Well, well, well, now we know what Jihan Wu’s been up to.

[anonymoustroll420]

screw it lets repeat myself and add some more  to it

using gmaxweles own mindset

so libsecp256k1 "efficiency gain" is also an attack because it improved efficiency by 5x
so segwit quadratic/linear 'fix' is an attack because it improved efficiency
so fibre making its own tier ring network around the pools is an attack for propagation efficiency
so diluting full node count using prunning is an attack for making home computer efficiency
so diluting fullnodecount by having segwit nowitness mode is an attack
so making LN is an attack

.. at this point i can feel the rage of blockstreamists ready to pounce with their blockstream defender responses

so when something developed by blockstream is used to get more efficient, its ok. but its not blockstream sanctioned=attack
. hmmm i see..

might be easier for blockstreams partners to become more efficient instead of spitting out the dummy because they are not as efficient.

P.S if blockstream are so perfect and have the best codebase.. there should be no reason for so many 'fixes' via segwit because utopia should already have been coded in 2013

An ASIC moving from 56nm to 28nm is an efficiency gain, because it does more work (more operations) for less electricity.

ASICBoost doesn't do more operations for less electricity. ASICBoost lets the ASIC skip doing some of the work. It does not contribute any additional security to the network. If everyone used it, it wouldn't make a difference, it would still require the same number of operations to attack the network prior to it existing.

This is why ASICBoost is a shortcut, not an efficiency gain, it does not contribute any additional security.

[aarturka]

amnotback,

Why altcoiners even manifest their alt heads in brought daylight? Your thread is altcoin discussion, I'd rather not see your crap in Bitcoin themes

[iamnotback]

An ASIC moving from 56nm to 28nm is an efficiency gain, because it does more work (more operations) for less electricity.

And although there is the same amount of silicon, it costs more because it has more human intellectual capital in it.

ASICBoost doesn't do more operations for less electricity. ASICBoost lets the ASIC skip doing some of the work.

Incorrect. It does more PoW hashes for the same amount of electricity (and time). And it can be sold for a higher price (if enabled) because ... although there is the same amount of silicon, it costs more because it has more human intellectual capital in it.


Edit: to add this:

If a signature is supposed to resist forgery against 2¹²⁸ operations, but you find a way to do it with 2⁸⁰ instead, this is an attack.

In that sense, it is not an attack as it still takes at least difficulty * 232 operations on average to find a solution. The inner loop got a bit optimized by reusing an intermediate result for many iterations, but if you call this an attack, then you may also call using the mid-state an attack.

Agreed. I had previously written the same.

[iamnotback]

amnotback,

Why altcoiners even manifest their alt heads in brought daylight? Your thread is altcoin discussion, I'd rather not see your crap in Bitcoin themes

Bitcoin will not be changed. If you don't agree, then you will learn by losing your BTC, that you are wrong.

So where are you going to transact when fees go to $100 with small blocks.

You're in a delusion of denial dude.

It is going to funny watch all you Bitcoin maximalists freakout when you can't transact on Bitcoin any more and you can't change Bitcoin and when you try to fork Bitcoin, the whales are going to kill your fork and take all your BTC. I love it. You get what you deserve for being an arrogant fool.

[anonymoustroll420]

Incorrect. It does more PoW hashes for the same amount of electricity (and time). And it can be sold for a higher price (if enabled) because ... although there is the same amount of silicon, it costs more because it has more human intellectual capital in it.

Right, it does more hashes, but uses less operations (work) per hash to achieve this.

[aarturka]

don't you know that all alts exist with only goal to prey for weak hands' bitcoins

[Finksy]

An ASIC moving from 56nm to 28nm is an efficiency gain, because it does more work (more operations) for less electricity.

ASICBoost doesn't do more operations for less electricity. ASICBoost lets the ASIC skip doing some of the work. It does not contribute any additional security to the network. If everyone used it, it wouldn't make a difference, it would still require the same number of operations to attack the network prior to it existing.

This is why ASICBoost is a shortcut, not an efficiency gain.

If it accomplishes the same task with the same level of security, why does it matter? None of this gives a rational explanation of how using it detracts from blockchain validation or security. It works within the framework as set out by PoW mining.  Everyone is bringing morality into the discussion (is it OK for the task to be accomplished with less work?), when that should have no bearing on the discussion of protocol.  Ethics do not belong in protocol debate, they belong in the decision-making process of supporting manufacturers.  ASICboost is an innovation, it's not the problem here. The hindrance to future improvements is the problem (if such improvements should be wanted by the community at large).  The same argument could have been framed around the move to ASIC from FPGA if it had been done in a patent laden, secrecy-veiled implementation, but it wasn't (and nor should it have!).

[anonymoustroll420]

The same argument could have been framed around the move to ASIC from FPGA

No because that made the network more secure, as it required more work to attack it, ASICBoost doesn't do that.

[aarturka]

It is going to funny watch all you Bitcoin maximalists freakout when you can't transact on Bitcoin any more and you can't change Bitcoin and when you try to fork Bitcoin, the whales are going to kill your fork and take all your BTC. I love it. You get what you deserve for being an arrogant fool.

If it happens all alts fall with bitcoin, what makes your alt different in this situation

[dinofelis]

screw it lets repeat myself and add some more  to it

using gmaxweles own mindset

so libsecp256k1 "efficiency gain" is also an attack because it improved efficiency by 5x
so segwit quadratic/linear 'fix' is an attack because it improved efficiency
so fibre making its own tier ring network around the pools is an attack for propagation efficiency
so diluting full node count using prunning is an attack for making home computer efficiency
so diluting fullnodecount by having segwit nowitness mode is an attack
so making LN is an attack

.. at this point i can feel the rage of blockstreamists ready to pounce with their blockstream defender responses

so when something developed by blockstream is used to get more efficient, its ok. but its not blockstream sanctioned=attack
. hmmm i see..

might be easier for blockstreams partners to become more efficient instead of spitting out the dummy because they are not as efficient.

P.S if blockstream are so perfect and have the best codebase.. there should be no reason for so many 'fixes' via segwit because utopia should already have been coded in 2013

An ASIC moving from 56nm to 28nm is an efficiency gain, because it does more work (more operations) for less electricity.

ASICBoost doesn't do more operations for less electricity. ASICBoost lets the ASIC skip doing some of the work. It does not contribute any additional security to the network. If everyone used it, it wouldn't make a difference, it would still require the same number of operations to attack the network prior to it existing.

This is why ASICBoost is a shortcut, not an efficiency gain, it does not contribute any additional security.

PoW is about RELATIVE security.  That's all PoW can deliver, because it is bad cryptographic security: the bad guys and the good guys are on equal footings.  Normally, in cryptographic security, the good guys have a (secret key) advantage over the bad guys (ignoring the secret key).  But PoW is the silliness incarnated when it comes to security.

Also, PoW is not even PoW: it is proof of hash result.  If by some magic, I can provide you with an input that hashes to a funny hash starting with a lot of zeros, you MAY BELIEVE that that is a proof of me doing a lot of calculations, but all I had to provide you with, cryptographically, was a specific input.  How I got to that input is my business.  The hypothesis is that I had to do a lot of trial-and-error, and hence that my providing with a solution is a "proof" of that lot of trial and error.  However, how one does that, is one's own affair.

You could think that it is only fair to do it with paper and pencil.  You might say that all those people using computers to compute "proof of work" are not being fair.  The work has to be done by the human, not by a machine.   Ok, then we allow for machines, but only CPU.  Until a smart guy comes along and does it with a GPU or an FPGA.  Then another guy makes an ASIC.  These are all tools to render more efficient the ways of transforming economic resources (human time, machine cost, design time, electricity....) into the demanded result.  Thinking of how to make a more efficient ASIC or thinking of how to do a smarter organisation of the calculation are just different ways of making the transformation of economic resources in resulting "difficulty solution" more efficient, that is, using less economic resources to produce more spectacular (higher-difficulty) results.

As PoW is only as secure as the waste of the economic resources of the enemy needed to do so (which is utterly stupid btw) any progress by "the good guys" that is kept secret for "the bad guys" is in fact an extra security feature.  By making more public the "secret efficiency" of miners, in fact, one has lowered somewhat the competitive advantage of the "good guys" securing the block chain.  But it wasn't a secret: a paper on the Arxiv lined out already how to make a smarter use of calculations to obtain results.

There is absolutely no difference between "making better asics", "using computers over pen and paper", "using asics over computers" or "improving the calculational scheme".  All these methods improve the efficiency of producing more "difficulty solutions" with less economic means.  For the good guys, and for the bad guys.  So the relative security remains the same, if both of them have access to the same tech ; it improves if the good guys keep it secret.

And in any case PoW is stupid security.

[Finksy]

The same argument could have been framed around the move to ASIC from FPGA

No because that made the network more secure, as it required more work to attack it, ASICBoost doesn't do that.

Lowering node size makes for lower power consumption from same hashrate. (or higher hashrate for given power consumption)

Optimizing Chip layout makes for lower power consumption from same hashrate. (or higher hashrate for given power consumption)

Running ASICs in string in order to reduce VRM losses makes for lower power consumption from same hashrate. (or higher hashrate for given power consumption)

ASICboost makes for lower power consumption from same hashrate.  (or higher hashrate for given power consumption)

Is this incorrect?

Why do the details of how this is accomplished matter (other than in hindering protocol updates) if it does not create vulnerabilities in the PoW process?  Ethics or morality DO NOT matter in this context (i.e. "is it fair that it performs same task from less work")

[iamnotback]

don't you know that all alts exist with only goal to prey for weak hands' bitcoins

I don't disagree. Even though I also think that is some propaganda that has been put in your head.

Buying and holding Bitcoins is simply not going to be plausible for small hodlers because the block size will never be increased and so eventually you will not be able to afford the transaction fees to move your Bitcoins.

So what you going to do? Fight the whales? You will lose.

I already wrote in this thread that the fundamental problem is fungible money. But you don't want to listen to my solution for that, so I can only ask you pragmatically how will you handle the reality? Getting angry won't help you. Fighting the whales will just bankrupt you.

What you really want is to be meaningful. And there is a solution for that:

At the most generative essence level, we are fighting because fungible money forces us to. It is a fundamental flaw in the nature of money. And I am proposing a solution to that.

[dinofelis]

The same argument could have been framed around the move to ASIC from FPGA

No because that made the network more secure, as it required more work to attack it, ASICBoost doesn't do that.

On the contrary.  Attackers having ASICS could now attack a network that cannot be secured any more with FPGA, while before, it could.  PoW is about the expected relationship of the waste of *economic resources* an attacker needs to spend to reach a difficulty level, versus the waste of economic resources that those wanting to secure the network, are ready to spend.  Any improvement in efficiency of this conversion of economic waste into difficulty works both ways: in the hands of attackers, it makes the chain more vulnerable ; in the hands of good guys, it makes the chain more secure.

[Killerpotleaf]

The same argument could have been framed around the move to ASIC from FPGA

No because that made the network more secure, as it required more work to attack it, ASICBoost doesn't do that.

Lowering node size makes for lower power consumption from same hashrate. (or higher hashrate for given power consumption)

ASICboost makes for lower power consumption from same hashrate.  (or higher hashrate for given power consumption)

Is this incorrect?

Why do the details of how this is accomplished matter (other than in hindering protocol updates) if it does not create vulnerabilities in the PoW process?

you're all missing the point

its not about if its a cheat or not

the problem with ASICboost, is that miners using it can't handle most upgrades, and they won't even care about loosing boosting capabilities if ALL miners were using it, since at that point we'd have a level playing field to being with ( every boosting ) and a level playing field after an upgrade ( no one can boost anymore )

[iamnotback]

The same argument could have been framed around the move to ASIC from FPGA

No because that made the network more secure, as it required more work to attack it, ASICBoost doesn't do that.

On the contrary.  Attackers having ASICS could now attack a network that cannot be secured any more with FPGA, while before, it could.  PoW is about the expected relationship of the waste of *economic resources* an attacker needs to spend to reach a difficulty level, versus the waste of economic resources that those wanting to secure the network, are ready to spend.  Any improvement in efficiency of this conversion of economic waste into difficulty works both ways: in the hands of attackers, it makes the chain more vulnerable ; in the hands of good guys, it makes the chain more secure.

And intellectual property expended is one of the capital costs "wasted" (expended) on security.

Bitmain has huge sunk costs invested in ASICs. They have in their vested interests to protect Bitcoin.

Bitmain is protecting Bitcoin from those who want to change it from "protocol is law" to "governance is law".

Blockstream thinks they own the protocol. So they had to be spanked and sent to an altcoin.

Bitmain depends on being on the same side as the whales. If Bitmain tried to HF with BU, they would commit suicide, because whales would be forced to fight them.

BU was merely a diversionary tactic to get more of Bitmain's miners spread out into the wild, so that they can destroy Blockstream.

[Finksy]

Lowering node size makes for lower power consumption from same hashrate. (or higher hashrate for given power consumption)

ASICboost makes for lower power consumption from same hashrate.  (or higher hashrate for given power consumption)

Is this incorrect?

Why do the details of how this is accomplished matter (other than in hindering protocol updates) if it does not create vulnerabilities in the PoW process?

you're all missing the point

its not about if its a cheat or not

the problem with ASICboost, is that miners using it can't handle most upgrades, and they won't even care about loosing boosting capabilities if ALL miners were using it, since at that point we'd have a level playing field to being with ( every boosting ) and a level playing field after an upgrade ( no one can boost anymore )

See part in bold.  It is a relevant discussion that is subject to opinions (mine being that we should have the ability to introduce these protocols if it is so decided (democratically or not, so save that criticism. Decisions are both necessary and inherent, stagnation is still a decision)

[Finksy]

And intellectual property expended is one of the capital costs "wasted" (expended) on security.

Bitmain has huge sunk costs invested in ASICs. They have in their interests to protect Bitcoin.

Bitmain is protecting Bitcoin from those who want to change it from "protocol is law" to "governance is law".

Blockstream thinks they own the protocol. So they had to be spanked and sent to an altcoin.

Bitmain depends on being on the same side as the whales. If Bitmain tried to HF with BU, they would commit suicide, because whales would be forced to fight them.

That is your opinion.  Mine is that they are protecting their investment in mining capital, which without their improvement could not have been economically beneficial (as witnessed by ex-KNC statement)

[dinofelis]

Immutability of the protocol, is not on the table, it never was.

Of course.  Without immutability of the protocol, any other immutability has no meaning.  The protocol is what gives meaning (rights to spend) to binary data.  If you keep the data "immutable" (the historical block chain is not modified), but you change the rules that tell you what the data mean, then that's just as much altering the rights of the subscribers unilaterally as modifying the data directly.

If I write you an IOU, which says "I owe you 3 months of labor" ; and then I redefine what "months" mean (namely 5 seconds) and I redefine what "labor" means (namely, smiling to you), then without modifying a letter of my IOU, I now owe you a 15 second long smile.

Imagine that tomorrow, the protocol changes, and allows every UTXO to be spend twice instead of once   That's a very tiny protocol change, isn't it ?  And we didn't touch the block chain 

In the same way, rights to spend are also implicitly included in mining, the fees one can expect etc....  Any change to those rules (by taking transactions off-chain, by increasing block size, by changing block reward....) changes the expected rights to spend.  If you do that, you are having a system that doesn't keep the implicit agreement people engaging in the system (by owning coins, buying mining equipment,...) assumed under the "immutability" concept.

Wanting to modify the protocol is similar to modifying the block chain.  There's no difference in practice.

[iamnotback]

And intellectual property expended is one of the capital costs "wasted" (expended) on security.

Bitmain has huge sunk costs invested in ASICs. They have in their interests to protect Bitcoin.

Bitmain is protecting Bitcoin from those who want to change it from "protocol is law" to "governance is law".

Blockstream thinks they own the protocol. So they had to be spanked and sent to an altcoin.

Bitmain depends on being on the same side as the whales. If Bitmain tried to HF with BU, they would commit suicide, because whales would be forced to fight them.

That is your opinion.  Mine is that they are protecting their investment in mining capital, which without their improvement could not have been economically beneficial (as witnessed by ex-KNC statement)

Once you add up of the facts, you'll realize I am correct (@dinofelis preceded me with this opinion and now I have identified the technical specifics of the factual support):

Bitmain depends on being on the same side as the whales. If Bitmain tried to HF with BU, they would commit suicide, because whales would be forced to fight them.

BU was merely a diversionary tactic to get more of Bitmain's miners spread out into the wild, so that they can destroy Blockstream.

[dinofelis]

And intellectual property expended is one of the capital costs "wasted" (expended) on security.

Bitmain has huge sunk costs invested in ASICs. They have in their interests to protect Bitcoin.

Bitmain is protecting Bitcoin from those who want to change it from "protocol is law" to "governance is law".

Blockstream thinks they own the protocol. So they had to be spanked and sent to an altcoin.

Bitmain depends on being on the same side as the whales. If Bitmain tried to HF with BU, they would commit suicide, because whales would be forced to fight them.

That is your opinion.  Mine is that they are protecting their investment in mining capital, which without their improvement could not have been economically beneficial (as witnessed by ex-KNC statement)

There is no contradiction.  It is as if you were saying:

"your opinion is that water evaporates because of heat ; mine is that fast molecules are just leaving the surface".

The dynamics is immutability, and the detailed phenomena are that every possible significant change is against the interests of some (by definition of change !), who try to counter it to protect their interests.