About efficiency

[juanlive]

One of the few things of bitcoin that makes noise in my head is the extremely unefficient system to mine and check transactions. If I understood well, the complexity of calculations is artificially sat up, so I wonder, could all these petaflops be used, at the time than checking transactions, for something useful too, like making calculations for the particle accelerator?

[DannyHamilton]

One of the few things of bitcoin that makes noise in my head is the extremely unefficient system to mine and check transactions. If I understood well, the complexity of calculations is artificially sat up, so I wonder, could all these petaflops be used, at the time than checking transactions, for something useful too, like making calculations for the particle accelerator?

There are 0 petaflops used in mining.  I'm pretty sure there are 0 flops used in mining.

flops stands for FLoating point Operations Per Second.

All mining is done with Integer operations.

Furthermore, mining consists of specially designed hardware that does only one thing, and does that one thing very well.  Specifically, it calculates the double SHA256 hash of a bitcoin block header as fast as it can.  It isn't capable of doing much of anything else.

Fortunately, this effort isn't wasted.  It is used for the very important purpose of securing over $15,000 million dollars worth of wealth and the global payment network where that wealth is used for highly reliable, high speed, irreversible, low fee transactions.

[juanlive]

What I mean is: is really necessary to use all that power?
I heard that complexity is artifically induced only to maintain a uniform metabolism.

So, if complexity is used just to find a code starting with many zeroes, couldn't be used another kind of data (from the particles accelerator or any other ongoing scientific experiment) in order to make the tests, so at the end of the process, two results are achieved: transaction confirmation and a scientific problem solved?

J

[Foxpup]

I heard that complexity is artifically induced only to maintain a uniform metabolism.

You heard wrong. The complexity is to make it extremely difficult to make fraudulent changes to the transaction record. To do this, you need a function that has the following properties:
1) It must be very difficult to calculate a solution;
2) It must be very easy to verify that the solution has been calculated correctly; and, most relevantly,
3) It must not rely on any outside source of data that could be tampered with to either allow fraudulent solutions, or prevent valid solutions from being accepted.

[DeathAndTaxes]

I would also add
4) Must be probabilistic
5) Must reference prior solution to prevent presolving.

[juanlive]

Cool! So, miners could get input from the particle accelerator (to name an option).

There is so much data to be processed from CERN that it could provide random data to miners to be processed (assuring the data accomplish the 5 points you have mentioned) and then take te results from the blockchain.

Using all the power of miners (the most powerful supercomputer ever created) just to find codes beggining with zeroes desn't make me feel very clever, especially when there are so many needs for computing power in many areas from physics to wheather forecasts.

[juanlive]

It would also provide an enormous intrinsec value added to Bitcoin, a value that relies not only in its users, but in the whole community.

[Foxpup]

There is so much data to be processed from CERN that it could provide random data to miners to be processed (assuring the data accomplish the 5 points you have mentioned) and then take te results from the blockchain.

Explain how this satisfies points 2 and 3.

[trell0z]

If it was possible to use that power at the same time to do things like that, yes it would be awesome. But if you're using that data to do it, there would always be people with access, and however good the intentions, where's there's people with access to information that can make you a lot of money.. someone will take that opportunity.
If you really want to help with calculations on different projects, get rich on coins and then buy dedicated hardware to help, can seldom have the best of both worlds.

[vlaladim]

I am wondering as long as there are companies offering 100Th cloud mining subscriptions after February... it is interesting for me what will happen... How much bitcoins will a 1TH miner/machine be making per day....
What you think?

[juanlive]

There is so much data to be processed from CERN that it could provide random data to miners to be processed (assuring the data accomplish the 5 points you have mentioned) and then take te results from the blockchain.

Explain how this satisfies points 2 and 3.

Well, points 2 could be achieved just by clustering data from CERN into chunks that would need to be force bruted in order to check, say, if certain trace could have been generated by a quark. Once a certain match is found, then it would be easy to check if it really matches or not, just by following the algorothm of the particle mentioned by the miner.

For the point 3, you just have to assure that the data provided by the CERN has not been processed before, may be just a direct flow from the detectors to the miners in a random way.

[juanlive]

If you really want to help with calculations on different projects, get rich on coins and then buy dedicated hardware to help, can seldom have the best of both worlds.

I'm not interested in doing that myself. What I'm propossing is that minery could be more efficient.

Actual method of mining is like saying: to protect data from being altered by anyone, we will rely on wasting collectivelly as much power as we can, so nobody for himself could reach the same potential.

All that power is wasted only to fight possible frauds. It's an intelligent method, but not an efficient one in terms of ecology and economy for the society, as it sounds to me.

And yes, that scientific data would be available to the public, but may be it only could make sense if you integrate at the CERN, and not just by looking at disperse chunks of data.

[DeathAndTaxes]

So CERN then becomes the central authority and you just violated rule 3.

Making a centrally controlled virtual currency is easy.  Just look at Linden dollars.  Making a decentralized currency is difficult.   You can't have a central authority in charge of issuing the work for proof of work and then pretend it is decentralized.

[juanlive]

So CERN then becomes the central authority and you just violated rule 3.

Making a centrally controlled virtual currency is easy.  Just look at Linden dollars.  Making a decentralized currency is difficult.   You can't have a central authority in charge of issuing the work for proof of work and then pretend it is decentralized.

I don't mean that CERN has central authorithy in any way. It would be only the (one of the) providers of algorythms. The check would be performed by the community, almost exactly as now, based on the own rules of each algorythm.

[DeathAndTaxes]

So CERN then becomes the central authority and you just violated rule 3.

Making a centrally controlled virtual currency is easy.  Just look at Linden dollars.  Making a decentralized currency is difficult.   You can't have a central authority in charge of issuing the work for proof of work and then pretend it is decentralized.

I don't mean that CERN has central authorithy in any way. It would be only the (one of the) providers of algorythms. The check would be performed by the community, almost exactly as now, based on the own rules of each algorythm.

Yeah that doesn't make any sense.  Maybe you should start with researching what miners currently do, then you can at least have a discussion which doesn't involve randomly placed buzz words.

[DannyHamilton]

Yeah that doesn't make any sense.  Maybe you should start with researching what miners currently do, then you can at least have a discussion which doesn't involve randomly placed buzz words.

+1

[juanlive]

Sorry, my English is not good.

I know what the miners does, I just can't express well in English.

I know that they look for a key, by brute force, that produces certain hashes. No much mystery there.

I just propose that some scientific centers could provide problems like these to be solved, which may be of value for them at the same time than serving to prove transactions.

Hashes providers would be benefited from that, and they would be glad to pay miners an extra fee for that job, and the whole society would benefit frmo all that computing power now dedicated to just find unsignificant hashes.

[DeathAndTaxes]

I know that they look for a key, by brute force, that produces certain hashes. No much mystery there.

That is not correct, or that isn't the whole answer.

Miners aren't looking for just a "key" (nonce) which produces a certain hash.    Miners are looking for BLOCKHEADER + nonce which when hashed is below the difficulty target.  The blockheader binds the block to the set of txs, the chain, and the prior block.  This means the input is both deterministic and can't be precomputed.   That doesn't exist if a central authority is handing out work.  Just using broad vague terms is useless.   This is a NON TRIVIAL PROBLEM it has been proposed (in similar useless broad terms) hundreds of times by brand new forum members who "have the answer" and five years later .... nope still doesn't work.

If you think it can work you need to do so research not use vague terms like "scientific cetners can provide problems".  Exact specific implementation details on how the "mining" would work and how it would secure the network.   The proof is in the details and when you start trying to construct such a system either you will realize it can't be done or others can point out specific attacks/flaws that you have no resolution for.  The reality is you can't, nobody can.  The proposed "solution" is flawed from the beginning.  

If scientific center(s) are providing the work then they become the central authority.
If there is a central authority then you don't have a decentralized system.
 

[juanlive]

If scientific center(s) are providing the work then they become the central authority.
If there is a central authority then you don't have a decentralized system.
 

Thank you for the explanation. You're right, I'll come with a more explicit solution if I find it.

But I don't mean the provider should become the authority in any way: just the producer of a kind of "salt" to be added to the chain.

A metaphore could be to try to use the mining system to cook some food for everyone. I'll come with a more serious explanation if I manage to make it.

But it's true that my idea is broad. I didn't mind to find a solution, but just to inspire others. May be one solution could be just to use the heat of the servers for something else.

Thank you for your attention.