Bitcoin miner - page 8

 
Andrey F. Zelinsky:

It feels like financial market 'experts' have fallen from the moon on this forum.

Yes, I have the same opinion about you as well. Maybe you have the wrong answer.

There is a lot ofpopular-scientific literature on how the global financial system and each individual country is organised.

There arevideo lectures on youtube, for instance Katasonov -- he explains it popularly.

If you dislike academic,specialised literature, read or listen to Starikov who for the slowest of us is explaining clearly what and how and why.

It's not serious for you, it's time to have your own opinion.

If there is no authority for you -- "whiskers, paws, and the tail of Matrosskin's cat" is the only authority -- then even the grave hump will not fix the hump.

And this is out of bounds! What have I done to offend you?

 
Sergey Chalyshev:

That's out of bounds! What have I done to offend you?

No offence meant in this day and age - no more than an amusing turn of phrase - you can quote it.
 
Andrey F. Zelinsky:
No hard feelings in this day and age -- no more than a funny turn of phrase -- you can quote.

I don't understand, a quote for what?

You mean wish you too?

No, I won't, you won't.

 
Sergey Chalyshev:

I don't understand, a quote for what?

You mean wish you too?

No, I won't, you won't.


I don't understand what you're saying.

 
anonymous:

1. Search the internet for "mining pool"; websites usually have instructions on how to join. In general it looks like this: the mining software needs to pass a parameter-link to the pool, from which it will take the initial data from which the new blocks will be mined.

2. I have no idea. At a minimum, we need to develop software that distributes mining tasks + if found, quickly transfers it to bitcoin network nodes + calculates hash rate of all participants in the pool + controls fairness of the participants.

3. Let's say 100 MBit / ddos protection / redundancy in case of channel failure.

About one block per day: for that you need to have ~1/144 of power of all bitcoin network in your hands. Just calculate how many S9 ASICs you need for that and how much power they consume. That's the minimum cost, so to speak.


I see that you're knowledgeable in the subject, let me ask - I have one question - hashes are enumerated randomly, and essentially can be repeated for different users, or there is a unified mechanism that informs that such hash with source data exists (on which it was obtained) and it should not be counted? Or just pools are strong enough to merge information about hashes already calculated by the pool, which rules out recalculation?

 
Aleksey Vyazmikin:

I see that you understand the subject, let me ask - I am puzzled by one question - hashes are searched randomly, and in fact can be repeated for different users, or there is a unified mechanism notifying that such hash with source data exists (on which it was obtained) and it does not need to be calculated? Or is it that pools are strong because they merge information about hashes already calculated by the pool, which prevents recalculation?

Mining was last done in February 2011 :D But you can think and guess how it should work...

Independent miners and different pools naturally repeat their calculations. They do compete with each other.

In case of one pool, it makes sense to give miners blocks for full search, limited by some range of extraNonce (the nonce field is fully searched, and very, very fast); if no response is received from miners in certain time, inversely proportional to their hash rate - their contribution to the pool is not counted. Also, protection is needed in case of such attacks:

1. A miner takes a task from the pool and returns a response at a given time - "there is no suitable solution for the block in this task" - without actually performing any calculations. Protection is possible by distributing overlapping blocks or duplicating tasks to different miners and comparing solutions. Anyone who is repeatedly caught in such an attack is banned.

2. A miner uses the same equipment to participate in multiple pools, inflating a non-existent hash rate. Protection is possible by requiring the pool ID to be included in the block.

Random overshoot and lack of synchronisation in the case of a pool reduces the efficiency of the hardware, as the computation will be repetitive. Such pools will have lower payouts and will not withstand competition from those using the equipment more efficiently. For a standalone ASIC pool that does not require protection from the attacks mentioned above, also the most profitable option is to split the task into blocks rather than randomly trying to do it at random.

Google "Stratum mining protocol"

 
anonymous:

Last time I did mining was in February 2011 :D But you can think and guess how it should be arranged...

Independent miners and different pools naturally repeat their calculations. They do compete with each other.

In case of one pool, it makes sense to give miners blocks for full enumeration, limited to some range of extraNonce (the nonce field is full enumeration, and very, very fast); if no response is received from the miner in a certain time inversely proportional to his hash rate, his contribution to the pool activity is not counted. Also, protection is needed in case of such attacks:

1. A miner takes a task from the pool and returns a response at a given time - "there is no suitable solution for the block in this task" - without doing any real calculations. Protection is possible by distributing overlapping blocks or duplicating tasks to different miners and comparing solutions. Anyone who is repeatedly caught in such an attack is banned.

2. A miner uses the same equipment to participate in multiple pools, inflating a non-existent hash rate. Protection is possible by requiring the pool ID to be included in the block.

Random overshoot and lack of synchronisation in the case of a pool reduces the efficiency of the hardware as the computation will be repetitive. Such pools will have lower payouts and will not withstand competition from those using the equipment more efficiently. For a standalone ASIC hangar that doesn't need protection from the attacks mentioned above, also the most profitable option is to split the task into blocks rather than randomly trying it out.

Google "Stratum mining protocol"


Thanks for the info. So, if miners could share the hashes they've already re-created, the process would clearly speed up. It turns out that their hashes are stored on the PC, but do they take up much? Maybe there is a sense of trading hashes in computed hashes?

 
Aleksey Vyazmikin:

Thanks for the information. So, if miners could share the hashes they have already redistributed, the process would clearly speed up. It turns out that their hashes are stored on the PC, but do they take up much? Maybe there is a point in trading hashes in computed hashes?

They are not stored and it is useless to store them. You have no media for that much data, and even if you do, the access rate will be lower than the rate of re-generation.

 
Sergey Chalyshev:

Your reasoning is boggling my mind ))

Do you have any idea what you are writing about?

If you do then please explain what"inflation in depth" means.

Do you (let's be on a first-name basis, for fuck's sake) see the inverted commas ?

My assumption is that since bitcoin inflation is strictly rationed and world currencies inflation is not, bitcoin will divide and divide into smaller and smaller chunks. That's what I called it.

 
Sergey Chalyshev:

I think you have overdone it here. The rouble is not even close to depending on the dollar, and has not spun around but only depreciated. So far, the ruble's road has only been one way.

Exactly... And in his words - his job is 'money related'.

In my view, the ruble also depends very little on the dollar. Much more the ruble's dependence on energy resources.

But, there is little difference - both the rouble and the dollar are just paper, there is virtually no use value in it (and no value at all in non-cash currency), so one cannot say that 'they are secured by something'. Collateral - implies the unrestricted possibility of exchanging currency for the object of collateral and back. This is not the case with the dollar, the rouble, bitcoin or any other modern currency.

Reason: