mad that you
scouted them early and saw they were going DTs and then leave
the game, as has happened to me before.
ZERG leaves me alone for a bit, I have no idea
whether he and TOSS are undergoing any
skirmishes although frankly I'd be surprised given the heavy losses
that ZERG has just undertaken. My two
expansions are up and operational, and it is of course imperative
that I keep building SCVs continually out of both of them. Because I
went 1 rax, 1 fact at home, I have tons of minerals at this point,
so I make several turrets at each island, enough to make any terran
Resist the temptation to just start mining with
your first SCVs at your early islands. It is CRITICAL that you
maintain control of each island; if you lose either of them,
you will not have forces anytime soon to reclaim them. And
since you do have spare minerals and most Terrans are always
gas-limited, you should be beefing up your defenses such that
you completely discourage any direct attacks on your
I build enough turrets and a couple of bunkers
such that not even a speed upgraded shuttle will be able to
get through, or even a mass hydra drop is guaranteed to suffer
tremendous casualties in taking either expansion. They'll need
late game island busters to take either island back
It is particularly imperative that you
defend your islands strongly and early because any good FFA
player will try to claim island expansions quickly on a land
map due to their strategic value. Your advantage is that you
get there first and deny them.
Confident that my islands are secure, I beef up more defenses at
my main, slap down a starport, sci facility, and 2 armories, and
start my mass BCs master plan. This again is credited to Ronin, he
also went mass BCs in his 2on2 although he eventually lost, but only
primarily because his partner got overrun and Ronin forgot about his
island expansions and stopped buliding SCVs on them.
I also make another CC at my main and expand to my natural. I'm
reconstructing defenses and have a tight wallin now with bunkers and
3 tanks at my main. At this point, it is critical that I
don't lose my main, since all my tech is there. I would build my
tech on islands, but I need every square inch of those for turrets
and bunkers. Hopefully my opponents do not have guardians or
carriers yet. Even if so, I should have enough BCs in relatively
short time to deal with taking back my expansions if I lose any of
One minor suggestion: Right after your sci
facility is done and you're waiting for your physics lab,
immediately start building SVs out of each starport if you
feel safe. You'll want your SVs as early as possible so you
can allow them to charge up manna.
There are some other minor skirmishes. ZERG sends a couple of lurkers to my base where
I have no turrets, but a quick comsat and tank shots take them out.
TOSS still hasn't harassed me at all,
although both parties are aware of my island expansions due to
observors and overlords. I'm starting to get worried from TOSS because he's doing as any good FFA player
should: lay low so you don't anger anyone, expand like mad, mass
your unstoppable fleet and attack by surprise. Comsat keeps tabs on
him, though, and it looks like he's just going for mass shuttle
drops with corsair support.
The rest of the game is
very straightforward. I max out supply at around 12 3/3 BCs, I have
around 6 to 8 fully upgraded SVs, and a dropship with 8 SCVs tags
along for repairs. If you're paranoid, you can also bring a few meds
with restore to counter plague and lockdown, and the meds also are
useful because they automatically heal your SCVs in case they take
any incidental damage.
By this time, my islands are starting to run dry, so I kill my
SCVs to free supply. The most efficient, yet somewhat cruel, way to
do this is to irradiate an SV and patrol it above your SCVs. I don't
have the luxury of just psi storming them. I can keep my CCs and
just float them to new expansions as I take them out with my BCs. I
also expanded to the old Terran's base (my partner who had left the
game), since often that is the last place people will recon.
Remember to queue up BCs when you are maxed out
on supply and have enough resources to do so. When I'm maxed
out, I also make sure that my starports all have BCs queued up
to build. This way, if one of my BCs dies or somehow I suffer
casualties that bring me under supply, one of the starports
will immediately starting bulding a BC without my having to
micro. Thus, I always hover near supply limit, and my BC fleet
will automatically replace itself even if I lose it in its
entirety. If you have enough money, which I did, you can be
outrageous and build 12 starports each with BCs queued waiting
for supply to free itself.
My fully upgraded doom fleet then goes to have a ton of fun.
There are innumerable things you can do with 12 BCs and 6 SVs. You
can orb (defense matrix) your BCs and mindlessly macro them. You can
keep your use comsat to spot targets and clone-yamato or
clone-irradiate them, then bring BCs to clean up. Things to
particularly watch out for are ghosts, arbiters, defilers, and dark
archons. A good stasis in particular can really ruin your day.
Should I yamato anything that threatens my
It's very important is to use yamato sparingly and
wisely. I only yamato things that pose a significant threat to
my BCs. Yamato is expensive and you don't want to be caught
without it when you need it. I don't waste yamato on spore
colonies, photons, or other meager defenses. Those will damage
my BCs, but e.g. a spore colony only does 6 damage against a
0/3 BC and I can certainly repair it afterward.
I should say that another effective counter against mass BCs is
scouts, of course. Fortunately, nobody builds those, since carriers
and corsairs also both beat BCs cost-effectively. Wraiths are great
anti-BC as well. If you do see either, you may want to consider
mixing in a few valkyries, since 3/3 valks just about break even
against 3/3/3 scouts and are marginally cost-effective against 3/3
wraiths, or you can just rely on EMP and defense matrix. Valks
aren't my unit of choice, but they can really cream the competition
if you're way ahead of your opponents in upgrades.
Here's some Starcraft nerd talk for you. The big
problem with using valks in large games is that they will tend
to suffer from the sprite
overload bug. This is not just a rare bug; you can
pretty much expect it to happen in any large games (i.e.
involving more than 2 players), particularly
Starcraft can only support a certain maximum number
of graphical entities (sprites) on the screen at once, and
valkyries use TONS of sprites when they fire. Each missile is
a sprite, the smoke trails also take up sprites. When there
are not enough sprites to accomodate your valkyrie fire, such
as when there are many other things firing at the same time,
your valkyrie will not fire at all. This can happen to other
units too, but it is primary evident with valks since those
uniquely require so many sprites just to fire a single shot.
I attacked TOSS first and cleaned out his peripheral
expansions to rob him of income, then hit his main, using comsat
judiciously to spot for potential threats. He didn't have nearly the
type of anti-air to do any significant damage, only some carriers
and corsairs, and I ended up wiping his base clean. I generally
proceed slowly: I'll engage in a major battle, then stop to unload
my SCVs and repair any damaged BCs completel, then move onto to the
Remember to repair your BCs after battles if
they're moderately damaged, especially after severe attacks
such as getting plagued. Also remember to keep enough
resources so that a Terran can have enough money to repair
Repairing is fairly simple: Select your 8 SCVs,
shift-queue them to repair all the BCs by right-clicking on
each one. I'll usually select all BCs, find the most
critically damaged, and move that away from the clump and have
the SCVs repair that first. Also, things tend to get pretty
cluttered, so it's useful to have your SCVs hotkeyed so you
can select them all easily. Remember not to hit the hotkey
while they're in the dropship, or you'll lose the control
group. For example, if have them hotkeyed to "1" and hit "1"
while they're partially unloaded, "1" will be reassigned to
include only the SCVs that have unloaded.
Another tip when using a BC doom fleet is to
build turrets in spots that you capture. This gives you a
"firebase" fallback in case your BCs encounter very heavy
resistance and you need some added firepower, and it also
gives you a spot of recon so you can spot any attempts from
your opponents to re-expand to those spots. Finally, it gives
marginal defense if you want to float one of your CCs from
your dried out islands or natural to captured locations.
Everything was fairly routine after this. I was surprised that 10
devourers can take down a BC in my group of 12 before I can
eliminate them, but I proceeded more carefully afterward and
comsatted ahead of my BCs to clone irradiate any devourers. I saved
yamato and used it on ZERG to first
mass-yamato his hive to death, then mass-yamato his greater spire.
If you're a Zerg player, don't get leave your
critical structures without redundancy, since e.g. losing a
greater spire can be a very bad setback.
hive or spire is one of the most demoralizing things you can
do to a Zerg player, and there is really no way to spot a
bunch of BCs from hit-and-running your hive. For kicks, you
can also destroy his queen's nest so that it takes him even
longer to tech back to hive level. If you can't destroy the
hive, just yamato the greater spire. This is worth doing
oftentimes even if you lose some BCs, since the Zerg player
will spend the rest of the game rebuilding his greater spire.
This is why it's important to have redundancy: it's a good
idea to have at least 2 spires in late game on different fault
lines (err, different bases) since you'll want dual upgrades
anyway if you're going air.
It was all cleanup from here on. I did lose a few BCs, but I'd
generally just I'd watch my control group and pull away any that
were getting critically red so I could repair, so in the end there
was little my opponents could have done at that stage to have
stopped my doom fleet at that stage. The real issue wasn't that a
BC/SV combo is unstoppable, but that I was able to get a big
resource advantage in mid game and convert it all into units
Remember to attack fairly early once you have
your BC fleet. Your resource advantage doesn't mean squat
after a while if you're maxed out on supply and have tons of
resources sitting around. There it loses its advantage, since
e.g. another Terran player can have the same sized fleet with
far fewer resources because you're both limited by supply.
Furthermore, since you invested in dual armories early and
made the sacrifice to get early upgrades, use that to your
advantage and don't sit around while your opponents catch up
in upgrades. In late late game, everyone is going to be at
3/3, so nobody will care that you plopped down 3 forges and
got 3/3/3 dragoons super early and sacrificed your econ as a
result, if everyone has 3/3/3 units by the time you decide to
attack with yours.
I hope you enjoyed this BR and
that it was educational to read. I try to focus on more unusual
types of play so that I don't risk boring you with the usual run of
the mill strats. Note though that consequently, you'll have to be
judicious about deciding when to use these, since e.g. quick
expansion does entail some pretty big associated risks.
I apologize for the lack of screenshots, but it is rather
difficult for me to import screenshots since I do not have ready
access to an FTP site where I can store the images. Also, someday, I
will post a report where I lose so that you guys don't think I'm one
of those pompous guys who posts to inflate his self-ego, even though
that may indeed be true ;-).