[Idea] P2P Chess Solving + Mining Pool

[EnergyVampire]

Just throwing the idea out there.

Assuming ASICs take more work away from GPUs, would it be possible to repurpose the GPUs to solve chess?

The pool(s) could make their money from Computer Chess Tournaments and distribute shares to miners for crunching the data. Of course, current pool and mining software would have to be modified but could this be doable?




Edit: The top-10 pools already have most of the resources in place, so I'm not proposing a new pool solely for this purpose. Besides, DeepBit already has a cool name which sounds like an old Chess Computer.

[bulanula]

Just throwing the idea out there.

Assuming ASICs take more work away from GPUs, would it be possible to repurpose the GPUs to solve chess?

The pool(s) could make their money from Computer Chess Tournaments and distribute shares to miners for crunching the data. Of course, current pool and mining software would have to be modified but could this be doable?




Edit: The top-10 pools already have most of the resources in place, so I'm not proposing a new pool solely for this purpose. Besides, DeepBit already has a cool name which sounds like an old Chess Computer.

You mean IBM's Deep Blue ? 

[Meni Rosenfeld]

I don't think GPUs are good for chess, but I could be wrong.

A chess pool would need a completely different structure than a mining pool, I don't think much can be reused from the existing code and infrastructure of mining pools.

I support the idea of (and had for a while an interest in) a distributed chess playing agent, if such a thing doesn't exist yet.

There are some chess-related grid computing projects, such as chess960@home.

[bulanula]

I don't think GPUs are good for chess, but I could be wrong.

A chess pool would need a completely different structure than a mining pool, I don't think much can be reused from the existing code and infrastructure of mining pools.

I support the idea of (and had for a while an interest in) a distributed chess playing agent, if such a thing doesn't exist yet.

There are some chess-related grid computing projects, such as chess960@home.

You are 100% right. CPU is good for chess not GPU but OP is clueless.

They use Ivy Bridge CPUs not dedicated hardware or GPUs anymore.

[grondilu]

I once thought about an alternative protocol where miners, instead of crunching numbers, compete in chess.  A tournament occurs every ten minutes and only the winner is allowed to publish transactions.

The problem with this is that it would require a lot of disk space to store all the games of all tournaments, in order to verify the block chain.  Yet this data would not be totally useless as there might be some interesting chess games in it.

[EnergyVampire]

Thank you all for the replies,

After reading the posts and having a few discussions on IRC/Skype about this idea, I believe it might be best to drop the topic. This topic might be better suited for a website dealing with distributed computing, chess, game theory, combinatorics, etc.

Thanks for your time.

PS: After checkers was solved in 2007, Jonathan Schaeffer made the prediction that chess would be solved before 2250.

[Bitcoin Oz]

For some reason Im reminded of "tic-tac-toe" and the movie war games.