Topic: experts and devs only please - POS3 or POSP are these the best yet of the POS .. (Read 787 times)

thanks for those explanations. Interesting details there I wasn't aware of previously.

Just letting you I might have quoted some of the posts in this forum here: https://bitcointalksearch.org/topic/m.14728246

thanks

Rat4 is definitely a pioneer in PoS development, he added code where after 500 blocks, Reorganization (Changes) can not happen.
So any transaction is locked in and safe from any history rewrites after 500 blocks. (Kind of like a rolling checkpoint like Nxt (720 blocks))
(One issue with blackcoin is economics ,it ignores the fact that it is competing with other investments that have a higher interest rate.)

Sunny King chose a different route, kept PoW with PoS and uses a Checkpoint Server which prevents any history rewrites and prevents the chain from forking. It centralizes control of the coin more than a rolling checkpoint which only blocks reorgs.


15 Days stake time has no effect on speed, Block Speed will determine its speed.
You can have a 30 second block speed or a 2 minute block speed .
 
Security wise it will keep the difficulty # lower, but coin age / checkpoint servers / rolling checkpoint can offset that issue.

It is just that if a person has a large amount of a coin, and their is a limited number of times per day the coin can stake to maintain its block speed.
1 minute block speed = 60 blocks per hour or 1440 blocks per 24 hours ,  which means only 1440 chances to stake per day.
1 day= 1440 chances to stake
5 days= 7200 chances to stake
10 days = 14400 chances to stake
15 days = 21600 chances to stake

More chances to stake means a fairer access to the compounding interest effect for everyone.
The longer the minimum staking time , the more chances the poor have to stake before the rich's coins are even eligible again.
http://www.basic-mathematics.com/simple-vs-compound-interest.html

 

thanks very much for the explanation last I checked this thread had just dropped away so I didn't realise you had replied.

Yeah I was enquiring from a kind of design/security angle than earnings right now. Although I see that earning more will induce more staking and perhaps greater security in that way.

So if we were to use blackcoin and increase annual % and increase minimum staking time then so far this would be the most ideal version of POS so far?
Sorry if this is a silly question but could blackcoin not have left a coin age checking piece of code in there too just for added security and maybe some kind of code that uses that and confirmations to define pos returns? or would that be totally pointless and counter productive?
Rat4 seems to be at the forefront of POS development. I wonder why sunny king never decided to make any of these adjustments or possible improvements.


Also would increasing minimum staking time to 15 days have any security or speed issues for transactions? You'll have to excuse my lack of knowledge but how does increasing minimum staking time help the rich? If the % is annual and both rich and poor kept their coins and wallets online for an entire year would then not get the 1%( in blackcoins case) return on their coins regarldless of whether they get it every hour in smaller chunks or every 15 days in larger chunks?

PoS verses PoS3 (Blackcoin)

This removes coin age and moves to number of confirmations, as to whether it is more or less secure than the original ,
currently unknown, it seems more secure, however without coin age it may make the coin more susceptible to a 51% attack where someone tries to purchase the coins outright, because it kind of removes the Chaos factor that protect the original PoS.
(Chaos Factor: Even thru you know how many coins there are, you don't know their weight or when they will be staking.)

Main Issue for the Blackcoin design is purely economic ,
1. interest rate is too low , at 1% per year it is a crappy investment
2. Minimum Time before staking is too short which allows the rich to monopolize the Staking Process and actually crowd out poorer investors delaying their stakes.


PoS verse PoSP (Crypto Bullion)

Both still use coin age , so security should basically be the same.
After looking at their Whitepaper.
http://cryptobullion.io/white-paper/

It has a similar issue with allowing staking to start too soon (1 hour)  
which allows the rich to monopolize the Staking Process and actually crowd out poorer investors delaying their stakes.
Also there is an underlying incentive to DDOS other PoSP users so that your own staking will receive a higher % of the staking.

PoS Specs do matter , longer minimum stake times are required to keep the wealthy from monopolizing the Staking Process
Which the current PoS3 & PoSP ignore, which will decrease their adoption rate by the majority of less wealthy users and hurt them in the long run.
(In one economic regard, they function like PoW, only the wealthy will see the most profit.)
(However if they increased their minimum stake age to 15 days or more , they could remove that issue.)
(They could also increase their interest rate to 5% per year and be a decent competition to Utility Stocks.)

 

FYI:
You forgot PoS II from MagiCoin & NobleCoin
They have a declining reward if you don't stake within 2 weeks or so .
Where if you have the wallet offline for more than 4 weeks, you have to send all of your coins back to yourself to reset their age , so you receive a higher stake.
Oddly enough , it was designed to make sure more people staked, but with only a few hours needed before staking ,it also favors the wealthy holders, but makes them stay online more so they don't lose stake rewards.

FYI2:
Of the 4 coins mentioned above , the top choice if you want to actually earn, is Noblecoin with a 8% per year interest rate.
You just need to stake every 2 or 3 days and keep between 1½ million to 2 million Noble per block to stake effectively.
Noble also has a nice difficulty #s , so it is secure from the history attack worries.
https://chainz.cryptoid.info/nobl/

Hybrid PoW/PoS like to claim that having both makes them more secure, but truthfully it just opens them up to being attacked on the PoW & PoS side.
Mint & ZEIT both dropped PoW after Rat4 did a demonstration showing that having a weak PoW in a hybrid coin was a security problem.
https://bitcointalksearch.org/topic/security-analysis-of-powpos-hybrids-with-low-pow-reward-551861

This issue of Permission verses Permission less is a joke, claiming that PoS people will not sell their coins is ridiculous, (Why are they even staking?)
We could just as easily say the Hardware Manufactures refused to sell you an ASIC and denied you permission.
It is also like saying that a Dairy Farmer will not sell you Milk, so it is permission based. (Then Why is he a dairy farmer?)
The Fact is every system requires permission from someone, including PoW.
Permission from the ASICS reseller, Electricity company , & ISP, so this myth is so busted.
Let's not forget, if your Government outlaws ASICS , then they just denied you permission.

PoS will always become more Decentralized over time as their user base increases,
PoW will always become more Centralized over time due to Economic Factors , such as ASICS upgrades, & Electricity Requirements.

And in regards to PoW security ,
Ghash has already owned BTC for ~24 hours with 55%
http://www.coindesk.com/bitcoin-mining-detente-ghash-io-51-issue/

Currently the top 3 BTC mining Pools are in China
If they combined , then China would be mining over 60% of BTC

So the only reason BTC is not completely owned at the moment is the Chinese keep their pools separated by nothing more than a mutual agreement, they could completely own BTC any time they choose.

In Regards to PoS Security,
Coin-age used correctly can help protect a PoS coin,
example: Even if someone purchased 51% of the coin, he would not have 51% until the age of his coins was also equal to or higher than the age of the other staking coins. That will be very hard to calculate the actual age of every single block, and due to the fact people stake at different times , no way for him to calculate exactly when would be best to use his coins to achieve the maximum result. Easier to hack a bank security by far.
Plus if he timed it wrong and uses any of his stake , it would be days before he could even make another attempt and during that time the other 49% has staked and may have a higher number than him, before he can even try again.
Diabolically Clever , when you examine it.  
By the way the more coins purchased the higher the price of the coin.

 

FYI:
At the End of the Day,
PoS will defeat PoW purely because of economic factors.

Thanks for that explanation. Yes POW is clearly the best way. I see that. However since I notice all recent releases are going POS after their intial pow or ico I was just wondering which of the current flavours are least worse if you see what I mean.

If however a developer was determined regardless of what we say to use POS and you could only persuade him to go POS3 or POSP which would you suggest he goes for and why?  Has anyone studied them closely to see if there are significant advantages of one over the other or if they are just similar to such a degree there is no clear winner?

It's already been tried.  Coins like Mintcoin had PoW and PoS at same time.  Once the PoW subsidy became low, you're earning a lot more coins by staking while PoW was less profitable.  This made the PoW difficulty go to nothingness while the PoS difficulty is high.  Some random joe (we'll call him BobSurplus), can just walk up and start mining with large hashpower and get like 20 PoW blocks in a row with no PoS in-between and instantly double spend.

Even if you work out the economic incentives through inflationary PoW block reward to make it so people are always mining PoW at high difficulty in comparison to PoS, at that point, PoW is a more secure model anyway (not to mention being a decentralized permisionless ledger while PoS isn't), so there's no point in having PoS at all at that point if you're going to have miners anyway.  The PoS weaknesses just negatively effect the PoW in that system.

thanks for that..that is interesting..

So really for now the fact the current normal POS coins have little buy support is stopping most people bothering.

The POS3 and POSP are just giving more incentive to steak your coins and therefore hopefully securing the network a little better? although if attacker has a huge proportion of the staking coins the method of inducing people to steak is not important. For POS to be strong then you want lots of coins staking in lots of different wallets and no collusion.

However, you do need a large % of the coins staking to make an attack. If you can not get a large % of the staking coins then you can not perform this kind of attack you are mentioning.

I wonder if combining POW + POS helps or if this kind of history attack is not effected.

NXT and Bitshares both have "rolling genesis block" or "decentralized checkpoints" to try and prevent history attacks, so it's not going to happen to either of those two coins.  But that "feature" does make trade offs for other problems that could potentially be as disastrous.

As for history attacks on "normal" PoS coins, anyone who can perform one likely already owns a huge percent of the coin themselves, and the buy walls for the things are so small, like 10-50btc, that they can't even sell the stake they own before doing the attack and trying to accumulate old keys from others.  I think Monsterer believes Eth would be the first big target for history attack since it has some buy support (at least for now).  Vitalik will probably just do the same thing to prevent it as NXT and Bitshares, which is just a bandaid that leads to other problems.

Rolling genesis block is immediate forced convergence.  If somebody doesn't like a change made by miners forcing a fork in Bitcoin (something crazy like raising coin count), Bitcoin forks off into two different chains with different hash power and users on each.  Then over time, the market will eventually kill off the "stupid" one.

When stake holders force a change with a "rolling genesis block" in a PoS coin, you're kind of stuck with it and have much less recourse as a user for what you can do afterwards.  Both problems and solutions in closed entropy systems tend to be permanent once they happen, while in open entrpoy systems, you're required to maintain a current user economic majority over the duration of it's existence to force convergence instead of everything being a dictatorship by whoever the richest guy was 10 years ago.

@ roach

I agree with what you said. I just do not want mention of specific coins here to start any problems in this thread.

I do agree with your points though both regarding that specific coin (scam) and pos not being as suitable as POW for crypto.


Or please repost without mention of any specific coin because your explanation was much better than my own.

What i really want is just to imagine there is a world with only the options described in the OP. POW alone does not exist.

I am not at all saying POS is the way to go over POW only that I would love to know which variations of POS are superior to others.

Please no mentioning from anyone specific coins. It will only turn in to an all out a war zone.

I have heard it said that POW+POS is less secure than POS alone.

then others devs have said the opposite.

Is there anyway to demonstrate clearly which is correct?
Are there variations in the levels of security that pos3 and posp or others offer?

In terms of security if POS is less secure than POW why have we not seen attacks on the larger POS coins?

Let's not get into a pump discussion of particular coins. Let's just discuss the tech.

POS3 (rat4)

POSP (elambert)

These are the best yet of the POS systems?

What are the pros and negs of each vs each other or vs vanilla POS.

Also what are the advantage of

POS + POW  vs Pure POS of any kind.

Are their other variations that are as good or have advantages over either of these?

Let's not just hear POS is trash compared to POW. That is not the discussion here. I would like to see if anyone knows which of all the POS variations is best and for what reasons.