New quest decks

[Agetian]

@ Sloth: Yep, I meant trading when blocking - the current default AI is generally very reluctant to accept trades when blocking.
Good idea about implementing a medium aggressive AI as well! I was thinking, actually, since the AI profiles will be used for various deck strategy types (which is actually very logical!), maybe we can consider the following:

- First off, in Constructed, let the user choose the AI profile per opponent in the lobby screen (as opposed to going to the game settings and setting an overall "system-wide" profile for all opponents or choosing to completely randomize the profile).
- Once that's done, maybe we can default to a certain "safe" (current "Default"?) profile in the game settings as opposed to defaulting to a random profile, because, as you mentioned, choosing a wrong profile for the deck type (e.g., as of right now, choosing a Reckless profile for a Control-style or at least more "defensive play" style deck) can be disastrous.
- Maybe we can then rename the profile "Reckless" to "Aggro" (so it makes immediate sense as to when this profile has to be chosen for the opponent), and continue adding profiles based on various deck types (then we can implement "Midrange", "Control", etc., and maybe ultimately even expand the idea to accomodate various "Combo" type decks where the profiles will contain definitions of various card combinations that the AI will want to "prioritize" - it might be a little too early to talk about advanced things like that at the moment, but still something to consider in the future).

What do you think?

- Agetian

[tojammot]

I think it's great and perhaps quest opponents can now include specific combo decks for once, like Goblin Welder+Intruder Alarm.

[tojammot]

Here's the miracle+scrying deck I promised, a hard version would focus solely on scrying.

Sanae Kochiya 2 | Open

Main:
2 Island
2 Mountain
1 Mystic Monastery
1 Plains
3 Swiftwater Cliffs
3 Temple of Enlightenment
4 Temple of Epiphany
3 Temple of Triumph
2 Tranquil Cove
2 Wind-Scarred Crag
4 Flamespeaker Adept
2 Keranos, God of Storms
1 Sigiled Skink
3 Sigiled Starfish
1 Thassa, God of the Sea
4 Banishing Stroke
4 Thunderous Wrath
4 Vanishment
2 Bonfire of the Damned
1 Entreat the Angels
4 Reforge the Soul
1 Reminisce
2 Serum Visions
4 Temporal Mastery

Your thoughts?

EDIT: I made a very hard version of the Robin 3 deck

Robin 4 | Open

Main:
1 Mox Sapphire
1 Mox Emerald
1 Mox Ruby
4 Invisible Stalker
3 Jhessian Infiltrator
1 Curiosity
1 Latulla's Orders
2 Curse of Stalked Prey
3 Sigil of Sleep
3 Destructive Urge
2 Naturalize
1 True-Name Nemesis
1 Spider Umbra
1 Wheel of Fortune
1 Timetwister
1 Ancestral Recall
1 Time Walk
1 Bident of Thassa
2 Lightning Bolt
4 Triton Shorestalker
4 Misty Rainforest
4 Scalding Tarn
2 Wooded Foothills
1 Forest
1 Island
1 Mountain
1 Taiga
3 Tropical Island
3 Volcanic Island
1 Basilisk Collar
2 Keen Sense
1 Loxodon Warhammer
1 Rancor

[Sloth]

Here's the miracle+scrying deck I promised, a hard version would focus solely on scrying.

Sanae Kochiya 2 | Open

Main:
2 Island
2 Mountain
1 Mystic Monastery
1 Plains
3 Swiftwater Cliffs
3 Temple of Enlightenment
4 Temple of Epiphany
3 Temple of Triumph
2 Tranquil Cove
2 Wind-Scarred Crag
4 Flamespeaker Adept
2 Keranos, God of Storms
1 Sigiled Skink
3 Sigiled Starfish
1 Thassa, God of the Sea
4 Banishing Stroke
4 Thunderous Wrath
4 Vanishment
2 Bonfire of the Damned
1 Entreat the Angels
4 Reforge the Soul
1 Reminisce
2 Serum Visions
4 Temporal Mastery

Your thoughts?

Sorry tojammot, i should have mentioned this earlier: The AI is not aware of Miracle cards when scrying. In fact the scrying AI is very underdeveloped. This is everything related to scry:

Code: Select all

    public static boolean scryWillMoveCardToBottomOfLibrary(Player player, Card c) {
        boolean bottom = false;
        if (c.isBasicLand()) {
            CardCollection cl = CardLists.filter(player.getCardsIn(ZoneType.Battlefield), CardPredicates.Presets.LANDS);
            bottom = cl.size() > 5; // if control more than 5 Basic land, probably don't need more
        }
        else if (c.isCreature()) {
            CardCollection cl = CardLists.filter(player.getCardsIn(ZoneType.Battlefield), CardPredicates.Presets.CREATURES);
            bottom = cl.size() > 5; // if control more than 5 Creatures, probably don't need more
        }
        return bottom;
    }

So the AI will almost never move a card to the bottom.

Drawing cards during the opponets turn (suggested by Nordos) is the only way i see the current AI use synergies with Miracle cards.

[Sloth]

Here is a UR Chasm Skulker deck:

Arjun 2 | Open

7 Island
1 Mountain
4 Scalding Tarn
2 Polluted Delta
1 Steam Vents
4 Sulfur Falls
4 Volcanic Island
1 Swiftwater Cliffs
2 Arjun, the Shifting Flame
2 Whirlpool Drake
2 Whirlpool Rider
3 Teferi's Puzzle Box
4 Desperate Ravings
1 Noggle Ransacker
1 Jhessian Thief
4 Accumulated Knowledge
1 Timetwister
1 Dragon Mage
1 Memory Jar
1 Time Spiral
1 Wheel of Fate
1 Wheel of Fortune
1 Brainstorm
1 Obsessive Search
4 Chasm Skulker
2 Jace's Erasure
2 Niv-Mizzet, the Firemind
1 Sphinx's Tutelage

[tojammot]

Man, I really ought to update my Forge, anyways here's a wolf deck. It still feels like there's something missing.

Princess Mononoke 2/3 | Open

Main:
8 Forest
4 Savannah
4 Sunpetal Grove
2 Temple Garden
4 Windswept Heath
1 Sword of Body and Mind
3 Chameleon Colossus
1 Darkthicket Wolf
1 Lone Wolf
4 Master of the Wild Hunt
3 Sacred Wolf
1 Tolsimir Wolfblood
2 Tundra Wolves
1 Wandering Wolf
4 Watchwolf
4 Witchstalker
1 Wolf-Skull Shaman
4 Woodland Changeling
2 Wren's Run Packmaster
4 Young Wolf
1 Raised by Wolves
1 Howl of the Night Pack

I'll give up on the miracle deck until Scrying AI has improved (Which is a long time)

[Nordos]

Maybe Mayor of Avabruck? Any other lords? (Adaptive Automaton, Coat of Arms?)
Why no Timberback Wolf? Why only one Raised by Wolves?
Wolf-Skull Shaman has ... 14 targets? Sure, it is no small number, but...
Dunno if I would go that path, or if I would prefer staying wolf only.
Master of the Wild Hunt is very good, but Wren's Run Packmaster can run into problems if no elf is preset (or one of the changelings).
You included the Changelings only to make these sub-tribals work, which is, IMO, pretty messy. Same with including White... You have 7 cards with white mana, the Tundra Wolf isn't necessary, the Tolsimir Wolfblood hits quite late and the Watchwolf can be dead cards as long as he hasn't white mana. Additionally did you include Dual- and Fetchlands in a deck, which is on the verge of medium.
Instead of going white, try using forest as a base (Howl of Night Pack and the likes).

Why no Wolfir Silverheart? It is a very powerful wolf, IMO better than Tolsimir Wolfblood ... Well.



just my 2 cents

[tojammot]

Maybe Mayor of Avabruck? Any other lords? (Adaptive Automaton, Coat of Arms?)

This is a wolf tribal, not werewolves. (Saving that for Angua von Uberwald, and AI support for Moonmist) Also, I wanted deck originality, I guess that backfired too.

Why no Timberback Wolf? Why only one Raised by Wolves?

Trying to keep the curb steady, hoping the AI with put out wolves before powering them up. I'm also experimenting with Obelisk of Urd.

Wolf-Skull Shaman has ... 14 targets? Sure, it is no small number, but...
Dunno if I would go that path, or if I would prefer staying wolf only.
Master of the Wild Hunt is very good, but Wren's Run Packmaster can run into problems if no elf is preset (or one of the changelings). You included the Changelings only to make these sub-tribals work, which is, IMO, pretty messy.

Now that I think about it, I'm inspired to make a changeling deck with tons of very powerful lords and Champions.

Same with including White... You have 7 cards with white mana, the Tundra Wolf isn't necessary, the Tolsimir Wolfblood hits quite late and the Watchwolf can be dead cards as long as he hasn't white mana. Additionally did you include Dual- and Fetchlands in a deck, which is on the verge of medium.
Instead of going white, try using forest as a base (Howl of Night Pack and the likes).

Well, Watchwolf is pretty good for its cost and Luke Skywalker 3 only runs four blue cards.

Why no Wolfir Silverheart? It is a very powerful wolf, IMO better than Tolsimir Wolfblood ... Well.

I thought it would be better to empower more wolves.

just my 2 cents

It's fine, it has been a while. I'm still tweaking the deck,

[Sloth]

Now that I think about it, I'm inspired to make a changeling deck with tons of very powerful lords and Champions.

A deck like this would have some awkward draws like playing Master of the Pearl Trident then Rageblood Shaman and then Lord of the Unreal.

Well, Watchwolf is pretty good for its cost and Luke Skywalker 3 only runs four blue cards.

I have nothing against including white, but did you consider red for Immerwolf?

Why no Wolfir Silverheart? It is a very powerful wolf, IMO better than Tolsimir Wolfblood ... Well.

I thought it would be better to empower more wolves.

Both are solid cards and Wolfir Avenger is also solid.

[tojammot]

A deck like this would have some awkward draws like playing Master of the Pearl Trident then Rageblood Shaman and then Lord of the Unreal.

Hmm, perhaps a challenge with Conspiracy as a starting card?

I have nothing against including white, but did you consider red for Immerwolf?

Strangely enough I didn't, although I did consider red for Taurean Mauler and Pyreheart Wolf.

Both are solid cards and Wolfir Avenger is also solid.

Gonna compete with Sacred Wolf and Witchstalker.

[tojammot]

I'll put that on the backburner for a while since I just discovered Willbreaker recently. It's a fun card, all I need is a flavorful character to name this deck with.

Unnamed 2 | Open

3 Nimbus Maze
2 Glacial Fortress
4 Tundra
4 Flooded Strand
4 Island
4 Plains
2 Hallowed Fountain
1 Celestial Colonnade
1 Blinding Souleater
1 Deepwater Hypnotist
1 Errant Doomsayers
1 Gideon's Lawkeeper
1 Goldmeadow Harrier
1 Heliod's Emissary
1 Kor Line-Slinger
1 Master of Diversion
3 Minister of Impediments
2 Niblis of the Urn
1 Sentinel of the Eternal Watch
1 Sidar Jabari
1 Stormscape Apprentice
4 Willbreaker
1 Curse of Inertia
1 Debtor's Pulpit
2 Flood
1 Ghostly Prison
1 Prison Term
1 Web of Inertia
1 Tamiyo, the Moon Sage
2 Disenchant
2 Divination
1 Force Spike
1 Rebuff the Wicked
1 Rethink
1 Spell Snare

[Nordos]

Dunno if the AI can use it, but maybe something like Retribution of the Ancients? As to speak, always using it by removing 0 counters after Willbreaker hits the board...

[tojammot]

Does the AI even use it like that? And adding a third color might strain the deck's performance.

[Nordos]

That was exactly my question, sorry if you misunderstood
I wasn't planning on adding a third color, but just asking whetever such cards would be possible (that one was the first I could think of which could be abused in such fashion)

EDIT: Cauldron of Souls would work so much better, but ... would the AI target opponents creatures with it?

[tojammot]

Ah. Knowing the AI, I don't think so. Same goes for the Cauldron.

Unless of course, specific AI interactions are coded for Willbreaker.