daum3ns's avatar
daum3ns 1 year ago
well the hard numbers are there, all we have to do is the math.. many have done this before and there are many parameters you could take in consideration.. you will also find many discussions about this in the internetz... but let's go: we take the following assumptions: * we completely ignore energy costs for simplicity i.e we calculate only the initial costs of the attack, not the costs to actually run the attack * i just couldnt find the price for the best cpu in the xmrrig comparison as it seems to be a custom cpu or something, so we take the second best.. XMR case: * according to [1] moneros current hashrate is 2.23 GH * according to [2] the second best cpu for mining makes 148132H/s (AMD Epyc 9654) * one of this costs 3650 $ [3] (i found also other prices and im not a cpu expert so i just took the lowest) => for 51% (1.115 GH) we need 7527 such cpus (2230000000/148132) => costs us 27'473'807 $ [1] [2] [3] https://www.serversupply.com/PROCESSORS/AMD%20EPYC%2096-Core/2.4GHz/AMD/100-000000789_368317.htm BTC case: * hashrate: timechaincalendar.com shows 672.5 EH [4] * best miner makes 335 TH, the bitmain antminer s21 hyd and costs 4800$ [5] => for 51% (336 EH) we need 1003 such devices (336'000'000/335) => this costs us 4'814'328'358 $ [4] [5] so its 175x times more expensive to just buy this stuff... is this calculation correct? i don't know tbh ... seems not too much 1003 such miners.. also there are prob a lot of other factors we should consider.. as said, many have thought and discussed this.. we're not talking about a coordinated attack here or a consliracy of bad actors or something but about one actor starting from scratch... but IMO thats exactly the point.. do you think AWS/MS/Google/NSA have those 7527 cpus (or an equivalent to the hashingpower they bring) lying arround right now? i think each of them have... at this very moment... and could simply do it because there's no need to invest in specialized hardware.. thats the protection that asics give..

Replies (2)

Are you factoring the current size of Monero VS Bitcoin? (mining difficulty would increase, so attack difficulty would increase as well if Monero was of similar size) Why have they not attacked Monero? Is it because it is too small to be on their radar? Is it not worth diverting resources away from their current intended purposes? When it becomes a big enough threat to finally consider attacking will the hashpower be larger and more challenging to attack then? (Monero is a fraction of Bitcoins size) With Monero the lower barrier to be able to participate in mining also allows for more decentralization (at least in theory) I think ASIC and ASIC-resistant PoW both have their own unique downsides and upsides. ASIC is harder to attack, but once an attack is underway it is harder to recruit new honest hash to push back. The opposite is true of RandomX. Other ASIC-resistant advantages: ASICs centralize and make good targets in several ways. -Much more discrete (power draw, heat, noise) -General CPU purchases mean you blend in VS buying an ASIC. Everyone knows what you're going to do with that ASIC. -Large concentrated mining farms are easy to co-opt and regulate for governments. -Over half the hashpower now requires KYC -No P2pool. Stratumv2 helps but isn't sufficient. Large pool operators control payout so are targets for governments.
daum3ns's avatar
daum3ns 1 year ago
yes you have some good points here.. i love moneros idea that everyone can participate easily and i totally see the need for more decentralizazion regarding btc mining..