This article will discuss three main topics . Although the three have not been fully discussed, they are becoming more and more relevant and important:
1. MEV after EIP-1559 and before the Ethereum merger (The Merge) (August 5th-early 2022);
2. MEV (Optimism, Arbitrum) of 2-layer Rollup;
3. Multi-chain environment based on MEV.
First, let us briefly understand MEV
The root cause of MEV is that within a certain time window, you cannot fully control the operations you want to perform. Specifically on Ethereum, when you submit a request (such as a transfer), it will enter the candidate list for miners to choose from. During this period of time, you basically cannot control who “takes the order” or whether it will Was added to the block on time. On ETH’s Rollup (such as Optimism or Arbitrum), once your request is submitted to a centralized sequence operator, you lose the dominant power.
The time window between submitting the transaction and confirming the transaction is an available opportunity for other people (miners), and the value of this opportunity is MEV.
MEV can be roughly divided into three types: benign MEV (included in the protocol), bad MEV, and catastrophic MEV. Examples are as follows:
1. Benign MEV: The operation of some agreements relies on MEV capture to some extent, such as the liquidation of Aave, Maker, and Compound, or arbitrage between Uniswap and SushiSwap to maintain market effectiveness.
2. Bad MEV: Robo-run, sandwich arbitrage trading, etc.
3. Examples of catastrophic MEV : Threats to the consensus layer caused by reorganization and time-bandit chain reorganization attacks-if we do not enforce the method to end this probability, they may always appear in some form.
In the following discussion, I will focus on the latter two types of MEV.
As an Ethereum user, in addition to using dedicated relay networks such as Flashbots, Archerswap, and MistX, or using the RFQ system ZRX, Hashflow, and Cowswap for requesting quotations, we can do our best to avoid becoming victims of large-scale “bad MEV”, It is to set a maximum allowable deviation, which DEX calls “slippage”.
But is this enough?
At least it’s not perfect for now. The private relay system of MEV-Geth running Flashbots requires permission (miners must be whitelisted in order to forward groups of bundled transactions). It is also not really private, because before the chain, miners can still see the bundled transactions, and the bundled transactions on the chain can also be used. Although we have not seen this problem for the time being, it may become a problem after EIP-1559, because miners will have to ban this behavior.
In fact, some “short-cut bribers” will choose to submit transactions through other (non-public) channels. An anecdote from this morning is: Cypto Punk’s #3860 NFT was sold for less than $0.01 due to a seller’s error, and a “cunning” buyer submitted a transaction through Flashbots and paid for 22 ETH. Miner fees to jump the transaction of snapped #3860 to the top block. It is a win-win situation for the bribers who have obtained priority and the miners who get high gas charges, but for ordinary users (other #3860 buyers) it is an unfair competition.
If you are interested in learning more about the future of Flashbots, a truly amazing white hat community. Please check Robert Miller’s research proposal on MEV-SGX to understand the future of private relay. On the other hand, the RFQ system, which is effectively controlling over-the-counter transactions through DEX, is based on the assumption that market makers provide effective quotations in the liquid market and generate MEV forms through transactions.
MEV solutions for existing Flashbots robots
It is worth mentioning that “bad MEV” is nothing new to financial markets at all-if anyone has read Michael Lewis’ book “Flash Boy” (the inspiration for Flashbots), they will find that traditional trading markets also happen. The same thing: If you’re interested, you can find some great papers here. They concluded: “These high-frequency traders are paying for expensive services such as hosting and data services. By grabbing Before trading instructions, high-frequency traders impose false taxes on other investors without providing any return.” It sounds very much like a bad MEV in the crypto world, except that the crypto world follows “code is the law” and is not regulated by the SEC.
As for the catastrophic MEV, I suggest you click here to read the article on Re-orgs written by Paradigm and Vitalik.
The bottom line is that although the community has united against the “bad MEV” and these patches have been adopted, we need a better solution. The problem of “catastrophic MEV” still exists before the merger.
What will MEV look like after the upgrade in London?
Most people are paying attention to the upcoming London hard fork EIP-1559. Let us summarize the changes brought about by 1559:
· Instead of a single gas fee, each request order after the upgrade will include “tips” and “cost caps”;
· The basic miner fee will only be included in the block if the upper limit of the charge is at least the basic fee of the block.
The calculation formula is as follows: g = gas upper limit, δ = prompt, c = cost upper limit, r = basic cost, p = Gas price (only relevant before EIP-1559 upgrade):
The MEV-Geth currently earned by miners is based on more than 85% of the hash rate, where δ is the “tip” of each Flashbots package (~. 3 ETH/block):
Therefore, assuming that the upper limit of the gas fee g remains unchanged, the income difference between the miners before and after the upgrade in 1559 is roughly:
Since the gas fee was very high before the upgrade, the miners’ subsequent income may be lower than before the 1559 upgrade. Although we have not considered what the “tips” of the Flashbots bundle will be after the upgrade (and the transaction volume between these “tips” may increase significantly), the possibility of changing the results is very small because the increased traffic through Flashbots also Will reduce the basic cost r… This also means that less than expected ETH will be burned.
Although it may be necessary to repeat some of the content of the previous article, I still want to emphasize:
1. As miners’ profits are squeezed, the capture of Layer1 MEV is expected to become more competitive.
2. According to EIP-1559, you cannot submit a transaction with a 0 Gas fee-this means that there may be more useless blocks that have to be paid for the basic fee to be included.
3. EIP-1559 was designed before MEV was widely mentioned. Its expectation of burning gas is relatively moderate; Flashbots pays to miners through smart contracts. With the introduction of Layer 2, it plays a role in reducing the basic cost of Layer 1. effect.
How do Layer 2 respond?
The chart I recently followed is the frequent occurrence of MEV transactions on Uniswap v3 on Ethereum, and the situation of Optimism is also here. Optimism’s current total locked-up volume (TVL) is only 7 million US dollars, but I don’t think anyone would expect it to be far behind the total locked-up volume of 2 billion US dollars on Layer 1, at only 1/10 of the cost (Optimism first generation There are only ETH, DAI, SNX, USDT and WBTC in the version).
So, what does this matter? Along the trajectory of funds, we finally found that MEV has become the most controversial point of DeFi adoption on Layer 2, whether it is from Layer 1 to Layer 2 or the Layer 2 solution itself.
On Layer 1, the extractor of MEV is a miner, but on the second layer of Rollup, such as Optimism or Arbitrum, the sorting right is transferred to the sequencer. This sequencing program can manipulate the sequence arbitrarily like the miners on Layer 1. At startup, both competing teams will run the same sequencing program. The idea is that the centralized sequencing program will be checked by a group of verifiers, very similar to BSC. If the sequencing program finds malicious behavior, they can submit fraud proofs, cancel any incorrect transactions, and force their deposits to be confiscated, thereby canceling invalid order books. For this reason, a 7-day dispute window was also set up to withdraw funds from Layer2. However, if a large amount of MEV is obtained on Layer 2, this may be technically delayed for a long time.
L1> L2 deposit withdrawal process
Assuming that these centralized sequencing programs run by Optimism and Offchain Labs are altruistic and do not pursue MEV profits, then transactions should be processed entirely through delays and the time picked up by the sequencing programs. However, it is understandable that this is a huge pressure on the permissionless spirit of Ethereum, so both protocols propose to target strong Layer 2 fairness, despite Optimism’s MEV auction (MEVA) and Arbitrum’s fair ordering service (FSS) Fairness is weak on Layer 1.
In theory, MEVA is quite simple: the right of sorting is a valuable commodity, so the best way of valuation is to provide the ability to re-rank the transaction as an auction, and then give it to the highest bidder. Similar to how the mining industry creates revenue sources for miners to align the network to a large extent, MEVA allows sequencing programs to monetize their contributions and guarantee the chain’s economics in a centralized, winner-takes-all manner. Safety.
But does it really make sense? I am not the first to be skeptical of this pattern (pmcgooha has a great article here). But in the final analysis, this still does not solve any economic motives behind bad and catastrophic MEVs.
MEVA is a complete version of 0xbunny’s request for reorganization, but it only applies to Layer2, not Layer1. They legitimize the preemption in the code, and in return, allow the preemptors to pay a small amount of their income to the sequencer, and the sequencer is actually an accomplice. (It’s like a corrupt country where the rich become richer at the expense of users.)
Let’s take a look at Arbitrum’s final solution-FSS. FSS uses Chainlink’s oracle network (DON) to first transmit the request to the network, and then forward the request to a designated smart contract that applies FSS. The oracle network uses a design called Aequitas to distribute these encrypted request orders on arrival time (which means low-latency requests will win), and they cannot be decrypted or viewed before the order is issued.
The overall FSS design of Arbitrum is implemented by Chainlink
Enforce the transaction sequence when a transaction is received
Then, the decentralization was effectively transferred to the Chainlink oracle. This is also the key application scenario of oracles currently in DeFi-at least to me, this feels like a very natural step for oracles used in the technology stack. This makes a lot of sense, and I hope it will work.
1.Layer2 Optimistic’s solution brings an unprecedented usability to ETH, but at the beginning, we trust the centralized sequencer run by the team to extract MEV (runs as efficiently as BSC).
2. We should be prepared. Once these solutions are closed, MEV may be flooded on Layer 2-solutions like MEVA and FSS will not be released on time, I am not saying that they are not optimistic before implementation In the initial stage, it is important to predict how Ethereum will change in the next year.
3. I think that the bidding war surrounding MEVA and the conspiracy that follow will become a major theme of Optimism and Arbitrum. This will be a technical competition to submit transactions to DON most effectively with the lowest latency (in the end, it should be much more difficult to withdraw MEV on Arbitrum).
4. We have created something bigger and better! It is great to see that zkrollup and their Optimistic counterparts have solved many problems together.
Finally, I want to talk about an argument that many people in the community were very resistant to earlier this year-cross-chain DeFi is more likely to be inevitable.
MEV proves that the block space has the ability to extract value: multi-chain (whether it is Layer 1 or different schemes of ETH Rollup) is only a time scale problem, because there are so many different ways to balance the block space to minimize it And democratization, while retaining some form of decentralization.
In addition to Ethereum, what are the other Layer 1 related to DeFi? Solana and Avalanche are also considered to be the basic public chains to minimize MEV, and there will be more information on this in the future.
Thanks to Blake Richardson, Avi Felman and Steve Lee for the overall feedback on the BlockTower team (please follow our new media page Medium) and countless other community members for their research.
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/mev-is-a-topic-worth-discussing-in-the-post-eip-1559-era/
Coinyuppie is an open information publishing platform, all information provided is not related to the views and positions of coinyuppie, and does not constitute any investment and financial advice. Users are expected to carefully screen and prevent risks.