Yet Another Shandalar Enemy Deck Set

[Hadsingrinsen]

I decided to overhaul all 55 enemy decks for Shandalar, to add more variety and make the game more challenging. This is the result of that time-consuming decision.

This is not a criticism of the earlier works by jiansonz and podbelski. Rather, I wanted a to have a different style of play and more variety. I made a more radical departure from the original decks than jiansonz did, and almost all of them are completely rewritten. There is more variation in the card choices than in jiansonz's version, but probably less than in podbelski's. The difficulty level is much higher than both, especially in the beginning, and I did not pull many punches, the AI being as poor as it is. I hope to extract some cleverness from the player, so that it will no longer be possible to carry the same powerful deck around everywhere. Perhaps you'll let me know the extent to which I've succeeded, since don't have any free time left to play the game myself.

I would like to consider this version 0.1, and will make further revisions as needed (if anyone gives me feedback.) Please do let me know if:

- You find AI bugs. I made an effort to remove those cards which I have observed misbehaving, but it's probable that more need to be expunged. I left a few cards in that I think are likely to be buggy, such as Lightning Bolt, in hope that people will tell me if they are. Even better, if someone could get to the bottom of the mid-combat targeting bug so that we could put cards like Giant Growth back in, that would be very nice.

- You find some opponents too easy. I playtested many of the decks but not all of them, and I only know my own strategies. I am particularly interested if you find upper level enemies easier than lower level ones, especially if you have ideas on how to make them harder. I am not particularly interested in making the lower level enemies easier.

- You have ideas about how cards should be distributed in the game. I made an effort to avoid placing cards that appear as dungeon prizes in decks that give ante, but it's possible there were some oversights. I also have not played the latter half of the game for many years, and don't recall if the dragons and genies give ante. If not, I may want to put some restricted cards in their decks.

- You want to tell me what strategies you used to beat the various enemies. This will help me improve them, and I may also copy your strategy with permission. I am particularly interested in reforming the great wizard decks, which currently are powerful but not highly polished, and players who have played a certain mono-color deck for a long time probably know better than I do how to build one.

If you play with these decks I suggest treating the following cards as restricted, that is keeping no more than one in your deck at any time.

Ali from Cairo
Ancestral Recall
Berserk
Black Lotus
Braingeyser
Candelabra of Tawnos
Channel
Copy Artifact
Demonic Tutor
Fastbond
Feldon's Cane
Forcefield
Fork
Hurkyl's Recall
Ivory Tower
Library of Alexandria
Mind Twist
Mishra's Workshop
All Mox Gemstones
Regrowth
Sol Ring
Time Twister
Time Walk
Wheel of Forture
All dual lands (Tundra, Taiga, etc.)

My decks abide by these restrictions in all but a couple cases. Other than Arzakon and the great wizards, the only decks that use restricted cards at all, except for dual lands, are the henchmen to the great wizards (Summoner, High Priest, etc.) The genies and dragons are allowed to treat the dual lands as unrestricted and take four of each, and the great wizards are allowed four copies of restricted cards. (However, I gave them all 80-card decks, so it's as if they have only three copies each.) Arzakon's deck contains mostly restricted cards and is quite formidable - sometimes it can even beat the Arch Angel.

I hope that you won't peek at the decks, as finding out what's in them is part of the game, so I will never write a blow-by-blow of how I designed them, like jiansonz did. I also hope you won't save and load, although perhaps it's not my business. I feel sorry for the poor AI, which lacks your advantage of human intelligence.

On that note, you should also treat Juggernaut as if it were unable to block, because the AI believes this, and if the Fungus Master shoots its 2/2 Fungusaur with Cuombajj Witches, be a gentleman and assign the other damage point to the opposing player. Siren's call apparently does not work when used by the player, so don't include it in your deck. If you'd like to add to this running list of trivialities, do not hesitate to contact me. I made this set for the original Microprose version, but it should work equally well with Shandalar 2012 (please let me know if otherwise.) There are some minor rule differences, the most notable ones involving creature types. For example there are only two zombies in the original - Scathe Zombies and Drowned. Walking Dead is a Walking Dead type. This means for example that the giant swampwalking army that the Necromancer used to build using Zombie Lord doesn't work in my version. It also makes kobolds much weaker. It's up to you if you want to abuse the new creature types, but my decks don't.

Incidentally, I was rather liberal with changing color combinations around. Many of the decks are now multicolor, or have splashes of another color. In part this was necessary in order to avoid having people stock up on Northern Paladins and Karmas and march into the swamp expecting an easy win, but more often I did it for purely thematic reasons, as in the case of the Cleric. I did make an effort to make all the decks thematically appropriate, sometimes sacrificing power to do so. A marginal exception was the Sedge Beast - I really couldn't keep on theme because I didn't know what a Sedge Beast was. (Try looking up "sedge" on wikipedia.) I decided it was a freak of nature. If I recall there was also a dragon deck that fell slightly short of my expectations, because there are simply not enough dragon cards.

Some advice on how to play the game: early on you will have to avoid most of the opponents, who are more likely to take your cards than give you theirs. You should probably go from town to town judiciously buying and selling cards and running away from everything in sight until you can put a passable deck together. Some of the early enemies are very powerful - don't fight the Witch. The Conjurer has an extremely good deck that the AI unfortunately mishandles, and he ends up being a moderate challenge. Your best bets for early battles are the Seer, the Druid, the Priestess, and maybe the Sorceress, but they are not exactly pushovers by the old standard. The Priestess's army of walls is no more.

With that, here it is.

To install it, first back up your Decks folder (...\MagicTG\Magic\Decks\) and then overwrite its contents with the unzipped files.

[jiansonz]

I decided to overhaul all 55 enemy decks for Shandalar, to add more variety and make the game more challenging. This is the result of that time-consuming decision.

Awesome! Really looking forward to trying this out. I have a couple of decks I like a lot, and I often try to assemble them (or variants of them). It will be interesting to see how Angelic Strike, Atlantis Armageddon and Destructo-Eater fare in this environment...

I left a few cards in that I think are likely to be buggy, such as Lightning Bolt, in hope that people will tell me if they are.

What's buggy about Lightning Bolt? I haven't noticed anything, other than maybe that the computer is a bit too eager to kill my little 1/1 creature with it, instead of holding on to it, wait until it gets 4 mana and kill two of my 1/1 critters with a Fireball and save that Bolt for my War Mammoth instead...

I made an effort to avoid placing cards that appear as dungeon prizes in decks that give ante, but it's possible there were some oversights.

With a few exceptions (e.g. Hurkyl's Recall), the cards in your restricted list are flagged by the game as special and won't appear in the ante anyway. But if the enemy decks have fewer Moxes and the like, we get fewer opportunities to get them via Bronze Tablet and such, so there can still be a point in making them more rare.

I also have not played the latter half of the game for many years, and don't recall if the dragons and genies give ante. If not, I may want to put some restricted cards in their decks.

They do give ante, but see my last answer.


So the ten creatures (OK, you may have changed that number...) that use two colours each have only one dual land in their deck?

On that note, you should also treat Juggernaut as if it were unable to block, because the AI believes this

Fair enough.

and if the Fungus Master shoots its 2/2 Fungusaur with Cuombajj Witches, be a gentleman and assign the other damage point to the opposing player.

You mean, not kill the Fungusaur, I presume? Also fair enough. But I take full liberty to spank that pesky Royal Assassin/Nettling Imp/Banshee/whatever if I want to...

A marginal exception was the Sedge Beast - I really couldn't keep on theme because I didn't know what a Sedge Beast was. (Try looking up "sedge" on wikipedia.)

As a botanist, I can tell you that sedges are grasslike plants (the genus Carex and close relatives). They grow mostly on moist/wet places.

Some advice on how to play the game: early on you will have to avoid most of the opponents, who are more likely to take your cards than give you theirs. You should probably go from town to town judiciously buying and selling cards and running away from everything in sight until you can put a passable deck together.

Pretty much like usual, but I suspect it´s even more important.


Great, off to play now.

[jiansonz]

OK, now I have played a Shandalar campaign against these enemy decks, and like to share some thoughts. My deck for the endgame was a green/red/black one that focused on land destruction, Hypnotic Specters, some decent creatures and a little direct damage. For the early (which turned out to be the main part) game, I used a black and green weenie deck.


I´ll put everything under spoiler cover:

| Open

* Not knowing the decks beforehand was very refreshing! I had a Glasses of Urza from the beginning that turned out to be extremely useful (kept it all the way until the endgame). Just seeing the enemies' ante card (even if I paid them off) was interesting. At one time, I even paid a green amulet to see the Dragon Lord deck!

* I am very impressed by the power (especially late game-) of some of the white decks, like Lord of Fate and Sainted One. I think I needed 9 attempts to beat the white castle.

* The Witch isn't so scary if you have a deck like I did, with cheap creatures, non-land mana (Birds and Elves) and some direct damage. She has only 10 life, you know.

* I do not know if the enemy decks generally had a little too few lands, or just that their large size made them inconsistent at drawing enough lands, but they got stuck at 0-2 lands in play far often than what I consider normal (note that I did not use any land destruction for most of the campaign). I had many easy wins becuase of this.

* What is Meekstone doing in the Goblin Lord deck? In the current version of Shandalar, the Goblin King boosts itself, so it is already 3/3 when it comes into play. Not a good thing with Meekstone in play.

[Hadsingrinsen]

Thanks a lot for your feedback. I can see why the early part of the game might be the bulk of it - it's pretty hard to make decks that are competitive against a human player who has the cards he wants (like four Hypnotic Specters.) I think some of the later decks might need to be overhauled - I may have made the mistake of putting some of the more aggro-type decks later in the game, because I regarded them as more powerful than the alternatives (like the Witch, who in retrospect is not really that tough with 10 life). However by the end of the game you can probably deal with this type of strategy. I'd be interested to see your save file(s), particularly to look at the record vs. each opponent, which could tell me where work is needed.

As for the White decks, I'm fairly happy with how they turned out. I thought that the deck to beat was the Arch Angel (hence my earlier comment that Arzakon can sometimes beat her) but after reading your comment I ran a few duels and found that her advantage over either Lord of Fate or Sainted One was only marginal. I'm also concerned there might be a bug in her deck. While I was playing Shandalar I noticed the game asking me for 5W mana to cast Pearled Unicorn, which is supposed to cost 2W. That could really screw her up since she has four of those. I wonder if you observed her casting Pearled Unicorn for 3 (or 6) mana?

Anyway, I'd like to bring the other late-game decks up to this level. A few of the dragons and genies need to be totally rebuilt, and I will give all of them some restricted cards thanks to your comments earlier. I think the Astral Visionary and Dragon Lord were a little off-kilter and probably wouldn't stand up to a deck like you described. Was there any palace you found particularly easy?

As for the land problem, that sounds like a significant issue. The decks were generally about 37% land, with some variance. I'll take a look and see if some need a smattering more.

In my version the Goblin King doesn't boost itself; that definitely needs correcting. I'll switch the Goblin Lord to a more straightforward strategy with no Meekstone. (The only reason I didn't do this to begin with was that I thought it would make him too similar to other decks, like the Crusader, and I wanted to try a new gimmick. Perhaps I'll transfer his gimmick for another deck where it could be more natural and effective.)

Finally, regarding my earlier comment that Lightning Bolt might be bugged, my issue is that frequently during combat the AI will use fast effects to kill its own creatures instead of mine. Likewise I nixed Giant Growth from every deck because the AI would almost always boost my creature instead of its own. Did you ever observe this sort of thing happening?

[jiansonz]

I see nothing wrong in having some of the more aggro-type decks for the bigger creatures.

It is too late to see the save files for the campaign I played, since I am currently trying out podbelski's overhaul, and all the save slots I used playing "your" Shandalar have been overwritten.

Yeah, the Arch Angel was also a tough bugger. I found her to be on par with how she plays in my overhaul, so yeah, I pretty much avoided her when I could...
Did not notice anything weird with the Unicorns, but I honestly can't tell if they worked normally for her. Pearled Unicorn has always worked fine when I play it in my version of Shandalar (I am using it in my current campaign). 3 extra mana cost for white spells is exactly what Gloom does. There was not one of these in play?

Never got to experience your Arzakon deck because the game crashed for me when that duel started. I am honestly not that interested in this fight, but the standard Arzakon deck is pretty fun both playing and playing against in a Gauntlet (outside Shandalar). My standard way (the capaigns I decide to care about it) of fighting the standard Arzakon deck in Shadalar is a large (72 cards) white/green/red deck that has the main goal of making Marsh Vipers unblockable, while I gain life, have a strong defense and use certain white and red spells to wipe out enemy threats.

I wiped out the Blue, Green and Black castles on the first try, but there was some luck involved (I think I started with a Shivan Dragon(!) in play against the green wizard). Needed 3 attempts against the Red castle and like I said 8-9 against the White. I had an 1-1 record against both Sainted One and Dragon Lord.

Even if Goblin King did not boost itself, if the computer got two Kings out, a Meekstone would still screw the power 1 goblins and both Kings.

I very rarely see the computer use Lightning Bolt wrong (then again, I generally see it being used outside combat - it is most often used near the end of my turn, after I have summoned something) but it happens more frequently with Giant Growth. The computer still uses it correctly more often than not but it does indeed target my fighting creature sometimes, or even a creature that is not involved in the fight.

Helm of Chatzuk (I noticed it in at least one deck) is not a good card for the AI, it does not know how to use it in a useful way. Banding in general is not handled well by the AI. It does OK with it defensively but misses many opportunities when it attacks. Say we both have a Serra Angel out and the AI also has a Mesa Pegasus. If I attack, the computer will be smart, band and kill my Angel and its Pegasus. However, it will not band and attack (something I would do in that situation).