Preventing Asic mining [fork] after next halving would solve a lot of problems.

[cr1776]

Is there a chance to bring back mining to the normal user now?
(Yes, very generic question, hope to have several points of view on that topic)

This assumes "normal" users are not miming now, I think plenty are in the pools. But...

1. Buy an ASIC miner :-) .  Anyone can.

2. Soon, the ASICs will hit the wall of silicon like CPUs and GPUs are so you won't see orders of magnitude improvements like between CPU -> GPU -> FPGA -> ASIC.  When that happens you'll see much slower hash rate increases like differences between GPU generations. Sure, the hash rate was growing, just slower.  Even getting a next gen ASIC might only buy you a small percentage gain which will make it easier to predict return vs orders of magnitude. Just like you don't see computers speed up two orders of magnitude in a year, hash rate growth will slow.  Perhaps then more "normal" users will mine.

[-ck]

Satoshi for sure never planned asic mining. 99% of the bitcoin community would benefit when asic mining is prohibited.

The current system where every user is a network node is not the intended configuration for large scale.  That would be like every Usenet user runs their own NNTP server.  The design supports letting users just be users.  The more burden it is to run a node, the fewer nodes there will be.  Those few nodes will be big server farms.  The rest will be client nodes that only do transactions and don't generate.

While not explicitly stating asics, he did describe very well the move to large farms doing all the mining, and asics are just a natural evolution all of us anticipated.

[MaxDZ8]

ASIC is not the problem, GPU mining would have ended up exactly the same given enough time. Mega GPU mining farms will inevitable rise.

I think it is worth noticing as long as computation stays in the real of mass-produced hardware, it keeps pouring money in a pressure cooker which will help other applications as well. At the very least, financing new process nodes in volumes. I think there's considerable value in keeping stuff on general hardware.

[yumei]

Quote from: yumei on October 05, 2014, 10:30:49 PM
Isn´t it theoretically possible to build a algorithm which makes actual asic miners worthless.

No.

An ASIC is just a computer. It's more specialized than a general-purpose computer, but in the end it's still just a computer.

Any algorithm that can be implemented in a general purpose computer can be implemented in an ASIC, there's nothing "magical" going on.

A relatively simple algorithm, like SHA-256, can be implemented in a general-purpose computer, or in an ASIC, with relatively little cost. This makes mining relatively accessible (though not necessarily profitable) to everyone.

A more complicated algorithm would take more effort to implement in a general-purpose computer, and likewise more effort to implement in an ASIC. In the beginning, this makes mining more accessible to those with general-purpose computers. If there's enough demand (if the coin in question becomes popular enough), then eventually the algorithm will be implemented in an ASIC, but due to it's complexity it will be accessible to fewer people.

In other words, the more complex the coin, the larger the start-up costs, and the less accessible mining becomes. This is exactly the opposite of what I assume is your goal.Quote from: yumei on October 05, 2014, 10:30:49 PM
Isn´t it theoretically possible to build a algorithm which makes actual asic miners worthless.

No.

An ASIC is just a computer. It's more specialized than a general-purpose computer, but in the end it's still just a computer.

Any algorithm that can be implemented in a general purpose computer can be implemented in an ASIC, there's nothing "magical" going on.

A relatively simple algorithm, like SHA-256, can be implemented in a general-purpose computer, or in an ASIC, with relatively little cost. This makes mining relatively accessible (though not necessarily profitable) to everyone.

A more complicated algorithm would take more effort to implement in a general-purpose computer, and likewise more effort to implement in an ASIC. In the beginning, this makes mining more accessible to those with general-purpose computers. If there's enough demand (if the coin in question becomes popular enough), then eventually the algorithm will be implemented in an ASIC, but due to it's complexity it will be accessible to fewer people.

In other words, the more complex the coin, the larger the start-up costs, and the less accessible mining becomes. This is exactly the opposite of what I assume is your goal.

But you can not mine litecoins with this generation of ASIC, am I right? Because litecoin is not SHA-256? So the ASIC we have are not simple computer, they are just working with SHA-256 system, with all SHA-256 crypto systems? Then there should be a possiblity to build an algorithm into the bitcoins sourcecode which prevent ASIC miner. There should be enough good scripter here to be able to answer this question.

Quote from: yumei on October 05, 2014, 02:37:33 PM
Satoshi for sure never planned asic mining. 99% of the bitcoin community would benefit when asic mining is prohibited.

Quote from: satoshi on July 29, 2010, 02:00:38 AM
The current system where every user is a network node is not the intended configuration for large scale.  That would be like every Usenet user runs their own NNTP server.  The design supports letting users just be users.  The more burden it is to run a node, the fewer nodes there will be.  Those few nodes will be big server farms.  The rest will be client nodes that only do transactions and don't generate.

While not explicitly stating asics, he did describe very well the move to large farms doing all the mining, and asics are just a natural evolution all of us anticipated.

I described before the difference between normal GPU farms and ASIC farms. Asic Manufactorer would primary produce for themself and mine thereself if its worth. See BFL, just imagine you are a ASIC manufactorer, you would mine yourself, you would sell to your partners (which pay more or buy anything you have) and maybe at least or the garbage you would sell to normal consumers. The  Asic Manufactorer have a monopol, they can do what they want. This is not free marke!!! With GPU manufactorer we don´t have such problems, because the manufactorers producing them for the whole world and that for more then
decades, every computer needs a graphic cards.

[kokojie]

ASIC is not the problem, GPU mining would have ended up exactly the same given enough time. Mega GPU mining farms will inevitable rise.

PoW concept itself is a doomed concept.

This is exactly what I am not thinking! Just imagine GPU mining farms, even if their are farms, you can let your servers or home computer mine while using them for something else as well. You can buy GPU hardware anywhere and there is no risk that nvidia mine thereself, there is no monopol. I bet we didn´t have the low prices we have right now with GPU mining only.

You don't get it, the specific algorithm of PoW mining doesn't matter at all, there will always be an expense associated with it, and it can not cheap in order to secure the network. It doesn't matter whether it's ASIC mining or GPU mining, hundreds of millions of dollars must be spent on PoW mining every year, in order to secure the network. We would have the exact same low price right now, if we were GPU mining, the expense does not change.

The only way to avoid this expense, is to fade out PoW mining altogether, and change to a PoS system.

[naplam]

The only way to avoid this expense, is to fade out PoW mining altogether, and change to a PoS system.

Which comes with its own set of problems and is not ready to be a real replacement for PoW as it is implemented now in several coins. Until PoS is improved, a PoW/S combination that lowers the mining expense while keeping a similar level of security would be beneficial though.

[yumei]

Quote from: yumei on October 05, 2014, 10:22:04 PM
Quote
ASIC is not the problem, GPU mining would have ended up exactly the same given enough time. Mega GPU mining farms will inevitable rise.

PoW concept itself is a doomed concept.

This is exactly what I am not thinking! Just imagine GPU mining farms, even if their are farms, you can let your servers or home computer mine while using them for something else as well. You can buy GPU hardware anywhere and there is no risk that nvidia mine thereself, there is no monopol. I bet we didn´t have the low prices we have right now with GPU mining only.

You don't get it, the specific algorithm of PoW mining doesn't matter at all, there will always be an expense associated with it, and it can not cheap in order to secure the network. It doesn't matter whether it's ASIC mining or GPU mining, hundreds of millions of dollars must be spent on PoW mining every year, in order to secure the network. We would have the exact same low price right now, if we were GPU mining, the expense does not change.

The only way to avoid this expense, is to fade out PoW mining altogether, and change to a PoS system.

O.k this is my last post for this thread, regardless of your bad behaviour, either I can not describe correctly or you don´t want to understand.

I will explain it again. Actually the amount of miners is a lot less of what we had before Asic mining, who wants to buy a asic without knowing he will get it before next years, without knowing if his asic miner is not already old and slow until he gets it, without knowing how much asic in the same time will be purchased by company xy with best contacts to asic manufactorer zy. This problems you won´t have with GPU.

So lets say the amount of people mining is actually only 10% of what we had with GPU mining. Lets say instead of 10000 miners we have only 1000 miners, so we have a centralization of mining. So a small amount of people have more power about the system, this is also called monopolization. So for instance 100 (10% of miners) people can do a lot more harm when they decide to sell their mined coins, then 100 (1% of miners) people could do before Asic mining.

[kokojie]

The only way to avoid this expense, is to fade out PoW mining altogether, and change to a PoS system.

Which comes with its own set of problems and is not ready to be a real replacement for PoW as it is implemented now in several coins. Until PoS is improved, a PoW/S combination that lowers the mining expense while keeping a similar level of security would be beneficial though.

Yes, a hybrid system probably is required for some time to eventually fade out PoW

[andytoshi]

The only way to avoid this expense, is to fade out PoW mining altogether, and change to a PoS system.

Proof of stake has been shown to be unusable as a distributed consensus mechanism.

[amaclin]

PoW concept itself is a doomed concept.

No.
Decentralization itself is doomed because of less efficiency of energy consuming.

[amaclin]

Proof of stake has been shown to be unusable as a distributed consensus mechanism.

Proof-of-work has the same flaw:
Some time miners are forced to switch off their asic-devices.
Because they are unprofitable.
After that PoW loses the security because it can be possible to buy switched-off hardware for zero price, switch it on for an hour and gain 51%
Sorry.

[naplam]

Proof of stake has been shown to be unusable as a distributed consensus mechanism.

Proof-of-work has the same flaw:
Some time miners are forced to switch off their asic-devices.
Because they are unprofitable.
After that PoW loses the security because it can be possible to buy switched-off hardware for zero price, switch it on for an hour and gain 51%
Sorry.

No, you can't mine at zero cost (electricity??). The W stands for work work costs
Sorry.

[amaclin]

No, you can't mine at zero cost (electricity??). The W stands for work work costs

I haven't been a miner. (OK, I switched on setgenerate option once or twice for an hour just for fun on my PC)
I am not a miner.
I will not mine. No expenses

[DannyHamilton]

No, you can't mine at zero cost (electricity??). The W stands for work work costs

I haven't been a miner. (OK, I switched on setgenerate option once or twice for an hour just for fun on my PC)
I am not a miner.
I will not mine. No expenses

Correct.  No mining, no POW expenses.

But as soon as you try to mine (for a 51% attack), expenses are involved.  The security model provided by POW is that it is more profitable to use your mining power to support the network than it is to use the mining power to attack the network.

[kokojie]

No, you can't mine at zero cost (electricity??). The W stands for work work costs

I haven't been a miner. (OK, I switched on setgenerate option once or twice for an hour just for fun on my PC)
I am not a miner.
I will not mine. No expenses

Correct.  No mining, no POW expenses.

But as soon as you try to mine (for a 51% attack), expenses are involved.  The security model provided by POW is that it is more profitable to use your mining power to support the network than it is to use the mining power to attack the network.

Ok, that really explains why all these PoW altcoins being 51% attacked to death, and all these PoS altcoin never attacked. Because PoW has a laughable and failed security model. It's just a matter of time Bitcoin will be attacked by a well funded and determined attacker. PoW security is a joke.

[amaclin]

The security model provided by POW is that it is more profitable to use your mining power to support the network than it is to use the mining power to attack the network.

This is true only while the difficulty is growing.  

[DannyHamilton]

Correct.  No mining, no POW expenses.

But as soon as you try to mine (for a 51% attack), expenses are involved.  The security model provided by POW is that it is more profitable to use your mining power to support the network than it is to use the mining power to attack the network.

Ok, that really explains why all these PoW altcoins being 51% attacked to death,

Yes, those scams, pump-and-dump schemes, and other trash all failed to secure enough "work" to secure their system prior to launch.  No surprise at what happened next there.

and all these PoS altcoin never attacked.

I wan't aware that there haven't been any attacks on any POS yet. Interesting. Once there actually is an attempted attack, let me know how it works out.

Because PoW has a laughable and failed security model.

Seems to be working fine for bitcoin for 6 years now. Bitcoin has survived many attempted attacks.

It's just a matter of time Bitcoin will be attacked by a well funded and determined attacker.

The longer that attacker waits, the more expensive that attack is going to be.  I look forward to seeing what happens.

PoW security is a joke.

Perhaps, perhpas not.  Either way, it doesn't have anything to do with the flaws in POS.

[amaclin]

Seems to be working fine for bitcoin for 6 years now. Bitcoin has survived many attempted attacks.

It is not a proof. Thousands of years people were told that an Earth is based on a tortoise.

The longer that attacker waits, the more expensive that attack is going to be.

This is true only while the difficulty is growing.

Perhaps, perhpas not.  Either way, it doesn't have anything to do with the flaws in POS.

The flaw is in decentralization itself. Once the cost becomes too high for supporting it for "good" user. And cheap enough for an attacker.
And it is the same flaw in PoS/PoW/Proof-of-anything

[DannyHamilton]

Seems to be working fine for bitcoin for 6 years now. Bitcoin has survived many attempted attacks.

It is not a proof. Thousands of years people were told that an Earth is based on a tortoise.

Certainly.  The "proof" is in the math that demonstrates the security model.  The same math that demonstrates the problem with POS.

The longer that attacker waits, the more expensive that attack is going to be.

This is true only while the difficulty is growing.

True.  And right now, it is still growing.  Therefore, right now, the longer that attacker waits the more expensive that attack is going to be.

Perhaps, perhpas not.  Either way, it doesn't have anything to do with the flaws in POS.

The flaw is in decentralization itself. Once the cost becomes too high for supporting it for "good" user. And cheap enough for an attacker.

It's an interesting experiment, isn't it?  I can't wait to see how it plays out.  I tend to believe that the market forces and incentives are such that it will always be more profitable to participate than to attack.  I guess all we can do is wait a few decades and see.

And it is the same flaw in PoS/PoW/Proof-of-anything

Certainly, but some systems (such as POS) have additional flaws that allow an attack with little or no cost at all.

[lucasjkr]

This is not capitalism, this is monopolism. You have some companys create new asic hardware, this companys mine thereself or sell their main parts to just 4-5 big companys.

There are only 2 companies producing high-end GPUs (AMD, Nvidia) and only 2 companies producing high-end CPUs (Intel, AMD) in large volume. In theory, if cryptocurrency gains worldwide adoption, the potential for mining to become dominated by a single manufacturer is far larger with CPU or GPU mining than it is with ASIC mining.

The reason AMD didn't corner the mining market back when GPU mining was all the rage is that it simply wasn't worth the effort back then. Who knows what the future would've brought in an ASIC-less world?

That argument doesn't work; even with the switch to ASICs, either intel or Amd could corner the mining market in a heartbeat if they so desired; no, they wouldn't be able to do so using off the shelf parts, but designing a processor to the the sha hashing is well within either companies abilities, and producig  said chips.... Either could produce enough capacity to dominate the mining market without batting an eye.