Author

OldTrees · « **Reply #276 on:** January 03, 2013, 12:27:17 pm »

Quote from: Vangelios on January 03, 2013, 10:05:30 am

Quote from: storyteller on January 03, 2013, 02:53:55 am
which is worse, a shield that stops creatures that dont do spell damage or have momentum for three turns, or a shield that stops everything including spell damage from creatures and momentum for only one turn?

there is a remedy for your ails. Dim shields is strong, but defeated in many ways. If anything, we should have shields with higher DR, not reducing the effectiveness of a useful, but hindered shield. Its simple. Out there in the realms of elementals, some of them can phase out for a short amount of time, and not be damaged by physical forces. Either A) wait them out, B) use Momentum to hit them anyway, or C) use spells and beat them at their own game.

I havent had trouble with a dim shield in at least 6 months. Its not that powerful in the end.

Quote from: OldTrees on January 03, 2013, 04:08:38 am
I would like to end that line of thought by admitting that I too feel that Dim Shield is OP. However I do not have a good argument to allow me to conclude that. I do know that Dim Shield can be balanced if it is not already. Intuitions from others that I trust have recommended a 2 turn duration for a cheaper cost.

I'm realizing that the course of the discussion, is going to "dim the shield can be defeated or no" and should be " shield dim is balanced this with others or not"

The cost of dim shield should be 12 , because 15 is a miracle, and not often in less than 3 turns our HP already running out again because the opponent has creatures with high damage, dim blocks DR infinite, and 3 turns is finite but it is very long. And even costing 12 , it is often better than miracle in some situations.

Dusk Shield doubles the till death doom clock.
Back when I played Mono Aether (Pre SoW) I usually stated chaining Dimensional Shields when I had ~4 turns remaining and I chained ~3 Dim Shields
So I increased my personal doom clock from 4 to 16 and paid 18

+ Shield slot + 3 cards
Dusk Shield would have increased the doom clock from 4 to 8 and cost 6

+ Shield slot + 1 card
So
x4/+12 cost 18

+ S + 3 cards ~= 27
x2/+4 cost 6

+ S + 1 card ~= 10
I think the number of turns is a more accurate measurement than the multiplier of the doom clock
27/12 vs 10/4
2.25 < 2.5
For Dimensional Shield to equal Dusk Shield in this equation it would have to cost
2.5 = 30/12 = 21

+ S + 3 cards = (7

+ 1 card) per Dimensional Shield + the shield slot
7

+ 1 card

Note: I used the weaker Dusk Shield instead of the ~67% DR 1 Ice Shield.

Cardplayer · « **Reply #277 on:** January 04, 2013, 02:50:05 am »

No need for a nerf to dim shield. I've hit mono-aether multiple times and carrying some PC in the deck usually deals with the shield quite nicely. Even if not readily available at the moment, momentum makes the shield virtually worthless without killing momentumed creature. It's fine the way it is and I feel a nerf would unnecessarily damage the power of stalls and mono-aether as the shield is already reasonably expensive for a shield.

kimham8a · « **Reply #278 on:** January 04, 2013, 03:08:58 pm »

Quote from: Laxadarap on January 02, 2013, 10:03:20 pm

Quote from: Jenkar on January 02, 2013, 10:00:17 pm
Quote from: kimham8a on January 02, 2013, 09:44:32 pm
All elements can make a mono deck and beat mono with a decent winrate (except ). Therefore we don't need to add soft PC because all elements already have options to beat mono . People just don't use them because it makes their decks less efficient vs other decks.
This being actually a potential problem.

Not a potential problem as far as I can see, even though they call can beat it, that is a mono deck going into mono aether knowing exactly what to build against, all aether has to do is change it's deck and it could beat 90% of the decks.

Aether's problem is low damage output. Any mono deck that can heal beats mono aether. Any mono deck with a decent shield beats mono aether (entropy, water). Air could beat mono aether stalling with fog and a one turn break in the dims with skyblitz. Earth can use earthquake, and

's cards are VERY high cost. Now what element doesn't beat mono aether no matter what mod of it you use? And the cards I say the monos should use aren't so bad against other decks either, it's just that nobody happens to use them when they see a mono aether.

Captain Scibra · « **Reply #279 on:** January 06, 2013, 05:36:52 pm »

Lol, the Salvagers now make this OP. Self PC greatly increases your 'phase count'.

OldTrees · « **Reply #280 on:** January 06, 2013, 07:37:49 pm »

Quote from: Captain Scibra on January 06, 2013, 05:36:52 pm

Lol, the Salvagers now make this OP. Self PC greatly increases your 'phase count'.

False.
Salvagers only work if the Opponent was destructive. Self PC does not work.

DarkBaron12390 · « **Reply #281 on:** January 07, 2013, 12:09:41 am »

Quote from: OldTrees on January 03, 2013, 12:27:17 pm

Quote from: Vangelios on January 03, 2013, 10:05:30 am
Quote from: storyteller on January 03, 2013, 02:53:55 am
which is worse, a shield that stops creatures that dont do spell damage or have momentum for three turns, or a shield that stops everything including spell damage from creatures and momentum for only one turn?

there is a remedy for your ails. Dim shields is strong, but defeated in many ways. If anything, we should have shields with higher DR, not reducing the effectiveness of a useful, but hindered shield. Its simple. Out there in the realms of elementals, some of them can phase out for a short amount of time, and not be damaged by physical forces. Either A) wait them out, B) use Momentum to hit them anyway, or C) use spells and beat them at their own game.

I havent had trouble with a dim shield in at least 6 months. Its not that powerful in the end.

Quote from: OldTrees on January 03, 2013, 04:08:38 am
I would like to end that line of thought by admitting that I too feel that Dim Shield is OP. However I do not have a good argument to allow me to conclude that. I do know that Dim Shield can be balanced if it is not already. Intuitions from others that I trust have recommended a 2 turn duration for a cheaper cost.

I'm realizing that the course of the discussion, is going to "dim the shield can be defeated or no" and should be " shield dim is balanced this with others or not"

The cost of dim shield should be 12 , because 15 is a miracle, and not often in less than 3 turns our HP already running out again because the opponent has creatures with high damage, dim blocks DR infinite, and 3 turns is finite but it is very long. And even costing 12 , it is often better than miracle in some situations.
Dusk Shield doubles the till death doom clock.
Back when I played Mono Aether (Pre SoW) I usually stated chaining Dimensional Shields when I had ~4 turns remaining and I chained ~3 Dim Shields
So I increased my personal doom clock from 4 to 16 and paid 18 + Shield slot + 3 cards
Dusk Shield would have increased the doom clock from 4 to 8 and cost 6 + Shield slot + 1 card
So
x4/+12 cost 18 + S + 3 cards ~= 27
x2/+4 cost 6 + S + 1 card ~= 10
I think the number of turns is a more accurate measurement than the multiplier of the doom clock
27/12 vs 10/4
2.25 < 2.5
For Dimensional Shield to equal Dusk Shield in this equation it would have to cost
2.5 = 30/12 = 21 + S + 3 cards = (7 + 1 card) per Dimensional Shield + the shield slot
7 + 1 card

Note: I used the weaker Dusk Shield instead of the ~67% DR 1 Ice Shield.

I hold a B.S. degree in Mathematics, and I honestly have no idea what in the hell your arithmetic is doing. Do you have any units and explanations? If your calculations don't make sense to a mathematician...

OldTrees · « **Reply #282 on:** January 07, 2013, 02:02:43 am »

@DarkBaron

To use a card you must:
1) Draw the card [Draw = 1 card]
2) Pay the quanta [6

]
3) Have an available slot [Shield Slot = S]

1 card is theorized to be approximately equal cost to 2

X

is theorized to be approximately equal cost to X

Using the shield slot is theorized to be approximately equal cost to 3

When I played Mono Aether (pre SoW) I used 3 Dimensional Shields on average to delay my death by 9 turns. (the +12 was an error)
This cost drawing 3 cards, 18

and my shield slot. [3 cards + 18

+ Shield Slot ~= 27

]
If I had played Dusk Shield I would have delayed my death by 4 turns.
This cost drawing 1 card, 6

and my shield slot. [1 card + 6

+ Shield Slot ~= 10

]

Benefit / Cost
Dim: +9 turns / 27

~= 0.33 turns/quanta
Dusk: +4 turns / 10

= 0.25 turns/quanta
Since 0.33 > 0.25, Dimensional Shield is more efficient.
If Dimensional Shield were just as efficient as Dusk Shield (0.25) then it would cost X

each
0.25 turns/quanta = 9 turns / (3X

+ 9

)
3X + 9 = 36
3X = 27
X = 9

So I should have concluded that Dimensional Shield should cost 9

DarkBaron12390 · « **Reply #283 on:** January 07, 2013, 02:20:18 am »

That is much easier to follow, and I was merely trying to follow the post than take a stance. There are two inconsistencies I see. Firstly, 4/10 = 0.40... secondly, it seems the shield constant is inconsistent for dark and aether, despite it being a function of only the shield slot itself? Why is it used one time for each quantum, in spite of the fact we used three aether shields and one dark shield? It seems like a random 3

thrown in. For three shields, there must be an available slot each time to me. Otherwise, it's a fixed constant meant simply to say you have a shield slot. So I feel it should be be a total of 33

played vs. 10

, or to nix the shield overall for 24

and 7

.

Your calculations yield 0.33 vs. 0.40, which are pretty similar, yet asserting darkness is more efficient.
The former of my methods yields, using your parameters, 9/33 ~=.273 vs. 0.25 which is similar enough.
The latter yields 9/24 = 0.375 vs. 4/7 ~=0.571, which says darkness is much more efficient.

Maybe I'm just not seeing where the information comes from and don't see why a shield slot constant ought to behave the same for three shields used vs. one shield used.

OldTrees · « **Reply #284 on:** January 07, 2013, 08:48:33 am »

Thank you for catching the copy paste error. 4/10 should have been 0.4. That would mean that Dusk is more efficient and thus Dimensional Shield need not be nerfed.

The shield cost is when you are prevented from using the shield slot for something else. (an opportunity cost) It took 3 Dimensional Shields to occupy the slot since they have a duration of 3 turns. Thus the 3 Dimensional shields resulted in preventing me from using my slot just as 1 permanent shield would have prevented me.

PS: Thanks for correcting my math errors.

jawdirk · « **Reply #285 on:** January 07, 2013, 10:48:23 am »

Ok, I did some actual science. Experimental setup:

The cards:
Pillar.new - (unupped

)
Creature.new - 5 attack for 5

DimShield.new(x,y) - y

reduces damage to 0 for x turns
DuskMantel.new - 6

- reduces damage by exactly half permanently

The decks:

Spoiler for Hidden:

Simplifying (a.k.a. lazy) assumptions:
I did impliment mulligan, but both players start with 7 cards in hand. The rush deck always goes first, the test deck always goes 2nd. Dusk Mantel always allows through exactly half of the damage. I didn't really tune any of the decks -- as far as I can tell it shouldn't matter. I did run numbers with -2 pillar +2 creature in all the decks (including the rush deck) and it the results were similar but slightly greater damage by the test decks across the board. AI is vastly simplified. Dim shield chaining begins at < 50 HP if the quanta is available. Dusk Mantels are cast ASAP. Shields are always cast before creatures if the quanta is available.

The Experiment:
How much damage can a 30-card deck using either Dusk Mantel or any of a number of modified dim shields do to a rush deck before the rush deck either kills the deck, or decks out. So that means that dim shield decks only get at most 22 turns to do damage, even if they chain perfectly through the whole game.

The results (100K Monte-carlo simulations for each, rounded to nearest damage point):

Spoiler for Hidden:

My interpretation:
I think this shows that dim shields are clearly OP compared to Dusk Mantels. Yes, this is a quanta-rich test, but as I said, I tried it with 2 fewer pillars everywhere, and nothing much changed. Note that even costing 10 and lasting only 2 turns, they are still more effective than dusk shield by a large margin. I'd love to see more results. Feel free to tweak the decks and run more simulations; the ruby code is below.

Possible future experiments:
Test the effect of adding deflagration to the rush.
Test the effect of adding poison to the rush.

Sample verbose of dusk shield:

Spoiler for Hidden:

Sample verbose of dim shield:

Spoiler for Hidden:

The ruby code (could have bugs, beware, could eat your cat, run at your own risk, etc.):

Spoiler for Hidden:

Code: [Select]

def log *strings
  # uncomment to output blow-by-blow
  #puts *strings
end

class Game
  attr_accessor :player1, :player2, :turn_count
  def initialize(player1, player2)
    @turn_count = 0
    @player1, @player2 = player1, player2
	@player1.game = self
	@player2.game = self
  end
  
  def turn
    @turn_count += 1
	log "Turn #{@turn_count}:::"
    @player1.turn(@player2)
	@player2.turn(@player1) unless over?
  end
  
  def over?
    @player1.hp < 1 || @player2.hp < 1 || @player1.cards.size == 0 || @player2.cards.size == 0
  end
  
  def to_s
    "#{@turn_count} p1hp: #{@player1.hp} p2hp: #{@player2.hp}"
  end
end

class Player
  attr_accessor :hp, :quanta, :quanta_per_turn, :damage_per_turn, :cards, :hand, :name, :shield, :destroy_shield_on_turn, :game
  def initialize(name, hp, cards)
    @name = name
    @hp = hp
	@shield = 1
	@quanta = @quanta_per_turn = @damage_per_turn = 0
    @cards = cards.shuffle
	@cards = cards.shuffle if cards[0..6].all?{|c| !c.is_a?(Pillar)}
	@cards.push(Dead.new)
	@hand = []
	7.times { draw }
  end
  
  def shield?
    shield < 1
  end
  
  def destroy_shield!
    self.shield = 1
  end
  
  def turn_count
    game.turn_count
  end
  
  def draw
    hand << cards.shift
  end
  
  def turn(enemy)
    destroy_shield! if destroy_shield_on_turn == turn_count
    draw
	log "? #{name}" + inspect, enemy.inspect
	@hand.sort_by{|c| c.cast_order}.each{|c| c.try_cast(self) }
	log "Boom!: #{enemy.hp} -= #{damage_per_turn * enemy.shield} (#{enemy.shield})"
	enemy.hp -= damage_per_turn * enemy.shield
	self.quanta += quanta_per_turn
	log "! " + inspect, enemy.inspect, "\n"
  end
  
  def deck_size
    cards.size - 1
  end
  
  def inspect
    #"~#{name} hp: #{hp}" # q: #{quanta} cs: #{deck_size} dpt: #{damage_per_turn}~"
    "~#{name} hp: #{hp} q: #{quanta} cs: #{deck_size} dpt: #{damage_per_turn} h: #{hand.inspect}~"
  end
end

class Card
  def try_cast(x)
    cast!(x) if cast?.call(x)
  end
  
  def cast!(x)
    shield.call(x) if shield
	x.quanta_per_turn += turn_quanta
	x.damage_per_turn += turn_damage
	found = false
	x.hand = x.hand.select{|x| out = (x != self || found || ((found = true) && false) )}
  end
  
  def to_s
    self.class.to_s
  end
end

class Dead < Card
  def cast?; ->(x){ true }; end
  def shield; nil; end
  def turn_quanta; 0; end
  def turn_damage; 0; end
  def cast_order; 0; end
end

class Pillar < Card
  def cast?; ->(x){ true }; end
  def shield; nil; end
  def turn_quanta; 1; end
  def turn_damage; 0; end
  def cast_order; 0; end
end

class Creature < Card
  def cast?; ->(x){ x.quanta -= 5 if x.quanta >= 5}; end
  def shield; nil; end
  def turn_quanta; 0; end
  def turn_damage; 5; end
  def cast_order; 2; end
end
 
 class DimShield < Card
   def initialize(lasting, cost)
     @cost = cost
	 @lasting = lasting
   end
   
   def cast?; ->(x){ x.quanta -= @cost if !x.shield? && x.quanta >= @cost && x.hp < 50}; end
   def shield; ->(x){ log "dim shield!!!!!!!!!!!!! #{x.turn_count}"; x.destroy_shield_on_turn = x.turn_count+@lasting; x.shield = 0 }; end
   def turn_quanta; 0; end
   def turn_damage; 0; end
   def cast_order; 1; end
 end
 
 class DuskMantel < Card
   def cast?; ->(x){ x.quanta -= 6 if !x.shield? && x.quanta >= 6}; end
   def shield; ->(x){ log "dusk mantel!!!!!!!! "; x.shield = 0.5 }; end
   def turn_quanta; 0; end
   def turn_damage; 0; end
   def cast_order; 1; end
 end
 
total = 0.0
reps = 10000
# Player.new('rush', 10000, [Pillar.new]*15 + [Creature.new]*15)
# Player.new('controlrush', 100, [Pillar.new]*15 + [Creature.new]*15)
# Player.new('dusk2', 100.0, [Pillar.new]*15 + [Creature.new]*13 + [DuskMantel.new] * 2)
# Player.new('dusk3', 100.0, [Pillar.new]*15 + [Creature.new]*12 + [DuskMantel.new] * 3)
# Player.new('dusk4', 100.0, [Pillar.new]*15 + [Creature.new]*11 + [DuskMantel.new] * 4)

# Player.new('dim3-6', 100, [Pillar.new]*15 + [Creature.new]*9 + [DimShield.new(3,6)] * 6)
# Player.new('dim3-10', 100, [Pillar.new]*17 + [Creature.new]*7 + [DimShield.new(3,10)] * 6)
# Player.new('dim2-6', 100, [Pillar.new]*15 + [Creature.new]*9 + [DimShield.new(2,6)] * 6)
# Player.new('dim2-10', 100, [Pillar.new]*17 + [Creature.new]*7 + [DimShield.new(2,10)] * 6)
# Player.new('dim1-6', 100, [Pillar.new]*15 + [Creature.new]*9 + [DimShield.new(1,6)] * 6)
reps.times do
  game = Game.new(Player.new('shark', 10000, [Pillar.new]*15 + [Creature.new]*15), Player.new('dim1-6', 100, [Pillar.new]*15 + [Creature.new]*9 + [DimShield.new(1,6)] * 6) )
  game.turn until game.over?
  total += game.player1.hp
end
puts 10000 - total/reps

OldTrees · « **Reply #286 on:** January 07, 2013, 11:26:05 am »

Good tests jawdirk. +99rep for playtesting. I have no criticisms of the experiment.

I think this is a sufficient argument to conclude the need for a nerf.

Wow. I honestly did not think the 2,6 Dimensional Shield was that powerful.

jawdirk · « **Reply #287 on:** January 07, 2013, 06:31:11 pm »

Quote from: OldTrees on January 07, 2013, 11:26:05 am

Good tests jawdirk. +rep for playtesting. I have no criticisms of the experiment.

I think this is a sufficient argument to conclude the need for a nerf.

Wow. I honestly did not think the 2,6 Dimensional Shield was that powerful.

Thanks, I appreciate that. I too am surprised at the effectiveness of a 2,6 dim shield. This code was all written last night, so it is still possible that there is some sort of bug. I am also curious to boost the percentage of the dusk mantel incrementally to 100% to see if we can get a nice curve.

Author

OldTrees

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1027996#msg1027996

Cardplayer

Re: dimension shield https://elementscommunity.org/forum/index.php?topic=42134.msg1028225#msg1028225

kimham8a

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1028385#msg1028385

Captain Scibra

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029301#msg1029301

OldTrees

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029345#msg1029345

DarkBaron12390

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029433#msg1029433

OldTrees

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029458#msg1029458

DarkBaron12390

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029465#msg1029465

OldTrees

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029520#msg1029520

jawdirk

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029532#msg1029532

OldTrees

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029542#msg1029542

jawdirk

Re: dimension sheild https://elementscommunity.org/forum/index.php?topic=42134.msg1029609#msg1029609