please delete | Bitcointalksearch.org

SapphireSpire

jr. member

Activity: 49

Merit: 38

Quote from: odolvlobo on July 28, 2015, 11:10:02 AM

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena

How do you detect whether a miner is actualy creating only a single hash and thus relying on luck or secretly creates many hashes in order to have a higher chance to win?

Only one hash would be allowed from every node so, you'd need a separate node for every hash you want to play. If the network uses IP addresses to distinguish nodes, how many IP addresses do you think it would take to increase your odds of winning when the odds of winning are 2²⁵⁶-1 to 1?
Answer: 1000x more than exist.

Having two nodes would double my chances, and double my income. Also, the odds are not 1 in 2²⁵⁶-1 because you wrote that the lowest hash wins. If there are 3 miners, then 1 of the three always wins -- the odds are 1/3 (unless I have two nodes, in which case my odds are 2/3).

Oops- you're right about that- my bad. So how do I address this issue? hmmm

Miners would have to pay a fee per block equal to some fraction of the block reward. If we decide to tolerate 'cheaters' playing up 1000 hashes per block, then the mining fee must be 0.1% of the block reward. If we decide to tolerate only 100 hashes per block, then the mining fee must be 1% of the block reward. But if we want to limit all miners to just one hash per block, the mining fee must always be half of the block reward. To keep the mining fee low, the block reward needs to be low. The block reward should change dynamically in direct proportion to mining participation in order to maintain a target level of participation.

SapphireSpire

jr. member

Activity: 49

Merit: 38

Quote from: shorena on July 27, 2015, 10:33:40 AM

How do you detect "spam TX" that are just put in the block by the miners to reach the maximum amount possible, e.g. send coins from miner A to miner A's 2nd (3rd...) address?

Why would a miner do that when there is nothing to be gained from it? There are plenty of natural transactions to process for fees?

Quote from: shorena

How do you detect whether a miner is actualy creating only a single hash and thus relying on luck or secretly creates many hashes in order to have a higher chance to win?

Only one hash would be allowed from every node so, you'd need a separate node for every hash you want to play. If the network uses IP addresses to distinguish nodes, how many IP addresses do you think it would take to increase your odds of winning when the odds of winning are 2²⁵⁶-1 to 1?
Answer: 1000x more than exist.

Quote from: shorena

But I have my custom wallet, that does what I want. How do you make sure that my cheating attemt is detected by the network?

This has already been addressed by other parts of the Bitcoin protocol. It has nothing to do with the change to the mining process I'm proposing here.

Quote from: shorena

For now you have not even explained how you want to make sure your system is not cheated.

You haven't yet explained how you're planning to cheat it.

Quote from: shorena

Its more expensive to set up several million nodes than it is to calculate the same amount of hashes.

That's good, it discourages this behavior. The potential return from winnings doesn't justify the investment.

Quote from: shorena

How do you make the distinction between 1st and 57845784th hash?

Math.

shorena

copper member

Activity: 1498

Merit: 1528

No I dont escrow anymore.

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena on July 27, 2015, 10:33:40 AM

How do you detect "spam TX" that are just put in the block by the miners to reach the maximum amount possible, e.g. send coins from miner A to miner A's 2nd (3rd...) address?

Why would a miner do that when there is nothing to be gained from it? There are plenty of natural transactions to process for fees?

Because your proposed rules make it beneficial to have a full block and there are times when there are not enough transactions.

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena

How do you detect whether a miner is actualy creating only a single hash and thus relying on luck or secretly creates many hashes in order to have a higher chance to win?

Only one hash would be allowed from every node so, you'd need a separate node for every hash you want to play. If the network uses IP addresses to distinguish nodes, how many IP addresses do you think it would take to increase your odds of winning when the odds of winning are 2²⁵⁶-1 to 1?
Answer: 1000x more than exist.

Why would the odds be 2²⁵⁶-1 to 1? You said lowest hash not a specific.

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena

But I have my custom wallet, that does what I want. How do you make sure that my cheating attemt is detected by the network?

This has already been addressed by other parts of the Bitcoin protocol. It has nothing to do with the change to the mining process I'm proposing here.

Yes it does. The network as it currently is does not know how many hashes a miners has created and it does not need to care, because due to the difficulty rules its enforced that finding the hash is hard.

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena

For now you have not even explained how you want to make sure your system is not cheated.

You haven't yet explained how you're planning to cheat it.

I have several times. I can just calculate more than a single hash, yet you just ignore this and answer that your version of the client will magically enforce this.

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena

Its more expensive to set up several million nodes than it is to calculate the same amount of hashes.

That's good, it discourages this behavior. The potential return from winnings doesn't justify the investment.

Correct, I could just cheat instead of following your rules.

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena

How do you make the distinction between 1st and 57845784th hash?

Math.

I think we are done here.

odolvlobo

legendary

Activity: 4466

Merit: 3391

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena on July 27, 2015, 10:33:40 AM

How do you detect "spam TX" that are just put in the block by the miners to reach the maximum amount possible, e.g. send coins from miner A to miner A's 2nd (3rd...) address?

Why would a miner do that when there is nothing to be gained from it? There are plenty of natural transactions to process for fees?

Because you wrote that the higher number of transactions would win. Anyway, if you assume that all the blocks are full then that criteria is not important.

Quote from: SapphireSpire on July 28, 2015, 07:23:41 AM

Quote from: shorena

How do you detect whether a miner is actualy creating only a single hash and thus relying on luck or secretly creates many hashes in order to have a higher chance to win?

Only one hash would be allowed from every node so, you'd need a separate node for every hash you want to play. If the network uses IP addresses to distinguish nodes, how many IP addresses do you think it would take to increase your odds of winning when the odds of winning are 2²⁵⁶-1 to 1?
Answer: 1000x more than exist.

Having two nodes would double my chances, and double my income. Also, the odds are not 1 in 2²⁵⁶-1 because you wrote that the lowest hash wins. If there are 3 miners, then 1 of the three always wins -- the odds are 1/3 (unless I have two nodes, in which case my odds are 2/3).

VirosaGITS

legendary

Activity: 1302

Merit: 1068

Quote from: BitUsher on July 27, 2015, 07:42:48 PM

Quote from: VirosaGITS on July 27, 2015, 06:57:46 PM

While we're at it, maybe the block reward should grant subsidy bonus to people who use renewable energy? Tongue

The principles of economics and laws of thermodynamics already incentivize miners to use renewable energy. Many mining farms are now using hydroelectric power and some even geothermal. The one temporary exception is a few are using cheap coal but that is only because certain states are both protecting polluters and even subsidizing their behavior.

Yeah, i love my hydro. Just pointing out the thread was going on a tangent. The waste of electricity really is significant if it is of non clean/non renewable source. Blaming BTC of creating pollution is like blaming Americans of eating at McDonalds.

BitUsher

legendary

Activity: 994

Merit: 1035

Quote from: VirosaGITS on July 27, 2015, 06:57:46 PM

While we're at it, maybe the block reward should grant subsidy bonus to people who use renewable energy? Tongue

The principles of economics and laws of thermodynamics already incentivize miners to use renewable energy. Many mining farms are now using hydroelectric power and some even geothermal. The one temporary exception is a few are using cheap coal but that is only because certain states are both protecting polluters and even subsidizing their behavior.

VirosaGITS

legendary

Activity: 1302

Merit: 1068

While we're at it, maybe the block reward should grant subsidy bonus to people who use renewable energy? Tongue

Since we're into far fetched and convoluted ideas that cant/wont be implemented anyways~ xD

jonald_fyookball

legendary

Activity: 1302

Merit: 1008

Core dev leaves me neg feedback #abuse #political

Quote from: ?? on ??

Quote from: jonald_fyookball on July 26, 2015, 11:22:26 PM

It will simply become an arms race of who can set up the most IP addresses and run the most mining scripts from them.

Better than the current arms race of who can consume the most Kwh.

Quote from: jonald_fyookball

The overall effect will not be a more secure network.

How would the network be less secure?

Well, assuming you're going to use IP addresses as the means
of determining a unique miner, then its all about who has the IP
addresses.

Since IP addresses are issued by IANA ( Internet Assigned Numbers Authority)
they are subject to control by central authority and everything that flows
down from that and the whole ecosystem of IP addresses.

This is inferior to the open, permissionless, non politicized system
that proof of work offers.

shorena

copper member

Activity: 1498

Merit: 1528

No I dont escrow anymore.

Quote from: ?? on ??

Quote from: odolvlobo on July 26, 2015, 11:00:11 PM

In the current system, the first miner to find a hash gets the block and the interval is on average 10 minutes. However, in your system, I assume you need to compare hashes and/or transaction quantities until a particular time is reached and then accept the best. Is it possible to synchronize time in a decentralized system?

My system isn't based on whose first so it isn't necessary to synchronize time. My system would encourage miners to wait as long as possible to accumulate as many transactions as possible before generating a hash which would delay processing. That could be handled by limiting the number of transactions in a block. All the blocks would then be the same size so whoever generates the smallest hash would win.

How do you detect "spam TX" that are just put in the block by the miners to reach the maximum amount possible, e.g. send coins from miner A to miner A's 2nd (3rd...) address?
How do you detect whether a miner is actualy creating only a single hash and thus relying on luck or secretly creates many hashes in order to have a higher chance to win?

Quote from: ?? on ??

Quote from: odolvlobo

My point was that if the block with the most transactions wins, then a miner will fill the block with "fake" transactions in order to have more transactions than anyone else.

If you're wallet is working properly, it won't allow fake transactions. For a new block to be valid, all transaction inputs must correspond with the unspent outputs of transactions in previously confirmed blocks.

But I have my custom wallet, that does what I want. How do you make sure that my cheating attemt is detected by the network?

Quote from: ?? on ??

Quote from: odolvlobo

Anyway, after reading your proposal again, it seems like you are suggesting that each miner gets to generate one and only one hash. That raises the issue of how to determine whether or not two hashes were produced by the same miner. One of the beauties of Bitcoin, is how it has side-stepped this issue.

My system would create an incentive for miners to setup as many wallets as possible so as to generate as many hashes as possible per block but I don't have a problem with this because it's more energy efficient and it wouldn't compromise security. Even if a miner controlled a vast majority of wallets on the network it wouldn't enable them to fork the blockchain.

For now you have not even explained how you want to make sure your system is not cheated. Its more expensive to set up several million nodes than it is to calculate the same amount of hashes. How do you make the distinction between 1st and 57845784th hash?

jonnybravo0311

legendary

Activity: 1344

Merit: 1024

Mine at Jonny's Pool

Quote from: ?? on ??

Quote from: odolvlobo

My point was that if the block with the most transactions wins, then a miner will fill the block with "fake" transactions in order to have more transactions than anyone else.

If you're wallet is working properly, it won't allow fake transactions. For a new block to be valid, all transaction inputs must correspond with the unspent outputs of transactions in previously confirmed blocks.

They don't have to be fake transactions. You can put as many real transactions as your heart desires, including transactions you make on your own to fill in the block, which every intelligent miner in your system would do. They'd fill their blocks to the size limit with whatever transactions they could. Since the person who finds the block also receives the transaction fees, the net result of a miner filling a block with "made up" transactions is that he gets the fees he spent to send the transactions right back into his wallet for producing the block.

BitUsher

legendary

Activity: 994

Merit: 1035

First of all Bitcoin PoW is not wasteful as it is a very valuable task.

Determining a suitable PoW algo that cannot be gamed and is both efficient, dynamic and cannot be manipulated is extremely difficult. This is evidenced by the fact that even the altcoins like foldingcoin and curecoin cannot depend on the security by means of computational cycles with protein folding alone. Primecoin is interesting PoW but some question the usefulness of finding Cunningham chains and bitwin chains and question its security and its risk from botnet attacks.

It is hypothetically possible that someone will find an PoR algo in the future that provides a reliable form of security but unlikely and we shouldn't assume that their is one.

Our efforts would be better spent on finding ways to recycle ASIC heat waste for a second purpose other than merely securing the blockchain.

TransaDox

full member

Activity: 219

Merit: 102

I am in favour of the clients doing more of the work but for slightly different reasons, i.e. not energy consumption.

If the clients verified and included transactions instead of miners, then all clients could share in the processing fees and at the same time dilute the 51% attack to all network nodes not just the smaller subset of miners that have specialist hardware. While dedicated mining farms would still get the lions share of fees and still get the mining rewards, everyone would get something for their participation and support of the network while decreasing the possibility for a cartel to capture the network.

Leave the big boys mining and let everyone else handle the day-to-day running of the network.

Amph

legendary

Activity: 3248

Merit: 1070

Quote from: ?? on ??

Quote from: Amph on July 27, 2015, 02:28:21 AM

there is no wasting energy, if that energy can be used to secure something so valuable as bitcoin, which is already true today and will be more in the future

this is another of those non-issues that continue to be brought back...

It's a waste of energy if there is a more efficient way to achieve the same (or better) degree of security.

but there isn't otherwise we might use that way, but if we were using that way, someone else would come up and say the same thing that was said in this thread, you see where we are going?

Amph

legendary

Activity: 3248

Merit: 1070

there is no wasting energy, if that energy can be used to secure something so valuable as bitcoin, which is already true today and will be more in the future

this is another of those non-issues that continue to be brought back...

jonald_fyookball

legendary

Activity: 1302

Merit: 1008

Core dev leaves me neg feedback #abuse #political

Quote from: odolvlobo on July 26, 2015, 11:00:11 PM

Anyway, after reading your proposal again, it seems like you are suggesting that each miner gets to generate one and only one hash. That raises the issue of how to determine whether or not two hashes were produced by the same miner. One of the beauties of Bitcoin, is how it has side-stepped this issue.

This.

It will simply become an arms race of who can set up the most IP addresses and run the most mining scripts from them.

The overall effect will not be a more secure network.

odolvlobo

legendary

Activity: 4466

Merit: 3391

Quote from: ?? on ??

Quote from: odolvlobo

1. You will need a time limit. That is a problem.

For what? There is already a block interval of ~10 minutes. That doesn't have to change.

In the current system, the first miner to find a hash gets the block and the interval is on average 10 minutes. However, in your system, I assume you need to compare hashes and/or transaction quantities until a particular time is reached and then accept the best. Is it possible to synchronize time in a decentralized system?

Quote from: ?? on ??

Quote from: odolvlobo

2. Who wins, the person with the lowest hash value or the person with the most transactions?

In keeping with the current consensus rules, the block with the most transactions should take priority.

Quote from: odolvlobo

3. Or is it the lowest hash value per transaction? That will only encourage miners to fill up blocks with "fake" transactions.

No. The idea is not about hashing transactions individually but it is possible for a block to contain only one transaction. It probably happened a few times in the very beginning but it would only ever happen again if Bitcoin dies out.

My point was that if the block with the most transactions wins, then a miner will fill the block with "fake" transactions in order to have more transactions than anyone else.

Anyway, after reading your proposal again, it seems like you are suggesting that each miner gets to generate one and only one hash. That raises the issue of how to determine whether or not two hashes were produced by the same miner. One of the beauties of Bitcoin, is how it has side-stepped this issue.

odolvlobo

legendary

Activity: 4466

Merit: 3391

First, the "source" of the random number is not important here. It takes the same amount of energy regardless of what it is based on.

Beyond that, a few things that need to be worked out:

1. You will need a time limit. That is a problem.
2. Who wins, the person with the lowest hash value or the person with the most transactions?
3. Or is it the lowest hash value per transaction? That will only encourage miners to fill up blocks with "fake" transactions.

SapphireSpire

jr. member

Activity: 49

Merit: 38

nothing to see

Topic: please delete (Read 1958 times)