17:23:45 Thread on the continuous decline of the BTC security budget: https://twitter.com/MorganTBennett/status/1482411627655933953 17:27:54 Half the numbers he mentions are irrelevant though 17:28:41 The security of the network is determined by the hashrate in relation to the number of devices actively mining 17:30:18 Don't recall the exact numbers I calculated, but the number of bitcoin asics is somewhere in the order of 1-10 million devices (assuming Antminer S9 or S19), for a value of several billion dollars worth of hardware 17:31:31 The mining incentive for for-profit miners is determined by the coin emission * price - and thanks to the massive pumps in price in the last year, the emission rate of BTC is extremely high 17:33:09 The only relevant thing in that post is the fees are not fully replacing the decreasing base reward - but that's only part of the picture 17:34:32 You could have an emission rate of 1 BTC/year, but if that 1 BTC were worth ten trillion dollars, mining would still be more profitable than it is today 17:36:25 A much bigger threat to mining is the increasing cost of electricity, which forces miners to move to places with cheaper (and often dirtier) electricity, or shut down their operations 17:39:29 is determined by the coin emission * price - and thanks to the massive pumps in price in the last year <--- did you mean what you typed? the emission is fixed, only the price increased. 17:43:02 Yeah sorry, I used the same term for raw emission in btc and for the "emission" in fiat (which is the sum of base reward + fees, multiplied by the price) 17:43:55 Btc has the highest "emission" in usd, in the range of 500$/second 17:44:00 Only matched by Eth 17:47:46 right. So if the price goes up 2x/4 years, then the total $ emitted will double as the block reward halves. HOWEVER it does mean that the $ securing the network stays the same while the market cap doubles 18:01:23 I'm guessing the price of asic's scales with the price of BTC? 18:15:03 Inge: that's right. Though I don't think the market cap is inherently relevant. The "amount" of security does not depend on it, nor would the network become less secure if the market cap went up while the fiat emission stays the dame 18:18:39 is the question, what is the worth of the network and what is required to secure it? 18:21:19 Mining security is not an on/off switch, it's a relative amount. The bigger the mining network, the more resources it takes to attack it. Given bitcoin's mining network size, I'd say it's pretty secure - for the time being 18:22:37 It would be extremely hard for anyone to come up with another 2 million antminer S19s (or more) to pull off a 51% attack on Bitcoin 18:23:49 no need to purchase 2 million miners, just take over 51% of what exists 18:24:28 "just" 18:24:58 would have to be state sponsored no doubt 18:33:08 Even with all the state sponsoring, it would still require a massive amount of raw resources to manufacture all that silicon 18:33:58 As for hacking the existing devices/pools, that's a whole other issue 18:36:46 Ah, found my previous estimates 18:36:59 * merope uploaded an image: (169KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/FzbzzLmVovncbCPHREmOuHYv/Screenshot_20220116-193541_Element.jpg > 18:38:22 As for ETH + ETC, their combiner networks are equivalent to roughly 18 Million Vega gpus, for a rough cost of 7-8 billion USD 18:39:41 By comparison, the Monero network is ~400k Ryzen 3600X cpus, or ~150k Ryzen 5950X 18:58:18 hmmm, just need to win a few hundred mil on the lottery, buy a warehouse, hook up some cooling, then buy every Ryzen, mobo and memory combo I can find 18:58:56 wonder how the network would react if you just dropped like 100K 5950Xs onto p2pool 19:00:29 That would be roughly 2GH/s. At that rate, you would be better off just solo mining 19:00:59 I'd still p2pool it, just setup my own sidechain :P 19:01:14 (And would be in control of nearly 40% of the total network) 19:01:47 If you're mining on p2pool on your own sidechain with nobody else, then you are effectively solo mining 19:01:58 Might as well connect the miners directly to the node 19:02:24 (Or through a proxy, to avoid ddosing your own node) 19:04:23 would still take a month or so to put together and rack up 100k machines 19:05:33 And also 100+M USD for the hardware alone 19:07:11 Plus the warehouse with industrial cooling, plus the people required to assemble all that, plus a cheap energy source that can supply ~25 MW 19:08:07 cover the roof with solar panels and put a claim in for hybrid power grants :P 19:08:35 Lol 19:08:56 That would be quite the big roof 19:08:57 could probably charge the UPS's with solar :) 19:10:30 oh and lighting etc... then just run everything else off the mains... so long as there's some kinda solar/renewable energy in use, the government is likely silly enough to give you a 25% discount for utilities 19:12:05 I assume you're talking about the US? 19:12:14 UK 19:12:31 thought I suppose US probably has something similar in the more urm climate friendly states 19:12:49 Good luck getting a large enough energy storage system that can supply enough power for the night though 19:13:21 looking at what? peak 300W per rig? 19:14:16 ~250W while mining steadily, from what I've gathered 19:14:28 (Assuming some efficient tuning) 19:15:35 my 3900X runs at 140W from the wall 19:16:01 Hashrate? 19:16:07 14,600 19:16:57 3800CHz at 1.025v I think 19:17:02 *GHz 19:17:16 Nice 19:19:10 pretty damn good 19:19:42 I haven't measured my 3600, but the overall power usage of the house hasn't gone up over the last year so, probably not terrible 19:20:20 my elec costs just went up 10% 19:20:55 gas/elec combined here, fixed rate recalculated annually 19:56:42 I don't think it's physically possible to just buy 100k Ryzen 5950X 19:56:52 not in one go at least