Which Proof of Stake System is the Most Viable

[clout]

Generally, yes.  

What do you want to nitpick about?

You have to understand that cryptography secures your wallet not the network. The network is secured via consensus. Bitcoin as with all proof of whatever systems is a resource democracy. It is the natural structure for decision making of all distributed organizations throughout history. If you have 51% of the capital in the world you run it. This is the same phenomena with bitcoin proof of work. Your say over the network is proportional to your share of the designated resource, with bitcoin that is hashing power. Miners delegate there hashing powers to mining pools and those are the people you effectively trust to run the network. They are like the board of directors of the company that is bitcoin.

Every resource democracy works the same and is subject to the same 51% attack. The problem with bitcoin in particularly is that the capital intensive nature of mining disproportionately advantages those with economies of scale and ultimately centralizing the decision making powers. These are not trustless systems, rather you are distributing  trust amongst your peers with hopes that at the very least 51% of them are trustworthy. Regardless of the system, you are not trusting math you are trusting people.

Agree about the 51% thing... it applies to every system.

Disagree about the economies of scale being a problem though. ASICS and hashing power are likely becoming increasingly commoditized, scalable, and distributed.  

1) thats not true 2) even if it were true how does that justify spending $500 million a year on securing the network, when that cost is not necessary.

[jonald_fyookball]

The watt per GhS ratio has been dropping and continues to drop
as ASICs keeps improving.

https://en.bitcoin.it/wiki/Mining_hardware_comparison

Some of the best rigs are getting close to .1 watt /gHs,
which would drop the $500 million figure by 99%.

So, overall, its going to get cheaper and more efficient
to do proof-of-work.

[DeathAndTaxes]

The watt per GhS ratio has been dropping and continues to drop
as ASICs keeps improving.

https://en.bitcoin.it/wiki/Mining_hardware_comparison

Some of the best rigs are getting close to .1 watt /gHs,
which would drop the $500 million figure by 99%.

So, overall, its going to get cheaper and more efficient
to do proof-of-work.

No it won't (or at least not for that reason).   The efficiency of GPUs was ~ 50x that of CPUs however the that just resulting in 50x the hardware and as a result 50x the difficulty.  The same dynamic played out in the form of ASICs vs GPUs.  More efficient hardware improves security but they don't reduce the cost.  It improves security because a well funded attacker would always pick the most efficient solution.  If the defenders were armed with CPUs when an attacker spent a couple million to deploy a 2 PH/s ASIC farm it would be like fighting tanks with slingshots.   The efficiency of defenders must continue to improve with technological progress to ensure an attacker can't "cheat" and simply use more efficient hardware to obtain 51% of the computing power at a reduced cost.

The justification for PoW is that it is proven to work.  In the long run hardware will be commoditized which means the miner cost will be slightly below the miner revenue.  If revenue rises then more hardware will be deployed.

[clout]

The watt per GhS ratio has been dropping and continues to drop
as ASICs keeps improving.

https://en.bitcoin.it/wiki/Mining_hardware_comparison

Some of the best rigs are getting close to .1 watt /gHs,
which would drop the $500 million figure by 99%.

So, overall, its going to get cheaper and more efficient
to do proof-of-work.

I agree with DeathAndTaxes. You are also not looking at the fact that the mining cost is fixed and does not correspond to the market price of bitcoin. So if the value of bitcoin rises so does the cost of securing the network. It currently cost $500 million to secure the network, but if bitcoin is going to succeed that cost will dramatically increase even given the halving of block rewards. 

[DeathAndTaxes]

Of course PoS isn't free either.  If Peercoin was worth as much as BTC (value of money supply) then its 1% inflation rate would be about $70M a year.  To make the comparison more direct lets assume that PPC is no longer mining a meaningful number of PoW blocks is operating as PoS only.

[jonald_fyookball]

The watt per GhS ratio has been dropping and continues to drop
as ASICs keeps improving.

https://en.bitcoin.it/wiki/Mining_hardware_comparison

Some of the best rigs are getting close to .1 watt /gHs,
which would drop the $500 million figure by 99%.

So, overall, its going to get cheaper and more efficient
to do proof-of-work.

No it won't (or at least not for that reason).   The efficiency of GPUs was ~ 50x that of CPUs however the that just resulting in 50x the hardware and as a result 50x the difficulty.  The same dynamic played out in the form of ASICs vs GPUs.  More efficient hardware improves security but they don't reduce the cost.  It improves security because a well funded attacker would always pick the most efficient solution.  If the defenders were armed with CPUs when an attacker spent a couple million to deploy a 2 PH/s ASIC farm it would be like fighting tanks with slingshots.   The efficiency of defenders must continue to improve with technological progress to ensure an attacker can't "cheat" and simply use more efficient hardware to obtain 51% of the computing power at a reduced cost.

The justification for PoW is that it is proven to work.  In the long run hardware will be commoditized which means the miner cost will be slightly below the miner revenue.  If revenue rises then more hardware will be deployed.

Lets be clear on what we're talking about.  I was referring to electricity costs (not hardware costs).

Also, what you are saying is true but different than what I'm saying.
There's 2 variables:

1. efficiency, more of which drops the electricity costs,
2. hashrate, more of which increases the electricity costs.

what i'm saying is that electricty costs will drop because
of the increased efficiency.   There may be an increase
in hashrate which will offset this, but thats unrelated, and
obviously would be accompanied by an increase in security
as you pointed out.

[DeathAndTaxes]

The overall cost won't drop.  Efficiency is per hash but there will just be more hashes.   If efficiency improved 10x, then yes the energy (and thus cost) per GH would be 1/10th but there would just be 10x as many GH/s.  The network is more secure but cost is the same.

There will never be a scenario where the hashrate won't rise when more efficient hardware becomes available. Changes in miner compensation is the only thing which will affect miner cost.  If hardware is more efficient then miners would make more profit.  Other users would see that and deploy hardware to gain access to those easy profits which would drive the hashrate up.

As pointed out we say that with the emergence of ASICs.  Now the exchange rate also rose and the block reward halves which makes it not an apples to apples comparison but even if the exchange hadn't rose as the network become more efficient then miners would be producing higher profits which would lead to more hardware being deployed and the cost would eventually be the same at 100 J/GH or 1 J/GH.


Simple put baring a change in miner compensation (which is based on block reward & exchange rate) if #1 improves by 10x then #2 will increase by 10x as well.  

[jonald_fyookball]

You're not taking into account the fact
that over time, older hardware will die.
It will become more expensive to maintain
it than simply to upgrade.  Plus it will
be cheaper to run because its more efficient.

[DeathAndTaxes]

You're not taking into account the fact
that over time, older hardware will die.
It will become more expensive to maintain
it than simply to upgrade.  Plus it will
be cheaper to run because its more efficient.

If old hardware goes offline then new hardware will be deployed to take advantage of the higher profitability.  The network cost in BTC terms is just as high today as it was when the network was secured by GPUs.   Efficiency of the network has improved by at least 100x yet costs remain roughly the same by your logic the cost should have declined by 99% yet it hasn't.   I am not forgetting anything.  The cost expended by miners will always be within some % of the revenue paid to miners.  It doesn't matter what hardware is used.    Miner profit is the gap between costs and revenue.  The higher the profit/return the more incentive there is to deploy additional hardware. 

[jonald_fyookball]

You're not taking into account the fact
that over time, older hardware will die.
It will become more expensive to maintain
it than simply to upgrade.  Plus it will
be cheaper to run because its more efficient.

If old hardware goes offline then new hardware will be deployed to take advantage of the higher profitability.  The network cost in BTC terms is just as high today as it was when the network was secured by GPUs.   Efficiency of the network has improved by at least 100x yet costs remain roughly the same by your logic the cost should have declined by 99% yet it hasn't.   I am not forgetting anything.  The cost expended by miners will always be within some % of the revenue paid to miners.  It doesn't matter what hardware is used.    Miner profit is the gap between costs and revenue.  The higher the profit/return the more incentive there is to deploy additional hardware. 

Fine.    you're basically saying the same thing I said, which is that efficiency has been increasingly tremendously, with the twist that the network is also growing tremendously.  I agree.

In any case, I don't see this as any kind of problem for Bitcoin. Do you?

[wizzardTim]

Qora has a new POS system!!!!! And if its true that BCNext created it, it would be NXT's evolution - the real BCNexts project. There are many indications for this. If you wanna find out, check the relevant thread.

[DeathAndTaxes]

Fine.    you're basically saying the same thing I said, which is that efficiency has been increasingly tremendously, with the twist that the network is also growing tremendously.  I agree.

In any case, I don't see this as any kind of problem for Bitcoin. Do you?

It isn't a "problem" but to say improved efficiency will reduce the cost of the network is an inaccuracy.  Reducing the revenue available to miners will reduce the cost expended by miners.  It would not be good for security but a 90% reduction in the block reward right now would reduce the "cost" of the network by 90%, a 90% reduction in the J/GH or $/GH would not.

[jonald_fyookball]

Fine.    you're basically saying the same thing I said, which is that efficiency has been increasingly tremendously, with the twist that the network is also growing tremendously.  I agree.

In any case, I don't see this as any kind of problem for Bitcoin. Do you?

It isn't a "problem" but to say improved efficiency will reduce the cost of the network is an inaccuracy.  Reducing the revenue available to miners will reduce the cost expended by miners.  It would not be good for security but a 90% reduction in the block reward right now would reduce the "cost" of the network by 90%, a 90% reduction in the J/GH or $/GH would not.

Ok, not "will" for sure but certainly "could".  There's no guarantee an increase in energy efficiency will be accompanied by a corresponding increase in cost efficiency.  Right?

[clout]

Fine.    you're basically saying the same thing I said, which is that efficiency has been increasingly tremendously, with the twist that the network is also growing tremendously.  I agree.

In any case, I don't see this as any kind of problem for Bitcoin. Do you?

It isn't a "problem" but to say improved efficiency will reduce the cost of the network is an inaccuracy.  Reducing the revenue available to miners will reduce the cost expended by miners.  It would not be good for security but a 90% reduction in the block reward right now would reduce the "cost" of the network by 90%, a 90% reduction in the J/GH or $/GH would not.

Ok, not "will" for sure but certainly "could".  There's no guarantee an increase in energy efficiency will be accompanied by a corresponding increase in cost efficiency.  Right?

The cost of mining from the perspective of the bitcoin network is the block reward provided to miners. None of the costs or efficiencies within the process of mining matter because the network is incapable of taking those variables into account. The cost of security increases with value of bitcoin. If the value of bitcoin increases proportionately more than given decreases in block reward then there is a net increase in the cost of securing the network.

At the end of the day, distributing crypto-equity through fixed algorithms is fundamentally flawed because it does not consider market forces.

[DeathAndTaxes]

Ok, not "will" for sure but certainly "could".  There's no guarantee an increase in energy efficiency will be accompanied by a corresponding increase in cost efficiency.  Right?

Of course there will be.   If tomorrow miners were earning 1000% annual ROI what do you think would happen?

Miner cost will always be within a few % of miner revenue unless you think people will simply allow 1000% ROI to exist and not deploy more hashpower to extract that profit.

[DeathAndTaxes]

At the end of the day, distributing crypto-equity through fixed algorithms is fundamentally flawed because it does not consider market forces.

That is nonsense.   If the exchange rate rises the market value of the money supply also rises.  If miners receive 1% of the money supply annually it doesn't matter if a BTC is worth $1 or $100,000.  The network isn't distributing "equity" it is providing compensation for securing the network.  The subsidy is merely a bootstrapping mechanism.  In the future users will pay for security and if they pay 1% to PoW miners or 1% PoS stakeholders they are still paying for security.

[Peter R]

At the end of the day, distributing crypto-equity through fixed algorithms is fundamentally flawed because it does not consider market forces.

It doesn't make sense to distribute equity this way, but it is quite possibly the only legitimate way to distribute money. 

And anywayz, like OMG.  PoS was sooo last week.  Nxt's secret sauce smells bad and the nothing-at-stake problem remains unsolved. The new alt-coin narrative is protocol-enforced privacy.  Get with the program LOL!   

[jonald_fyookball]

Ok, not "will" for sure but certainly "could".  There's no guarantee an increase in energy efficiency will be accompanied by a corresponding increase in cost efficiency.  Right?

Of course there will be.   If tomorrow miners were earning 1000% annual ROI what do you think would happen?

Miner cost will always be within a few % of miner revenue unless you think people will simply allow 1000% ROI to exist and not deploy more hashpower to extract that profit.

This is an oversimplification.  ROI within what time frame?  A miner expects to get his money back for the rig in a certain amount of time but there are many different rigs being sold at different times to different miners at different prices all while the difficulty is increasing.  People are making predictions on what the difficulty will be with differing degrees of accuracy.  Then there is also risk and cashflow considerations that each miner has to contend with.

Say the fyookball-1000 mining rig (a fictional machine) gives you 1TH and runs at 1000w.  Now the fyookball-2000 comes out and gives you 2TH but runs at 100w.  Well, now you doubled the hash but cut the electric bill. By 90%.  

If every miner owned 1 FB1000 and suddenly replaced with the FB2000 , each miner would be making the same when the difficulty changes anyway.....but the power consumption is down.

[DeathAndTaxes]

Say the fyookball-1000 mining rig (a fictional machine) gives you 1TH and runs at 1000w.  Now the fyookball-2000 comes out and gives you 2TH but runs at 100w.  Well, now you doubled the hash but cut the electric bill. By 90%.  

If every miner owned 1 FB1000 and suddenly replaced with the FB2000 , each miner would be making the same when the difficulty changes anyway.....but the power consumption is down.

Except why buy 2 why not buy 10 and make even more money?  If it is profitable to deploy 2 it is profitable to deploy 10 and people will.  The sooner and the more you buy deploy the higher your return.  It actually doesn't matter if you don't do it, someone else will.  The hashrate will rise until there is no (or miminal) return available just like it was before the new more efficient tech came along.  Once again we have seen this happen on a continual basis for almost five years now.  It has always followed this trend and it always will.  If margins are high then miners will deploy hardware until the margins aren't.  More efficient tech is very profitable at the current difficulty at the time is is released however collectively miners will eventually drive the difficulty right back up to near break even when using the more efficient gear.

[jonald_fyookball]

Say the fyookball-1000 mining rig (a fictional machine) gives you 1TH and runs at 1000w.  Now the fyookball-2000 comes out and gives you 2TH but runs at 100w.  Well, now you doubled the hash but cut the electric bill. By 90%.  

If every miner owned 1 FB1000 and suddenly replaced with the FB2000 , each miner would be making the same when the difficulty changes anyway.....but the power consumption is down.

Except why buy 2 why not buy 10 and make even more money ?  If it is profitable to deploy 2 it is profitable to deploy 10 and people will.  They will keep deploying more and more and more hashing power and until the ROI is right back down to that low level it was before the more efficient hardware came along.

This is the same as ANY commodity business.

Yeah but the thing is the rig is the main cost not the electricity...at least for now.

I'm not trying to troll.  I think you are proving my point....it is the costs of the rigs
trying to keep pace with the difficulty.... What's this got to do with the electrical
Efficiency?  They are different things.

Anyway, I'm tired of this debate...you are the bitcoin expert here so I'll concede
The argument, I'm probably missing something. 

The important thing is the efficiency is getting better and better, and we are
Getting more security for the dollar...and bitcoin is in good shape.