Why are miners scared of POS?

24

Comments

  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @Eastwind I should also mention that the block time goal for PoS is around 8 seconds, compared to 14.25 - 14.5 seconds now. So that constrains the amount of time available to reconcile the block bets as well.
  • cascadia_coincascadia_coin Member Posts: 67

    If you want to stake you're going to need around 1,500 ETH.

    This statement does anger me a little that you say you need at least $18,000 USD to even participate (current market rate). Where does this figure come from?
    Can't say. But someone else has kinda made it public here so I guess the NDA will be lifting soon (Sept/Nov):
  • cascadia_coincascadia_coin Member Posts: 67

    Techman34 said:

    Good point - I suppose an Exchange could run both chains to get the fees from both while playing the card that they are neutral in all this... But still, who's going to buy the coins on our rogue chain if they can't be used anywhere else?

    Exchanges would look at the health of the POS system to even see if they would implement the fork. From what I read from @cascadia_coin people would have to have at least 1,500 ETH to participate. Exchanges don't like that low volume trading as it cuts a lot of people out.
    /0
    It might be 1,500 to stake but that doesn't mean you won't be able to trade lesser amounts on exchanges as you can now. Proof of Stake just means to gather more ETH you have to have a substantial stake in it already since mining will be over. You will only be able to get ETH (other than buying or trading for it) will be staking or using your existing ETH (1,500+) to accumulate more just by leaving it alone in your cold storage wallet.

    No more mining. Price of ETH will go up as a result probably as well.


  • EastwindEastwind Member Posts: 107
    edited July 2016
    if there are 249 validators active it becomes 311,250 ETH)

    If it has only 250 validators, if you want to attack it, you just need to buy some Ethereum to be a validator, that will be much less than 10% of total Etherum and much cheaper.
  • Techman34Techman34 Member Posts: 405 ✭✭✭
    I know there is the argument against PoW from the waste of energy etc, but 250 parties dies not seem like a lot to secure things as the risk of collusion/manipulation would be much greater with PoS of this style, is it not?
  • greenusergreenuser 50.8862°N 4.5537°WMember Posts: 439 ✭✭
    I think you have to be a member of R3 to become a validator....
    Anyhow, why not try to get VB to go Proof-of-Capacity (PoC), then anyone with a computer can mine it, & from a phone even
  • cascadia_coincascadia_coin Member Posts: 67
    greenuser said:

    I think you have to be a member of R3 to become a validator....
    Anyhow, why not try to get VB to go Proof-of-Capacity (PoC), then anyone with a computer can mine it, & from a phone even

    He did in fact mention that anyone with a laptop or tablet would be able to. So maybe that's it?
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @cascadia_coin No, that's not it. PoS is completely different. You may run your validator on a laptop, but it still costs you a minimum os 1250 ETH for a stake, and there's a maximum for 250 stakes (validators) for the whole network.
  • cascadia_coincascadia_coin Member Posts: 67
    dlehenky said:

    @cascadia_coin No, that's not it. PoS is completely different. You may run your validator on a laptop, but it still costs you a minimum os 1250 ETH for a stake, and there's a maximum for 250 stakes (validators) for the whole network.

    Thanks for the clarification.
  • indytimindytim Member Posts: 123
    Question : as more people write scripts for Ethereum smart contracts, and those scripts become progressively more "fully featured" and complex, where does the computational power to execute them come from? The validators?
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    The Ethereum clients (geth, parity), i.e. full nodes. Validators don't even exist yet; that role comes into being with PoS.
  • indytimindytim Member Posts: 123
    dlehenky said:

    The Ethereum clients (geth, parity), i.e. full nodes. Validators don't even exist yet; that role comes into being with PoS.

    I'm just thinking, perhaps a liitle wishfully, that the computational power to execute scripts and validate tx will be considerable, even with POS. Might n't there be a role for mining power?

  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @indytim GPUs don't run transactions, nodes (geth, parity) run transactions on CPU. GPUs would add nothing to transaction processing capacity/capability. As the network grows, nodes may need to be a bit beefier, but that's about it. My parity v 1.2.2 node uses 4% of the CPU (i7, 6 core-2 threads) once it has synced, and that's with 9 ethminers run against it solo.That's on a CPU that can run at 1200% (cores/threads). When there are lots of transactions, it may hit 25% temporarily. When it's doing a full sync, it maxes out around 200%+. So, what do you think?
  • OilidOilid Member Posts: 68
    Hi guys,

    Sorry i am really a beginner in the ethereum world and run a rig of 95MH/s.

    Just to make sure if i did understand well actually i have 25 ethers on my wallet:
    - after the hard fork (if there is one) those ethers will be accepted only on the "old" blockchain ? So it means that they will have no value because everyone gonna be using the new one?
    - Nothing made to allow miners like me with my 25 ethers to trade them to 25 on the new blockchain ? Only people with massive amount of ether can ?

    So the best would be to trade those ether to bitcoin to don't lose their value ?


    From a concerned little miner ...
    Thanks !
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @Oilid They will be accepted on the majority blockchain, whichever that one turns out to be.
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    edited July 2016
    @Oilid The exception to that would be *if* exchanges decide to support an "alt-ETH" coin on the minority blockchain, but the chances of that happening are about zero. In that case, your ETH would be honored, but new blocks mined on the minority blockchain would also be honored.
  • greenusergreenuser 50.8862°N 4.5537°WMember Posts: 439 ✭✭
    edited July 2016
    dlehenky said:

    @Oilid They will be accepted on the majority blockchain, whichever that one turns out to be.

    hmmm... I am running two instances.... (two different machines) Well... I will be when the 2nd one finally syncs. It's exciting isn't it?

    So if we could end up with a darketh as wel as an eth. Cool, another coin!
    I have started mining eth again, i maybe one of the first to catch a darketh block
  • OilidOilid Member Posts: 68
    dlehenky said:

    @Oilid They will be accepted on the majority blockchain, whichever that one turns out to be.

    Thank you very much for the information. I was not sure at all ... SO that matter of hard fork isn't even a problem. I mean the only problem is the trust in the chain ? If one hard fork had been made, another one can be too no ?
  • SmokyishSmokyish Member Posts: 203 ✭✭
    Oilid said:

    dlehenky said:

    @Oilid They will be accepted on the majority blockchain, whichever that one turns out to be.

    Thank you very much for the information. I was not sure at all ... SO that matter of hard fork isn't even a problem. I mean the only problem is the trust in the chain ? If one hard fork had been made, another one can be too no ?
    It's technically possible of course, but socially not likely to happen again any time soon.
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @Oilid @Smokyish Hard forks are used to implement all network updates, e.g. when we transitioned from Frontier to Homestead mid-March. That will happen again in the fall when we move from Homestead to Metropolis. The network can't introduce new foundational features without a hard fork. BTC would/will have to use a hard fork to increase the block size, if that every happens.
  • SmokyishSmokyish Member Posts: 203 ✭✭
    @dlehenky Yes, i am well aware :)
    As s/he mentioned trust in the chain, i was making the comparison to this current situation with a hard-fork that isn't a normal road-map fork.
  • DividesByZeroDividesByZero Member Posts: 58
    Random musing about POS. What i can glean from reading posts and code on git
    • You can set a range min/max deposit (between 0-249) and go 9 rounds of bare-knuckle boxing for dibs. Note - I did get a chuckle that 250 would be dividing by zero.
    • Vast majority of people will be cut out and essentially every new person wanting to stake would be the 250th, always (in the long term).
    • At a minimum, with 250 guardians there would be 1,905,211.015 million ETH tied up. Liquidity on the exchanges would dry up. On the flip side there would be a lot of volume of buy orders while people try to secure enough ETH to stake.
    • Would you trust putting your ETH into a POS pool. Right now they hold a tiny fraction and you hold the rest, going forward they would hold all of your ETH and pay you a fraction.
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @DividesByZero Personally, I have no interest in participating in PoS, whether it be individually or on a pool. I'm also not convinced, at this point, that PoS will prove to be viable, but only time will tell.
  • Techman34Techman34 Member Posts: 405 ✭✭✭


    • Would you trust putting your ETH into a POS pool. Right now they hold a tiny fraction and you hold the rest, going forward they would hold all of your ETH and pay you a fraction.
    ^^Seems like a very risky proposition.

  • HeKazHeKaz Member Posts: 9
    "Hey, we lost all your ether trying to bet on stupid block. Sorry about that. Please deposit more."

    Yeah, I don't think PoS pool without a guaranteed deposit will work. And then they better stake by themselves if they have to cover them with a guarantee.

    I feel like the reward to stake will be too low to tie 1250 eth there. If the stake reward is high enough, then it becomes a race of "who is the richest here" to stake. It's a lose-lose proposition for me.

    The network will be less secure and way more centralized. Also, nobody but those 250 validators can hold back developers changes and the developers are the one that hold the more ethereum of the bunch so they are going to hold back themselves? Yeah, no...
  • BiodomBiodom Member Posts: 693 ✭✭✭
    edited July 2016
    personally, it would be too risky to bet 1250 eth to earn 125 max in a year.
    however, POS pools will come together and then it might make more sense, although what would be a difference between such pool and a bank? Not much.
    however, if there will be, say, 20 major POS pools and you spread your 1250 eth equally between them, then even if one crashes completely, you are still 125-62.5 eth ahead after a year and since pool ID can rotate each year (basically, reassembly of the pool), why not? 10% beats negative interest rate hands down.

    All of this is theoretical for now, of course.
  • DividesByZeroDividesByZero Member Posts: 58
    How popular is ETHC? I have a lot of peer attempts that are not pro-fork.

    I0722 22:16:20.939827 eth/handler.go:295] Peer 662e229870706239 [eth/63]: timed out DAO fork-check, dropping
    I0722 22:17:45.551786 eth/handler.go:295] Peer c66557801f1695fa [eth/63]: timed out DAO fork-check, dropping
    I0722 22:17:53.635962 eth/handler.go:295] Peer 662e229870706239 [eth/63]: timed out DAO fork-check, dropping
    I0722 22:18:38.249391 eth/handler.go:295] Peer 6cc2e00818fa9cf3 [eth/63]: timed out DAO fork-check, dropping
    I0722 22:19:02.835216 eth/handler.go:295] Peer b02d4cee47de189d [eth/63]: timed out DAO fork-check, dropping
  • DividesByZeroDividesByZero Member Posts: 58
    edited July 2016

    How popular is ETHC? I have a lot of peer attempts that are not pro-fork.

    I0722 22:16:20.939827 eth/handler.go:295] Peer 662e229870706239 [eth/63]: timed out DAO fork-check, dropping
    I0722 22:17:45.551786 eth/handler.go:295] Peer c66557801f1695fa [eth/63]: timed out DAO fork-check, dropping
    I0722 22:17:53.635962 eth/handler.go:295] Peer 662e229870706239 [eth/63]: timed out DAO fork-check, dropping
    I0722 22:18:38.249391 eth/handler.go:295] Peer 6cc2e00818fa9cf3 [eth/63]: timed out DAO fork-check, dropping
    I0722 22:19:02.835216 eth/handler.go:295] Peer b02d4cee47de189d [eth/63]: timed out DAO fork-check, dropping

    Second thought. Couldn't this be a clever way to do a denial of service attack. Spin up a couple thousand VMs with a non-fork client and aggressively try to peer. The default is 25 peers in Geth.
  • dlehenkydlehenky Member Posts: 2,249 ✭✭✭✭
    @DividesByZero I get the same thing on the main fork. I think it's just because of the connection limit for geth (or parity) peers. The p2p peers don't know you have a limit, so ...
  • dont12dont12 Member Posts: 60
    edited July 2016
    G416G said:



    Ethereum foundation is bunch of young guys wrapped up in technical but no clue what happens outside the server room, and judging from the DAO no clue about the code either.

    my 2 cents.

    True bunch of young guys, but you are confusing slockit which was the DAO with the Guys involved with Ethereum, which are the people that pulled off the hard fork.

    I would not bet against them getting Casper and Sharding off the ground. Then the Network is entirely internal and can theoretically get to sub 1s block times.

    The thing about 250 Validators its not like they are the same ones. They will put some kind of round robin or Random choice thing in place so its not always the same people. The Deposit is to ensure that you don't cheat, the idea is this; if you mine you are betting your electricity cost on getting a block. If you don't get a block you lose your electricity cost. But as we know over time, mining is profitable. Same thing with POS you deposit ether some of it may be lost but on average over time you generate a return. The deposit ensures you have skin in the game, the same way a miner does with equipment and electricity cost. Its not 1500 ether and oops sorry its all gone try again. Idea is to get away from the centralization that POW leads too. But to maintain the same economic rationale behind mining.

    They have already built in a difficulty bomb, its already steadily increasing the difficulty, all be it slowly, but in about 6m it will take the present hash power of the network like 6m to mine a block. So you can fork ethereum, but then need to hard fork it to remove difficulty bomb, so its not the original chain anymore.

    Buying new mining equipment is a bet on POS not coming online, before you recover your capital or betting that ether goes up so much in price you get your capital back, so seriously why not take your capital and just buy ether and if it goes up in price sell it, no need to buy build maintain rigs.
Sign In or Register to comment.