How to eliminate Large mining pools

[Sherlock.Holmes]

Members A-D would work on the first phase, each on their own as if they were mining on their own account. Once one of the above members solved the first phase, they send the solution to a server, which sends it to one of members E-F who works on the second phase.

the idea of sharing work (pooling) is to have multiple workers that work in parallel to each other and not reply on other's solution. each worker gets its own work and submits the result of that work then gets the next parallel job.
if a worker has to wait for other worker to finish then it defeats the purpose and the optimization gained by parallel work.

The OP describes the first process to take longer than the second process. If for example, the first process takes twice as long as the second process, a mining pool can allocate 2/3 of miners to work on the first process and 1/3 on the second process, and given enough workers, there will be little 'downtime'.

As an alternative, workers could complete both the first and second processes, periodically send some type of verification they are working. Or a single entity could own many computers, each of which is a miner, and each of which is programmed to send the block reward to the same address controlled by the same entity.

I think you just misunderstood what I said. The difficulty of the first phase is greater than that of the second phase. It means that the difficulty of the first phase is greater than that of the node with the best luck in the second phase, not the node with medium luck.

The first phase of cooperation is not cost-effective. For example, if a miner has a node A, and he adds a new node, when the new node joins in a collaborative manner, their computing power is improved, but luck is not high. When new nodes join in a non-cooperative manner, their luck will increase. If the marginal effect of improved luck is greater than the marginal effect of increased computing power, then non-cooperation is the best choice for miners.

The cooperation in the second phase is meaningless. The work in the second phase needs to wait for all nodes in the first phase to be completed by an algorithm to select the best luck parameters before carrying out the tasks of the second phase. The difficulty of the second phase is very low for the best luck parameter, and the cooperation is meaningless.

[Sherlock.Holmes]

Once you see what they have done you realize you can only beat them with governments that hurt them.
No algorithm will work to beat them.
And as long as they exist 3 to 10 large pools will always be.

The leverage they can do by having cloud mining as part of their business model is far greater than most understand.

Look at their bitdeer plans
https://www.bitdeer.com/pricing/?id=1

Try to understand there is zero account and inspection for what they offer.
Try to understand that the s19pro they offer does 110th and 29watts.
I am 99% certain they can be tuned to do 100th and 22watts
In their case they build the gear so that loss of 10 th is meaning less.

Now see that they offer.  Many complex contracts based on s17pro

Imagine that they simply use 100th 22 watt gear in their cloud and charge you for. 2 s17 pros at 38 watts a markup

this is a huge power over charge.

In fact, we are building a network without mining pools. We improved on the underlying consensus mechanism, making the mining pools unable to operate and manage effectively.

A lot of independent miners will be left in the network, not pools.

[Sherlock.Holmes]

Thank you very much for your participation in the testnet.
After nearly a week of testing, a total of 38 independent miner nodes have participated in the test. The current block data is as follows:

1 The testnet stable run for 10 days

2 Generate about 20961 blocks

3 The block parameters are verified to meet the logical expectations

4 We have received emails from 17 independent miners, here we reply to the following related questions:

  4.1 Are the test tokens included in the mainnet?
  The tokens of the testnet are only used for testing and are not included in the mainnet.

  4.2 When does the test end?
  The test will end in about a week or two.

  4.3 When is social media established?
  We will still run in a free and open manner. In principle, we do not actively establish social media channels. All fans are welcome to actively establish social media channels and participate in the communication, thank you very much.

  4.4 What happens when the mining process crashes occasionally?
  After our testing, the mining process occasionally crashed due to insufficient memory. There will be no crashes on mining devices with memory up to 4G.

  4.5 How did a large-scale outage occur in the entire network?
  There was a large-scale interruption of about 30 minutes on the entire testnet, which was caused by the fluctuation of the network where the discovery server of the test network is located. Before the main network goes online, we will deploy multiple discovery servers in different places.

In addition, the testnet will continue to run for about a week or two, and will soon run the mainnet after completion of the test. Everyone is welcome to focus on it.

The description of the block parameters and the logic behind are being collated and will be published as soon as possible.

[hextobig]

I have a suggestion. If your two-phase proof-of-work is correct, are three-phase or more phases of proof of work safer? If so, why not use more phases of proof of work?

[angrybirds.allen]

A small number of large mining pools, or more small miners, which one is safer? 

[hextobig]

A small number of large mining pools, or more small miners, which one is safer? 

My point of view.
According to game theory, vested interests will strive to maintain the security of the network, whether it is a mining pool or a small miner, so the security of the two is equivalent.
From a practical point of view, if someone wants to maliciously damage the network, attacking the mining pool may be easier than attacking more scattered miners.
In addition, decentralization is more fair to miners, but the operation and maintenance costs of miners may be slightly higher than mine pools.
Scattering may be more conducive to scenes with higher fairness requirements such as random number, election, etc.

[Sherlock.Holmes]

After a period of testing, the luck project is stable.

The mining process and important parameters (see block explorer for details) are now explained.

Luck adopts a 2POW mechanism, in which the difficulty goal of the first phase is expressed as DifficultyAlpha. Calculate Luck after the completion of the first phase, Luck value range 1---1e+9, record lmax=1e+9. The calculation formula of the difficulty target in the second phase is as follows
DifficultyBeta=Basis*((lmax/(lmax-Luck))^2).

As the number of participating nodes increases, Basis will decrease sharply. The difficulty target in the second stage is more sensitive to the change of Luck, which is not conducive to alliance mining.

In the block explorer, you can see several new parameters, explained below.

Luck: The luck value of this round of mining
Basis: Basic control parameter of the mining difficulty in the second phase
FirstNonce: Random value of the first phase
DifficultyAlpha: Mining difficulty target value of the first phase
DifficultyBeta: Mining difficulty target value in the second phase
SecondNonce: Random value of the second phase

[Sherlock.Holmes]

A small number of large mining pools, or more small miners, which one is safer?  

My point of view.
According to game theory, vested interests will strive to maintain the security of the network, whether it is a mining pool or a small miner, so the security of the two is equivalent.
From a practical point of view, if someone wants to maliciously damage the network, attacking the mining pool may be easier than attacking more scattered miners.
In addition, decentralization is more fair to miners, but the operation and maintenance costs of miners may be slightly higher than mine pools.
Scattering may be more conducive to scenes with higher fairness requirements such as random number, election, etc.

The probability of fork in 2POW is smaller than that of traditional POW. This is because the difficulty of each node is different, and the probability of generating a block at almost the same time will be relatively small.

More small miners forming 51% attacks will be more difficult than large mining pools.

Once the 51% attack condition is formed, to successfully launch a selfish mining attack, 2POW is also more difficult than traditional POW.

[vjudeu]

It is still possible to create mining pools in this coin, because rewards do not have to be shared in the coinbase transaction or even in the same block. For example a mining pool can use stratum-like protocol and give only block header to each miner, as it is done today. Then, all rewards could be received by creating normal, non-coinbase transactions and including them into any blocks. And then it does not matter if it is one, two or N phases of proof-of-work, because it is all about how often you receive coins. If there will be a lot of miners, block difficulty will be high, so any miner will have two options: join some pool and get reward every day, or (for example) do solo-mining and get reward every year.

Edit: Maybe I missed something, but where I can find source code? There is a GitHub page, but there are only images for whitepaper and how to mine instructions. How can I get sources to build binaries from scratch?

Edit2:

balance: 14.999979000000002

Is there any floating-point type used? If not, what is the smallest unit of this coin?

[Sherlock.Holmes]

It is still possible to create mining pools in this coin, because rewards do not have to be shared in the coinbase transaction or even in the same block. For example a mining pool can use stratum-like protocol and give only block header to each miner, as it is done today. Then, all rewards could be received by creating normal, non-coinbase transactions and including them into any blocks. And then it does not matter if it is one, two or N phases of proof-of-work, because it is all about how often you receive coins. If there will be a lot of miners, block difficulty will be high, so any miner will have two options: join some pool and get reward every day, or (for example) do solo-mining and get reward every year.

Edit: Maybe I missed something, but where I can find source code? There is a GitHub page, but there are only images for whitepaper and how to mine instructions. How can I get sources to build binaries from scratch?

Edit2:

balance: 14.999979000000002

Is there any floating-point type used? If not, what is the smallest unit of this coin?

Two reasons for the formation of the mining pools: the advantage of power merger is far greater than the scattered power, that is, the marginal contribution of miners to the mining pool will be far greater than the output of their independent mining, which is the internal reason for the formation of the mining pools. The fluctuation of the income of individual miners is the external reason why independent miners join the mining pool.
In 2POW, when miners join the mining pool, their marginal contribution will be lower than their independent mining output, and will increase the management cost of the mining pool, which will allow miners to mine independently.
The fluctuation of the income of individual miners can be adjusted by increasing the number of nodes. The number of independent miners and nodes will be a market behavior.

The source code of the project will be opened within one to two weeks after the mainnet started.

The smallest unit is 0.000000000000000001.

[Sherlock.Holmes]

As conceivable once it is set up, you may extravagant putting it some place extraordinary. At the point when the opportunity arrives to move your signage about, recollect  how much do custom neon signs cost   that glass cylinders can be delicate, so attempt to keep away from contact with different surfaces where conceivable.

I think you should have sent it in the wrong place, please delete it, thank you

[vjudeu]

In 2POW, when miners join the mining pool, their marginal contribution will be lower than their independent mining output, and will increase the management cost of the mining pool, which will allow miners to mine independently.

Why their contribution will be lower? Mining pool can behave exactly as one, single, huge node. Mining pool will calculate luck and difficulty, giving each miner the same block header to mine. From miner's perspective it will be exactly as it is today, because mining pool will calculate everything that will be needed (no matter how many sub-layers of calculating difficulty will be there) and mining pool will give each node the same header, it is needed to calculate only once! If the pool will give each miner a different header, it will be pointless, as they can generate such without joining the pool. Pools are not created to be only network proxies, they are created to give more frequent rewards to each miner!

The fluctuation of the income of individual miners can be adjusted by increasing the number of nodes. The number of independent miners and nodes will be a market behavior.

If overall block difficulty will be high, a single miner will still create less artificial nodes than the whole mining pool, so any pool will provide more frequent rewards than such miner could achieve alone. Also, if the block time is so short, maybe someone would prefer getting rewards every hour or every day, instead of every 20 seconds. Then, when joining the pool (being one, huge, single node) it is equal to creating "1/n node" (n=number of miners connected to the pool). Each miner can create one, two, three or more nodes, but alone it is impossible to create "1/2 node" or "1/3 node". To lower fluctuaction of the income everyone is forced to join some pool or create it and invite someone.

Edit:

For example, we can change the puzzle from "Find a block whose hash value is lower than a specific target" to " A block was found, and the hash value of the digital signature of this block is lower than a specific target.". Such a puzzle-solving problem would prevent the mining pool administrator from outsourcing work to other miners.

But remember that anyone can still create a transaction. And since there will be Turing-complete smart contracts, it could be even easier than in Bitcoin. For example: a mining pool owner will give rewards only if you put some coins as a collateral and then if you mine new coins using key given by the server. Then, you can get your coins back only after mining pool owner will move these coins somewhere else, to an address only under its control.

So, yes, mining pool operator still can outsource this work to other miners, because at the beginning these miners will give them transaction they can get coins from if a miner will take these newly created coins.

Edit2: Also note that mining pool operator can safely outsource even signature mining to other miners, because they are creating blocks below target difficulty. So they cannot simply send such block to the network, because it will be rejected. It is worth something only inside the pool! Also, the pool can accept only blocks containing coinbase and a transaction sending something to the pool. Such transaction can be also send over lightning network, so no one will notice that the pool is actually receiving something!

[hextobig]

In 2POW, when miners join the mining pool, their marginal contribution will be lower than their independent mining output, and will increase the management cost of the mining pool, which will allow miners to mine independently.

Why their contribution will be lower? Mining pool can behave exactly as one, single, huge node. Mining pool will calculate luck and difficulty, giving each miner the same block header to mine. From miner's perspective it will be exactly as it is today, because mining pool will calculate everything that will be needed (no matter how many sub-layers of calculating difficulty will be there) and mining pool will give each node the same header, it is needed to calculate only once! If the pool will give each miner a different header, it will be pointless, as they can generate such without joining the pool. Pools are not created to be only network proxies, they are created to give more frequent rewards to each miner!

As OP pointed out in the paper, if the pool calculates the tasks of the first phase, then put the luck value and the difficulty of the second phase into the head, and then distributes the tasks of the second phase to other miners. In this case, the overall luck value is relatively low, the difficulty of the second phase will be very large, and the advantages of cooperation will be offset, as E2<E1 in the paper.

[autumngrass]

If  luck  has no mining pools, does this mean that all participants will compete fairly?

[Paperweight]

According to game theory, vested interests will strive to maintain the security of the network, whether it is a mining pool or a small miner, so the security of the two is equivalent.

Let's take this to the logical conclusion. Communism.

No mining rewards at all. Purely deflationary currency. Transaction fees go to nobody. Hodlers are incentivized to mine blocks of transactions because it increases the scarcity value of their own hodlings.

However, there's a tragedy of the commons because your hodlings go up in scarcity whether you mine or not. There's nothing new under the sun. Perhaps this can be solved by purging any account that doesn't contribute enough signed mining shares.

[Sherlock.Holmes]

According to game theory, vested interests will strive to maintain the security of the network, whether it is a mining pool or a small miner, so the security of the two is equivalent.

Let's take this to the logical conclusion. Communism.

No mining rewards at all. Purely deflationary currency. Transaction fees go to nobody. Hodlers are incentivized to mine blocks of transactions because it increases the scarcity value of their own hodlings.

However, there's a tragedy of the commons because your hodlings go up in scarcity whether you mine or not. There's nothing new under the sun. Perhaps this can be solved by purging any account that doesn't contribute enough signed mining shares.

We are not trying to build communism. We just continued the design of Satoshi Nakamoto's competitive consensus, and we tried our best to make the mining process fairer in order to realize the concept of "one-CPU-one-vote" proposed by Nakamoto.

[beneri]

Thank you

[mikeywith]

That's a great effort you put there, but I have to sadly inform you that this isn't going to work IMO at least.

You are trying to solve an economical issue rather than a technical issue, this whole "one-cpu-one-vote" is highly misunderstood and I am having a very hard time believing that Satoshi didn't see large pools coming, according to Satoshi:

Only people trying to create new coins would need to run
network nodes.  At first, most users would run network nodes, but as the
network grows beyond a certain point, it would be left more and more to
specialists with server farms of specialized hardware

https://www.mail-archive.com/cryptography@metzdowd.com/msg09964.html

Everything that has to do "money" will always end up in the hands of a few players, this is just how the universe works, bitcoin mining has even a worse issue where the less profitable it becomes to mine -  the less "decentralized" it gets, and as block rewards keep going down, it only gets worse.

Mining is pretty fair the way it is now, it doesn't favor anybody and nobody has any patented/exclusive methods that give them any advantages over other miners.

Another thing that many people seem to be missing is that running your own node is freaking difficult and cost a lot of money despite of the small block size, let alone having to run a dozen of them.

Forcing miners to run their own nodes will only lead to fewer miners (as people) and eventually, mining will not only be "pool" centralized, the hashpower itself will be centralized, that is an issue that we currently don't have and it would be best not to create.

I don't want to hijack this thread with my own "proposal" to solve this economical problem, but as the old saying goes "if you can't beat them, join them", and to do that, we should push for a "community mining pool" that is competitive enough as a way of protection against large pools.

Facts:

1- The majority of individual miners want to make the most money most simply and cheaply. (current plug-n-play system works perfectly)
2- The majority of individual miners DON'T want luck to be in their way and they want steady and stable payouts. (PPS pools do the job perfectly)

You take that away from them, and they will be gone.

There is, however, a large portion of miners who would be willing to shift their hashrate to a "trusted" pool in order to give it more power to compete against all other "untrusted" pools given that the "trusted" pool works exactly as the other pools, the majority of them are NOT willing to set up their own node/s and go through all the trouble, a good percentage of them don't know how to do it in the first place, but you know who has the money, skills and skin to do all that? large cooperations and farms, so your proposal might just be a great push towards more centralization.

[Kakmakr]

Just a question... Could something be developed or the Bitcoin Protocol be changed to split mining? So, let's say you have more than X amount of hashing power ....you get pushed to a "Big pool" mining side of Bitcoin, where the big mining farms compete with each other with higher difficulty and then if you have less than Y amount of hashing power, you get placed in another Pool where the difficulty is much lower.

This will enable 2 groups to continue mining, with ASIC mining specialists on the one side and your smaller CPU/GPU group on the other. 

"Luck" will make it difficult for miners to plan profitability and future scaling of their mining farms, so I think that will just over complicate the whole process. 

[Sherlock.Holmes]

That's a great effort you put there, but I have to sadly inform you that this isn't going to work IMO at least.

You are trying to solve an economical issue rather than a technical issue, this whole "one-cpu-one-vote" is highly misunderstood and I am having a very hard time believing that Satoshi didn't see large pools coming, according to Satoshi:

Only people trying to create new coins would need to run
network nodes.  At first, most users would run network nodes, but as the
network grows beyond a certain point, it would be left more and more to
specialists with server farms of specialized hardware

https://www.mail-archive.com/cryptography@metzdowd.com/msg09964.html

Everything that has to do "money" will always end up in the hands of a few players, this is just how the universe works, bitcoin mining has even a worse issue where the less profitable it becomes to mine -  the less "decentralized" it gets, and as block rewards keep going down, it only gets worse.

Mining is pretty fair the way it is now, it doesn't favor anybody and nobody has any patented/exclusive methods that give them any advantages over other miners.

Another thing that many people seem to be missing is that running your own node is freaking difficult and cost a lot of money despite of the small block size, let alone having to run a dozen of them.

Forcing miners to run their own nodes will only lead to fewer miners (as people) and eventually, mining will not only be "pool" centralized, the hashpower itself will be centralized, that is an issue that we currently don't have and it would be best not to create.

I don't want to hijack this thread with my own "proposal" to solve this economical problem, but as the old saying goes "if you can't beat them, join them", and to do that, we should push for a "community mining pool" that is competitive enough as a way of protection against large pools.

Facts:

1- The majority of individual miners want to make the most money most simply and cheaply. (current plug-n-play system works perfectly)
2- The majority of individual miners DON'T want luck to be in their way and they want steady and stable payouts. (PPS pools do the job perfectly)

You take that away from them, and they will be gone.

There is, however, a large portion of miners who would be willing to shift their hashrate to a "trusted" pool in order to give it more power to compete against all other "untrusted" pools given that the "trusted" pool works exactly as the other pools, the majority of them are NOT willing to set up their own node/s and go through all the trouble, a good percentage of them don't know how to do it in the first place, but you know who has the money, skills and skin to do all that? large cooperations and farms, so your proposal might just be a great push towards more centralization.

Sorry to see your comments so late.
First of all, we have not done any form of IXO, and we do not plan to do IXO in the future. We just conducted a social experiment of a new mining method.

Our project has been online for nearly a month. According to our estimations, there are already nearly thousand of nodes in independent mining now. There have been hundreds of nodes in joint mining, and independent nodes can still get normally and proportionally every day, which is impossible under the traditional POW mechanism.

Next, I will respond to your comments as follows:
1 "one-cpu-one-vote" is indeed the vision that Satoshi Nakamoto first proposed, and the mining pools have already appeared before Nakamoto left. Obviously the mining pools violated "one-cpu-one-vote".
2 Joining a mining pool is not the true intention of most miners. In fact, it is just a choice that miners are forced to make under the monopoly of power. Miners joining the mining pool may not be able to simply obtain stable income. Under the condition of mining by luck, it may not be impossible to obtain stable income simply.
3 Decentralization is not an end, it is just a way to obtain network security. more independent nodes = more decentralized = more network security.