Author Topic: Doing it wrong - full conquest on 120 planets vs 7/7  (Read 31515 times)

Offline AnnoyingOrange

  • Jr. Member Mark II
  • **
  • Posts: 71
Doing it wrong - full conquest on 120 planets vs 7/7
« on: December 14, 2018, 12:34:26 pm »
DOING IT WRONG

My first game in AI war after the tutorial was nothing too crazy, a 5/5 on a 40 planet encapsulated map.
It was incredibly easy: towards the end, I had captured more than half of the map, yet my AIP was barely past 200, so I could just waltz into the last homeworld and the AIs barely tried to stop me.

I decided to up the challenge: the tutorial was very clear on how dangerous AIP increases were, I like capturing planets and unlocking new toys, and Steam showed me there was an achievement for capturing 110 planets in a single game, so I got an idea.

This is the result of said idea, after one failed attempt and 41 hours and 22 minutes in real time:





The red AI is 7 Chivalric/Bouncer, because I dislike wormhole guardposts, and the orange one is 7 Chivalric because I realized how annoying protecting irreplaceables would be on such a campaign, and because I did not fear CPAs: both have a -50% handicap to resources, to keep ship spam slightly more reasonable than in my first attempt.
I gave orange the hybrid plot, and red the preemption plot, IIRC.
I chose to go without fog of war, since I found scout management to be mostly a chore even on a 40 planet map, I enabled a bunch of minor factions, the important ones being golems, botnet, spirectaft, and fallen spire (all hard), then dyson sphere, zenith trader, and spire civ leaders, and I chose the X map because I completely failed to notice the "needs a planet with 2 wormholes" requirement on the spire city I built in my first game.
I started as player+champion with alt progress and nemesis waves both on 5, shield bearers because I like tanky units, and very low ship caps in the futile hope that would help my framerate in the late game.

With that out of the way, the start of the game was pretty simple: secure the area near homeworld, check nearby goodies, and plan how to reach and liberate spire civ leaders.
The last step is vital, as civ leaders are one of the strongest AIP reducers in long games, and I would need to increase AIP by at least 2400 to reach my goal of full conquest: on the flip side, not dealing with the ouposts ASAP would punish me with the equivalent of 1 AIP every 6 minutes.

I was lucky to find the dyson sphere only 3 hops away from my homeworld (it's the unconquered planet near the center), giving me free defenders patrolling my territory, and I was even more lucky to find the botnet golem close by.
I rushed through the middle the X and up the top-right branch to liberate the two CLs there and collect some fabricators, especially nice were the decoy drones and the speed boosters.

At that point I had most turrets up to mk3 unlocked including gravity ones and tractors, military stations mk3 to hold the chokepoints, economic stations mk2 filling the cluster above my homeworld, mk3 metal extractors, a champion with nearly everything fully unlocked, three golems (artillery, hive, and regenerator) besides the botnet, and speed boosters to bring any unit wherever I wanted in less than a minute.
The AI response was still... disappointing, all things considered. The biggest waves were easily contained without assistance from one of my fleets, the botnet golem turned most lingering threat into allies, and my economy could afford a full fleet rebuild without having to dig into the stockpile.
Time to build some Spire cities.

The cities are marked as priority 0 on the map, the first one to be built was placed right next to the center of the X and filled with reactors for defense, shortly followed by the one behind my homeworld for ship production and then the one in the top right to hold the chokepoint near the Super Terminal.

The AIP was starting to get a bit high, so I decided to get into the lower left branch of the map and conquer that tasty cluster with 3 Civilian Leaders, an advanced starship fabricator, and a black widow golem: but to do that, I had to pass through an extremely annoying planet with six wormholes, plenty of defenses, and the unusually long name Asimoattune.
At the time, I decided that conquering Asimoattune and then putting a warp jamming station on it would be good enough: in hindsight that might have been a bad call, as the wormhole layout makes static defenses far less effective than elsewhere.
The cluster got its own spire city, and I decided the fifth was going to form a chokepoint near the homeworld of the orange AI: a couple armored warheads later (Seriously, 5 imposion IV guardposts overlooking the wormhole?), and I unlocked the big Spire ships: amusingly, even the Dreadnought can fit in transports, which makes it the nastiest fast response unit I have in my arsenal by far.

I capture the two clusters below my homeworld, as their normal waves were starting to get a bit too much for their chokepoints. One of them has a Coprocessor, I pop it without a care, and replace it with the sixth Spire city.

All is well, and things still seem fairly manageable even though AIP is now past 900, then I get a 10k+ normal wave on my first Spire city, and this while the bulk of my Spire fleet is cleaning up 5k threat
coming in from the Super Terminal cluster: all my other assets combined are enough to stop the assault and drop my framerate in the single digits, but it's now clear than I need at least two chokepoints in a row to handle waves safely.
I conquer Iosvu, in front of my first city, and Admtic, the planet next to Asimoattune and the orange homeworld (Lyvarcha): too late I realize Lyvarcha has a core raid post on it.

>120 Enemy ships (III), 5 Starships to Admtic blah blah Core Raid Post blah blah

Much less than 5% of the average wave at this point, guess it's free scrap at least: also Admtic has a wonderful wormhole layout, making it super easy to defend against anything but the biggest attacks.
Admtic gets a warp jammer post too, and I start corrupting the saboueur design on a nearby planet: of course, a 7k wave is announced on Iosvu, scheduled less than half a minute after the hack is done.




Not a big deal, as Iosvu is essentially ablative armor to absorb the bulk of a wave, I can always rebuild it later as there's nothing too important on it.
My main fleet is not fast enough to reach it anyways after the hack is done, and neither are the golems, but the Spire fleet is on Asimoattune and can make it: let's see how well they fare against carrier spam.
A few minutes later, the carriers go up in flames mere seconds after warping in, and I sit back, relax, and enjoy the show: there's something weird though, the number of attacking ships doesn't seem to go below 3k, I wonder why.




OH, THAT'S WHY.
I guess pulling out my only defense fleet from a planet with almost six digits of threat fleetpower lurking at the border might leave said planet vulnerable to attacks, who could have guessed?
I order the Spire fleet to regroup and head back ASAP, it looks like there's only one heavy bomber starship left on Iosvu and my champion + implosion spirecraft can handle that no problem.

While I defend the least defendible planet in my entire empire, I am informed of the destruction of the command station on Iosvu: turns out the one heavy bomber starship actually was fifty-nine HBS stacked exactly on top of each other.




Oof, lesson learned: always check ship counts.
On the bright side, the SC implosion artillery almost managed to stop them:I'd add those ships to the main fleet in a heartbeat, but they are so fragile and the AI unlocked Railpods at some point.

In any case, at 42 hours realtime the attacks are repelled, rebuilding is almost complete, and there's still 1k threat next to my stuff.




I think it may be time to neuter the bottom left clusters to make Asimoattune slightly safer, it won't be easy nor quick as even home station reinforcements alone can be 600+ units by now but at least I'll have some AIP decrease in the meantime from the Civ Leaders.
I may even pop the 3 remaining coprocessors, 2 of them are close to friendly territory thus I only need to work to reach the one on the planet marker priority 3, in the bottom left.
Not sure how I feel about going for the Super Terminal, even with all this firepower I doubt I'll manage to get -100 AIP or more from it, maybe waiting for more spire cities would be for the best.


« Last Edit: December 14, 2018, 12:42:54 pm by AnnoyingOrange »

Offline Worblehat

  • Newbie Mark III
  • *
  • Posts: 29
Re: Doing it wrong - full conquest on 120 planets vs 7/7
« Reply #1 on: December 15, 2018, 02:53:29 am »
amusingly, even the Dreadnought can fit in transports, which makes it the nastiest fast response unit I have in my arsenal by far.

That is a good trick! It never occurred to me that Spire ships could be transported, but I guess if they couldn't they'd say "Immunity: Transport" or something like that. Slow speed seems to be their one weakness and this largely fixes it...  :D

I'd ask what the Golem/Spirecraft/Botnet exo waves are like, but I guess once you're facing six-city FS exos nothing else really compares. My tendency is to turn on options one at a time, one step at a time, rather than everything at once. If you remember the relative strengths of those compared to one-city FS exos, that would be an interesting point of comparison though.

Thanks for posting this; nice to see a new AAR! Especially with such a wildly different playing style from mine; to me 200 AIP is "high"...  :P

Offline AnnoyingOrange

  • Jr. Member Mark II
  • **
  • Posts: 71
Re: Doing it wrong - full conquest on 120 planets vs 7/7
« Reply #2 on: December 15, 2018, 10:30:37 am »
I'd ask what the Golem/Spirecraft/Botnet exo waves are like, but I guess once you're facing six-city FS exos nothing else really compares.

As far as I can tell there are at least 2 kinds of exo strikes still strong enough to be noticeable: one that spawns Spirecraft ships, and one that spawns dire guardians, both also get generous amount of normal units of course.
I've not pinpointed which exo is which, though, as many of them charge up in sync.

Offline AnnoyingOrange

  • Jr. Member Mark II
  • **
  • Posts: 71
Re: Doing it wrong - full conquest on 120 planets vs 7/7
« Reply #3 on: December 15, 2018, 06:17:23 pm »
Time to deal with the coprocessors.

Only one of them is out of immediate reach, so I send a small task force to strike. And by small task force I mean an Artillery golem, a Hive golem, my champion, 20 mk3 engineers, 49 speed boosters, and the Spire Dreadnought on board of the assault transport I've nicknamed "Party van".
The target coprocessor is 3 hops away from Asimoattune, so I neuter the adjacent planet with the Dreadnought and the champion, then empty the hive golem on the second, and only the artillery golem steps foot on the third, just long enough to take the shot: thanks to the speed boosters the whole operation can be carried out hastily, with the speed of my golems being boosted well above that of a raid starship.

I don't care when a 3k wave on Iosvu is announced, that's manageable even without the Spire fleet, but right as I'm about to pull out the last units I am warned about an upcoming 18k CPA.



That's 50% bigger than the last one, and almost twice the size of the largest wave I've handled so far.
At least there is time to prepare.
First thing first, the Spire fleet rushes to Iosvu to handle the wave in the safest way possible: as expected, they crush the opposition without effort, no matter how many mini rams the AI spams their way.
Those things would be nasty if they weren't so slow, good thing my enemy doesn't have any unit to boost their speed.



Remembering the sneaky HBS stack of last time, I check on the apparently lonely HBS making its way towards the command station: at least this time it's only 9 of them.



With 10 minutes remaining I pop the two remaining coprocessors sending AIP near 830, neuter a couple more planets near Asimoattune, and conquer one adjacent planet to act as a crumple zone: I intend to expand in that cluster soon-ish, as it holds one of the last three spire civ leaders, a zenith power generator, and three different golems (regen, hive, anc cursed).
The CPA hits, and threat breaks past 20k for the first time in this game.



Almost 10k threat piles up in front of Iosvu, no big deal given the Spire fleet defending it, less than 3 k is behind my newly acquired planet, and a good 7k goes for Asimoattune, 4k of it deciding to strike almost immediately.
The botnet golem was already in position, less than 5 minutes later the attack is crushed and 2k new zombies join my cause.



That being dealt with, I decide to acquire one mk3 planet near Admtic to benefit from its spire archive: not exactly a massive strategic asset given how abundant knowledge is, but the less threat can spawn near that area, the better, and even though the planet is already neutered it has accumulated almost 500 ships, all piled up on the only wormhole.
It's already so dangerous that I must send in my Botnet golem together with Decoy drones and Shield bearer stacks to absorb some of the alpha damage, I really need to find a better way to break those threat plugs: unfortunately I need to do it so frequently that both warheads and Martyrs are not viable options.

The archive collects knowledge for half an hour, in the meantime I work my way into the cluster with 3 broken golems I mentioned before.
I bring along a full stack of Penetrators, overkill but you never know where they could be useful, and half an hour of downtime is no big deal in such a long game.
When capturing the middle planet in the cluster, Theryomik, I face my first Hunter/Killer: the Spire fleet annihilates it before I can get a good look at it, the bugger still takes out a big chunk of the mk3 forcefield covering my beachead.
130k alpha damage, 1 M health, immunity to black hole machines, immunity to artillery... they are nasty even at mark 1, good thing it spawned at the best place and time I could have asked for.
With the help of the Penetrators, three more planets and two more golems join my empire: as of now, I have one artillery, one hive, two regen, one cursed, one black widow, and one botnet golem, and there are only two spire leaders outside my control.
Time to collect a bit more knowledge and decide what to do next.



Offline AnnoyingOrange

  • Jr. Member Mark II
  • **
  • Posts: 71
Re: Doing it wrong - full conquest on 120 planets vs 7/7
« Reply #4 on: December 20, 2018, 05:05:45 pm »
In most strategy games, once you've built up your economy and disrupted the enemy's infrastructures, you're already won: ending the game is but a chore by then.
AI War is designed from the ground up to avoid such a situation, so I was a bit surprised to see I've reached it anyways.
Last post ended at 47 hours real time, this one goes from there to 66 and a half, most of them spent letting the game run idle at +6 speed and waiting for the Spire Leaders to lower AIP.

With two dreadnoughts in a transport, even CPAs are a mild annoyance at worst.



Speaking of annoyances, 10k+ waves are frequent and a source of massive stutter as the bulk of my Spire fleet melts through carriers: to try and keep this problem under control I've decided to keep AIP below 800, down from 950ish.



After clearing out two clusters in the bottom left branch, I decide to go and conquer a part of the bottom right branch to acquire an advanced factory (not like I use fleetships much by now, I barely touch starships), a spot for my tenth spire city, and a piece of Titanite for two SC jumpships mark V.



As the operation goes on, I see a handful of hunter killers (slain by static defenses in most cases), my first EMP guardians (I already hate them a bit), the warp gate guardians (I will never let one on my planets, too spooky), some CPAs, too many CPU-melting waves, and enough exo strikes to notice how only the Fallen Spire exos are beefy enough to warrant any attention on my part: a dozen spirecraft ships plus as many dire guardians cannot be taken lightly no matter your defenses, a single dire implosion guardian in the wrong place at the wrong time can eat a couple golems before you notice what happened.
At least I have enough regen golems, and so much metal I don't know what to do with it: at one point my income was 100k/s thanks to salvage, and even when reactivating a golem it only goes negative for a second or two.



Yep, that's a logistic station 3 on my toughest chokepoint.
I don't see the point of military 3 here: tachyon coverage is not worth giving up the planet-wide speed effects (grav turrets get instakilled by that many carriers, can't rely on them), and the mil 3 weapons are irrelevant when compared to the Spire fleet stationed there in FRD.

Those beam weapons are ridiculous, even frigates and destroyers can effortlessly outdamage the best fleetship deathball I could put together: as a bonus, the beam doesn't waste any power when burning through stacked carriers filled with hundreds and hundreds of ships, and those targets are too large and too clumped together to be missed.
The dreadnoughts are even more insanely powerful and can take down an Eye with direct fire in a matter of seconds: eliminating guardposts is mostly optional at this point, and now that I have the SC jumpship I could easily cleanse all non-homeworld planets remaining in less than an hour.
The only downside of the jumpship is how buggy it is: occasionally enemies can get a shot in after a warp, couple that with near-zero health and immunity to repair and you have the perfect recipe for frustration and savescumming.

I think I'll do some test runs at the super terminal now, mostly for bragging rights on max AIP reduction and to see how far in the negatives I can go: hacking points are probably better spent on design hacks, at least for this game.

Offline AnnoyingOrange

  • Jr. Member Mark II
  • **
  • Posts: 71
Re: Doing it wrong - full conquest on 120 planets vs 7/7
« Reply #5 on: December 29, 2018, 12:50:53 pm »
Eighty conquered planets, ten fully developed Spire cities, and exactly 150 Spire ships: two thirds of the way to total domination, and everything is going smoothly.



Having three Dreadnoughts means I can delete wormhole guardposts (and Eyes) in two main beam firing cycles, which combined with the Spirecraft Jumpship is going to make the final third of my expansion a breeze.
With so few enemy planets left, I have to carefully consider how to approach design backups: I don't want to end up with all of them on a single planet, and no choice besides downloading or corrupting them all in bulk, at least not yet.
Better to try and remove some of the annoying ones ASAP (teleptorting stations, maybe reprocessors given their enormous health), and then conquer some that could help against the AI's final stand (anything high dps or fast enough to build).
As a precaution, I let my AIP fall even lower than usual before starting the raids.



The first round of hacks is mostly a success, even if I lose one of my two hive golems without realizing: no big deal, by now even a full swarm is not really noteworthy.
Next step is to expand into the two clusters near Asimoattune, to get two more Spire cities, gain control over the last civilian leader, and remove one rather uncomfortable front from this war.



By now I've spent around 260 hacking on corruptions and 240 on downloads, with both costing 150+ points next time, and I can expect to gain 700 more HaP this game: a couple more, and it will be time to handle all remaining backups in one final hack.
I decide to try running the superterminal, and I easily manage to deplete over 1600 HaP for almost exactly 200 AIP reduction.



12k attack is barely a challenge, especially if I get the bulk of my fleets in position instead of only bringing the 3 strongest ships I have, the ST itself is easily taken out by a sniper turret salvo, and the post hack response is nothing too crazy either: half a million strenght is an average wave on Iosvu.
Still a colossal waste of hacking and a pointless increase of the AIP floor, but that's why it was a test run: now I know for sure I can hack the ST for far longer than I'll ever want to.

Disappointed by the ST, I noticed I got rid of the last CSG-A and decided to go poking at the homeworld I already cornered.
That was a bad idea.


10 H/Ks, 26 Spirecraft ships, 12 Dire guardians, 12 golems, and one mothership.
Defense is succesful (not too hard when there's ten or so planets to fortify on the way), but I take way too many losses, in part thanks to a peculiar mechanic regarding womrholes and in-flight projectiles: ships going through a wormhole while there are projectiles coming for them will take all the damage of said projectiles as soon as they are through the wormhole.
Time to reload and come up with a plan more elaborate than "go in and shoot".

Offline AnnoyingOrange

  • Jr. Member Mark II
  • **
  • Posts: 71
Re: Doing it wrong - full conquest on 120 planets vs 7/7
« Reply #6 on: April 11, 2019, 01:26:25 pm »
THE END

Taking an AI homeworld at 700 AIP is not a smart idea, but that hasn't stopped me from starting this campaign, has it?

After a few tries I figure out that the AI doesn't bring out the big guns unless I attack the homeworld with spire ships, and the guard post mix means even my golems can't survive for long let alone get close to any valuable target: time to dust off the starships.
They're so weak compared to the superweapons I'm used to by now, but they're disposable and that's all that matters for such a prolonged assault.: the plan is to set up a dock to continuously send decoys into the wormhole, and then move in a blob of MK3+ flagships, neinzul carriers, zeniths, and heavy bombers, in theory the decoys and neinzul spawns should absorb the bulk of the damage while the rest can work on the guardposts.

A few millions metal later, the wormhole is safe enough to bring in a mobile builder and put up a small beachhead, and I add spire, lighting, and plasma starships to the mix now that my docks aren't working so hard.
With a particularly nasty implosion guardpost dealt with, the golems join the fight, the two artillery make short work of the remaining forcefields and the enemy fortress while the cursed provides some much needed long range attrition.

Five hours after the start of  the offensive, the home command station of AI 2 falls.


Amusingly enough, the AI homeworld makes for a pretty good checkpoint to tank waves, even without the bulk of my spire fleet: it can tank normal waves and all kind of exos without issues, and I haven't even brought in the implosion artillery yet!


Shortly after, AIP falls below 600, and I decide to leave the game running AFK overnight given how trustworthy my defenses have been and how time-consuming the civilian leaders are.


The rest of the campaign is kind of disappointing: it's simply a matter of sending the dreadnought jumpship in to wipe a planet clean, then moving the main spire fleet one jump forward, occasionally building one more spire city...
The real time clock is past 100 hours by the time I own 119 planets, and for the second homeworld I just turn all remaining asteroids in the galaxy into Martyrs and then throw them in en masse, followed immediately after by the full spire fleet my 16 cities can sustain, 235 ships including 5 dreadnoughts: AI 1 calls in a Mothership only to see it explode within seconds, and the rest of its resistance is unremarkable.

One last colony ship is built, and the deed is done.

Offline Cyborg

  • Master Member Mark III
  • *****
  • Posts: 1,957
Re: Doing it wrong - full conquest on 120 planets vs 7/7
« Reply #7 on: April 13, 2019, 03:51:17 pm »
Awesome story so far!
Kahuna strategy guide:
http://www.arcengames.com/forums/index.php/topic,13369.0.html

Suggestions, bugs? Don't be lazy, give back:
http://www.arcengames.com/mantisbt/

Planetcracker. Believe it.

The stigma of hunger. http://wayw.re/Vi12BK