Detailed explanation of EIP-1559 and off-chain proof

IPFS Force Zone (ID: IPFS-FORCE)

, author: Hu Feitong, published with authorization.

There are two things worth mentioning recently. One is the heated debate on EIP-1559 in Ethereum, and the other is that Filecoin's space-time proof has been changed to off-chain verification. EIP-1559 was first implemented on Filecoin to good effect, providing a good time base for Ethereum adoption. And off-chain verification is a bold attempt. In the world of blockchain, Filecoin continues to lead in innovative design.The Ethereum chain is congested and Gas is soaring. During the nearly 6 years of the development of Ethereum, the transaction fees earned by miners have exceeded the block rewards, which is probably not thought of by the original designers of Ethereum. How to solve the network security and high transaction fees of Ethereum is imminent.Currently, the Ethereum community is actively promoting the implementation of EIP-1559, but, unsurprisingly, this proposal has met with strong opposition from some miners. The main reason is that the BaseFee was burned and a large chunk of income was lost.

EIP-1559 is not a new proposal. In fact, EIP-1559 has been implemented on Filecoin for half a year. Then, I suggest that the Ethereum community and miners can refer to the implementation of EIP-1559 in the Filecoin network. Despite many doubts before implementing it, in my article

• Let’s talk about Filecoin’s transfer fees

• This has also been discussed, and there are indeed many factors that need to be fully considered and considered clearly. But in general, EIP-1559 has performed very well in the Filecoin network, as Juan said: So far, EIP-1559's "basic fee" mechanism has done a good job of keeping the network capacity at 100%. very good.

• Filecoin boldly took the lead in adopting EIP-1559 when the network was launched, without historical burden, and despite some doubts, it still advanced very smoothly and smoothly. At present, the stability and security that EIP-1559 brings to the Filecoin network, as well as the network utilization, are very effective. To sum it up roughly:

• Since the Gas BaseFee fee is closely related to the network bandwidth consumption, through the adjustment of BaseFee to encourage or restrain the growth of computing power, the Filecoin network has been in a reasonable high-load operation state since its mainnet launch, avoiding excessive load operation and ensuring network security. Guaranteed, stable operation

The adjustment mechanism of Gas BaseFee encourages miners to adopt a more intelligent mechanism to control the growth of computing power. At the same time, it encourages miners to configure mining pools more flexibly from the perspective of cost. For example, recently, about 70% of the computing power growth of the entire network comes from The 64G sector is an adaptation of the miners to the network

The Base part of the Gas fee is completely burned and the existence of an over-estimated burning mechanism makes it meaningless or worthless for miners or network participants to artificially manipulate the Gas price. Everyone thinks more about how to save costs and prevent the increase Gas fee to increase the channel of income, the network is more secure

The burning of a large amount of Gas benefits the holders of the entire network. At the same time, the burning of BaseFee suppresses the private transactions between Filecoin customers and miners

At present, one of the main dissatisfaction of miners in Ethereum is that the benefits have been reduced. However, this situation is the same in the Filecoin network. The difference is that the original design of Filecoin is like this, but from the perspective of network operation and development, the basic fee is burned to the miners. Miners are more willing to accept the benefits it brings than the benefits it brings. Especially the improvement of stability is very important for Filecoin, a chain with mortgages and penalties.

Of course, there is a big difference between Ethereum and Filecoin. The important point is that Filecoin includes a mortgage and penalty mechanism, which requires higher stability. How to avoid community splits on the adoption of EIP-1559, the Ethereum community and miners may need more discussions to reach a consensus.

• Although EIP-1559 has brought many benefits to Filecoin, one point still needs to be considered, that is, EIP-1559 has greatly reduced the transaction income of miners, so after the Filecoin network has been running for many years, without additional issuance, the district Block rewards will become less and less, so in addition to ecological applications, it is still a question whether transaction fees are needed. In theory, this is still needed. Because Ethereum includes an additional issuance mechanism, this problem is not serious, but Filecoin is not. Of course, that is many years later, and the community can come up with new proposals to solve it according to the situation.

• The adoption of EIP-1559 improves the stability and security of Filecoin. However, when Filecoin only does basic proof, the network is already in a high-load operation state, so what about data transactions, real data storage and retrieval? What about ecological applications? The problem of low TPS has been mentioned many times in the community. So how to improve TPS? One thing that must be done is to reduce the network bandwidth consumption of proof of storage. Filecoin v1.5.0 is the first step to solve this problem: move the verification of space-time proof to off-chain.

• This is another bold innovation of Filecoin. Assuming that most of the miners are rational miners, the security of the chain is guaranteed through the reporting mechanism, rather than the verification of the entire network. This mechanism has already been adopted in Filecoin, that is, the destruction verification of the consensus mechanism. This time it is proof verification, which is once again the first of its kind in the blockchain world. Why does this work? There are some reasons:

Filecoin's proofs are not transactions and do not directly involve the exchange of funds between users; therefore, late detection of false information will not have an immediate security impact

Strict reward and punishment measures are used to ensure that the cost of miners submitting dishonest certificates is extremely high, so they are unwilling or afraid to submit false certificates to ensure network security

Someone will always perform off-chain verification to ensure that every proof will not be missed

Protocol Labs has promised that they will run at least one node for off-chain verification, so as to ensure that dishonest proofs are discovered in the first place; some miners in the community also said that they will verify every space-time proof message.

• However, there is still a problem here, that is, in a blockchain world, an ideal design should not rely on individual or organizational commitment, but on mechanisms, profit-driven, and incentive-compatible. If the benefit to miners running off-chain validation is less than the cost, then there is insufficient incentive. In the current situation, it cannot be proved that the reward amount currently designed can be attractive enough for long-term verification. In a completely decentralized world, the effect will be greatly reduced. This area is still a topic worthy of study. But we believe that before a more perfect mechanism emerges, even if driven by interests, there will still be some people in the ecosystem who will conduct verification to ensure the health and safety of the chain, thereby protecting the value of the chain.

• The upgrade of Filecoin v1.5.0 has greatly reduced the gas consumption of wdPoSt, but this part of the gas consumption of the entire chain is not too high. Therefore, the improvement of network TPS is limited. It is roughly estimated that the optimization of this part (FIP-0010) plus the optimization of HAMT/AMT data structure and access of FIP-0007 may bring about 15% improvement of the entire network TPS. This is not enough to lower the current Gas BaseFee.

The good news is that, as mentioned earlier, this is only the first step in reducing the gas consumption of basic proofs. The next step is two other optimizations that you can look forward to:

FIP-0008: Add the PreCommit method for miners to submit in batches