Q1: How to understand Ethereum Merge in an easy-to-understand way?
@pan: Merge is actually upgrading the consensus mechanism of Ethereum from PoW to PoS, replacing the traditional system of miners and mining machines. There are two core changes after the upgrade.
The first is that Ethereum has become more “green”. It does not have such a large computing power. It only needs some ordinary node operation services. This is one of the advantages and the most important adjustment.
Another adjustment is in the confirmation of blocks, that is, the so-called finality will be more certain, which is different from the “probabilistic confirmation” of the Nakamoto consensus. Other aspects, such as capacity expansion and gas fee reduction, will not be covered by Merge. These changes will not come until the future Surge Verge stage.
Q2: What are the minimum requirements for running a node after Merge? Can we have our own computer?
@pan: In fact, the operating cost of the node has not been reduced.
Miners that consume a lot of power have been removed, but now Ethereum nodes are still very heavy, so there are still requirements for SSD, network speed, and CPU. Although a home computer should be able to meet the demand, the cost of SSD is still very high, and if it is an archive node, it will even be more expensive, requiring a capacity of 4TB or 8TB, and ordinary hard drives are not enough.
Q3: Does Celer have running nodes? Does Merge have a big impact on Celer?
@Dominator008: I personally don’t feel much. We also run nodes for each chain. I understand that after Merge, the nodes of Ethereum will not change much. Including his entire evm layer, JSON RPC layer, there is no incompatibility, it is completely the same as before. It’s just that after switching to PoS, its finality will be more certain, it is no longer a probabilistic one, that’s the difference.
Q4: As a project party, do you need to prepare a lot?
@Dominator008: Basically no preparation required. And now, if you are using, for example, the Ropsten testnet, it is basically what it looks like after the merge, that is, the bottom layer is PoS, but contracts and everything can be used normally on it. So this should have no effect on the current situation.
But it actually needs to run two clients on the node, right, now it is divided into two Eth1 Client, Eth2 Client, or Consensus and Execution client. These two clients need to run on the same machine at the same time, so the Consensus Client may have some extra consumption, compared to the Eth1 stage. But it should be okay, not the kind that can’t be tolerated.
Q5: Looking back on the technical roadmap of Ethereum merge, what debates will there be in the Ethereum community about the PoW-to-PoS solution? Why do this thing?
@pan: The Merge should not have been a relatively high priority in the early Ethereum roadmap, especially around 18/19/20, it was not even a single thing. The priority at that time may be, sharding – beacon chain – sharding – and then perform promotion.
But since the end of 2020, after Vitalik proposed the rollup-centric solution, the entire technical route has actually been decoupled. The expansion of Ethereum is not so urgent to adopt sharding sharding technology, but it can actually be solved by rollup in the early stage. So the entire roadmap of Ethereum has undergone a very big change. Then The Merge can be introduced relatively early, and the burden of PoW can be abandoned faster – the Ethereum community must think that PoW is a burden, and then build more of this high-performance on the foundation of PoS as soon as possible. Or go in more directions.
So the main core is that after this rollup-centric roadmap, The Merge has become a priority for the entire community. At the same time, many routes are no longer completely dependent, it may be a form of a linear series, you have to do one thing and then another. But now it is in parallel, with various roadmaps. In fact, it is carried out in this synchronization-Merge is a piece, and the subsequent route has surge verge spurge, which may have several stages. This stage is actually in parallel development. Whichever is faster to develop or which is more important may be given priority to launch. Yes, it may be in such a state now.
Q6: Excluding carbon neutrality and energy savings, why would PoW be a burden?
@pan: You said this must be an important reason. Another reason, I think, is because Ethereum has already set this roadmap at the earliest stage, which is the route from PoW to PoS. So everyone already knows where the future is. So when everyone decides this roadmap, they will definitely take the migration between these two as a very important and a key time node.
@starit1992: Actually, the Ethereum community, including Vitalik, came up with the idea of switching from PoW to PoS. Then one of the consensus algorithms mentioned at that time was Casper. In fact, Casper has also had different branches, and finally everyone adopted an algorithm of this Casper ffg plus another LMD-GHOST for branch selection. Then the final consensus algorithm is Casper+LMD-GHOST, so it is called Gasper.
Although we call this fork ETH PoS, in fact, the last consensus algorithm it uses is called Gasper. In fact, it has undergone a very long-term debate and research in the community, and then finally reached such a stage. .
In the end, whether it can really carry such a transition from PoW to PoS, that is, whether from the community or from various aspects, I think there are still some doubts.
We may talk about it later, including from the perspective of the recent model, from the perspective of incentives, including the fact that after the miners are really removed, he is very important for the long-term healthy development of Eth. , is not a better option, I think this is subject to consideration.
Because from the perspective of the Ethereum community, PoW was first adopted because there was only PoW work, and now PoW is to be removed. It is not a purely technical reason, but more of a reason that is mixed with some political factors. Whether it’s political factors or the choice of the community, in fact, I think from a longer-term perspective, it remains to be seen.
Q7: What are the other considerations besides energy consumption to support POS?
@Dominator008: Compared with the underlying algorithm of the underlying formula, everyone will look at the core team. Core developers including Buterin are not opposed to ETC. In fact, it is more about environmental protection. There are also some proposals related to ESG and blockchain environmental protection in the parliament. ETC has also copied most of the upgrades of ETH at present, but it is uncertain where it will develop in the future, so we can only wait and see.
Q8: Will miners still exist in POS?
@pan: For miners, it is easy to convert ETH1.0 to 2.0, but there will be some adjustments to the architecture of the cloud server.
@starit1992: The earliest POW was a single miner, not the current structure where the mining pool is separated from the miners. This structure emerged because of the emergence of protocols like Stratum to connect miners and mining pools. One of the things that Stratrum does is workload splitting and distribution. The mining pool will split a large workload into multiple small workloads and distribute them to different miners, accept the workload submitted by them, and screen out qualified ones. Submit to the chain. This forms a two-layer structure. The current state of POS is a one-layer structure, and all nodes are solely responsible for producing blocks, but will there be a role like a mining pool-miner in the future, will a POS mining pool appear, and what way will it appear? Looking forward to the question.
Q9: Does POS change much for ordinary users?
@pan: Nothing has changed other than the number and timing of blocks confirmed by exchanges.
Q10: Does POS have any impact on the development language of developers?
@Dominator008: The smart contract layer is exactly the same. The impact on cross-chain is that it was difficult to do cross-chain verification due to the high gas fee before, but after POS, it may bring certain benefits to cross-chain. It is possible that the security can be improved. Time will not have much impact; for the average developer, the impact is relatively small.
Q11: Do you think there are any hidden dangers or risks in this merger, or is it a very smooth state?
@Dominator008: Assuming miners don’t band together to get things done, there should be a relatively smooth transition. At present, many people have been tested, and it seems that the probability of success is still very high. Of course, the exchange will definitely carry out some operations to suspend deposits.
@starit1992: I think there is a high probability that nothing will happen, but if something goes wrong, it will only be a short-term blow to Ethereum, and there will not be much change in the long run. As long as there is no such fundamental level, for example, a consensus algorithm problem, it should not be a problem.
I have a bit of confusion about the POS mechanism. Its finality is not finalized. Under a certain mechanism, it is possible for him to revert finality, but it has to pay a relatively high cost, about one-third of the capital. . There are reasons for this mechanism. For example, if you control more than one-third of the funds, there may be problems with launching certain attacks.
The first point is that I think this protocol is very complicated. The second point is if this mechanism is really triggered, will there be many unexpected situations, especially for bridges, will there be security risks.
@Dominator008: This is a bit like a long-range attack. According to my understanding, if most people don’t recognize it, I’m not sure that this rule will be successfully triggered. If this happens, it must be a big crisis. Regarding the two finalities, everyone is still not ready.
@pan: The test environment is still unable to simulate such a large transaction volume. There are a large number of various forms of smart contract transactions on the mainnet. However, on the testnet, I am not sure whether the test can be so comprehensive, and the complexity is actually quite high. of.
Q12: What will the on-chain state look like when the Merge occurs?
@starit1992: For the mainnet, there is no need to worry about replay attacks, it is the forked chain that needs to worry about this problem. If you want to fork, everyone must follow the chain approved by the Ethereum Foundation. In this case, it is another chain that needs to be worried about replay attacks. In the context of the existence of defi, it is difficult for any forked chain to survive for a long time.
Q13: The next big upgrade after Merge is EIP4844, please introduce EIP4844?
@pan: Let’s talk about history first. Layer2 is a new paradigm, which is to convert some transactions of layer2 into a storage form, and store it on layer1 through call data, so layer1 is the storage place for layer2. Layer2 also does the computation, and state changes happen at layer2. This paradigm is done without any adjustments to Ethereum. Ethereum can only use call data to store transaction state changes of layer2.
Ethereum’s roadmap has this in mind, to design a new model for layer2 rollup transactions. Therefore, under the big framework of danksharding, the Proto Lambda of the OP Labs team and danksharding made a combination, and proposed a scheme called 4844, which is a combination of the names of proto-danksharding.
This scheme turns the original two-layer structure of Ethereum into a three-layer structure. The storage layer is specially taken out, and a new transaction mode is designed called blob transaction. In the 4844, independent of the EVM layer, the transaction or state of the rollup can be stored exclusively. The difference between this layer and the permanent storage of Ethereum in the future is that for rollup, it does not need longer range data, but only needs to ensure the reliability and accuracy of data within a month or a few months. Previous data can be discarded. At least ensure that the data is accurate in the process of state change, but not necessarily guaranteed to be retroactive.
Ethereum specifically took out the storage layer and abstracted it into a 4844 special proposal. All rollup, layer2, EVM or general VM can do more expansion based on the data of this layer of blob transaction. This is 4844.
@todd: I think the advantage of this is that the nodes store less stuff, and it will be relatively easy for everyone to run the nodes. However, it will have an impact on the market status of the node. In some scenarios, if a hack occurs, it is still necessary to be able to check the previous data, but after the Ethereum Danksharding, the data 3 months ago will be thrown away. After the Ethereum upgrade, there will still be It is required to let some nodes do complete storage.
@pan: The good thing about this is that Ethereum is layering data again. Everyone’s understanding of the full node will change. Ethereum may provide data services to third parties in the future. In the future, the index of full data may become a special service. Other full nodes can at least determine that the full node data is accurate.
@todd: After 4844, a new solution will be used to replace call data to store layer2 data, which will be a relatively efficient way. Does that mean that after 4844, the gas of layer2 will be further reduced?
@pan: 4844 will not see any reduction in transaction volume for layer1. Because the scale of layer1 is small, the gas of layer2 will be relatively significantly reduced.
Proto Lambda has a 50-page technical point of view ppt, a comprehensive explanation of the 4844. There is a picture inside that the data layer structure does not belong to the execution layer, but belongs to the consensus layer. So blob transactions are handled at the consensus layer.
@todd: I think it’s hard to say that the first half has been run, 4844 is more of an improvement on rollup, and there is no obvious improvement on layer1. If the original danksharding can be done completely, the efficiency of the entire network will be better, and it will be in a better state.
Q14: Is post EIP-4844 the right time to reduce the complexity of the ETH network?
@pan: The first question to consider is, do all transactions on ETH require global consensus? If it is, the expansion of Layer 1 is definitely the most important thing; but if not, the current Pro-Danksharding plus some off-chain expansion methods, such as AnyTrust and zkporter, are actually enough to meet the needs of most users.
@todd: Insert a question, in AnyTrust’s solution, the practice of transferring the DA layer to the off-chain is actually risky. Because ETH Layer1 needs to conduct global consensus according to the DA layer under the chain, if the DA layer under the chain is dishonest, it will cause risks.
@pan: KZG technology may solve this problem. With the regular deletion of transaction data in Pro-Danksharding, it can have a very large expansion effect.
Q15: Is EIP-4844 the core next upgrade after Merge?
@starit1992: There is also the Shanghai fork.
@pan: In the previous Shanghai fork plan, EIP-4844 was counted as part of it. However, if the development of 4844 is not as fast as expected, the timing of the Shanghai fork may be missed. If the development progress is fast, it may also delay the Shanghai fork and wait for 4844 to go online together.
Q16: How is the Shanghai fork and the launch time of EIP-4844 determined?
@pan: The upgrade of ETH actually has a set of standards. First, the name of the upgrade will be defined, according to the order of Devcon; second, there will be 1-2 hard forks a year (no more). ETH has a large number of EIPs in simultaneous research and development. If an EIP is developed, it will be deployed to the hard fork of the corresponding time window.
Q17: After the Merge, a large amount of pledged ETH may be released, will it have an impact on DeFi?
@pan: There are already many financial derivatives on and off the chain. For large-scale ETH pledgers, it is highly probable that the impact of unlocking has been hedged through financial instruments.
@starit1992: ETH local currency interest rates will rise, which may lead to changes in the market’s risk-free interest rate, which in turn affects the market.
@todd: In the long run, the liquidity of the locked ETH can be released through Lido, so unlocking will not have much impact; in the short term, it is necessary to determine how much ETH is pledged through Lido. If the ratio is high, it will not have much impact on the market. influences.
Q18: Will Lido’s yield increase after this merge? In the past, the fund smashed stETH, which caused the anchor to fall off the anchor. Therefore, there was arbitrage space in the middle, so is stETH worth investing in? Is it safe? (DYOR)
@todd: It is very convenient to withdraw money after opening withdraw. Institutions smashed stETH because it had a liquidity shortage at the time. At that time, the market environment was quite special, and now there is no such market environment after the withdrawal is opened, so stETH and ETH will be equivalent, so there is no arbitrage.
@pan: There may be products like term structures. As the custodian, I hope that it will not appear for a certain period of time now, so it may appear that DeFi zero-coupon bonds may be issued and then traded in the secondary market. For most users, it will be reflected after a certain period of time without affecting their discounted products, which will have a certain opportunity to do it.
Q19 : Regarding MEV, we used to bribe miners to operate on-chain. What will happen to MEV after switching to POS?
@starit1992: Now structures do not form binary structures. Because nodes are randomly selected, there is currently no way to form MEVs.
@todd: MEV will still be there, but how MEV is extracted, or who has certain performance extraction will change. The essential reason for the existence of MEV is that one node has the right to decide the content of the next block. After switching from POW to POS, there may be no economic benefit to owning such a node. The POW architecture can have such a node with the greatest probability. After switching to POS, the latter concept is gone, the nodes may be more dispersed, and the probability that MEV will fall to a specific node becomes lower. All in all, the opportunity for MEV will exist simply by the block producing node, and the transition from POW to POS will not change this logic.
@pan: MEV problem is definitely there. If the short-term or medium-term problem is bigger, and the market mechanism can handle this problem, the node will have additional income. In the long run, everyone wants to solve it, but it must be based on the premise of maintaining consensus.
Q20: If there is a risk in Ethereum merge, what is the worst outcome?
@starit1992: Hope there are no problems. If something goes wrong, the worst-case scenario would be a consensus problem, or a massive node drop, which would have serious consequences. There may be some minor issues, but they won’t affect many users.
Posted by:CoinYuppie，Reprinted with attribution to:https://coinyuppie.com/ethereum-turns-to-pos-what-will-be-the-impact-on-project-parties-miners-and-users/
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.