5 ms·

Proof of stake on Ethereum is right around the corner. As soon as that happens there will be plenty of cheap GPUs back on the market.

raverbashing 9y ago
The funny thing about the bitcoin fans is how they don't realize (or think it's good) that the BTC network does not become faster or scales as more hardware is added, quite the contrary
- nkrebs13 9y ago
  I don't think many people think that. They think more hardware means more money in their pockets.
- clarkmoody 9y ago
  It does, however, become exponentially more difficult to double-spend, which is the whole reason we have proof-of-work security in the first place.
  - TeMPOraL 9y ago
    Which is why cryptocurrencies won't convert into more energy-efficient schemes - after all, they see the exponential waste as feature - and we'll be stuck with this energetic disaster until some authority finally puts a stop to this charade.
    - clarkmoody 9y ago
      "Some authority" will use violence to stop mining locally, and it will immediately move elsewhere. This is a nearly pure market system, driven by economic incentives. Cryptocurrency mining is the best and highest use for the scarce resource of energy. Once it becomes unprofitable to mine, then the hash rate will stabilize and the energy will be used for better purposes. Marginal producers of hash rate will be priced out of the market.
  - xg15 9y ago
    Yes, however, it already is enormously difficult to double-spend. If this were the actual motives of miners adding ever-more hash power, we ought to discuss at some point how secure is "secure enough" and at which point the added security has a higher cost than benefit. Otherwise the whole thing sounds like the paperclip maximizer justifying its actions: "But if I burn this forest, we will have even more paperclips!"
    - clarkmoody 9y ago
      The only motive of miners is profit. This is by design. How else could you incentivize a bunch of random people to find partial SHA-256 collisions? Security is ensured by the economic incentives baked into the protocol. Security is the by-product of economically selfish behavior.
vanderZwan 9y ago
Oh wow, I never thought of that... If it works out the way you suggest, that will be fun for the undergrad projects.
IntronExon 9y ago
Translation: Massive centralization right around the corner. It’s just amazing how quickly all of those Libertarian principles died in the face of a quick buck.
- kernelPan1c 9y ago
  Yes! PoS poses a much greater risk of centralization
- TeMPOraL 9y ago
  The principles were never there. That is, in cryptocurrency space, you have a small minority of "true libertarians" (who'll likely stay with PoW scheme), and then the vast majority made of scammers, greedy people, and clueless ones who just want to get rich quick.
caioariede 9y ago
Those guys will probably switch to another PoW coin
- crypt1d 9y ago
  some will. But ETH has the majority of the hashing power, so any significant moves to other coins will result in major difficulty increases. Which will further decrease mining profitability and hence even more GPUs on the market.
- danmaz74 9y ago
  Probably ETC?
xbkingx 9y ago
Yup, and it'll be interesting to see how other cryptos adapt. Proof of Stake, though, worries me. It's a very "the rich get richer" solution trying to prevent the rich getting richer. Less electricity, sure, but it feels like it deepens the economic wedge between classes. If you stake 1 coin and I stake 100 coins and there's a 10% reward, after payout you'll have 1.1 and I'll have 110. How is that decentralizing? And you can't put tiers in the system because people will just install a thousand wallets/nodes and split their coins until they fall into the highest paying bracket. I really like what the Monero folks are doing. They announced that they'll be making small changes to their PoW algorithm a few times per year in order to deter ASICs (no point to spending time and money on a super specialized machine that will be worthless in 2 months). It doesn't deter GPU mining too much, though I'd imagine that it's more of a hassle keeping a swarm or systems up to date than one rig. Ideally, they would release tweaks that reduce GPU efficiency, like highly serialized computations with lower difficulty (so the payout is similar). And then you have Gridcoin, Curecoin, and the AI focused coins that give something back to the world, forcing a charitable donation. It's a shame that they aren't more popular/valuable. They're kneecapped by academia's schizoid relationship with computing in too many ways, but seem to have good teams and communities behind them. My amateur opinion is that the flexibility of shown by Monero (changing their PoW algorithm) should be standard for all coins, as should the usefulness of work, like Gridcoin and Curecoin (not just 'securing the network'). Maybe something like a carbon tax, where exchanging the currency carries a contract to provide some proportional amount of work in a scientific project. Use a reverse Proof of Stake with a generational memory so that stake weight follows a coin for a while after each trade. The generational weight can be reduced by performing useful work. That useful work itself could be bought and sold, but disappears when spent maturing the traded coin. I guess I'm trying to brainstorm a system that requires everyone to participate in useful work of unpredictable efficiency, where stability is added by punishing frequent exchanges, and holding the coin heals the wealth gap. Basically, if you're just acquiring money to 'have money' the system will deem you less important than someone acquiring money to 'use money', but forgives somewhat if you give back in computing power. Just spitballing. I'm sure I'm missing a ton of details, but it was a fun exercise for me.

raverbashing 9y ago

The funny thing about the bitcoin fans is how they don't realize (or think it's good) that the BTC network does not become faster or scales as more hardware is added, quite the contrary

nkrebs13 9y ago
I don't think many people think that. They think more hardware means more money in their pockets.
clarkmoody 9y ago
It does, however, become exponentially more difficult to double-spend, which is the whole reason we have proof-of-work security in the first place.
- TeMPOraL 9y ago
  Which is why cryptocurrencies won't convert into more energy-efficient schemes - after all, they see the exponential waste as feature - and we'll be stuck with this energetic disaster until some authority finally puts a stop to this charade.
  - clarkmoody 9y ago
    "Some authority" will use violence to stop mining locally, and it will immediately move elsewhere. This is a nearly pure market system, driven by economic incentives. Cryptocurrency mining is the best and highest use for the scarce resource of energy. Once it becomes unprofitable to mine, then the hash rate will stabilize and the energy will be used for better purposes. Marginal producers of hash rate will be priced out of the market.
- xg15 9y ago
  Yes, however, it already is enormously difficult to double-spend. If this were the actual motives of miners adding ever-more hash power, we ought to discuss at some point how secure is "secure enough" and at which point the added security has a higher cost than benefit. Otherwise the whole thing sounds like the paperclip maximizer justifying its actions: "But if I burn this forest, we will have even more paperclips!"
  - clarkmoody 9y ago
    The only motive of miners is profit. This is by design. How else could you incentivize a bunch of random people to find partial SHA-256 collisions? Security is ensured by the economic incentives baked into the protocol. Security is the by-product of economically selfish behavior.

vanderZwan 9y ago

Oh wow, I never thought of that... If it works out the way you suggest, that will be fun for the undergrad projects.

IntronExon 9y ago

Translation: Massive centralization right around the corner. It’s just amazing how quickly all of those Libertarian principles died in the face of a quick buck.

kernelPan1c 9y ago
Yes! PoS poses a much greater risk of centralization
TeMPOraL 9y ago
The principles were never there. That is, in cryptocurrency space, you have a small minority of "true libertarians" (who'll likely stay with PoW scheme), and then the vast majority made of scammers, greedy people, and clueless ones who just want to get rich quick.

caioariede 9y ago

Those guys will probably switch to another PoW coin

crypt1d 9y ago
some will. But ETH has the majority of the hashing power, so any significant moves to other coins will result in major difficulty increases. Which will further decrease mining profitability and hence even more GPUs on the market.
danmaz74 9y ago
Probably ETC?

xbkingx 9y ago

Yup, and it'll be interesting to see how other cryptos adapt. Proof of Stake, though, worries me. It's a very "the rich get richer" solution trying to prevent the rich getting richer. Less electricity, sure, but it feels like it deepens the economic wedge between classes. If you stake 1 coin and I stake 100 coins and there's a 10% reward, after payout you'll have 1.1 and I'll have 110. How is that decentralizing? And you can't put tiers in the system because people will just install a thousand wallets/nodes and split their coins until they fall into the highest paying bracket. I really like what the Monero folks are doing. They announced that they'll be making small changes to their PoW algorithm a few times per year in order to deter ASICs (no point to spending time and money on a super specialized machine that will be worthless in 2 months). It doesn't deter GPU mining too much, though I'd imagine that it's more of a hassle keeping a swarm or systems up to date than one rig. Ideally, they would release tweaks that reduce GPU efficiency, like highly serialized computations with lower difficulty (so the payout is similar). And then you have Gridcoin, Curecoin, and the AI focused coins that give something back to the world, forcing a charitable donation. It's a shame that they aren't more popular/valuable. They're kneecapped by academia's schizoid relationship with computing in too many ways, but seem to have good teams and communities behind them. My amateur opinion is that the flexibility of shown by Monero (changing their PoW algorithm) should be standard for all coins, as should the usefulness of work, like Gridcoin and Curecoin (not just 'securing the network'). Maybe something like a carbon tax, where exchanging the currency carries a contract to provide some proportional amount of work in a scientific project. Use a reverse Proof of Stake with a generational memory so that stake weight follows a coin for a while after each trade. The generational weight can be reduced by performing useful work. That useful work itself could be bought and sold, but disappears when spent maturing the traded coin. I guess I'm trying to brainstorm a system that requires everyone to participate in useful work of unpredictable efficiency, where stability is added by punishing frequent exchanges, and holding the coin heals the wealth gap. Basically, if you're just acquiring money to 'have money' the system will deem you less important than someone acquiring money to 'use money', but forgives somewhat if you give back in computing power. Just spitballing. I'm sure I'm missing a ton of details, but it was a fun exercise for me.