Author Topic: Red is not My Favorite Color  (Read 14733 times)

Offline Faulty Logic

  • Hero Member Mark III
  • *****
  • Posts: 1,194
  • Bane of the AI
Red is not My Favorite Color
« on: April 10, 2014, 08:12:54 pm »
80 planet simple clusters fireflies 14849409

10/10 Heroic/Spire Hammer // Starfleet Commander/Techno Parasite

2 Champions, complete visibility, mono waves.

Resistance 4
Rebellions 4
Dyson 4
Wardens 4
Enclaves 4
Golems 5
Fallen Spire 5
Spirecraft Moderate

And for the AI:
Hybrids 4/4
Advanced Hybrids 7/0
Avengers
Hunter 0/10
Trains 0/1
Shark B 4/0

===Operation Back in the Game====

My homeworld is the sole entrance to a 15-planet cluster. I could possibly fit 3 full cities behind it. But if wanted more, I'd have to allow more than one entrance.

It's a pretty nice cluster, with 2 ARSs, a FactIV, missile and sniper controllers, and tiger and teleport station fabs.

Preparing for the first wave: unlocked military stations mkIII, enclave starships mkIII (I think these still need a gentle tap with the nerfbat. Maybe every six seconds instead of every five?). I have six enclaves and a cap of fireflies mkI when the wave is declared.

Uh, did the wave logic change recently? The wave announced with a laughable 14 frigates, 4 starships, at 3:30. I get two spire frigates, a neinzul shadow frigate, and a bomber starship mkII.

I used the salvage to finish the missile and sniper turrets by the time two mkIV flagships, a mkIV leech starship, and 14 parasites arrive.

Their metal builds the rest of the turrets and fleetships.

The nearest ARS is two hops away. A quick science lab in transport reveals zelecs/planes/commandos. No hack.

I popped the nearest two data centers with the Negotiators, then conquered the ARS.
I took the teleport station fab next to the ARS planet. Yeah, the enclaves are still a little too good. They and the champions were all I needed.

Unlocked fireflies II/III.

New wave announced. It had a spire cruiser. This is somewhat worrying for future endeavors.

I built the rest of the turrets and starships with the salvage. It's really awesome when you're just starting out, cutting the boring waiting down to almost nothing.

New wave, with 76 fighters and 9 starships.
9 Zenith starships. Mk III.

Projected shields and two riots literally pushing the ships back to buy time saved the day.

And now I'm rolling in money, so I built the mercenary enclave.

After scouting the next ARS, I pulled the fleet together to hack for shield bearers. No problem. I quickly conquered it (and a decoy fab), the sniper turret controller, and the advanced factory. Then I gate raided so that waves must come to the homeworld.

I pulled the fleet back home to clean out the six-destroyer threatball.

Unlocked Forts mkI, gravityIIIs, HBC mkIs. Waves fund the fort construction.

Jumpship-raided all data centers.

AIP 50
Gametime 1:40
I've conquered five planets.

To move forward, I'll have to destroy warp counter posts. Eventually I'll want to conquer the RudeGesture world in my cluster.

I set up the counterwaves to arrive five minutes apart. A normal wave happens first, and I start building mercenaries.

Then the first counterwave arrives. With a spire battleship. Fortunately, it kept trying to engage my enclave drones.

A new normal wave announces. 14 high-mark starships. None of which are immune to gravity.
I'm having trouble spending all this money fast enough. But unlocking engineerIIs and building a second merc doc help with that.

The second counterwave had a spire dreadnought. It wiped a lot of my fleet, but they are quickly replaced.

Gametime 1:54

State of Empire:
If warheads can't solve it, use more warheads.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Red is not My Favorite Color
« Reply #1 on: April 10, 2014, 08:54:57 pm »
10/10 Heroic/Spire Hammer // Starfleet Commander/Techno Parasite
Hunter 0/10
Shark B 4/0
This... should be interesting.


Quote
enclave starships mkIII (I think these still need a gentle tap with the nerfbat. Maybe every six seconds instead of every five?)
For all marks, you mean?


Quote
Uh, did the wave logic change recently? The wave announced with a laughable 14 frigates, 4 starships, at 3:30. I get two spire frigates, a neinzul shadow frigate, and a bomber starship mkII.
The wave code's gone through multiple overhauls in recent weeks, yes, but with an eye to maintaining fairly similar overall fleet ship strength. 

Just tested it myself, and yea, that's just what the math comes out to on ultra low:

On 7.018 (summary: 15 MkII bombers plus starship) :
Code: [Select]
4/10/2014 8:23:39 PM (7.018)
-----------------------------------
Starting CreateWaveToPlanet at Game Time: 0:04:01 ; Player.AIType: Bouncer ; Player.AIDifficulty: 10 ; EffectiveAIP: 10 ; AITechLevel: 2 ; IsSchizo: False
MultiplierFromWaveInterval = 1
MultiplierFromHumanHomePlanetCount = 1

FleetShipBudget = (base value, equivalent to a single mkI fighter on high caps) = 1
FleetShipBudget *= AITechLevel = 2
FleetShipBudget *= MultiplierFromWaveInterval = 2
FleetShipBudget *= MultiplierFromHumanHomePlanetCount = 2
FleetShipBudget *= EffectiveAIP = 20
FleetShipBudget *= player.GetHandicapMultiplier() = 20
preMinStep_DifficultyMultiplier = player.AIDifficulty / ( 13 - player.AIDifficulty ) = 3.33
FleetShipBudget *= preMinStep_DifficultyMultiplier = 66.67
FleetShipBudget *= rand( 0.8 , 1.1 ) = 56.92
FleetShipBudget *= player.AITypeData.WaveSizeMultiplier = 56.92
postMinStep_DifficultyMultiplier = (if <= 3 then 1, <= 4 : 1.5, <= 5 : 1.75, <= 6 : 2, <= 7 : 2.25, <= 9 : 2.5, <= 9.3 : 2.9, <= 9.6 : 3.3, <= 9.8 : 4, 10 : 5) : 5
FleetShipBudget *= postMinStep_DifficultyMultiplier = 284.58
= For a final FleetShipBudget for this wave's fleet ships of 284.58

percentChanceParasiteWave (from AI type) = 0.03
percentChanceBomberWave (from AI type) = 0.06
percentChanceStealthWave (from AI type) = 0.06
= actually chosen WaveType: Bomber

AIP is between the current tech level's AIP threshold and the next tech level's, so promotionRatio (percent of fleet ships to be bought from next tech level up) = 0.02
Bomber waves try to spend 90% of the strength budget on types from the Bomber group
* called PickUnitsForWave with: BomberII
** Picked 13 BomberII @ 20 each = 260.06
Some strength budget remaining from previous pick phases (if any), so spend what remains on types from the Normal group
* called PickUnitsForWave with: FighterII, BomberII, MissileShipII, EyeBotII, MicroFighterII
** Picked 2 BomberII @ 20 each = 40.01
= After picking fleet ships, FleetShipBudget is now -15.49

player.AITypeData.StarshipBudgetMultiplier: 0.25
= StarshipBudget = originalFleetShipBudget * player.AITypeData.StarshipBudgetMultiplier = 71.14
* called PickUnitsForWave with: Flagship, AIRaidStarship, DreadnoughtII, LeechStarshipII, BomberStarship, ZenithStarshipI, SpireStarshipI
** Picked
1 SpireStarshipI @ 192 each = 192
= After picking starships, StarshipBudget is now -120.86


== Wave total ships: 16
TypesForDirectAdd count by type:
BomberII : 15
SpireStarshipI : 1
TypesForCarrierAdd count by type:
== For a grand-total Strength value of 432

On 7.009 (summary: 22 MkII Fighters, plus 3 starships)
Code: [Select]
4/10/2014 8:37:18 PM (7.009)
-----------------------------------
Starting CreateHomogenousWaveToPlanet at Game Time: 0:06:41 ; Player.AIType: Vanilla ; Player.AIDifficulty: 10 ; AIProgressionLevel: 10 ; AITechLevel: 2
WaveSize = MultiplierFromWaveInterval * MultiplierFromHumanHomePlanetAndChampionCount: 1 * 1 = 1
workingShips = ( effectiveAIP * player.AIDifficulty ) / ( 13 - player.AIDifficulty ) : 33.33
workingShips *= FInt.FromParts( 0, Configuration.NonSimRandom.Next( 800, 1100 ) ) : 31.7
workingShips = Max(workingShips,34 * handicap_multiplier) :34
Inside AdjustNumberShipsFromAIType, multiplier: 1
after AdjustNumberShipsFromAIType call, workingShips :34
numberShips = workingShips.IntValue :34


4/10/2014 8:37:19 PM (7.009)
-----------------------------------
Receiving AddInboundWave from Host at Game Time: 0:06:41
WaveSize factor: 1
Raw Units Dictionary Entries:
FighterII => 34
ZenithStarshipI => 1
BomberStarship => 2


4/10/2014 8:37:18 PM (7.009)
-----------------------------------
Performing first CheckWave with size factor of 1 on wave at Game Time: 0:06:41

CheckWave: populating count of FighterII with base magnitude of 34
numberUnits = kv.Value * this.WaveSize : 34
after applying the ship-type-specific cap multiplier (which includes the unit-cap-scale multiplier), numberUnits : 4.25
after applying UsefulnessInAIWaveMultiplier if any, numberUnits : 5.1
after applying Mark-based multiplier if any, numberUnits : 4.59
after applying difficulty-based multiplier (if <= 3 then 1, <= 4 : 1.5, <= 5 : 1.75, <= 6 : 2, <= 7 : 2.25, <= 9 : 2.5, <= 9.3 : 2.9, <= 9.6 : 3.3, <= 9.8 : 4, 10 : 5), numberUnits : 22.95
after applying at-least-one rule, numberUnits : 22.95
numberUnitsInt = numberUnits.IntValue : 22

CheckWave: populating count of ZenithStarshipI with base magnitude of 1
skipping most calc step due to starship type
after applying at-least-one rule, numberUnits : 1
numberUnitsInt = numberUnits.IntValue : 1

CheckWave: populating count of BomberStarship with base magnitude of 2
skipping most calc step due to starship type
after applying at-least-one rule, numberUnits : 2
numberUnitsInt = numberUnits.IntValue : 2

Wave total ships: 25
TypesForDirectAdd count by type:
FighterII => 22
ZenithStarshipI => 1
BomberStarship => 2
TypesForCarrierAdd count by type:
totalStrengthRepresented = 736

So on 7.009 it rolled a 0.95 on the random roll, and had a 1.2 usefulness-in-wave multiplier (because fighter), and generated 22 triangle ships.  22 / 1.2 / 0.95 ~= 19.3

And on 7.018 it rolled a 0.85 on the random roll, and had a 0.8 usefulness-in-wave multiplier (because bomber), and generated 15 triangle ships.  15 / 0.8 / 0.85 ~= 22

So it was definitely worth a look to make sure it hadn't gone sideways somewhere, but there's actually had a somewhat higher base magnitude now.

Not sure how you managed 14 frigates, though, as those have the same usefulness-multiplier as fighters.

Of course, it came out kind of anemic on the starships, as I did heavily change how those go.  However, I suspect that the new starship code will make up for its early laxity later on...


Quote
9 Zenith starships. Mk III.
Ah yes, there it is.  Though that's due to the Starfleet Commander, I'm sure.


Quote
The second counterwave had a spire dreadnought. It wiped a lot of my fleet, but they are quickly replaced.
Nonetheless, a DN sounds... uncomfortable.
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Faulty Logic

  • Hero Member Mark III
  • *****
  • Posts: 1,194
  • Bane of the AI
Re: Red is not My Favorite Color
« Reply #2 on: April 12, 2014, 07:27:46 am »
Quote
For all marks, you mean?
Yes.

Quote
kind of anemic on the starships
I don't consider 4 mkII starships at 5:30 anemic. Also, see later for why you should not, for the love of god, make it send more starships.

Quote
Nonetheless, a DN sounds... uncomfortable.
It was. All fleetships and most starships were wiped out. It died before it melted through the home forcefields, though.

===Operation Parallel Processing====
Jumpships + bomber starships knock out the co-processors very quickly.
AIP = 73 (actual AIP is -30ish)
1 massive AI ships (3960 S) are en route to your planets
? ? ?

But I checked for raid engines, and deepstrike just adds threat...

Turns out this was my first encounter with a Hunter/Killer factory. And before I can hunt him down, a CPA announced with 800 ships.

And a 300-bomber wave from Heroic Spire Hammer, with 19 starships. Resolves into:
8 siege starships mkII
2 neinzul frigates
2 neinzul destroyers
1 neinzul cruiser
3 spire frigates
2 spire destroyers
1 spire DN        <=======  Outranges gravity turret mkIII

The CPA and H/K join the fun. Turns out there was also a hybrid swarm, and six heroic human destroyers hanging out from Heroic's spawn-champs-into-threatfleet behaviour.

I used a martyr on the bombers, and a continual stream of fire, enclave drones and fireflies on the dreadnaught. Once that fell, the other ships had to deal with gravity. They did fight all the way to outside the home forcefield net, but could never quite kill anything what with the space dock cranking out decoy drones, supported by all my engies.

A nerf, I tell you. Enclave starships need a nerf. GravIIIs could use one, too.

I rebuild quickly, and also reach merc caps. I need more defenses.

=== Operation Homeland Security====
And I want to make room for three cities. I'm leaving two planets for now, one a RudeGesture (waves to anywhere could be a straight game-over at the moment) and the Superterminal (which I'll hack once I pull a decent FS fleet together).

I quickly conquer the other 8, bringing AIP to 170. This gives me the missile turret controller and mkV tigers.

All Unlocks:
Engi II
Mil III
Enclaves III
FlagshipII
Zenith I
Leech II
FortressIII
Mod fort
Gravity III
HBCIV

Gametime 2:45

Waves are nasty.
Really nasty.

Like that 50-Zenith starship scenario? Happened.

96 assorted starships? Repelled.

How about 3 spire dreadnoughts?
Yeah, that happened too (thank goodness for cloaked rams).

I sent out the Negotiators to become destroyers before the coming exo.
Then, I smash a path to and conquer the nearest artillery golem.
Repaired and returned, decolonized.

A brief aside:

Golems need a huge buff. They die fast (even heavily escorted, it got down to 12%).
I think a good starting point would be 10x health and 2x damage (1/2 reload for arti). Most golems have trouble taking down a mkII planet.

I'm only ever getting edge-case use out of them. In this particular case:
%#&*! Kill the spire dreadnoughts! NOW!

I have the spire modular forts built by the time the first exo arrives.
But thing's didn't start to go wrong until the wave arrived. Three dreadnoughts again (used another martyr on the 700 bombers), and because of how they're built, it takes nine artillery shots to kill them all. By which point they had shredded my gravity line all the way to the home ff net.

Hybrids and threatfleet champs pour in, and a new horror is revealed. An exo group not targeted for home:

It just cruises past while the defenses are otherwise occupied, and kills a command station.

Quote
Shark B 4/0
Oh. Right.
The sharkwave also cruises past, killing a different command station. And another.

At this point (primary exo almost completely dead, heroic hammer wave down to starships, some enemy chaff still milling about my homeworld, and no gravity net), the starfleet commander joins the party, announcing 300ish parasites and 82 starships. Mk IV.

Time to call in the mercs. And various stolen ships. And the dyson gatlings I won. They kill the second and third sharkwaves, and I'm able to get the gravity network back on in time for the next wave.

Wave molassified and killed. Situation stable. The plan is to do two more nebula scenarios and become a cruiser, build three cities worth of FS, and ride the Superterminal for a while.

Observations
The lost Zentih tech used to make golems: tissue paper.

Shadow battleship (7000) > Spire f#$%ing Dreadnought (4950) ?

Decoy drones: surprisingly actually a bit useful on occasion.

Gravity: perhaps a bit more nerf? I suggest flattening the ranges at the mkII's, and buffing their health a bit to gentle the blow a bit. Or add logic like tractors: anything in the grav well can shoot at the grav turret.

Waves: Oh the pain.
But aside from the AI types, I think the starship scaling is a bit overboard right now. The starship portion is more threatening than the fleetship portion, despite getting only ~1/4th the budget. This suggests tuning internal starship costs.

I don't think I realized how important the Zenith trader was to really robust chokepoints. I'd have 600,000 more energy, a couple OMDs, and, most importantly, a BHM. If I lose, I'm going to try the same settings except for enabling him, and upping the trains to 4 to compensate somewhat.

State of Empire:
« Last Edit: April 12, 2014, 07:30:34 am by Faulty Logic »
If warheads can't solve it, use more warheads.

Offline Toranth

  • Hero Member Mark III
  • *****
  • Posts: 1,244
Re: Red is not My Favorite Color
« Reply #3 on: April 12, 2014, 09:04:33 am »
Observations
The lost Zentih tech used to make golems: tissue paper.

Shadow battleship (7000) > Spire f#$%ing Dreadnought (4950) ?

Decoy drones: surprisingly actually a bit useful on occasion.

Gravity: perhaps a bit more nerf? I suggest flattening the ranges at the mkII's, and buffing their health a bit to gentle the blow a bit. Or add logic like tractors: anything in the grav well can shoot at the grav turret.
It's hard to do the same comparisons with the Spire Dreadnaught as with the Shadow ships, because so much of the Spire stuff has 0.1 Command Grade hull modifiers.  Still, I tried.
I used a fully deckout out Spire Deadnaught, with 2 Mk IV shields, 2 Mk IV HBCs, 4 Mk III HBCs, 2 Mk IV Lasers, and of course, the 48,000,000 damage main gun.

Against a Mk I H/K, the DN wins, losing both shields (80 million HP) in the process.
Against a Mk II H/K, the DN loses, doing less than 50% damage.

Unfortunely, those 0.1 modifers matter here.  The full Mk IV HBC can supposedly do 4,000,000 Damage per shot.  However, anything smaller than a Golem or Forcefield will be missed by most of the 40 beams.  An H/K is only big enough to catch 20 beams - 2 million damage.  The exact count varies on range, etc, as well.  But at the preferred standoff range for both units, 20 it is.  4 sec reload.
Mk III HBCs have fewer beams that are more tightly clustered, so all 12 beams can hit easily.  That'd be another 1.2 million damage per gun.  5 sec reload.
Mk IV lasers do 80,000x5, with an 8 second reload.
And the Photon Lance's 48,000,000 damage with an 8(10) second reload.

So, in theory and without modifiers, the DN does 48,000,000 (lance) + 800,000 (2 x Lasers) + 4,800,000 (4 x Mk III HBC) + 4,000,000 (2 x Mk IV HBC w/20 beams each) = 57,600,000 damage in the intial volley.
In 10/12 seconds, after every module has fired twice =  115,200,000 damage = enough to kill an H/K Mk I.
During that time, the H/K will fire 3 times, doing 520,000 x 25  = 13,000,000 per volley times 3 volleys = 39,000,000 damage.  Shield modules have a "Structural" hull type, so H/Ks get a 2x modifier, and that explains why the DN loses all 80,000,000 HP of shields during a 1-on-1 fight.

Without modifiers, a DN does 4,800,000 DPS (lance) + 2 x 50,000 DPS (lasers) + 4 x 240,000 (Mk III HBCs) + 2 x 500,000 (Mk IV HBCs w/20 beams each) = 6,860,000 DPS.
100,000,000 / 6,860,000 = 14.6 seconds to kill an H/K (15 seconds) which is probably about correct.
160,000,000 / 4,333,333 = 36.9 seconds for an H/K Mk I to kill a DN without modifiers, 27.7 seconds with modifiers.  This matches with the 11 volleys I counted during testing.

6 volleys for an H/K Mk II to kill the DN, 20 seconds or so.  The time oddness is because at max H/K range, its shots have an almost 3 second flight time.

20 seconds of DN fire without modifiers would be 137,200,000 damage by DPS, or 2 x Lance + 2 x Lasers + 4 x HBC IIIs, + 5 x HBC IVs
= 2 x 48,000,000 + 2 x (800,000) + 4 x (4,800,000) + 5 x (4,000,000) = 136,800,000 damage.  Close enough.


So, by combat strength, a Spire Dreadnaught is worth more than an H/K Mk I, but less than an H/K Mk II. 
H/K Mk I = 3960 Strength, and 4950 Strength is 1.25 times that.
Not unreasonable.  The performance difference comes from the fact that HBCs are so good at taking out fleetships, and fleetships are about all the Human player has to counter AI threats.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Red is not My Favorite Color
« Reply #4 on: April 12, 2014, 04:06:02 pm »
Quote
kind of anemic on the starships
I don't consider 4 mkII starships at 5:30 anemic. Also, see later for why you should not, for the love of god, make it send more starships.
My comment there was in the context of the 10/10 test I did on 7.018 where it only gave me 1 starship in the first wave.  As opposed to the 10/10 test I did on 7.009 where it gave me 3 starships in the first wave.

But, as I also said, "However, I suspect that the new starship code will make up for its early laxity later on...".

And... um, yea, indeed it did.

But, in making observations on the starship counts, bear in mind one very important fact: the Starfleet Commander AI type doesn't get 1/4 of its fleet ship budget for starships.  It gets 100% of it.

Do you know offhand how many starships (not counting champs or spire cap ships, just the stuff that can be built at a starship constructor) the non-starfleet-commander AI was sending per wave?


Anyway, with extra wave components scaling, this:

Quote
10/10 Heroic/Spire Hammer // Starfleet Commander/Techno Parasite

when translated into binary, actually means:

Quote
Please take a Masterwork Adamantine Warhammer and slam it into my face.  Repeatedly.


And yet I note it has not yet won this game, which is impressive.


Quote
but could never quite kill anything what with the space dock cranking out decoy drones, supported by all my engies.
Ah yes, the Limburger Drones.  I've seen them used cheesily more often by the AI than by the humans, but the potential is there.


Quote
Golems need a huge buff. They die fast (even heavily escorted, it got down to 12%).
I think a good starting point would be 10x health and 2x damage (1/2 reload for arti). Most golems have trouble taking down a mkII planet.
Yea, they tend to pale a bit on high difficulty or against bigger superweapons.  If they're the only superweapons on the board they can still be meaningful centerpieces, like the absurd things RockyBst's been doing with that Cursed Golem in that 10-planet 10/10 thread.  But yea, in a game with spire cap ships flying around... golem?  There was a golem here?

On the other hand, the Armored Golem cannot get a 10x health buff.  It currently has 500 million hp, and 5 billion > Int32.MaxValue (we can't use Int64 for a lot of stuff because it wouldn't fit into our FInt fixed-point math; also if we start throwing billions of hp around on a single unit it's just kind of silly).

This is one of the reasons we need to downscale all the health values, so its possible to have individual units with more durability.  Of course, I could just give them some kind of super armor that reduces damage to 5% instead of 20% or whatever, but trying to avoid things like that.

But, of course, that kind of downscaling is going to involve a fairly large amount of work, and other things beat it in the 8.0 poll.  But we're making progress towards it.


Quote
It just cruises past while the defenses are otherwise occupied, and kills a command station.

Quote
Shark B 4/0
Oh. Right.
The sharkwave also cruises past, killing a different command station. And another.
I boggle that you survived.


Quote
Gravity: perhaps a bit more nerf? I suggest flattening the ranges at the mkII's, and buffing their health a bit to gentle the blow a bit. Or add logic like tractors: anything in the grav well can shoot at the grav turret.
Hmm, probably the range reduction and health buff.  The can-shoot-at-it rule would be significantly more complex (a ship keeps a reference to its tractor-er as there's only ever one, but not to gravs affecting it).

Possibly even a very substantial health buff, such that a MkIII would in many ways a MkII that can survive moderate attack.


Quote
I don't think I realized how important the Zenith trader was to really robust chokepoints.
Yea, the Traitor goodies are pretty gamechanging when used en masse.


On the Spire DN's strength cost I'm quite happy to have those use special costs like the champ units do for the Heroic.  The main questions are: what values, and why?
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Red is not My Favorite Color
« Reply #5 on: April 12, 2014, 04:10:13 pm »
It's hard to do the same comparisons with the Spire Dreadnaught as with the Shadow ships, because so much of the Spire stuff has 0.1 Command Grade hull modifiers.  Still, I tried.
Thank you for the investigation, the numbers are very interesting. 

I wouldn't have figured a DN to lose to a MkII H/K, but I suppose that's because I largely only ever see DNs amongst large fleets with tons of screening elements (which naturally get shot at first), and H/Ks tend to melt even under frigate beams when there's dozens of frigates.

I'd think a Shadow BB could take a Spire DN because it gets (iirc) a lot more modules and is generally pretty beastly.  Without even getting into the extra flexibility.  But I might be wrong on that matchup.
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Toranth

  • Hero Member Mark III
  • *****
  • Posts: 1,244
Re: Red is not My Favorite Color
« Reply #6 on: April 12, 2014, 05:57:19 pm »
It's hard to do the same comparisons with the Spire Dreadnaught as with the Shadow ships, because so much of the Spire stuff has 0.1 Command Grade hull modifiers.  Still, I tried.
Thank you for the investigation, the numbers are very interesting. 

I wouldn't have figured a DN to lose to a MkII H/K, but I suppose that's because I largely only ever see DNs amongst large fleets with tons of screening elements (which naturally get shot at first), and H/Ks tend to melt even under frigate beams when there's dozens of frigates.

I'd think a Shadow BB could take a Spire DN because it gets (iirc) a lot more modules and is generally pretty beastly.  Without even getting into the extra flexibility.  But I might be wrong on that matchup.

A normally equipped Shadow BB can do about 7.3 million DPS to the Spire Dreadnaught, meaning it would take about 23 seconds to deal the 160,000,000 damage to kill it.  However, during that time, the Shadow BB will take about 150,000,000 damage from the DN - well beyond the base 60,000,000 HP + 18.75 million HP shields.
A Shadow BB using Shadow Shields gets an extra 100,000,000 HP and could probably take a DN in a fair fight... especially if the BB kept moving, and thus could dodge much of the DN's Photon Lance.
A perfectly equipped Shadow BB (Zenith, with 12 Mk V Lasers and 8 Mk V Missiles) will do 9,465,000 DPS, taking 16.9=17 seconds to kill the DN.  It'd take about 134,000,000 damage during that time, or twice its HP (that build has no Shields).  Using Shadow Shields, again, it'd work, but not without.

I think it comes down to that the Shadow ship's modules fire much faster and hit a little harder, but the DN has a lot more HP.  Also, the H/Ks can dodge half the Mk IV HBC's beams, while the Champions cannot.  The DN's shield modules being 40 million HP each vs 18,750,000 for the Shadow ships is another big difference.  Even if the Shadow BB goes all shields and the DN all weapons, the DN still wins.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Red is not My Favorite Color
« Reply #7 on: April 12, 2014, 06:22:57 pm »
Ah, ok.  So the Shadow BB just isn't in the same weight-class HP-wise as the Spire DN, though its DPS is comparable (indeed, looks like it's generally better).

But the Spire DN does worse vs H/K's than a Shadow BB, right?  Because the Spire's weapons are massively hampered vs command-grade hulls?

Anyway, overall this would seem to argue for a Spire DN being priced at least the same as a Shadow BB (7000, when the modules aren't counted, as in wave-spawning), no?


Basing these costs off of how it compares to an H/K seems reasonable for a rough pass, but FYI the unit of strength is actually "high-cap mkI fighter equivalents".  In other words, 1 MkI fighter on high-caps = 1 strength.  On normal caps it'd be 2 strength each.  And on high caps a half-size-cap ship type (like a zenith beam frigate) would be 2 strength each.

So a mk*cap is 192 strength, though I generally round it to 200.

So, in theory, a 7000-strength cost ship like a Shadow BB should be very roughly equivalent in utility (as AI ships in a wave, in this case, as that's when the 7000 cost is used for that ship) to 35 mk*caps, or 5 caps of mkV ships.  I'm curious if that's even close to true.
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Toranth

  • Hero Member Mark III
  • *****
  • Posts: 1,244
Re: Red is not My Favorite Color
« Reply #8 on: April 12, 2014, 06:43:04 pm »
Basing these costs off of how it compares to an H/K seems reasonable for a rough pass, but FYI the unit of strength is actually "high-cap mkI fighter equivalents".  In other words, 1 MkI fighter on high-caps = 1 strength.  On normal caps it'd be 2 strength each.  And on high caps a half-size-cap ship type (like a zenith beam frigate) would be 2 strength each.

So a mk*cap is 192 strength, though I generally round it to 200.

So, in theory, a 7000-strength cost ship like a Shadow BB should be very roughly equivalent in utility (as AI ships in a wave, in this case, as that's when the 7000 cost is used for that ship) to 35 mk*caps, or 5 caps of mkV ships.  I'm curious if that's even close to true.
I tried doing an actual combat simulator for "How many fighters to kill an X" a while back, but I kept running into problems.  One of these days I'll finish it...

A quick test, using a bunch of Spire Stealth Battleships Mk V (cap 5 on normal) shows that the DN can beat 30 SSBs every time, and 35 about 50% - a lot depends on positioning of the ships.
A Shadow BB is equally matched against about 25.  I'll need to try some more loadouts to see if something else works better.

Offline keith.lamothe

  • Arcen Games Staff
  • Arcen Staff
  • Zenith Council Member Mark III
  • *****
  • Posts: 19,505
Re: Red is not My Favorite Color
« Reply #9 on: April 12, 2014, 07:21:12 pm »
I tried doing an actual combat simulator for "How many fighters to kill an X" a while back, but I kept running into problems.  One of these days I'll finish it...
If you find it entertaining to do :)  I have various ways of doing math models.  There's even a really old mode in the AIW executable for just simulating fights, but it's pre-unity and I'm sure I've broken it to smithereens over the years with various shenanigans.  I'm also not terribly confident in the results of fights nobody was actually watching, as there are lots of little details like moving around mid-fight that would be glossed over.  Though even such results can be useful for doing a rough pass.  Which is better than numbers which bear no correlation to reality.


Quote
A quick test, using a bunch of Spire Stealth Battleships Mk V (cap 5 on normal) shows that the DN can beat 30 SSBs every time, and 35 about 50% - a lot depends on positioning of the ships.
A Shadow BB is equally matched against about 25.  I'll need to try some more loadouts to see if something else works better.
Very interesting, it honestly hadn't occurred to me that 35 fleet ships of any description could take on a Spire DN and possibly win.  But I guess SSBs are pretty tough-as-nails.   I imagine the Spire Railclusters are fairly nasty too.
Have ideas or bug reports for one of our games? Mantis for Suggestions and Bug Reports. Thanks for helping to make our games better!

Offline Faulty Logic

  • Hero Member Mark III
  • *****
  • Posts: 1,194
  • Bane of the AI
Re: Red is not My Favorite Color
« Reply #10 on: April 13, 2014, 02:12:42 am »
Quote
Do you know offhand how many starships (not counting champs or spire cap ships, just the stuff that can be built at a starship constructor) the non-starfleet-commander AI was sending per wave?
24 normal starships in a 1022-fighter wave.

Quote
Ah yes, the Limburger Drones.
They do nothing against photon lances.

Quote
But, of course, that kind of downscaling is going to involve a fairly large amount of work, and other things beat it in the 8.0 poll.  But we're making progress towards it.
I was hoping a golem buff could be done easily, before the rework. I thought you had said stat changes were fairly trivial at some point, though I could be mistaken.

Mr. Armored golem could go up to 2 billion HP, and quadruple/quintuple the dps, to fit within int32.

A cursed golem dies instantly to an OMD. A hive has less health than a core flagship, and can barely take a typical mkIII planet with a full swarm. Even in the absence of other superweapons, they are so expensive and fragile as to almost never be worth it.
If warheads can't solve it, use more warheads.

Offline Kahuna

  • Core Member
  • *****
  • Posts: 2,222
  • Kahuna Matata!
Re: Red is not My Favorite Color
« Reply #11 on: April 13, 2014, 03:22:23 am »
OMDs need a nerf.
EDIT: Buff Spirecraft too.
set /A diff=10
if %diff%==max (
   set /A me=:)
) else (
   set /A me=SadPanda
)
echo Check out my AI War strategy guide and find your inner Super Cat!
echo 2592 hours of AI War and counting!
echo Kahuna matata!

Offline Faulty Logic

  • Hero Member Mark III
  • *****
  • Posts: 1,194
  • Bane of the AI
Re: Red is not My Favorite Color
« Reply #12 on: April 13, 2014, 04:07:47 am »
Quote
OMDs need a nerf.
EDIT: Buff Spirecraft
I disagree. OMDs are rare, and easily counterable. Spirecraft are more complicated. Some of them are practically useless, some are close to right, some are OP.
If warheads can't solve it, use more warheads.

Offline Kahuna

  • Core Member
  • *****
  • Posts: 2,222
  • Kahuna Matata!
Re: Red is not My Favorite Color
« Reply #13 on: April 13, 2014, 05:06:17 am »
OMDs one shotting the big stuff is pretty stupid.
set /A diff=10
if %diff%==max (
   set /A me=:)
) else (
   set /A me=SadPanda
)
echo Check out my AI War strategy guide and find your inner Super Cat!
echo 2592 hours of AI War and counting!
echo Kahuna matata!

Offline Faulty Logic

  • Hero Member Mark III
  • *****
  • Posts: 1,194
  • Bane of the AI
Re: Red is not My Favorite Color
« Reply #14 on: April 13, 2014, 05:14:04 am »
OMDs one shotting the big stuff is pretty stupid.
Why? It's what they were designed to do. Though I agree they shouldn't one-shot golems, because golems are supposed to be massive and tough, but for everything else, they're fine.
If warheads can't solve it, use more warheads.

 

SMF spam blocked by CleanTalk