Bitcoin Forum

Bitcoin => Bitcoin Discussion => Topic started by: boounce on June 06, 2011, 11:53:06 PM



Title: bitcoin competitor that folds proteins?
Post by: boounce on June 06, 2011, 11:53:06 PM
Back in the day I used to donate extra CPU cycles for gene folding and SETI.   Is it possible to create a cryptocurrency that does this instead of mining?  I've been searching the forums, but I don't see much discussion on the topic. 
 


Title: Re: bitcoin competitor that folds proteins?
Post by: FreeMoney on June 07, 2011, 12:03:26 AM
Lots of problems.

One problem: If the work exists before it is time to solve it then people can "save up" solutions and cause short term double spends eventually regardless of how small their power is. With Bitcoin the problem is "made up" when the previous block is found.


Title: Re: bitcoin competitor that folds proteins?
Post by: boounce on June 07, 2011, 12:27:58 AM
Lots of problems.

One problem: If the work exists before it is time to solve it then people can "save up" solutions and cause short term double spends eventually regardless of how small their power is. With Bitcoin the problem is "made up" when the previous block is found.

Wouldn't they still need to devote the power to solve the problem?  Perhaps you could break the problems into smaller chunks to be randomly checked out by single solvers.



Title: Re: bitcoin competitor that folds proteins?
Post by: Dobrodav on June 07, 2011, 01:10:11 AM
There is some hard argument`s against using hash power to something else than just to confirming transaction.
I am forgot them :) Well you should start looking in Satoshi paper.


Title: Re: bitcoin competitor that folds proteins?
Post by: gigitrix on June 07, 2011, 02:34:43 AM
You need to be able to verify the answer for it to be useful to Bitcoin.

If computers can easily verify the answer, it's highly unlikely that it can be of any use to science. Simple as.


Title: Re: bitcoin competitor that folds proteins?
Post by: FreeMoney on June 07, 2011, 05:34:10 AM
It takes trillions of hashes to solve a block and 1 hash to verify and the difficulty scales easily on a whim. How are you going to get an endless supply of protein problems easily verifiable of arbitrary difficulty that can't be solved beforehand?