Nobody is still using GPUs. The best GPU on the market is around 1 GH/s. A decent ASIC (the Antminer S7) is at 5,000 GH/s, for around the same price as the GPU. GPUs are now completely negligible as far as mining goes and they have been for multiple years.
Residential power is the same thing. Yes, there are some people who are mining out of their homes (and in the past I was one), but they are completely swamped out by the people harvesting economies of scale who are running datacenter-scale mining operations. Losing all miners on residential power will not have a meaningful effect on the network hashrate.
> the best-situated miners using the "most efficient" hardware is a factor of 6. The halvening would make that 3.
At which point it would rise again over time to find its natural economic equilibrium. So long as it doesn't go below one you will not find a huge crash occurring.
> That doesn't sound like much margin to me.
3X over seems like a huge margin to me. You don't see that much margin often in your everyday life.
> What if we see another 50% crash in BTC value? That too would take us right down to the edge of failure, right?
We have seen huge crashes in the price of Bitcoin before, and what happened is that the network hashrate stopped most of its short term growth, which is fine. It has occasionally actually gone down as people switched off their miners temporarily. Certainly never catastrophic. You can look up nice charts showing the entire history of Bitcoin's price, overlaid with network hashrate and mining efficiency -- it's interesting.
> But the concerns certainly sound legitimate to me, and what you posted honestly seems like apologism.
No offense intended, but if you didn't even know that GPU mining is years obsoleted, you need to go learn more before attempting to make judgments on what is valid critique and what is apologism. All the information on hashrates of current mining rigs and power efficiency is out there. Go spend a few minutes looking them up and running your own calculations rather than arguing from a base of zero information. You're falling victim to arguments that are phrased in a way that is seemingly persuasive, but that fall apart when you actually look at the history of the network hashrate and how it reacts to price changes.
Keep in mind that the halvening can be modeled exactly as a doubling of difficulty, which isn't that bad, and has occurred many times over short timeframes over the past. A halvening isn't nearly as bad as, say, a collapse by half in the price of Bitcoin (which affects all of your holdings as well, not just your future earnings), and that too has happened several times before. And a halvening has ALREADY HAPPENED BEFORE, and nothing catastrophic whatsoever happened. The network didn't even blink. You need to have credible arguments as to why the next one will be different from the previous one, and I haven't seen a single one. I mentioned that in the post you responded to and you completely ignored it.
I vaguely remember the days when people mined at home using their CPUs and people would write their own GPU miner and hoard them to get am edge. These days a long gone.
ASIC miners have been great but they cannot scale indefinitely. Currently even the larger pools have electricity costs somewhere between 30% to 50% percent of their average payout. Previous halfing are handled more gracefully because they occured when the value of BTC was going up, and it won't be the case in July. And it's not like we can somehow manage to make miners with 100% more performance per watt either.
Residential power is the same thing. Yes, there are some people who are mining out of their homes (and in the past I was one), but they are completely swamped out by the people harvesting economies of scale who are running datacenter-scale mining operations. Losing all miners on residential power will not have a meaningful effect on the network hashrate.
> the best-situated miners using the "most efficient" hardware is a factor of 6. The halvening would make that 3.
At which point it would rise again over time to find its natural economic equilibrium. So long as it doesn't go below one you will not find a huge crash occurring.
> That doesn't sound like much margin to me.
3X over seems like a huge margin to me. You don't see that much margin often in your everyday life.
> What if we see another 50% crash in BTC value? That too would take us right down to the edge of failure, right?
We have seen huge crashes in the price of Bitcoin before, and what happened is that the network hashrate stopped most of its short term growth, which is fine. It has occasionally actually gone down as people switched off their miners temporarily. Certainly never catastrophic. You can look up nice charts showing the entire history of Bitcoin's price, overlaid with network hashrate and mining efficiency -- it's interesting.
> But the concerns certainly sound legitimate to me, and what you posted honestly seems like apologism.
No offense intended, but if you didn't even know that GPU mining is years obsoleted, you need to go learn more before attempting to make judgments on what is valid critique and what is apologism. All the information on hashrates of current mining rigs and power efficiency is out there. Go spend a few minutes looking them up and running your own calculations rather than arguing from a base of zero information. You're falling victim to arguments that are phrased in a way that is seemingly persuasive, but that fall apart when you actually look at the history of the network hashrate and how it reacts to price changes.
Keep in mind that the halvening can be modeled exactly as a doubling of difficulty, which isn't that bad, and has occurred many times over short timeframes over the past. A halvening isn't nearly as bad as, say, a collapse by half in the price of Bitcoin (which affects all of your holdings as well, not just your future earnings), and that too has happened several times before. And a halvening has ALREADY HAPPENED BEFORE, and nothing catastrophic whatsoever happened. The network didn't even blink. You need to have credible arguments as to why the next one will be different from the previous one, and I haven't seen a single one. I mentioned that in the post you responded to and you completely ignored it.