Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[bigg_nate]

Here's the puzzle: design a kingdom that will let you end a solo game as quickly as possible.

Clarifications:

• You do not get perfect shuffle luck. You're trying to end the game as quickly as possible in an actual game, not just if everything falls perfectly your way.
• You may end the game either by buying out the provinces/colonies or by exhausting 3 piles.
• You may choose any ten kingdom cards you want (11 if you want young witch, I suppose). You can use any of the 10 sets and any promos.
• You may choose whether you want colonies and platinums, and whether you want estates or shelters.
• You may not use any Event cards from Adventures.

I searched and found a couple similar ideas, but not this exact puzzle, I don't think.

I've got my best answer so far, but it could be beatable. In my two test runs I ended the game on turns 7 and 8, respectively, though I feel like I got kind of lucky. I'll give you guys a little time to think before I post my kingdom.

[Kirian]

• You do not get perfect shuffle luck. You're trying to end the game as quickly as possible in an actual game, not just if everything falls perfectly your way.

Note that this implies you must assume the absolute worst shuffle luck possible.  Note that this is a very difficult proposition.

[Zappie]

"worst shuffle luck" /// "actual game"

[GendoIkari]

Back in the Isotropic days, there was a solo challenge for people to end a 1 player game in as few turns as possible. So this would be assuming "average shuffle luck", and since people would try over and over on the same board, it really was assuming "better than average shuffle luck"... but still no where close to perfect shuffle luck, or worst possible shuffle luck.

With that challenge, I think 7 was the fasted someone had. This was also before Dark Ages, Guilds, or Adventures. I think the general strategy was simply Chapel, Ironworks, and Great Hall, and Island.

[bigg_nate]

No, don't assume the absolute worst shuffle luck either. Mathematically, what you're trying to optimize is the average case.

Think of it this way: set your kingdom, then play 100 games. Maybe you got really lucky one game and finished in 5 turns, and really unlucky in another and finished in 12 turns. But on average it took you, say, 7.5 turns. That 7.5 is the number that matters, not the 5 or the 12.

[bigg_nate]

I tried out chapel/ironworks/great hall/island. I could be missing something, but it took me 10 turns. Are you sure this was a solo match?

Open Chapel/ironworks
Turn 3: trash 4 cards
Turn 4: gain an ironworks, buy an ironworks (this was lucky)
Turn 5: gain 2 ironworks
Turn 6: gain great hall, trash 4 cards
Turn 7: gain 2 great halls, 1 island, and 2 ironworks, trash remaining coppers
Turn 8: put chapel on island, gain remaining ironworks, 1 great hall, 1 island
Turn 9: put great hall on island, gain remaining great halls
Turn 10: gain remaining islands

Shoot, I just realized I had 10 copies of great hall and island in the supply, but I can't imagine the extra 2 copies made a 3-turn difference.

[DG]

You might want to try a governor + king's court + bridge sort of thing, possibly with chapel. Throw in a baker coin, shelters, and an event like seaway. Emptying the province pile will be as fast as emptying 3 other piles.

[JW]

You might want to try a governor + king's court + bridge sort of thing, possibly with chapel. Throw in a baker coin, shelters, and an event like seaway. Emptying the province pile will be as fast as emptying 3 other piles.

One of my submissions to this season's kingdoms championship match kingdoms was an attempt to find a kingdom that let a (edit: solo) player empty the Provinces the fastest on average: Stonemason, Masquerade, Bridge, Fortress, Procession, Quarry, Baker, Governor, Border Village, King's Court. Provinces/Shelters.

I figured it would be too straightforward strategically for the championship match but would be fun for the after-tournament, which is exactly what happened. Masquerade was chosen instead of Chapel because it helps the Procession chain to have a 3-cost, and because you can sometimes cause a great reshuffle by trashing overgrown estate with masquerade on turns 3/4.

[liopoil]

You might want to try a governor + king's court + bridge sort of thing, possibly with chapel. Throw in a baker coin, shelters, and an event like seaway. Emptying the province pile will be as fast as emptying 3 other piles.

One of my submissions to this season's kingdoms championship match kingdoms was an attempt to find a kingdom that let a player empty the Provinces the fastest on average: Stonemason, Masquerade, Bridge, Fortress, Procession, Quarry, Baker, Governor, Border Village, King's Court. Provinces/Shelters.

I figured it would be too straightforward strategically for the championship match but would be fun for the after-tournament, which is exactly what happened. Masquerade was chosen instead of Chapel because it helps the Procession chain to have a 3-cost, and because you can sometimes cause a great reshuffle by trashing overgrown estate with masquerade on turns 3/4.

Wait, surely that is a 3-pile 90% of the time between equally matched opponents.

As for the challenge:

http://www.gokosalvager.com/static/logprettifier.html?/20150103/log.50a43f07e4b0addfc4f4304a.1420319023082.txt

6 turns, on my second try way back in January (see here)

EDIT: I'm fairly sure this is the kingdom to use. Inn can be replaced by some cool adventures card. Maybe magpie? This kingdom relies on develop-fortress-catacombs synergy with watchtower to empty all 3 at once, and could empty all three even if there were 32409381409128595071 of them. Are there other such combos? Rats and Magpie can empty their own piles, but are there combos which can gain any number of things from 2+ piles? Even KC-bridge is limited by buys (there is no way to generate an arbitrary number of buys).

Maybe this should be moved to the puzzle subforum or solo challenges.

[GendoIkari]

I tried out chapel/ironworks/great hall/island. I could be missing something, but it took me 10 turns. Are you sure this was a solo match?

Open Chapel/ironworks
Turn 3: trash 4 cards
Turn 4: gain an ironworks, buy an ironworks (this was lucky)
Turn 5: gain 2 ironworks
Turn 6: gain great hall, trash 4 cards
Turn 7: gain 2 great halls, 1 island, and 2 ironworks, trash remaining coppers
Turn 8: put chapel on island, gain remaining ironworks, 1 great hall, 1 island
Turn 9: put great hall on island, gain remaining great halls
Turn 10: gain remaining islands

Shoot, I just realized I had 10 copies of great hall and island in the supply, but I can't imagine the extra 2 copies made a 3-turn difference.

Well 2 extra copies of each is 4 more gains needed; that could easily be an extra 2-3 turns. And if I remember right, 7 turns was the best anyone managed to do; I don't know how many tries it took them to do it. I seem to remember usually getting 8 turns. I also don't remember what the optimal play was; it probably wasn't gaining all Ironworks before moving on to Island and Great Halls though.

[sudgy]

I think I get what he's saying.  I'll rephrase it this way:

"Find the fastest strategy in a simulator over the course of several games (say some thousands) given [all of the rules in the OP]."

[bigg_nate]

My kingdom was pretty similar to JW's. Instead of masquerade and governor, I had storeroom and band of misfits. I don't know that storeroom is that important; I added it because storeroom/quarry is such a good opener. If you're lucky enough to draw the storeroom on turn 3, you pick up stonemason/border village/border village/band of misfits/band of misfits AND trigger a reshuffle! But I'm surprised you left out band of misfits -- it makes it so much easier for your combos to line up. AFAIK, this is still the fastest way to clear out the provinces.

But liopoil's combo is much faster to empty 3 piles. Nice one!

GendoIkari, regarding 4 extra cards in the supply, realize that you've got exponential growth going on and by the end you're gaining 10 cards per turn. So 4 extra cards is not 2-3 more turns; it's less than an extra half turn.

Apologies for posting in the wrong forum, and thanks for moving the thread.