?

Log in

No account? Create an account
Jeu de Semaine, no. 8 - Suit up, son! You're going to Mars! [entries|archive|friends|userinfo]
Suit up, son! You're going to Mars!

[ userinfo | livejournal userinfo ]
[ archive | journal archive ]

Jeu de Semaine, no. 8 [Jun. 15th, 2011|09:13 pm]
Suit up, son! You're going to Mars!

Catch-up Round-up

In which I discuss all the games I've played but not blogged about.

Saboteur

This has been around for a while now. A beer-and-pretzels path-building card game where everyone is a happy dwarf who works cooparatively to dig towards the gold, except for one or two secret saboteurs who, are presumably, the gold's current owners. It's interesting game in that it tries to balance the asymmetry of the team size—few bad guys versus many good guys—against the asymmetry of the team goals—entropy vs. order. There's another interesting dynamic where certain cards let you see some of the game's hidden state, and those cards essentially become a loyalty test, where anyone who lies will get found out eventually.

Magical Athlete

Ok, start with Chutes and Ladders. Take out the Chutes and the Ladders. Then give each character a special ability. Then give each player a team of multiple characters, each of gets used only once over the course of multiple races. Then add a draft mechanic for determining who gets what team of characters. That's Magical Athlete.

This game is also very much a beer-and-pretzels game. The dice are dominant enough that you can't really care about winning. Some of the interactions between the characters are wacky enough to bait the rules lawyers into silly arguments. On the whole, it's a pretty fun filler game. It might be a good game to play with kids, starting with Chutes and Ladders and adding the additional layers as you go.

Chaos in the Old World: The Horned Rat

This is an expansion to Chaos in the Old World. In addition to adding a fifth player, it also provides a alternate set of card decks for each of the original players. So the game is, in some sense, part expansion and part patch, which is an interesting idea in itself.

Every other time I've played Chaos I've felt like it's really hard to figure out which of the multiple victory conditions one should be striving for. This time I played Khorne, which is the most straighforward of all of them. I ended up winning, but perhaps only because of an early clerical error on bakedweasels' part. I guess I haven't played enough of the unexpanded game to really understand how the expanded game changes things.

New Phyrexia

This is the latest Magic expansion. New Phyrexia is the third expansion in the Scars of Mirrodin block, which tells the story of the invasion of the world of Mirrodin by the corrupting blight of Phyrexia, the ensuing war, and, finally the defeat and occupation of Mirrodin by the Phyrexians.

Most of my Magic play over the last couple of years has been seal deck and booster draft formats, so I can't really speak to how these cards will affect the larger deck construction metagame of Magic. As a draft format, New Phyrexia is tricky. All the card sets in this block are artifact-heavy, and New Phyrexia adds a bunch of cards like Mutagenic Growth that can be paid for with either colored mana or life. In the greater scheme of things, this is pretty interesting; Mutagenic Growth's power is very green, but because it can be paid for with life, it can go in any color of deck. This allows players to put off-color powers in their decks, without much dilution to the feel of each of the colors. Mutagenic Growth still feels green, no matter how you pay for it. This also fits well with the fiction of the set. Phyrexia is a plane of demons, and paying life for off-color effects feels like a Faustian bargain.

The other thing that's going on in this block is lots of creatues that deal poison damage to players. (You lose the game at 10 poison counters.) This, combined with all the artifacts and all the cards that can go in any deck, turn the draft format into a game that's less about color and more about choosing your victory condition.

Leela

Leela is a relatively new Go-playing AI that I've been playing against. It uses the hip new monte-carlo tree search that all the hot Go programs are doing these days. It's quite strong, 1kyu if you believe its own claims.

One of the things that people compain about with monte-carlo Go AIs is that they are inhumanly ruthless when it comes to trading margin of victory for probability of victory. Once Leela decides she's ahead, she will give exactly enough ground as is necessary to settle the entire game in her favor. But the time when I find this to be really irritating is when Leela is losing. She will start to take stupid risks and fall on her sword trying to eke out any longshot chance of victory. So games against Leela are typically won by a mile or lost by an inch. Leela will gladly resign if the going gets too tough, but I wish I could throw the AI into a "give up trying to win and just maximize score" mode, just so I can play out the endgame and see how close the game really was.

As one reviewer pointed out, Leela tends to play for moyo, favoring influence over territory in the early game. I also noticed that she seems to have a blind eye for snap-backs. On multiple occasions I have made a snapback threat that was very answerable, yet went unanswered. Taking the snapback after that often causes her to resign on the spot.

linkReply

Comments:
From: bakedweasels
2011-06-16 08:12 pm (UTC)
You never told us how your weekend Formula De session turned out.
(Reply) (Thread)
[User Picture]From: algorithmancy
2011-06-16 09:11 pm (UTC)
I think you meant Formula Duh.
(Reply) (Parent) (Thread)