Pages:
Author

Topic: Chess + Crypto (Read 2977 times)

member
Activity: 84
Merit: 10
Correct Horse Battery Staple
April 08, 2014, 04:19:39 PM
#46
Hhmm... so how would the proof of work?
I've been thinking about this and it is quite difficult, but probably not impossible.\

If I am Kasparov, and you beat me at chess, I think that is good proof of work. If you lose, then it doesn't prove anything. So winning chess games should be linked to block creation.

You'd need a way to score players so that you know if someone beat a good player (= proof of work) or a bad or unknown player (= proof of nothing).

You'd need a way to prevent or discourage fake opponents too, that lose on purpose to give score to others.

I am thinking of something along the lines of pools of players.

Pool 1 has 10 players, Pool 2 has 100 players, Pool 3 has 1000 players etc.

And you get promoted or dropped down from pool to pool based on your play, with rewards going to winners in the higher pools.

But the details are not completed in my mind yet.
member
Activity: 116
Merit: 10
April 08, 2014, 12:14:49 PM
#45
Hhmm... so how would the proof of work?
legendary
Activity: 2212
Merit: 1199
April 08, 2014, 10:01:43 AM
#44
it is possible to spot use of software (cheating). Read up on it. Cheating players need to be banned of course.
integrate ilo for players and allow only games between players of roughly the same strength and it will work. Nice idea.

No it won't work.  Cheating can't be proved, and you could always bet a larger amount on games where you cheat more.

Cheating cannot be proved as there are so many ways to build a trailer / smthing....
You will spot one - they will create another ... this is how it works.
legendary
Activity: 1428
Merit: 1001
getmonero.org
April 08, 2014, 08:51:31 AM
#43
I had the same idea recently. However i lack the skills of making it happen...I would support this idea however.
legendary
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
April 08, 2014, 08:50:12 AM
#42
it is possible to spot use of software (cheating). Read up on it. Cheating players need to be banned of course.
integrate ilo for players and allow only games between players of roughly the same strength and it will work. Nice idea.

No it won't work.  Cheating can't be proved, and you could always bet a larger amount on games where you cheat more.
member
Activity: 112
Merit: 10
April 08, 2014, 07:28:35 AM
#41
it is possible to spot use of software (cheating). Read up on it. Cheating players need to be banned of course.
integrate ilo for players and allow only games between players of roughly the same strength and it will work. Nice idea.
sr. member
Activity: 406
Merit: 250
April 08, 2014, 07:01:31 AM
#40
ChessCoin would never work, even if it did work.  Cheesy


IBM:
"What makes you think we have over 51% of the hashing power?"




Grin
sr. member
Activity: 406
Merit: 252
April 04, 2014, 07:19:29 PM
#39
ChessCoin would never work, even if it did work.  Cheesy


Kasparov and Carlsen:
"What makes you think we have over 51% of the hashing power?"

legendary
Activity: 1918
Merit: 1570
Bitcoin: An Idea Worth Spending
April 04, 2014, 04:04:41 PM
#38
If you had some software that could analyse skill level and then match players accordingly could that deal with cheaters? If there is software out there people can use to cheat then we must be able to analyse skill level too.

Yes there are algorithms which can detect likelihood of cheating based on your rating.
They work great, but you can never prove 100% if someone is cheating or not,
and proof is required when money is involved and therefore I don't think it would help.

Like I said, let 'em cheat, but, in essence, there is no cheating because Ccc controls the clock. At the end of the game, a player is either rewarded with 0 Ccc due to a lose, 0.5 Ccc for a stalemate, or 1 Ccc for a win.

A player possessing a great chess program may be able to amass more wins, but due to how long it takes Ccc to make its next move, Ccc controls the payout schedule. After X blocks/games, the difficulty, read clock setting, is adjusted determined by how many players, how many wins, and/or other parameters.

For shits and giggles, a tally is kept on the main site to show how games were played to day, how many are in progress, along with total loses, wins, and stalemates (possibly, further broken down by countries).

Of course, at first the games are relatively easy to win, along with being played relatively quickly. But, as more and more blocks of wins occur, the games get more difficult to win, coupled with Ccc taking more time to make its next move.

Shortly, down the road, few will be physically playing the game, though that's always an option, dependent on moves taking ~1 minute or ~5+ minutes. Later, as the difficulty increases, 10/20/etc. minutes between moves become problematic, thus would already be in the bot-vs-bot stage.

Proof of Work, is a given with Ccc. Thus, no issues there.

No pre-mining/playing, with the exception of showing proof-of-concept prior to the official beta release.

I'm just thinking out loud, hoping to see Ccc come to fruition.

~Bruno Kucinskas

yup, quickly becomes computer chess.

...Of which, is not a bad thing.

What if one were able to pick any completed game, picking some move round as either black or white, then resume play, either playing sans a bot or bot-vs-bot.

Or, what if a player picks some completed game that was lost, resuming play at some point where the score was relatively close (within some %, that is), and turns it into a win, thus enabling the player to earn 2 Ccc for h/his efforts.

If it still results in a lost, some future player can pick it up at the exact same point to see if s/he can turn it into a win to now earn 4 Ccc. Of course, each iteration would definitely take more time between moves.

Again, just thinking out loud, for I love this project.

~Bruno Kucinskas
legendary
Activity: 2968
Merit: 1198
April 04, 2014, 03:51:08 PM
#37
Anyone can download a computer program that can make a move in a second that would beat 99% of players out there.

You will always have cheating in online chess.  Sad

It is however a (possibly) good way to incentivize creating technology to win chess games, mostly against other technology, if you think that is interesting. (In 2014, probably not.) Go might be better.
legendary
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
April 04, 2014, 03:49:27 PM
#36
If you had some software that could analyse skill level and then match players accordingly could that deal with cheaters? If there is software out there people can use to cheat then we must be able to analyse skill level too.

Yes there are algorithms which can detect likelihood of cheating based on your rating.
They work great, but you can never prove 100% if someone is cheating or not,
and proof is required when money is involved and therefore I don't think it would help.

Like I said, let 'em cheat, but, in essence, there is no cheating because Ccc controls the clock. At the end of the game, a player is either rewarded with 0 Ccc due to a lose, 0.5 Ccc for a stalemate, or 1 Ccc for a win.

A player possessing a great chess program may be able to amass more wins, but due to how long it takes Ccc to make its next move, Ccc controls the payout schedule. After X blocks/games, the difficulty, read clock setting, is adjusted determined by how many players, how many wins, and/or other parameters.

For shits and giggles, a tally is kept on the main site to show how games were played to day, how many are in progress, along with total loses, wins, and stalemates (possibly, further broken down by countries).

Of course, at first the games are relatively easy to win, along with being played relatively quickly. But, as more and more blocks of wins occur, the games get more difficult to win, coupled with Ccc taking more time to make its next move.

Shortly, down the road, few will be physically playing the game, though that's always an option, dependent on moves taking ~1 minute or ~5+ minutes. Later, as the difficulty increases, 10/20/etc. minutes between moves become problematic, thus would already be in the bot-vs-bot stage.

Proof of Work, is a given with Ccc. Thus, no issues there.

No pre-mining/playing, with the exception of showing proof-of-concept prior to the official beta release.

I'm just thinking out loud, hoping to see Ccc come to fruition.

~Bruno Kucinskas

yup, quickly becomes computer chess.
newbie
Activity: 22
Merit: 0
April 04, 2014, 03:47:15 PM
#35
Or we can just add a CAPTCHA at the beginning of a game...
legendary
Activity: 1918
Merit: 1570
Bitcoin: An Idea Worth Spending
April 04, 2014, 03:33:47 PM
#34
If you had some software that could analyse skill level and then match players accordingly could that deal with cheaters? If there is software out there people can use to cheat then we must be able to analyse skill level too.

Yes there are algorithms which can detect likelihood of cheating based on your rating.
They work great, but you can never prove 100% if someone is cheating or not,
and proof is required when money is involved and therefore I don't think it would help.

Like I said, let 'em cheat, but, in essence, there is no cheating because Ccc controls the clock. At the end of the game, a player is either rewarded with 0 Ccc due to a lose, 0.5 Ccc for a stalemate, or 1 Ccc for a win.

A player possessing a great chess program may be able to amass more wins, but due to how long it takes Ccc to make its next move, Ccc controls the payout schedule. After X blocks/games, the difficulty, read clock setting, is adjusted determined by how many players, how many wins, and/or other parameters.

For shits and giggles, a tally is kept on the main site to show how games were played to day, how many are in progress, along with total loses, wins, and stalemates (possibly, further broken down by countries).

Of course, at first the games are relatively easy to win, along with being played relatively quickly. But, as more and more blocks of wins occur, the games get more difficult to win, coupled with Ccc taking more time to make its next move.

Shortly, down the road, few will be physically playing the game, though that's always an option, dependent on moves taking ~1 minute or ~5+ minutes. Later, as the difficulty increases, 10/20/etc. minutes between moves become problematic, thus would already be in the bot-vs-bot stage.

Proof of Work, is a given with Ccc. Thus, no issues there.

No pre-mining/playing, with the exception of showing proof-of-concept prior to the official beta release.

I'm just thinking out loud, hoping to see Ccc come to fruition.

~Bruno Kucinskas
legendary
Activity: 1302
Merit: 1008
Core dev leaves me neg feedback #abuse #political
April 04, 2014, 11:29:24 AM
#33
If you had some software that could analyse skill level and then match players accordingly could that deal with cheaters? If there is software out there people can use to cheat then we must be able to analyse skill level too.

Yes there are algorithms which can detect likelihood of cheating based on your rating.
They work great, but you can never prove 100% if someone is cheating or not,
and proof is required when money is involved and therefore I don't think it would help.
hero member
Activity: 742
Merit: 502
Circa 2010
April 04, 2014, 02:43:12 AM
#32
The major problem I can see with this idea is that people will simply use chess engines against one another and then it won't be a game of skill anymore - much rather a game of who has the best engine and access to large amounts of computational power. Would be nice to see Bitcoin integration with some other skill based games though.
member
Activity: 94
Merit: 10
April 04, 2014, 02:25:24 AM
#31
If you had some software that could analyse skill level and then match players accordingly could that deal with cheaters? If there is software out there people can use to cheat then we must be able to analyse skill level too.
hero member
Activity: 870
Merit: 500
Trading will make me rich)
April 04, 2014, 01:46:50 AM
#30
Great idea! There are also other fun turn-by-turn games, that can "live" over blockchain like Huntercoin. HUC is little bit slow for action game, but something like chess, monopoly etc will perfectly fit!
member
Activity: 84
Merit: 10
Correct Horse Battery Staple
April 03, 2014, 05:18:43 PM
#29
I'm up for coding this. Anyone want to help?

All for fun not profit of course  Grin

newbie
Activity: 22
Merit: 0
April 03, 2014, 04:53:11 PM
#28
I was thinking of having 2 parts to mining.

* While your playing, transaction processing is running in the background using your CPU. (can it be limited to smartphones?)
* After the game is done, the network checks your game log against the games previously played. If your game is unique, you earn a coin.

So in essence you are always mining, but only get rewarded probabilistically (instead of probability and effort).

Betting on the other hand is something optional.




member
Activity: 84
Merit: 10
Correct Horse Battery Staple
April 03, 2014, 04:43:40 PM
#27
There is a way to circumnavigate around the cheating issue: Let 'em!

If Crypto-chess-coins is going to dole out coins based on some exponential growth system, then along with each game getting harder and harder to play, assuming one is playing against a computer which Ccc controls, Ccc will also control the clock as to how long it'll take before it makes its next move against each and every "miner". Thus, it doesn't matter how long it'll take for a miner to make s/he's move.

The onus will then be on the miners to obtain the best chess program available to garner more wins over time. At some point, when such is more difficult to obtain, more powerful chess programs will need to be created. Imagine, in the future ordering said programs/systems/rigs from Rook Labs or Bishop Studios or Knight Mills or Queens with Clubs.

Who knows! At the end of the day some über-chess program my be devised that wouldn't have been created earlier due to the advent of Crypto-chess-coins.

Now, one's able to envision Solitaire Coin, Backgammon Coin, or even Puzzle Coin, all pegged against a computer opponent with a dual difficulty component incorporated, adjusted depending on the payout schedule.

During the penning of the last paragraph, I envisioned something pertaining to Puzzle Coin, but I see that http://puzzlecoin.com/ is already taken, but from the looks of the site, it could easily be re-branded.

Picture sometime like below that has to be solved via moving the pieces into place, then rewarded once completed. See the beauty of the example? Advertiser's logos (with generated links once solved, then also emailed) are the puzzles.



I agree!

It now becomes a contest of who can either write the best chess program and/or throw computing power at it.

One way is to use the chess scoring system, and the nodes accept the block created by the highest scoring address at the time.

Nodes can verify all games and scores quickly from the genesis game of course.

Any newcomer who is good can quickly rise to the top due to how the chess scoring works, but as they get near the top they need to beat the best continually to get the top score.

There are a lot of security things to be thought of e.g. shill players who lose on purpose to boost another players score.

There is no need to have difficulty adjustment because it is always as difficult as whoever is playing at the time. And there is an financial incentive for miners as with bitcoin so plenty of "hash power". Or should I say "horse power".

I'm a coder and be interested in helping develop something like this, a prototype at least.

Edit:

Man this is really interesting -E.g.  I just thought would you build a program to try and win all the time or not? Winning all the time would take more resources, so you play less games. Or do you play more games and win 90% of the time. Which would give you a higher score? So much scope for miners to think rather than just throw money at asics like you do with bitcoin. Infact asics would be a disadvantage because you can't change the algorithm when a better one comes out next week.

Pages:
Jump to: