??????
You are not logged in.
Sometimes network outages or server outages happen. This is the reality of servers and networks, sadly.
The other sad thing is that the server can't tell the difference between a "real" outage and a "false" outage that someone is trying to use to save themselves from a loss. You know, make a bad move and then pull the plug to save yourself.
What happened here is that you joined the game, and the server got your join instruction, and then you lost connection. By the time you reconnected, the minute timeout had happened in that game. So your opponent saw the game start up and then saw you essentially go "away from keyboard" and abandon the game.... they waited for a minute for you to make your first move, but you never made the move. They were awarded the ante and the 6 chip penalty as a result.
Then, when you finally reconnected, your client was still waiting for the game to start, so you saw that waiting screen. But at that point, you were waiting for a fresh game, because the first one had already ended.
Whenever you CANCEL from that screen, it executes the leave_game action on the server. That action will show you the last filed "leaving report" for the last game you were part of, if there is one. In this case, it was from the game that ended because of the connection loss.
I agree that it would be even more bothersome at higher stakes, but again, I can't see any way around it. It is a two-player game, after all, and we need to be fair to both. In the case of a network outage, we favor the player who is still connected.
That doesn't mean it doesn't suck when it happens, and I'm sorry that it happened to you!
Fortunately, it's a pretty rare occurrence.
http://www.amazon.com/666-Jellybeans-Al … B000QO2DLY
Love that activity book and the story behind it (handing it out in schools on the same table where the bibles were being handed out).
No, I'm counting draws as losses. Win rate of 65% means win rate.
I'm actually playing my "bot" against a random bot, turn by turn. My bot is looking at the available information each turn (what we know about what the opponent picked so far) and picking the move that leads to the highest expected win percentage. That's giving me a 65% win rate against the random opponent.
Also, different boards have slightly different win rates for this method, but the average across a bunch of boards is 65%.
A naive reactive maximizing opponent against a known random opponent can achieve a win rate of 65% according to my simulations.
If the reactive maximizing opponent picks randomly on turn 1 instead of considering the best choice, the win rate goes down to 60%.
I think that doing full-blown expectimax will achieve even better than 65%. Though Josh is saying perfect play will win 2/3, so maybe not.
Whoops... Monte Carlo isn't needed. There are only 720 possible ways your random opponent can play. So, just need to simulate each of your 720 non-reactive strategies against each of those opponent strategies and see which one wins the most.
There's always an edge when you do this. Rarely it's as high as 3%, rarely as low as 0.5%, and usually between 1 and 2%.
Okay, interesting preliminary result, which was unexpected to me:
Monte Carlo simulation of a random opponent, and picking a pre-ordained, non-reactive move list that does "best" against the simulation on a given board, does not produce a move list with anything that looks like a statistically significant edge against said random opponent. I'm no expert in stats, but a 5% edge on 1000 experiments that erodes down to a 2% edge on 10,000 experiments and down to a 1% edge on 100,000 experiments looks like noise.
Also, taking a non-reactive move list that wins the most over 1000 games and running it for another 1000 often produces a worse win rate or even a majority loss (rarely---so there's probably a slight edge to be had).
So, being reactive is clearly necessary to defeat a random opponent.
Open question: when playing a random opponent, can you do better than picking randomly yourself on turn 1? I think you can.
One question is: Does random mean random all the way through, or random only on the first pick?
As Jere has demonstrated, there are situations on the third pick where you can force a win by picking right. If you're picking randomly even then, clearly you are throwing away victories.
Looking back at my tests with various AI bots, I recall that I had to serialize the game so that regular game tree minimax could be applied to it (regular minimax doesn't handle hidden information). So, the 82% win rate that I saw for tree mimimax vs a random opponent doesn't tell us much, because the minimax AI had too much information (it knew what its opponent picked turned 1 before making its pick turn 2).
A kind of blind Monte Carlo method against a random opponent would be to simulate the random opponent against all possible picking orders that we could use. There are only 720 to test, which seems pretty reasonable. Then you'd just pick the picking order that has the best average outcome. I'll experiment with this and report back. Maybe that's what Josh has already done to get his 2/3 figure.
Also added a feature to the EXISTING ACCOUNT login screen where you can specify that account details are NOT saved locally. This will be released as part of v16.
The bug with the revealed square leaking if one player is operating faster than the other has been fixed on the live server. There was a similar bug with bets leaking if one player is operating faster---that's also been fixed.
Sorry for the confusion here.
When you hit refresh, it shows you ALL open games in existence. If there's more than 9, they will spill onto multiple pages with NEXT and PREV buttons that will appear. If you're only seeing one on the list, there's only one open game.
If you're seeing some that are out of your price range, well, that's all there is. When there are more open games, the ones that are out of your price range will be on later pages, so you won't see them as the first result.
Yes, that is likely a bug where information might leak. I'll take a look. There's also concern about a modded client that could keep calling get_game_state over and over, waiting for the opponent's reveal to finally be posted even before we post our reveal. It's shouldn't let any reveal go through until they're both received, regardless of the timing of the calls.
Regarding the other questions that weren't answered yet...
I don't want TOO much chiming going on. Yeah, you suggested that it could be turned off separately. I think waiting for an opponent over the long term will be much less of an issue in the future when the player base grows. Currently, it only chimes when the game REALLY needs you to wake up.
Two people on the same computer just need separate CordialMinuet folders to have separate accounts.
There's no plan for user passwords in this game. It's my belief (as explained elsewhere) that passwords are usually the weakest link in any security system. They're used everywhere, with everyone asking you to create a password for almost everything, even when passwords make no sense.
For the vast majority of players, their computer is secure, and asking for a password every time the game starts isn't necessary. Most people use very weak passwords that I wouldn't trust to protect real money transactions anyway.
If your computer is not secure, just clear the account details before you quit the game. Look for the CLEAR button on the Deposit screen. Maybe I could add an option to ALWAYS clear the account details upon quitting the game, because I suppose people who would want to clear it always would often forget.
No plans for user agreements. Companies have to throw that kind of stuff at you because they can't represent themselves in court, so they are extremely cautious legally. I'm not afraid of frivolous lawsuits, because I can represent myself, so a frivolous lawsuit will cost me nothing. A company will spend hundreds of thousands of dollars on legal fees, even if they win in the end.
Baby, it's not my bot.
You could also set up a simple web proxy that scrapes the square out of the protocol (the protocol is all HTTP).
For median, you're taking the average of the middle two values (there are an even number of values in a row, so there is no middle value). But what value is the median? What does it tell you?
No screenshots yet. I realize that different people learn in different ways (visual vs. auditory, etc), so I should probably make a guide with diagrams. Maybe someone else is inspired to take this on?
Here are the rules:
Both players see the same 6x6 grid of numbers, from 1 to 36, arranged in a magic square (so that each row, each column, and each diagonal sums to 111).
You are green, your opponent is red. You pick columns, your opponent picks rows.
Each round lasts for three picking turns. You pick two columns and your opponent picks two rows. You pick one column for yourself in green, and one for your opponent in red. "One for me, one for you. Green for me, red for you."
Your opponent is doing the same thing, but picking rows, taking one for themselves and giving one to you. You both pick simultaneously, in secret.
The green row that they give to you will intersect with the green column that you give yourself. The cell where the green row and column intersect is added into your score.
You don't get to see what row your opponent picked for themselves, or what number they scored this turn.
There's a betting step after each pick. During the betting, players can optionally bet more coins into the pot. Bets happen simultaneously, with the person who bet less getting to see the higher bet and match, raise, or fold. The person who bet more does NOT get to see the lower bet unless the lower better folds. The person who bet more only sees the match or the raise.
On subsequent turns, you pick from the remaining columns, and your opponent picks from the remaining rows, until none are left. There are 6 columns to chose from on the first turn, 4 columns to chose from on the second turn, and only 2 columns left to chose from on the third turn. "On for me, one for you." Until the whole board is full.
By that point, each player will have scored three numbers at their three intersection points. You still don't know what cells your opponent scored, but you have quite a bit of information.
Then, there's one final reveal turn, where you get to pick one of your numbers to show your opponent. Your opponent does the same, and you pick simultaneously in secret.
After that, there's one final betting step.
If both players make it to the end without folding, then all scores are revealed, and the player with the highest total score (their three numbers added together) wins the coins in the pot.
If one player folds along the way, no information is revealed, and the player who did not fold wins the pot.
Play continues across multiple rounds until one player runs out of coins or leaves. There is currently a 6 coin penalty for leaving, paid to the other player. If you have fewer than 6 coins left when you leave, your remaining coins are given to the other player.
An ante is put into the pot by each player at the start of each round. The ante starts at one coin and goes up by one in each subsequent round.
Saw the 100 year old headlines---that's a great idea!
Yeah, that's quite a fortunate name combination.
Humbit is your project? Seems pretty cool. A 1000x1000 pixel image. Right now, it's pretty sparse. Seems like it will be a tableau of noise once you get a million people.
If you shout with radius 500, does it really go to everyone?
Who's the voice in your trailer? Are you British?
American Horror Story... here it is:
https://www.youtube.com/watch?v=omu0lDUcMP4
Never saw this before, but yeah, that's the feeling I was going for. Also similar to the opening credits of Se7en. As I thought about the trailer, I had a particular song in my head that I wanted to write. Clicking/ticking ratchet noise, then a loud, angry synth hit. Then I watched the Se7en credits again, and realized I was thinking of Mr. Reznor's music there. Hadn't seen the movie since it came out, but somehow, the music stuck with me.
Ended up writing my own music, of course. Used Justin Frankel's Reaper DAW, which was amazing.
I also don't know how twitter bots are implemented, but I'm assuming you're hosting this on a third-party web server somewhere and invoking the Twitter API from there.
Anyway, have you thought about just giving people direct access to a given graph by name from your web server? Like, go here, see a list of pseudonyms, click on one, see the graph. I mean, in addition to requesting the graph from the twitter bot.
Check the Active Users graph, linked in a sticky post, here:
Sure, you can do this. You should link to this page:
http://cordialminuet.com/foyer.php
Maybe let them know that this is a secret link, exclusively leaked to Fark.
Thanks for asking first, and everyone else, please ask too. Not quite ready to have this posted all over the place yet. There are some "exclusive" media coverages in the works that are depending on the game not being officially released yet.
Okay, v14 posted to fix this. Give it a try and let me know if you notice anything else.
I ended up fixing something else about the graph. If you hadn't dragged the final red slider, but had somehow nailed down the final green column (through mouse over or slider), the green bars on the graph would act like your opponent didn't know their final column yet (because it hadn't officially been picked as red yet), but given that there's only one column left for them, might as well nail it down too when drawing the green graph lines (the red lines reflected it, because you have information about your last column, so the green lines should probably simulate that pick as well). This is fixed now. Dragging the final green slider, or mousing over an uncolored square while betting post-turn-2, now properly uses the process of elimination to simulate what your opponent would know, given the only column that would be left for you to give them.
Thanks for holding my feet to the fire on this one.
Yeah, doing voice stuff is hard.
That said, hiring a pro voice actor seems totally fake to me.
You're right.... there's a bug in how it's computing future possibilities at that point.... working on fixing it now.
Suppose.... you're... BIRD... GRADUATION. This... is your coin stack. This... is your score graph.