4 ms·

I know miners will hate me for saying this it will bring up the lightning network but I think the transaction costs are too high. They have to come down (close

I know miners will hate me for saying this it will bring up the lightning network but I think the transaction costs are too high. They have to come down (close to zero if not zero) even if it means miners and node operators are working for no gain. I don't understand why it means that Bitcoin is insecure for people to be able to mine in their gpu. Maybe we need to leave Bitcoin alone and come up with another currency for low value transactions that happen frequently? I'm thinking like buying coffee or groceries.

AgentME 9y ago
>I don't understand why it means that Bitcoin is insecure for people to be able to mine in their gpu. What do you mean? People can mine with a gpu, but they won't make much because of the many miners using much faster ASICs. If you think Bitcoin should have been made so that ASICs weren't possible: a number of altcoins agree with that idea, but there's an argument to be made that optimized hardware is always possible, and by making it hard to make, it makes it more likely that a single small group who manages to make optimized hardware will be the only one to do so for a large time period, will be able to make huge amounts of profits funding more hardware, and at that point will easily be able to get up to >50% of the network hash rate (which breaks Bitcoin's decentralization and allows rewriting the blockchain). By using a proof-of-work that's relatively straight-forward to optimize means that when the possible mining rewards eventually incentivize development of ASICs, it's likely that multiple groups will be able to build them.
- fapjacks 9y ago
  Yes, but this way you're trading an elite made of hardware specialists and people that can afford to hire hardware specialists, with an elite made of people that can afford to fill datacenters with the most specialized hardware. Mining quickly becomes a game of "who can throw the most money into the problem" and ends up skewing the pool significantly because you are rewarded with money for winning the game, which means you can afford even more of the specialized hardware. That's the big thing with Bitcoin: All the miners are rich folks (these days, mostly Chinese) with huge datacenters and tons of cheap electricity. I have 12 of the USB ASIC miners that are many times unable to even pay for themselves in electricity costs because the difficulty is so high, which is the problem that squeezes out grandma's GPU miner and pushes even more people out of the mining game (not to even mention the ballooning size of the blockchain).
  - ghthor 9y ago
    Sort of. There is a potential that if you take over the network that it all becomes worth nothing because users understand the trust problem and cash out in the lead up. Or the users elect to not use your currency, it could still hold value, as the blockchain until that point is valid, but it remains paused until the 51% is resolved. This encourages you to work covertly to organize the 51% stake and you have to hold it in secret or all your coin/power may become worthless. It's essentially the cold war and in the same way, it's in everyone's best interest to avoid destroying the value of the coin by being selfish in an attempt for differential advantage.
    - fapjacks 9y ago
      Sure, and to this point, there have been at least a handful of times when the 51% attack could have happened, as well-known, related mining pools could have combined their power to achieve the majority. It has been plausible before, and I'd imagine what you mention is the reason it still hasn't happened. It would destroy the coin, because who knows what happened while they had majority. There would probably be some kind of fork, and I'm sure the value would diminish. It would be messy for everybody.
xorcist 9y ago
It doesn't need to be another currency. You could have a lower security blockchain tethered to the main one, with higher capacity and lower fees but still carrying Bitcoin transactions. In general those are called sidechains, and there is a particular proposal right now called drivechain that might be worth a second look. There are several other ways of transacting in Bitcoin outside the main replicated-everywhere-forever blockchain, Lightning perhaps being the most well known (with the defining feature of instant transactions). Most of these systems are hard to implement without unmalleable txids, so interest will likely pick up now that we have segwit on the main chain. Those systems never left the idea stage while transactions were free but there's significantly more interest now. There are lots of interesting developments going on.

AgentME 9y ago

>I don't understand why it means that Bitcoin is insecure for people to be able to mine in their gpu. What do you mean? People can mine with a gpu, but they won't make much because of the many miners using much faster ASICs. If you think Bitcoin should have been made so that ASICs weren't possible: a number of altcoins agree with that idea, but there's an argument to be made that optimized hardware is always possible, and by making it hard to make, it makes it more likely that a single small group who manages to make optimized hardware will be the only one to do so for a large time period, will be able to make huge amounts of profits funding more hardware, and at that point will easily be able to get up to >50% of the network hash rate (which breaks Bitcoin's decentralization and allows rewriting the blockchain). By using a proof-of-work that's relatively straight-forward to optimize means that when the possible mining rewards eventually incentivize development of ASICs, it's likely that multiple groups will be able to build them.

fapjacks 9y ago
Yes, but this way you're trading an elite made of hardware specialists and people that can afford to hire hardware specialists, with an elite made of people that can afford to fill datacenters with the most specialized hardware. Mining quickly becomes a game of "who can throw the most money into the problem" and ends up skewing the pool significantly because you are rewarded with money for winning the game, which means you can afford even more of the specialized hardware. That's the big thing with Bitcoin: All the miners are rich folks (these days, mostly Chinese) with huge datacenters and tons of cheap electricity. I have 12 of the USB ASIC miners that are many times unable to even pay for themselves in electricity costs because the difficulty is so high, which is the problem that squeezes out grandma's GPU miner and pushes even more people out of the mining game (not to even mention the ballooning size of the blockchain).
- ghthor 9y ago
  Sort of. There is a potential that if you take over the network that it all becomes worth nothing because users understand the trust problem and cash out in the lead up. Or the users elect to not use your currency, it could still hold value, as the blockchain until that point is valid, but it remains paused until the 51% is resolved. This encourages you to work covertly to organize the 51% stake and you have to hold it in secret or all your coin/power may become worthless. It's essentially the cold war and in the same way, it's in everyone's best interest to avoid destroying the value of the coin by being selfish in an attempt for differential advantage.
  - fapjacks 9y ago
    Sure, and to this point, there have been at least a handful of times when the 51% attack could have happened, as well-known, related mining pools could have combined their power to achieve the majority. It has been plausible before, and I'd imagine what you mention is the reason it still hasn't happened. It would destroy the coin, because who knows what happened while they had majority. There would probably be some kind of fork, and I'm sure the value would diminish. It would be messy for everybody.

xorcist 9y ago

It doesn't need to be another currency. You could have a lower security blockchain tethered to the main one, with higher capacity and lower fees but still carrying Bitcoin transactions. In general those are called sidechains, and there is a particular proposal right now called drivechain that might be worth a second look. There are several other ways of transacting in Bitcoin outside the main replicated-everywhere-forever blockchain, Lightning perhaps being the most well known (with the defining feature of instant transactions). Most of these systems are hard to implement without unmalleable txids, so interest will likely pick up now that we have segwit on the main chain. Those systems never left the idea stage while transactions were free but there's significantly more interest now. There are lots of interesting developments going on.