Author Topic: Why DPOS is better than other POS  (Read 24685 times)

0 Members and 1 Guest are viewing this topic.

Offline 天籁

  • Hero Member
  • *****
  • Posts: 744
    • View Profile

merockstar

  • Guest
ooooooooh ok it just clicked for me.

thanks bm

Offline bytemaster

This is going to sound incredibly stupid, but I think what I'm having trouble with is wrapping my head around the definition of "validation cost."

Could you define that for me? When I think validation cost, I think (in a proof of work scheme) the cost of the resources expended to validate the network. In the case of PoW that means electricity and mining hardware.

Is that correct?

If so, then the validation cost in a PoS network would just be the electricity the server takes to run a node? no?

At 2000 TPS validation costs equals the cost of running server which include 100 MB/sec network connection, a server with 256 GB of ram or more, high availability (redundant backups), DDOS protection, etc.   These kinds of systems can easily cost $5000 / month or more to operate at scale.   If you don't earn that much from block production then you will not do it.   
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

merockstar

  • Guest
This is going to sound incredibly stupid, but I think what I'm having trouble with is wrapping my head around the definition of "validation cost."

Could you define that for me? When I think validation cost, I think (in a proof of work scheme) the cost of the resources expended to validate the network. In the case of PoW that means electricity and mining hardware.

Is that correct?

If so, then the validation cost in a PoS network would just be the electricity the server takes to run a node? no?

Offline bytemaster

Does this actually apply to peercoin? Transaction fees in Peercoin are destroyed and stake rewards are pure inflation (somewhat balanced by the destroyed tx fees) set at ~1%

in peercoin, if people don't spend enough the supply gets diluted and the value of everybodies peercoin goes down, encouraging spending. Then when everybody is spending more tx fees are being destroyed which increases the value of everybodies peercoin, encouraging them to hold and establishing an equilibrium.

if I'm understanding this correctly, this argument relies on tx fees being redistributed (which is the case in nxt)

This is pure economics.  The only thing that peercoin does is transfer the costs from those making transactions to all shareholders.   You still have a constant cost-per-validator.

I would guess that peercoin at least has a means of paying for it until the block reward falls below the fixed validation costs.   It will eventually get to the point where you must produce X% of the blocks to cover validation costs (as the system scales).

The challenge is that most of these issues are only a big issue at scale. 
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.

merockstar

  • Guest
Does this actually apply to peercoin? Transaction fees in Peercoin are destroyed and stake rewards are pure inflation (somewhat balanced by the destroyed tx fees) set at ~1%

in peercoin, if people don't spend enough the supply gets diluted and the value of everybodies peercoin goes down, encouraging spending. Then when everybody is spending more tx fees are being destroyed which increases the value of everybodies peercoin, encouraging them to hold and establishing an equilibrium.

if I'm understanding this correctly, this argument relies on tx fees being redistributed (which is the case in nxt)
« Last Edit: July 12, 2014, 02:16:59 pm by merockstar »

Offline Jack

  • Newbie
  • *
  • Posts: 4
    • View Profile
微博主页http://weibo.com/248742737
优酷主页http://i.youku.com/bitshares

Offline bytemaster


Green is Income   Yellow is Profit  Blue is Cost

Assuming a fixed validation cost per transaction and a fixed fee per transaction, there is a limit to the amount of decentralization that can take place. 

Assuming the validation cost exactly equals the fee a network is completely centralized and can only afford one validator.
Assuming the fee is 100x the cost of validation then the network can support 100 validators.

What you will quickly see is that systems like Nxt and Peercoin must have very excessive fees if they would like to maintain the illusion that everyone can be a validator and earn fees at scale.    What this means for Nxt / Peercoin is that anyone with less than 1% stake cannot validate profitably unless their fees are higher than our DPOS chain. 

So if these chains assume 100 is to centralized and start promoting they have 1000 validators then their fees must be 10x.  If they grew to be the size of bitcoin ($10 B) then only those with $1M worth of coin could validate profitably and most would consider that an elite club.    If they reduce the minimum stake to be a validator to $1000, then their fees would be 10,000 times higher than our fees.   

Now we assume that every one with less than the amount required to validate doesn't participate and we assume a "reasonable" distribution of wealth, you will quickly see that unless they have insane fees, there will only be a handful of people with enough stake to validate profitably. 

What we can conclude from this is that the only way for POS to work is to delegate it.  In the case of Nxt they can pool their stake by some means and ultimately this will end up like DPOS prior to approval voting with a variable number of delegates.   Those doing the delegating wouldn't actually receive any income like you see with mining pools because the validation expenses will be consuming the vast majority of the transaction fees.

End result is that decentralization has a cost proportional to the number of validators and that costs do not disappear.  At scale these costs will centralize any system that does not support delegation.   We should design this kind of centralization in from the beginning so that it can be properly managed and controlled by the users instead of evolving in some adhoc manner as an unintended consequence.

 
For the latest updates checkout my blog: http://bytemaster.bitshares.org
Anything said on these forums does not constitute an intent to create a legal obligation or contract between myself and anyone else.   These are merely my opinions and I reserve the right to change them at any time.