The Absolute Best Battleship (Crystal Forge Edition!)

Larknok1 · Feb 2, 2017

Hi guys, I'm the dude behind the original thread. Since my original post, user Demiare pointed out that a Crystal-Forged Type-12 beats the Type-42.

For the uninitiated:

When describing a Battlehip or Cruiser's type, the first number gives the number of shields (with everything else as armor), and the second number gives the ship's plasma weapon number (with everything else a Tachyon Lance or Kinetic Artillery.)

So a Type-42 -- for example -- has 4 shields and 2 Plasma, with everything else Tachyon / Kinetic / Power Generators.

What Demiare pointed out is that adding in Accessory combinations throws a monkey-wrench into this operation,
and de-thrones the Type-42 as supreme king.

So, in order to make sense of the new combinations, here's the new terminology to account for accessories:

A third number at the end of the Type. That number gives the number of Shield Capacitors, with everything
else as Crystal Forged Armor.

So a Type-422 has 4 shields, 2 plasma, 2 shield capacitors.

And a Type-120 has 1 shield, 2 plasma, 0 shield capacitors (2 crystal forged).

----

A couple of notes for possible combinations:

1) I'm going to rule out everything with less than 2 Plasma (more than 2 Kinetic Artillery.) Previous tests (see original thread) have shown this to always be a sub-optimal weapon build.

2) I'm going to rule out all 0 shield builds that reach max armor. They won't receive any extra armor with repeatable techs, or armor % bonuses from starports.

This leaves:

100 series: 120, 121, 122, 130, 131, 132, 140, 141, 142
200 series: 220, 221, 222, 230, 231, 232, 240, 241, 242
300 series: 320, 321, 322, 330, 331, 332, 340, 341, 342
400 series: 420, 421, 422, 430, 431, 432, 440, 441, 442

---

That's 36 viable Battleship designs with (incidentally) 666 possible fleet match-ups.

Now, I'm not about to go 1 by 1 through them.

So let's make a few more assumptions to limit the options:

---

3) The 100 series optimal builds will all have 0 shield capacitors (2 Crystal Forged).

4) The 200 series optimal builds will all have 0 shield capacitors (2 Crystal Forged).

5) The 300 series optimal builds will all have 1 shield capacitor (1 Crystal Forged).

6) The 400 series optimal builds will all have 2 shield capacitors (0 Crystal Forged).

---

This leaves:

100 series: 120, 130, 140
200 series: 220, 230, 240
300 series: 321, 331, 341
400 series: 422, 432, 442

---

That's 12 viable Battleship designs with a still-too-large 78 Combinations.

We can do better.

Let's assume:

7) 2 or 3 Plasma builds are optimal. A 4 plasma build is unwieldy and far too easily countered by shields.

---

This leaves:

100 series: 120, 130
200 series: 220, 230
300 series: 321, 331
400 series: 422, 432

---

That's 8 viable Battleship designs with a mere 36 Combinations.

Now we've entered the range where things are testable.

It doesn't seem like we can reduce the number of test subjects any further without assuming one ship type will be better than others.

We can, however, pre-guess certain match-ups that should be fairly one-sided without needing to further test:

---

8) 100 and 200 series ships should beat ships with 2 plasma.

9) 300 and 400 series ships should beat ships with 3 plasma.

---

So the full catalog of match-ups becomes:

---

120 vs 130: Free win to 130
120 vs 220: ???
120 vs 230: Free win to 230
120 vs 321: Free win to 120
120 vs 331: ???
120 vs 422: Free win to 120
120 vs 432: ???
130 vs 220: Free win to 130
130 vs 230: ???
130 vs 321: ???
130 vs 331: Free win to 331
130 vs 422: ???
130 vs 432: Free win to 432
220 vs 230: Free win to 230
220 vs 321: Free win to 220
220 vs 331: ???
220 vs 422: Free win to 220
220 vs 432: ???
230 vs 321: ???
230 vs 331: Free win to 331
230 vs 422: ???
230 vs 432: Free win to 432
321 vs 331: Free win to 321
321 vs 422: ???
321 vs 432: Free win to 321
331 vs 422: Free win to 422
331 vs 432: ???
422 vs 432: Free win to 422

---

This tallies to 2 free wins for every fleet type, which leaves just these 12 match-ups to test:

---

120 vs 220: ???
120 vs 331: ???
120 vs 432: ???
130 vs 230: ???
130 vs 321: ???
130 vs 422: ???
220 vs 331: ???
220 vs 432: ???
230 vs 321: ???
230 vs 422: ???
321 vs 422: ???
331 vs 432: ???

Demiare · Feb 2, 2017

*With devil's smile* We also have blue crystal-infused plate that more easily to get. Then after initial results we should adjust them by resources and basic levels of repeated tech...

And I'm advice to use at least 50 BB (400 fleet size) per test.

Larknok1 · Feb 2, 2017

Demiare said:
*With devil's smile* We also have blue crystal-infused plate that more easily to get. Then after initial results we should adjust them by resources and basic levels of repeated tech...

And I'm advice to use at least 50 BB (400 fleet size) per test.

It's very dubious whether or not blue crystal-infused would beat standard builds.

Also, the main point of using 40+ BB is to reduce the RNG spread of Tachyons. Even then, it still remains true that most fleet battles are settled by whichever fleet rolls higher in the first couple Tachyon volleys.

Larknok1 · Feb 2, 2017

Hey Demiare, I'll tell ya what, you do the first six tests (~5 trials each):

120 vs 220: ???
120 vs 331: ???
120 vs 432: ???
130 vs 230: ???
130 vs 321: ???
130 vs 422: ???

and I'll do the next six tests (~5 trials each):

220 vs 331: ???
220 vs 432: ???
230 vs 321: ???
230 vs 422: ???
321 vs 422: ???
331 vs 432: ???

doing twelve tests would be a pain in the ass.

Demiare · Feb 2, 2017

Larknok1 said:
Also, the main point of using 40+ BB is to reduce the RNG spread of Tachyons. Even then, it still remains true that most fleet battles are settled by whichever fleet rolls higher in the first couple Tachyon volleys.

Noooo. It's have completely different reason. Targeting mechanic.

AI tend to maximize each shot damage. For guns with +% shield bonus like KA it's mean they will ignore hull and go for shield. In same time AI tend to try keep some shields on targets, or have very optimistic evaluation of damage.

In same time plasma & tachyon have -% shield. Game will do everything to avoid them deal damage to shield - mean that they will focus target with lowest shield (and if will appear other target with lower shield that current have - they will switch).

What's this mean? This mean when we have enough battleships to ensure that KA will breach enough (sorry for tautology) shields - anti-armor weapon will deal with only hull and by it's amount will kill ship regardless of it's regen (so all that will protect that ship will be armor). And in next combat round one side will have lower amount of ships = deal less returning fire = have less chances to win.

More ships get involved - stronger that effect will became. With larger fleets - armor&hp builds became effective then shield regen-based. So we need to choice our "baseline" fleet size.

Demiare · Feb 2, 2017

Larknok1 said:
Hey Demiare, I'll tell ya what, you do the first six tests (~5 trials each):

120 vs 220: ???
120 vs 331: ???
120 vs 432: ???
130 vs 230: ???
130 vs 321: ???
130 vs 422: ???

NP.

I'm don't on friendly terms with imagehostings so I can post screenshots proofs only on Steam service. You don't mind if I will merge 5 results in one picture via paint?

Larknok1 · Feb 2, 2017

Demiare said:
NP.

I'm don't on friendly terms with imagehostings so I can post screenshots proofs only on Steam service. You don't mind if I will merge 5 results in one picture via paint?

Do what you have to do.

Larknok1 · Feb 2, 2017

Demiare said:
Noooo. It's have completely different reason. Targeting mechanic.

AI tend to maximize each shot damage. For guns with +% shield bonus like KA it's mean they will ignore hull and go for shield. In same time AI tend to try keep some shields on targets, or have very optimistic evaluation of damage.

In same time plasma & tachyon have -% shield. Game will do everything to avoid them deal damage to shield - mean that they will focus target with lowest shield (and if will appear other target with lower shield that current have - they will switch).

What's this mean? This mean when we have enough battleships to ensure that KA will breach enough (sorry for tautology) shields - anti-armor weapon will deal with only hull and by it's amount will kill ship regardless of it's regen (so all that will protect that ship will be armor). And in next combat round one side will have lower amount of ships = deal less returning fire = have less chances to win.

More ships get involved - stronger that effect will became. With larger fleets - armor&hp builds became effective then shield regen-based. So we need to choice our "baseline" fleet size.

I'm aware of why stacking more than 2 KA is bad.

But adding more BBs doesn't fix that -- there's a fundamental cap on how many Kinetic Artillery weapons will target a single full HP/shield ship. You reach that cap somewhere between 5 to 10 KA targetting the same ship. After that, the fleet targets a second ship, and then a third ship, etc.

So increasing BB count beyond 20 or 30 or so should have no impact on this, because then you just have a linear relation of BB count to ships targetted by KA.

What matters is the ratio of ships targetted by Plasma to ships targetted by KA, which is usually optimal at 2 KA / 2 Plasma, unless you're fighting extremely high armor builds.

What larger BB count will do, however, is round out Tachyon rolls to approximately the average roll.

Demiare · Feb 2, 2017

Lamahorse · Feb 3, 2017

This is God's work.

Demiare · Feb 3, 2017

Lamahorse said:
This is God's work.

It's VERY unprecise - we don't including resources (and player will get them in +/- same time as fully-geared BB) or level 1-2 repeated. At least I do second iteration with resources&repeated after we get winners from stage 1.

LordMagus · Feb 3, 2017

Sorry i won't be around to help test this out, going on holiday. Will probably just confirm assumptions and automatic wins, as i hate making any assumptions at all.
The only suggestion i can make is to also keep track of ships destroyed, as losses and wins arent equivalent. So 5/5 wins isnt always the same as another 5/5 if the ships lost differ significantly.

zanaikin · Feb 3, 2017

I feel like there is a problem here because these tests are performed between PURE battleship fleets. But BBs don't exist by themselves.

I just waged a war against a xenophobic fallen empire with my most recent experimental fleets and noticed some interesting behavior in mixed fleet combat:

The FE had a kinetics-focused armament set (Giga Cannon, KA, and Flak).
My fleet was a mixture of all 4 classes:
- neutron torp corvettes
- PD destroyers (just got flak so haven't retired them)
- armored cruisers with 88% armor, 2x crystal-forge, KA + Flak + L+M Plasma
- BBs with 84% armor, 1 shield capacitor, tachyon + kinetic artillery + 1 plasma

The interesting parts:

(1) Once both fleets' capital ships entered range of each other, the FE focused down my BBs while mostly ignoring my better protected cruisers.
(2) The BBs almost never fired their plasma weaponry. The cruisers did only a little. Hull output of plasma damage was only around 5-7%. By the time the cruisers even brought their medium plasma guns to bear, the battle was essentially over.

I did outnumber the FE in this engagement (72BBs + 60CAs vs 60BCs + 80 escorts). So it was shorter than an equivalent-fleet battle.

So my wonder comes back down to: in mixed fleet conditions, would it not be better for BBs to focus on long-range weaponry and let the CAs bring plasma cannons? Since:

CAs clearly cannot serve as a damage sponge for BBs in a prolonged battle. In fact, they would have outlasted the BBs due to targeting priorities.

The longer a battle drags, the better plasma cannons are in comparison to kinetic artillery due to their lower range and better armor piercing.

The longer a battle drags, the better CAs become due to their higher HP. 2 armored CAs (8 fleet capacity) generally has 30-40% more HP than a BB (accounting for shields). Their higher evasion also helps against X and L sized weapons, especially when given Precognition computers. At reduced combat ranges, 4M +4L slots of weapons (2 CAs) also has higher DPS output than the battleship's XL + 4L.

CAs have both higher speed and combat behavior that allows them to close the range faster. While holding your BBs back at 100 range forces the enemy BBs to close range in relation to your CAs.

Also... I thought I would need more cruiser yards than battleship yards to keep my endgame fleet adequately replenished from combat losses. Apparently it's the other way around.

Demiare · Feb 3, 2017

zanaikin said:
(1) Once both fleets' capital ships entered range of each other, the FE focused down my BBs while mostly ignoring my better protected cruisers.

Actually this confirm need to determine best BB design because it will be always target of enemy.

zanaikin said:
Total output of plasma damage was only around 5-7%.

Total output or vs hull output? First is irrelevant.

Also your data is different from my :

Let's test. Our fleet : 52.5k fleet power, no resources, no admirals, all tech with 0 repeated.
*32 Armoured BB
*50 Armoured Cruiser (main combat build with 2KA + 2 medium plasma)
*20 Flak Cruiser (5 flak cannons and 1 autocannon to make sure that AI will move them at point blank - I'm hoping that they will soak some damage too).

vs Holy Guardians: 64.3k fleet power + some transports. 20 battlecruisers and 39 escorts. (don't know where they are lost it)
Won, remaining 31.2k fleet power.
15/32 BB, 33/50 armoured cruisers, 20/20 flak cruisers.

vs Keepers of Knowledge: 72.2k fleet power alone. 20 battlecruisers and 40 escorts.
Won, remaining 43.8k fleet power.
26/32 BB, 43/50 armoured cruisers, 17/20 flak cruisers.

vs Enigmatic Observers: 67.5k fleet power. 20 battlecruisers and 40 escorts.
Won, remaining 37.4k fleet power.
26/32 BB, 30/50 armoured cruisers, 13/20 flak cruisers.

vs Militant Isolationists: 101k fleet power. 30 battlecruisers and 60 escorts.
Won, remaining 24.3k fleet power.
28/32 BB, 0/50 armoured cruisers, 4/20 flak cruisers.

In any case my cruisers died on pair with battleships. My proposal that you have several levels of repeated techs and able to destroy most of FE on approach by tachyons power.

Also - many S- and M-sized weapon prefer cruisers over battleship at point blank (test data).

zanaikin · Feb 3, 2017

Hull output. I corrected. Total output isn't that useful anyway.

As noted, most M-sized weapons (discounting destroyers -- which are generally agreed to be inefficient for their low HP) weren't brought to bear until it was too late (outside flak; they performed beautifully and made my PD destroyers look useless)

I don't disagree that it's important to figure out a BB's defensive loadout. But offensively, I disagree with the notion that any less than 2 plasma is bad. It would serve the fleet better to put those plasma cannons on faster ships.

As I note: https://forum.paradoxplaza.com/forum/index.php?threads/the-use-of-bbs.996836/#post-22391022

Demiare · Feb 3, 2017

zanaikin said:
I don't disagree that it's important to figure out a BB's defensive loadout. But offensively, I disagree with the notion that any less than 2 plasma is bad. It would serve the fleet better to put those plasma cannons on faster ships.

I'm almost sure that even in terms of fleet power - your fleet had higher power then FE. I'm right?

You're running into oblivious trap - fleet optimisation for combat with equal or more powerful enemy is different from fleet optimisation for inferior enemy.

vs equal/superior enemy : you're going to lose either cruisers, either battleships, so remaining ships should be able to stand in fight (this why we're spreading KA).
vs inferior enemy : yes, your idea will work better.

zanaikin · Feb 3, 2017

That's why I noted that the longer the engagement goes, the more my BBs take casualties. In a prolonged, truly equal fight, my projections state that my BBs will have been wiped out while at least half (if not most) of my CAs remain. That implies that CAs should be optimized for prolonged battles (more plasmas) while BBs should focus on the ability to deal maximum punishment in shorter engagements.

Have you tried running an engagement yet where:
Fleet 1:
BBs + CAs with mixed KA + Plasma armaments on both classes
vs
Fleet 2:
CAs with full plasma loadout + BBs with only 0 or 1 plasma.

I should really figure out the console commands to run these tests...

EDIT:
Also one more note:

You may want to consider taking into account some of the strategic resources.

By endgame when one can field these full-tech fleets, there is almost no excuse why one does not have access (either mining or trade) to say... Garanthium and Neutronium (which synergizes to boost armor/hull) whereas shields only receive Lythuric gas.

Similarly, kinetics are improved by Teldar crystals, while energy weapons get... nothing.

Living metal is much more rare.

Demiare · Feb 3, 2017

@zanaikin

Test #1:

60 BB (all shield, regen, 2KA+2P) & 120 CR (all shield, regen, 2KA+2medium plasma)
60 BB (all shield, regen, 4KA) & 120 CR (all shield, regen, 6medium plasma)

Results: Mixed loadout wins with 97-98 cruisers remains.

Test #2:

60 BB (all shield, regen, 2KA+2P) & 120 CR (all shield, regen, 2KA+2medium plasma)
60 BB (all shield, regen, 3KA +1P) & 120 CR (all shield, regen, 6medium plasma)

Results: Mixed loadout wins with 2-4 battleships and 94-95 cruisers remains.

Oblivious (and predictable results), don't seem any need for additional experiments here.

Es57 · Feb 3, 2017

When I read threads like this, I realize that some people have wayyyyyyy too much time on their hands...

Interesting, though.

zanaikin · Feb 3, 2017

Those of us with science or engineering backgrounds are naturally curious =P

@Demiare
I don't really get why you equipped 6M plasma instead of 2L+2M (since L>M against all CAs/BBs), but huh. I guess I'll have to figure out how to run a test so I can see what's the behavior pattern that contributes to this.

Also: http://www.dictionary.com/browse/oblivious?s=t (since you kept using it)

The Absolute Best Battleship (Crystal Forge Edition!)

Major

Corporal

Major

Major

Corporal

Corporal

Major

Major

Corporal

Functioning Alcoholic

Corporal

Chief Scientist

Captain

Corporal

Captain

Corporal

Captain

Corporal

Lt. General

Captain