Dominion Strategy Forum

Please login or register.

Login with username, password and session length

Show Posts

This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.

Messages - GeoLib

Filter to certain boards:

Pages: 1 2 [3] 4 5 ... 46
51
Dominion General Discussion / Re: New to the Board and Game.
« on: April 14, 2015, 01:56:11 pm »
I am failing to get a good grasp on what to buy and when. It seems like I tend to fall behind vs my wife and never catch up. I just never sure what cards I should keep buying on a regular bases. Card I got caught on recently and figure it truly not a great card is Ghost Ship.

Ghost ship is very strong and a pretty brutal attack, but that doesn't necessarily mean you should just keep buying them mindlessly. If you want advice on specific games you can post in the Help! board with the kingdom and your explanation of what you did and where you thought you might have gone wrong.

52
General Discussion / Re: Brag Board
« on: April 13, 2015, 08:31:30 pm »
I'm the Major General in the Pirates of Penzance!

Glad you're very well acquainted, too, with matters mathematical.

Yeah, everybody freaks out when they hear I understand all of the math mentioned in the song...

Isn't it just the quadratic equation, binomial theorem, and Pythagorean theorem? Or are there more math verses I'm forgetting?

53
Rules Questions / Re: "Naming" a card
« on: April 13, 2015, 11:14:05 am »
So when you name a card, you're not forced to name an actual card that's being used in the current game; this comes up, for example, when you play Mystic but and want to make sure you don't draw anything. You don't even have to name a card that could in principle be in your deck; you're free to say "the Knights randomizer" or "the Three of Clubs" or "Black Lotus" or whatever.

This has led to an assumption on these board that the instruction "name a card" is actually equivalent to "say some words", whether it's the name of a card or not—and if the words you say happen to be the name of a card, then you get the intended effect.

But with the introduction of Events, "name a card" can't just be "say some words", right?

If you play Contraband, and your opponent "names a card" by saying the word "Expedition", that doesn't actually block you from buying Expedition that turn, right? Because Expedition, the Event, isn't a card, and so your opponent hasn't actually succeeded in naming a card to block you from buying.

So I think, as a board, we need to revise our interpretation of what's covered by the instruction "name a card".

I don't think anything really needs to change.

Quote from: Contraband
Worth $3
+1 Buy
When you play this, the player to your left names a card. You can’t buy that card this turn.

Clearly this can only prevent you from buying cards, and since when you buy Expedition you don't buy a card, but rather an event (Donald made this clear somewhere in the events preview thread, I'm pretty sure) Contraband cannot possibly prevent you from buying it. Or perhaps I am misunderstanding what you think is the problem. I don't think Contraband is supposed to prevent you from buying events.

54
Goko Dominion Online / Re: Features Thread
« on: April 09, 2015, 08:36:15 pm »
While I agree it's not a high priority, I feel like being able to manipulate how kingdoms are picked is something that would take a relatively small amount of time to implement but have a high reward for those who will use it.

But is also a very contentious issue for ranked games.

55
Dominion General Discussion / Re: Probabilities of a hand
« on: April 09, 2015, 03:22:58 pm »
Well the first problem that I see is that you are using the "choose" function which gives you the total number of combinations of cards possible. This gives you faulty percentages because the order matters in that calculation.

This is not correct. The choose function gives you the number of combinations in which order does not matter. Permutations are for situations where the order does matter.

56
General Discussion / Re: Random Stuff Part II
« on: April 06, 2015, 09:04:37 pm »
If Random dies, what will I talk about? Where will I share my funny pictures? Where will we talk about normal stuff?

Why not have a funny pictures thread? That way people who dont normally see the pictures might then see and maybe even contribute
And im not saying get rid of completely, but just leave it for the completely random non discussion stuff

Wait. We're arguing with Ozle? Guys, clearly this must be trolling :P

57
General Discussion / Re: Random Stuff Part II
« on: April 06, 2015, 05:42:54 pm »
Long live random! This is my favorite thread on the forum.

58
Goko Dominion Online / Re: Features Thread
« on: April 05, 2015, 04:25:49 pm »
tl;dr It's being rebuilt on Unity ... which will get us on web and mobile. From what I know about Mobile, the only hitch is for Chrome users, which frustrates me a bit as I use a Chromebook for the great majority of my online work and play, and prefer Chrome browser when elsewhere.
The other hitch I can see from Unity's site and other searching is that Unity web player is only supported for Windows and Mac, so a desktop Linux user like me is out of luck if you're only doing web and mobile (i.e. no native desktop builds).  It's not exactly encouraging when old threads like these are the top google hits:
http://answers.unity3d.com/questions/15227/is-there-an-unity-webplayer-for-linux-.html
http://answers.unity3d.com/questions/20888/what-can-i-do-for-linux-support.html

While I'm here asking technical things, I've brought up the idea of providing an API for bots and/or alternate clients a few times, and still wonder if that would be considered at all.  It would let the community address the Linux thing if that's not enough users to be worth your time, and let folks like the Salvager contributors continue to, well, contribute (afaik there's not much new work being done on Salvager now, it doesn't seem worth it when we know the upcoming release will make the extension stop functioning, but if the upcoming release allows for any kind of community contribution, we've already shown that we're willing to put time into that).

I love this idea. Please please please do this.



There is not one player hosting a game. The table you get after matching is "neutral". When a game starts at the table the cards available are the cards anyone brought to the table  (like it would do if they met afk and combined what they had). A basic card like Smithy that everyone "has" doesn't get a higher probability of appearing than a card that only one player at the table has – every possible card gets the same probability.

This way there is better reason to start buying cards even if you're not ready to buy them all. With the current setup there is not much reason to buy a single set if you don't play robots, because with few cards you will often play against someone with more cards anyway, so your cards won't be used. But with this scheme there is better reason to buy your favourite set to make sure that cards from it can appear in all games you play.

You wouldn't have to bring all your cards to the game. If you hate that Possession, move it to the cards-I-keep-at-home pile. (This avoid people not buying a particular set because there's one card there they don't want to get in their games as often.) You're not safe though – someone else might have brought it to the table.  Probably only few players will bother taking away cards they've bought like that, but it can be very useful for some.

This seems like a great way to address the "I don't want to play with cards I don't like" issue. Certainly I like it much better than veto mode or Donald's blacklists idea. I agree that it would also encourage people to buy single sets more frequently rather than either buying everything and hosting or searching for someone who's bought everything.

59
General Discussion / Re: Sample Uniformly from ~Normal Distribution
« on: April 04, 2015, 04:51:45 pm »
What you can do if you want to be exact is http://en.wikipedia.org/wiki/Rejection_sampling. But might be inefficient as hell, also because there are not enough values at the edge.

RS is a technique to sample from a distribution, starting from another one which is easy to sample. Given you know the density between those two.  And the density has an upper limit.
In your case, you can easily sample from your distribution which is approximately Gaussian, by the technique Titandrake already mentioned: Just draw the sample at a random index.
Your target distribution is uniform.
You also know the (approximate) density between those two: phi(x) := 1/sqrt(2pisigma²)exp(-x²/sigma²).  On a bounded intervall, it is also bounded (but the constant might be bad, which would make the algo inefficient). Call the constant M.

So what you do to draw one sample:

1.) Draw uniform sample z from distribution x:
2.) accept z with probability phi(z)/M. Otherwise, goto 1)

To draw 1/3N samples, you do this 1/3N times.

Hmmm... This does look like approximately what I want, though I think the method I came up with is actually easier to implement. Thank you though!

Have you looked into psuedorandom or quasirandom sequences?  The idea is that they're low discrepency, so you better span the space.  I think that's what you're trying to do with the bins.

Well python's random module uses pseudorandom numbers, but I think quasirandom is actually what you're talking about with low discrepancy. That might be worth checking out.


I've gotten a sample with x approximately uniformly distributed from the method I mentioned. I'm grumpy though because it means our model can't train on anything very far from the mean because they're all in the test set >:(

Thank you to everyone who made a suggestion

60
Dominion Articles / Re: Coin Tokens
« on: April 04, 2015, 12:18:00 pm »
Do you think Butcher would be stronger or weaker if it dealt in Coins instead of Coin Tokens? It'd often be easier to get good upgrades.

How could the card get stronger? I don't see any advantage having coins instead of tokens on butcher (excpept if you'd get more, obviously)

I assume he means if you could spend coins on the upgrade (a la storyteller) and bucher gave coins.

61
Dominion: Adventures Previews / Re: How about we tell Goko NOW...
« on: April 04, 2015, 12:00:39 pm »
People bring this up every so often. The first person Goko approached to code their implementation was Doug. He turned them down (and I can't blame him).
RGG, not Goko.

Ah, thank you for the correction. My mistake.

62
Dominion: Adventures Previews / Re: How about we tell Goko NOW...
« on: April 04, 2015, 11:25:27 am »
I really wish they'd pay doug to help.

People bring this up every so often. The first person Goko approached to code their implementation was Doug. He turned them down (and I can't blame him).

Edit: Ninja'd

63
General Discussion / Re: Brag Board
« on: April 04, 2015, 11:23:09 am »
I won the Goldwater Scholarship!

(I figure this will be meaningful for enough people here, but if you don't know what it is and are curious: http://goldwater.scholarsapply.org/)
That is beyond awesome man.  Are you a sophomore or a junior?  What's your major?

Thanks! I'm a Junior and a Chem-CS double major. I got an honorable mention last year.

64
General Discussion / Re: Sample Uniformly from ~Normal Distribution
« on: April 04, 2015, 11:21:49 am »
So obviously my explanation was unclear.

I have a bunch of chemical compounds with a specific property I'm trying to predict (real-valued). The set of compounds is approximately normally distributed in this variable x (not sure I'm saying this right. If I were to bin them and plot frequency vs. x, I get something that looks apprximately like a normal distribution). I want to select a test set of these compounds, but I would like it to be approximately uniformly distributed in this variable (so if I plotted frequency vs. x of the test set, it would be approximately uniformly distributed in the range of possible values). This isn't actually possible to do, since I want a third of the compounds, and there aren't enough at the ends, but something closer would be good.

So I think Titandrake's solution gives me a uniform random sample of the list, which is what I was doing previously.

The reason for wanting a different set is that the compounds we're actually interested in are the ones at the extreme values, so we want to make sure that when we say our model is effective, it isn't just good at predicting things with middling values because those are the bulk of the test set.

Does that make sense?

The solution I came up with last night was to generate a uniform random variable and select the compound with the x value closest to that value, which I think is probably fine. If someone has a better idea though, that would also be good since it's somewhat awkward as is.

65
General Discussion / Re: Brag Board
« on: April 04, 2015, 01:05:36 am »
I won the Goldwater Scholarship!

(I figure this will be meaningful for enough people here, but if you don't know what it is and are curious: http://goldwater.scholarsapply.org/)

66
General Discussion / Sample Uniformly from ~Normal Distribution
« on: April 04, 2015, 01:02:57 am »
So I'm trying to do this thing for my research and struggling with finding the best way to do it. But I thought maybe some of the people here might know of a good way (especially since we seem to have so many statisticians).

So I have some data that is approximately uniformly distributed wrt to a variable x. I would like a sample of that data that is approximately uniformly distributed wrt x on the range from the minimum to maximum. Specifically, I want to sample without replacement about a third of this data. I tried to bin things, but in order to have enough samples in each bin, I can only use 3 bins, which is not ideal. Is there a good way to do this? Obviously, since I want to draw without replacement I can't really get a uniform distribution, since the ends will quickly be depleted, but something "sort of uniform" would be good. Any thoughts?

67
Dominion Articles / Re: The Haiku Card Challenge
« on: March 26, 2015, 06:12:30 pm »
So counting words in this thread reminded me of counting syllables in haikus, and I thought, why not explain what a card does in a haiku?  Because that would be awesome, right?

Rebuild
Beats most strategies
All but the strongest engines
But not with Shelters

Okay I'm expecting much better from all of you.

Oh man. Prepare yourself for the retort from SCSN and MicQ

(optimized simulation suggests that rebuild is at most marginally worse with shelters)

I'm happy as long as the retort comes in the form of a haiku!

Lonely monolith
Stopped only by strongest engines
Even with shelters

Edit: damn. ninja'd. Also I miscounted the syllables. We'll just try that again

Lonely juggernaut
Beats all but strongest engines
Even with shelters

68
Dominion Articles / Re: The Haiku Card Challenge
« on: March 26, 2015, 05:58:35 pm »
So counting words in this thread reminded me of counting syllables in haikus, and I thought, why not explain what a card does in a haiku?  Because that would be awesome, right?

Rebuild
Beats most strategies
All but the strongest engines
But not with Shelters

Okay I'm expecting much better from all of you.

Oh man. Prepare yourself for the retort from SCSN and MicQ

(optimized simulation suggests that rebuild is at most marginally worse with shelters)

69
General Discussion / Re: On the Job: f.ds edition
« on: March 23, 2015, 11:25:48 am »
There are two Java developers at our office, a colleague and myself.

I use Eclipse and Maven and the other guy uses VI (not even syntax highlighting) and Ant (no Ivy). The irony is that the other guy is once hired as our main "Java expert", while I'm a generalist who happens to be a Java expert as well. I'm slowly convincing my team and superiors that using VI and Ant is not really efficient, nor is putting all of your libraries in SVN and keeping around old libraries on your classpath when you upgrade to a newer version of the same library.

Our manager recently visited an old friend who has set up a medium-sized (70 people) company with mostly (Java) developers. To his surprise, most of them used Eclipse and Maven (and lots of other useful tools)! I've often told him "well, I don't use them because I want to be different, I use them because every serious programmer is using them (or some other tools based on the same ideas)!" Of course it's hard for the non-technical guys at our office to understand what we're talking about, so I'm glad he got to see an actual serious company using those tools.

Man I hate Eclipse though. Using plain old VI is crazy, but Eclipse just feels so bloated. I'm a fan of Geany.

Maybe when you're doing serious development Eclipse's tools prove more useful, but my CS department uses Eclipse on the lab computers and my professor this semester keeps on giving us instructions on how to set up a git repository, commit, etc that are over a page long every time (because they keep changing) and people still keep having strange issues. Meanwhile I'm merrily just using git through the command line on my laptop rather than pecking through all the silly menus.

70
Dominion Articles / Re: Hunting Grounds
« on: March 21, 2015, 02:40:04 pm »
Oh my god. Calm your tits people! We are friends, not enemies. Correct with kindness, not with harsh criticism, even if the article is not up to your standards.

I like harsh criticism a lot more than kindness though. It's nice to receive harsh criticism. When you receive harsh criticism, you learn how to improve whatever it is that you're working on, and when you've finally managed to do a great job and receive positive feedback from the same guys, you know that they're not just trying to be polite, but your work was genuinely good. Then you'll have more confidence in the future when you're doing more of that stuff, which helps you make better decisions and get better results.

When your work sucks and you receive feedback from people who are trying to be kind, you don't learn anything useful, your work will continue to suck, you'll keep making more bad works in the future, and you'll never improve yourself until someone breaks it down for you. In the worst case scenario, you have something else to blame for your mediocrity (such as, not being the best Dominion player ever, not having the most expensive top-end equipment for whatever it is that you're doing, etc), and then the wishy-washy feedback makes you think that you're doing a good job, but the other stuff is holding you back from creating excellent works, and then you just "accept" that your works can never be as good as the works by people who don't have the other stuff holding them back, or maybe you'll start trying to improve the circumstances in hopes of getting rid of your imaginary obstacles, but it will never happen because you're not fixing the main problem which is yourself. I have totally suffered from people trying to be overly kind to me in this way, and it sucks.

So yeah, if you want to be helpful, offer harsh criticism. If you want to be harmful, offer kindness. If you don't want to be harsh or harmful, just do what I've done here for the most parts and shut up.

This is not quite accurate, I think. It really depends on who you're talking to. Things vary by person, but in general when people are just getting into something, it's good to focus on what they're doing well so that they feel encouraged to keep working at it. When someone is already pretty good at something, it's much better to help them find mistakes so they can continue improving. I recall reading a study in which novice and experienced musicians were told to either focus on their strengths or on their weaknesses. Novices improved more with the former and experts more with the latter.

I do in general agree with you that honesty is better, but being mean or rude rarely helps.

I think part of the problem here is that it doesn't really seem like there's that much to say about HG. I mean, it draws a ton, is expensive, and has a TfB effect. It works pretty much how you'd expect such a card to work. I think the most interesting things I've seen are Stef's comments that you may want to overbuild a little more in the presence of TfB and that its close to twice as good as smithy.

71
Goko Dominion Online / Re: Goko Dominion Salvager Discussion
« on: March 19, 2015, 01:27:33 pm »
The problem was that my code didn't address the possibility of a player scoring more than 32,767 VP.  My database strongly objected to this game.
It's not the only one.

Is that someone spending all their zaps?

Probably hacking adventures starting hands like in that post by SerakFalcon, but going a touch overboard.

72
General Discussion / Re: More Maths That I Could Use Help With
« on: March 17, 2015, 11:59:55 am »
I don't know about excel, but this would be easy to program in any sort of scripting language: just loop through all the possible dice rolls and record how often each outcome appears.

Yeah, this seems like way more trouble than it's worth in excel. I would use python.

73
Simulation / Re: Introducing DominionSim
« on: March 12, 2015, 04:06:57 pm »
It will only run on Windows, but you can run it on a VM or with Boot Camp on a Mac.

For CountOf(Gold, Dominion), I debated this a lot because the term "Deck" is overloaded and am open to suggestions. The Dominion rules refer to "Deck" as the pile you draw from. At the end of the game, you put all cards in your deck and calculate VP.  There isn't a term for "all your cards" so I thought your "Dominion" was the appropriate term. Also what's missing is an enum for describing the draw pile deck to count of cards in there (I just didn't get around to implementing it yet).

So options could be:
  1. Dominion for "all cards" and Deck for draw pile during game.
  2. Deck for "all cards", DrawPile for draw pile during game.
  3. AllCards for "all cards" and Deck for draw pile during game.
  4. ??

I'm about a 3rd of the way through the Intrigue cards. Goal is to get to them all eventually. There's some that I will likely procrastinate on (Possession). But I also want to focus on getting the tools right too. The tools decide how the cards are authored and I've re-authored the cards like 3 times since starting. So I didn't want to accumulate a huge set of cards I would have to redo again. I think the architecture is pretty good right now, so I'm going to get Intrigue done and go from there.

Yes, I want to release the source code. One step at a time...

Since deck is ambiguous, why not just avoid using it entirely? So AllCards, and DrawPile

74
General Discussion / Re: Random Stuff Part II
« on: March 11, 2015, 12:30:04 am »
They're bringing back Zoombinis! Any other 90s kids with nostalgia on here?

https://www.kickstarter.com/projects/zoombinisgame/zoombinis

75
General Discussion / Re: The Best Games ever
« on: March 11, 2015, 12:29:03 am »
Zoombinis!

Pages: 1 2 [3] 4 5 ... 46

Page created in 2.456 seconds with 18 queries.