As BCH and BSV have completed halving the output, and the computing power of related networks has plummeted, the topic of 51% attacks has been discussed again, and some people will quote data from crypto51 or NiceHash to demonstrate the need to attack these networks. the cost of.
As shown in the figure above, the data displayed by crypto51 is that the cost of one hour to attack the BCH network is only $ 8983, and the cost of one hour to attack the BTC network is $ 540,000. Is that really the case?
- Getting started with blockchain | What is a 51% power attack?
- What laws could be violated by launching a 51% attack?
- Six of the most notorious 51% attacks: the survival of the fittest is the jungle law of cryptocurrency
- Anti-51% attack: Harvard MIT scholar proposes new theory of fighting double spend
- 51% of attacks will increase the value of ETH? See how Vitalik explains the security of Casper FFG
- Launching 51% of attacks to stop evil, is this choice reliable?
In fact, there are 2 major problems:
- The data source NiceHash's available computing power for leasing is very limited. For example, the sha 256 algorithm's computing power for leasing is only 298 PH / s, accounting for about 16% of the current BCH network computing power;
- The data source NiceHash's computing power price is not fixed. Assuming that someone hires computing power to launch a 51% attack in the short term, the rental cost of computing power will also increase (this has happened in history);
Therefore, the actual situation is that if you want to launch a 51% attack on the BCH and BSV networks, the real cost is much higher than the value shown above, let alone changing the target to BTC. Basically, the NiceHash market cannot cover the attack For the large-scale PoW coins that require computing power (selfish mining 1/3 or slightly lower than 1/3, double spend 1/2), if you want to simply attack through the computing power leasing platform, it is basically a dead end. One.
What is the real cost of the double spend attacking the BCH network? This is divided into three situations, and depends on whether you are a new miner or an old miner. If you are a master without a miner, then you need to rent out all the available computing power on NiceHash (the cost is relatively negligible) and purchase A large number of mining machines, taking S19 PRO as an example, this cost may exceed 100 million yuan, or you can persuade some miners to collude with you to let them provide computing power to attack, then the cost is much less than buying mine The cost of the machine (but still greater than the above-mentioned value of $ 8983).
And if you are an old miner, and you have a lot of computing power, and the cost of the mining machine has been recovered, the cost of the attack is actually the electricity fee required to run the mining machine. Although there is such a possibility in theory, in reality , Miners with sufficient resources (such as Bitmain) will not do things that are detrimental to them.
Of course, you might say that no matter what, I just want to attack, even at all costs.
(Picture from: tucheng.com)
Well, of course, it is still possible to attack, but this does not necessarily mean that there is no way to solve it.
In fact, even if a 51% attack actually occurs, it is possible to recover.
According to researchers at Harvard University and MIT, in February 2020, there were cases of anti-51% attacks on the Bitcoin Gold blockchain. After the attacker completed the attack, the defender continued to mine on the original chain, and The attacker competes and eventually forces the attacker to give up double spending.
Assuming that this mechanism is effectively used, the cost of double spending can be greatly increased, and the attacker may be frightened.
Well, the purpose of this article is not to show how safe BCH and BSV are, but to clarify some misunderstood concepts.
Related reading: " Anti-51% Attack: Harvard MIT Scholars Propose New Theory to Fight Double Flowers "