CORDIAL MINUET ENSEMBLE

Oh, jeeze... this happened for TCD bigtime too.

These people are trying to increase the google pagerank of the sites that they link in their bios.

Looks like hundreds have been added.  Time for a DB query to clean them out.  Sorry to all the people who made accounts long ago but never posted....

Well, the other constraint is that I don't want any non-permanent marks on the paper part of the display.

I don't like video game displays that look like fake wood or stone or parchment or watercolor but then behave digitally.  That's why I stuck with pixel art for so long (it looks and behaves consistently digital).

Here, I have a simulated real-world asset in the center that behaves totally consistently.  Ink is ink, and is permanent.  The whole thing fades out at the end of the round and a new one fades in, clear of ink.  But you never see the ink flash on and off, etc.  The board itself never reacts to mouse-overs, but all the clearly-digital stuff around the board does.

So, it seems like multiple click-to-place-guess markers around the edge would get confusing pretty quickly.  For consistency, they would have to be on the edges only.  Let's say you have two red squares clicked.  That would mean two red markers on the top edge, and two on the right side.  But the intersection points of those four markers is ambiguous:

 v v 
000000
000000 <
000000
000000 <
000000

Which means you'd have to use different shaped markers to convey which one pairs with which:

 + v 
000000
000000 <
000000
000000 +
000000

Seems pretty confusing.

Also, I sometimes wonder about what counts as "skill" in the game and what the game should just hand you.  The original motivation for the score graph was that "I don't want fast arithmetic to be the main skill in this game."  Many online poker rooms have a text display that tells you what your combined hand is (like, "Hey you have 10-high straight in case you didn't notice.")  But beyond that, they show you nothing, strategy-wise.  They don't compute pot-odds for you, etc.  There is third-party software that does this stuff, but the poker board is easy to enter quickly into third-party software (number-suit, number-suit, number-suit, etc.)

Here, the board is pretty much impossible to enter into third-party software.  Though maybe software could "discover" the board if you entered the top line and the diagonal or something.  I don't know.

I feel like this is something I should table for now.  The game will be a success or flop regardless of the presence of extra strategic tools like this.  If it's a big success, we'll have lots more input from lots of hardcore, serious players, and I'll have plenty of time to add stuff like this as-needed.  If it's a flop, I've already been working on it for more than a year, and it's a pretty simple game...

Yeah, seems like this is easy enough to just have people "upgrade" from copper if they win gold later and their copper goes to the second-place person on that day, as needed.  I would just figure all of that out by hand when I'm shipping out the amulets.

Also, as far as dual accounts winning two, I guess there's no for-sure way around that, but I'll be in touch with each winner for shipping instructions, so that will make it less appetizing for people... lying directly to me via email, etc.

Jere, yeah, 36 is a lot.  I don't know if I'd have any skin left on my fingers after all that polishing.  Also, that number came from a time when I imagined them passing around and each held by one person, so there needed to be a lot of them.

Anyone with a negative score would just create a new account to get back to 0.  So, obviously, we can't let scores go negative.  We can still let people lose points after gaining them.  As long as they don't go below zero, they're never motivated to ditch their account.

Jere, you point out the difference between loss-win and win-loss if we floor the scores at 0.  However, there is no way around this, because someone who has loss-win should create a second account before playing the win, right?

I also worry about the "after day 1, I stand no chance" problem.  It seems like spacing the prizes out over many days could help with this.  It could be two per day.  I'm currently leaning toward 12 amulets (3, 4, 5 gold, silver, copper), so it could be six days, with the following pattern:

copper-silver
copper-silver
copper-silver
copper-gold
copper-gold
silver-gold

Or it could be 12 days, starting with copper and ending with gold.

But how do we deal with one person from winning more than one?  If you win a copper on day 1, are you blocked from winning gold on day 6?

Yeah, we should probably discuss this point.  Getting negative on the leaderboard is pretty harsh.

The problem is, you have to have some way to discourage reckless play....

There were a total of four cabal members this time.

Of course, I cannot say whether any of them won a prize or not.

It seemed to work better than the contest where it was just me, as there were 16 names on the leaderboard as opposed to only seven last time, with roughly the same number of people trying.  So, clearly more cabal games took place this time, which is good!

Yes, there is meaning to that amulet symbol... everyone sees the same symbol, so it doesn't have gameplay meaning.  Just thematic meaning.  It actually now has a double meaning.

Oh, man, that's really bad!

Thanks for telling me.

Prizes have been paid.  Than you all for playing and helping me test this!

Well, what you're talking about there is prop players.  Card rooms employ these, and it's something that I've thought about.

However, this game is exclusively heads up (poker term for 2-player only), so prop players don't make as much sense.  This game will ALWAYS have a "parity" problem, where there is generally at most one unmatched opponent at each stake level (the odd opponent).  A prop player could give that odd player a partner, but is also taking that odd player away from the next even player who joins (essentially passing the buck).

In 9-seat poker, a prop player sitting at an empty table can serve as an opponent for up to 8 other players, and then can drop out, leaving an empty seat for the 9th non-prop player.

When picking the amulet guardians, I'd be picking people who would likely be playing a lot anyway and would likely be candidates to win prizes themselves.  Which makes me think that they should probably NOT be compensated (it shouldn't be a chore). 

In fact, I don't know that I even have to tell them about it...  Well, that would probably encourage them to try to figure it out, maybe employing an alt account to help them figure it out.  And that might open the door to collusion.  So I think I would need to talk to them about it.  I do know three CM players here in town, for example.

Anyway, I wouldn't want it to be a "YOU MUST PLAY X GAMES A DAY" demand.  I'd just want them to play how they'd normally play.

Yeah, maybe so!

Yeah, that leaderboard should be cached.  Ooops.

Agreed that there should be something special happen for you post-game to let you know.

One concern that I had is how FEW games I actually ended up playing.  I was pretty much playing or waiting for a game constantly, but in three hours, I only repeat-played a given person on two occasions.

This seems pretty frustrating for the players who are trying, especially if we have way more.  Seems like it creates a kind of "let down" cycle for most people.  "I was trying all day, but NEVER got into a game with Rohrer."  I may be wrong about this, but I think people could descend into hopelessness if it's just too rare.  It's like the vast majority of people don't even get a chance to participate in the contest.  We had about 12 people playing at the peak, and some of them never played me.  Imagine 100 people.  It just gets worse the larger the pool gets.  You may try for a few hours, but you won't try for 5 days straight if you never get a bite.

One idea is to "elect" amulet guardians from the current leaderboards.  I shot that idea down before because those people wouldn't be able to win an amulet themselves, but Caravan points out that the rule could just be that they can't win their own amulet (obviously).  These people could be trusted folks that I speak to directly to ensure that they won't collude with each other to give each other points.

The other idea is to elect people automatically throughout the contest without telling them, and keep mixing it up by changing people depending on who's active.

This is like a dozen random, unwilling me surrogates.  The number could be adjusted based on the player population.

Though this seems vulnerable to collusion again (pairs of accounts that fast-play each other are more likely to get elected because they're always online, and even if they don't know they're elected, they'd be scoring points by fast-playing each other).

Getting back to a fixed set of people, which fast-playing colluding partners have no effect on, it could just be a straight-up point leaderboard like it is now instead of specific amulets, so they're not even out of the running for any amulet (they're at a slight disadvantage because they can play one less person for points than non-guardians, but I'd give them each a bonus of some kind).

Obviously, the list of people could be secret, and the chosen would have no motivation to tell each other (giving each other an advantage).  Maybe I've already chosen them.  Maybe they were already operating in the most recent test contest....

So few people use Linux, sadly, that this is the first I've heard of the problem!

Linux is my main daily work system, so it's important to me, but I've always felt like a source distribution with an automated build system was the best way for me to target Linux.  I recently started building binaries, but every extra binary target that I add involves quite a bit of overhead (booting into that system, building, uploading from there, etc.)  I already have 3 binary targets.  Maybe at some point, with enough clamor, I'll start doing it for 64-bit linux.