Babbitt Column | The computing power war is a paper tiger

Bitcoin mining output has halved, but the price has not doubled, so many mining machines will inevitably be idle. These idle mining machines will not be used to launch a 51% attack on the Bitcoin network, including BTC, BCH and BSV.

Let me first state the conclusion: absolutely not.

I used to think that computing power had a huge impact on the network and could pose a fatal threat. But after several years of observation, I found that even a successful 51% attack would not have a qualitative impact on the Bitcoin network.

Currently, the total hashrate of SHA256 in the entire network is very large, exceeding 120E. The hashrate is mainly concentrated on BTC, exceeding 110E. BCH is only more than 2E, and BSV is only more than 1E.

In terms of attack costs, launching a 51% attack on BCH and BSV is very cheap. According to theoretical calculations, the cost of reorganizing 6 BCH blocks, that is, launching a 1-hour 51% attack, is less than $10,000. The cost of attacking BSV is even cheaper, only about $5,000. Currently, there are more than 10 SHA256 mining pools that can launch a 51% attack on BCH and BSV.

BCH and BSV have been running at such a low computing power for a very long time, more than a year, but there has never been a 51% attack.

Another object of observation is LTC. Last year, when LTC halving was in full swing, the price once rose to more than $100, and the computing power was also very high, exceeding 400T, but now it is less than 200T. The computing power at its peak last year was more than doubled from what it is now. In other words, the computing power of LTC that has been idle has exceeded the computing power of the entire network.

LTC fell from a high of $100 to a low of more than $20. The hash rate also fell in the same proportion. In other words, the hash rate of LTC once reached less than half of the idle hash rate.

But the objective fact is that any 51% attack on LTC occurs.

In the face of such objective facts, I revised my previous view that computing power plays a decisive role in the Bitcoin network.

Another auxiliary fact is that Bitcoin Gold (BTG) once suffered a 51% attack, but the entire BTG system is still doing well.

What is the use of a 51% attack?

From the perspective of attacker benefits, the first effect of a 51% attack is double-spending transactions.

The principle of double spending is as follows. The attacker holds more than 51% of the computing power of the entire network, and constructs two conflicting transactions tx1 and tx2 before the attack. Tx1 is to recharge to the exchange, and tx2 is to transfer to the attacker's own wallet.

The attacker broadcasts tx1 and hides tx2. tx1 is deposited to the exchange, sold for ETH or other, and withdrawn to his own wallet.

The attacker then uses his own computing power to generate blocks to package tx2, and keeps hiding the blocks until they are longer than the chain that packages tx1, and then releases the chain that packages tx2.

Based on the principle that the longest chain is the valid chain, tx1 will be invalidated and tx2 will succeed.

The attacker successfully double-spended the money, but did not spend BTC, but got ETH.

The 51% attack on BTG two years ago was a double-spending attack on the Bittrex exchange. The attacker successfully stole millions of dollars of BTG. Bittrex eventually delisted BTG.

If combined with futures short selling, a successful 51% attack on an exchange and double spending should be able to make a lot of money.

So far, no 51% attack has been found to double-spend BCH and BSV. I am talking about double-spending on exchanges. This kind of double-spending is not the same as zero-confirmation double-spending. Zero-confirmation double-spending has a smaller impact.

Now exchanges are all aware of the risks, and basically 51% computing power attackers cannot steal money.

In addition, even if an exchange is attacked by 51%, only the exchange will be damaged, and the impact on the entire currency system will be small.

A 51% attack can also be used to monopolize mining.

Mining consumes electricity and hardware. The greater the mining power, the higher the mining cost, and the less the profit per unit of computing power. If a miner can monopolize mining and prevent other miners from coming, he can use lower mining power to maximize the profit per unit of computing power.

The basic principle is this.

Because SHA256 mining machines can mine multiple coins, BTC, BCH and BSV. If a large miner decides to monopolize the mining of a chain, he decides to launch a 51% attack on the chain and use his superior computing power to reject all other miners' blocks. The result is that other miners will not mine this chain. Then, the attacker can withdraw his computing power to mine other coins, and the attacked chain only uses a very small computing power to mine, so as to maximize the unit computing power revenue.

Once the attacker withdraws the computing power, other miners will definitely come back, so the attacker must always pay attention and frequently use 51% attacks to reject other people's blocks.

This strategy of monopolizing mining has been found in small currencies, especially graphics card coins.

BCH and BSV have such a small computing power now, especially compared to BTC, both are less than 3%. But there has been no mining monopoly. I think the most important reason is that it is too difficult to monopolize. The attacker needs to have more than 51% of the computing power of the entire network when attacking. After the attack, after withdrawing most of the computing power, other miners can use small computing power to harass the attacker. This offensive and defensive battle always requires a large cost for the attacker.

Another reason may be the real-name system. Now, those who can mobilize 1E computing power in the entire network are basically open. Even those who hide in the dark can be found out with a little investigation. Under such circumstances, the legal cost of launching an attack is too high.

In addition, for mainstream currencies, centralized mining will cause prices to fall, which makes monopoly mining even more meaningless.

What can’t a 51% attack do?

In addition to double-spending transactions, 51% attacks cannot use other methods to modify existing block ledgers. Coins cannot be created out of thin air, and historical transactions that do not have the private keys of the attacker cannot be tampered with.

A 51% attack can only modify the attacker’s own transactions. Therefore, a 51% attack cannot destroy the credit of the entire chain, it is just a default.

This is a bit like globalized business. The United States can default on its debts once or twice because of its strong military force. North Korea can also default on its debts because it is shameless, but the credit of the entire globalized market will not be greatly affected.

A 51% attack cannot be used to modify the protocol.

Modifying the protocol requires consensus with the vast majority of nodes in the entire network, not just mining nodes. Although mining nodes can modify the protocol individually, other nodes do not follow you, which is just an attacker issuing a new coin.

As long as a chain is decentralized, a 51% attack cannot modify its protocol. This is because the dominant computing power can only use the rule that the longest chain is the valid chain to force other computing power to follow, but cannot force other non-mining nodes to follow. For example, exchanges can directly shut down nodes and directly define the name of the coin. Even if the longest chain is the valid chain, other miners can refuse to recognize it. They can work with other developers and exchanges to modify the rules.

Using a hash war to modify the protocol was the biggest controversy when BCH and BSV split in 2018. At the time, some people believed that the hash war would definitely eliminate one of the chains and let the entire network follow the winner, but the fact is that apart from the war of words between the two sides, nothing else happened.

In history, there have indeed been cases where miners have modified the protocol of a chain through a 51% attack, but these are centralized coins, the developers and mining pools are integrated, and users have no idea what a complete node is.

In summary, on mainstream coins (mainly small coins with strong decentralized properties but not centralized), the so-called 51% attack, or hash rate war, has never successfully destroyed the credit of the chain in history. Even now, the cost of attacking BCH and BSV is very low and the potential benefits are very high, but no 51% attack has occurred.

The effect of a 51% attack is limited. It can only double-spend the attacker's own coins and cannot modify other people's bills.

Using a 51% attack to monopolize mining is uneconomical.

A 51% attack also cannot be used to modify the protocol.

So I think the so-called hash rate war is a pseudo-concept that has no impact on mainstream currencies. As for small currencies, they cannot survive without a 51% attack.