eth2 update 018 Special Edition: Everything you need to know about the Merge and tentative Mainnet Merge date announced
August 12, 2022
Heyo! It’s Ben or b17z (bits) here again with a special edition of the eth2 updates. It’s special in the sense that it’s going to be longer… but we’re going to be talking about fun stuff such as:
What the Merge does and doesn’t do
Ingredients of what needs to happen leading up to the Merge
How rewards change right after the Merge
But it will also contain some of the same info like in other updates:
Goerli merged with some issues
Growing DAG Size concern on Mainnet
Sepolia testnet post-Merge upgrade details
Beacon Chain Traction
412,534 Active Validators
13,200,976 Staked ETH, or about 10-11% of the total supply, worth roughly $22-23 billion as of Wednesday, August 10, 2022
Average Validator Balance of 33.73 ETH
Average Daily Validator Income for August 9: 0.003915 ETH
Total Validator Income for July 17: 1616.88 ETH
Goerli Merged…Successfully…on Aug 10 at around 9:45 PM EST
But it wasn’t without any issues, which you can conveniently track using this issue tracker.
So, what were the issues?
Lodestar had a configuration issue where some of the nodes didn’t have an EL configured.
Nimbus also had a configuration issue where some of the nodes were using an older version of geth that broke their setup with the Engine API.
Ben’s sidebar: I’ve noted before that there needs to be better awareness and tighter processes around configurations.
The main one was that Besu experienced a very large 30 slot reorg. What happened was that Besu incorrectly invalidated the canonical chain transition block. The Teku node then proposed its own head, which was to be expected. A Teku restart fixed the issue. Teku re-issued all its requests and Besu responded correctly allowing Teku to reorg onto the correct chain.
However, this didn’t stop them from tentatively announcing the Mainnet Merge date – currently September 15/16. You’re going to see me link that over and over in here because it’s such a big deal. It’s unofficially officially on the calendar!
Ben’s sidebar: I know all those issues might be cause for concern. I do want y’all to keep in mind that this is all very technically complex. Bugs will happen; issues will happen. They were confident enough to put a Mainnet Merge data (tentative) on the calendar so I don’t expect this to happen on Mainnet, but if the worst comes to pass, we can resort to social/community recovery, which I do think is a value prop.
Concerns for Growing DAG Size on Mainnet
First of all, what is the DAG size? It dictates hardware requirements for mining. When it exceeds certain thresholds, that mining equipment is no longer usable on Ethereum.
What is the concern? Ethereum is scheduled to exceed the 5GB threshold in about two weeks. There is concern that there might be a large drop in hash rate if it is exceeded because it’s a popular ASIC configuration. This brought into question how and when to schedule the TTD for Mainnet.
However, this also didn’t stop them from tentatively announcing the Mainnet Merge date – currently September 15/16. They believe the odds of a big drop in hash rate is low.
Sepolia Testnet post-Merge Upgrade
The Ethereum Foundation announced that Sepolia will undergo a post-merge Execution Layer upgrade on August 17. This upgrade will cause EL clients on the network to disconnect from peers which have not transitioned to Proof of Stake.
This upgrade will happen on Goerli and Mainnet as well.
What the Merge Does and Doesn't Do
What The Merge Doesn’t Do
Reduce gas fees
The Merge is just a change from a Proof of Work consensus mechanism to a Proof of Stake consensus mechanism. The Merge does not expand network capacity. As blockspace demand grows, so do gas fees. Same old; same old.
Increase transaction speeds
It is true that block/slot time is 12 seconds down from 13.6 seconds (if that’s how you measure speed) but that’s about it. That’s about a 10% difference in frequency of block production. This is not significant enough for users to notice.
But if the Merge won’t reduce gas fees and it won’t increase transaction speeds, what’s the point of all this? Don’t worry dear reader, we will get into this in the section after this on what the Merge actually does.
Give you the ability to withdraw staked ETH which would lead to a mass exodus event
The Merge doesn’t enable withdrawals. This is actually reserved for the Shanghai upgrade after the Merge.
In addition, you might be thinking that there might be one big mass exodus event after withdrawals are enabled. That would also not happen because withdrawals would be rate limited. Only 6 validators may exit per epoch. An epoch is 6.4 minutes. So some quick math says that would be around 1350 validators per day max.
But if it’s not one mass exodus event, wouldn’t it just delay the inevitable and become a cascade of the maximum number of validators to leave per epoch? Since we don’t know how many validators will be around at the Merge as that number is still growing, we will use the number of validators as of this writing on August 10, which is 412,534. So some quick math would say that would be around 306 days until all the validators exit Ethereum if the Merge would happen today.
I caution everyone to think about how unlikely an event like that would happen. It’s all about incentives. First, the APR of the protocol is dynamic. The APR is a function of how much is staked on the network. The more percent of the total circulating supply is staked, the lower the APR and the less, the higher. So, as the days go by with this theoretical cascading event, the APR would go up leading to an increased incentive to stake onto the network. That’s the beauty of having a market of stakers balancing the pros and cons of staking.
What The Merge Does
Moves from Proof of Work to Proof of Stake
Yeah, yeah, this has been said a bunch of times but this move actually sets the stage and foundation for continued protocol work to scale Ethereum namely danksharding. I wrote a tweet thread about this based on Delph Digital’s Hitchhiker’s Guide to Ethereum.
This provides a scalable base layer for Ethereum’s rollup-centric roadmap.
Ben’s sidebar: There is a lot of working being done on Ethereum currently that it’s too much to write in one update. But we will continue to explore these different topics as time progresses. For now, you can check out those resources as they’re absolutely phenomenal. I love discussing these things so feel free to DM me on Twitter as well.
In addition, Proof of Stake comes with a lot of benefits.
Makes Ethereum more energy efficient
It’s much more energy efficient – 99.95% more. This is a pretty big value prop. Because PoS secures a blockchain through capital not energy, the energy usage would be comparable to running a typical computer.
Ben’s sidebar: There are a lot of controversies around how much energy Proof of Work mining uses. Not to use “whataboutism” to take away from the real fact about PoW energy usage, but stuff like machine learning and the massive data centers we take for granted everyday use a lot of energy as well. So, I offer a question for you to answer for yourselves regarding energy usage of technology: “Is the energy usage worth the value prop the technology is offering?” So, in the case of PoW mining – “Is its energy usage worth it to secure a network, and worth it for decentralized, global, permissionless money and its benefits?”
Lowers barrier to entry to participate in the network
It lowers the barrier to entry such as reduced hardware requirements. PoW mining is accessible in theory, but often not in practice, due to the immense startup costs, difficult hardware acquisition, and reliance on incredibly cheap electricity. Transitioning to PoS lowers the barrier to entry for an individual to participate in consensus and securing the chain. If you want to run a validator on PoS Ethereum, it requires a startup cost of 32 ETH, which at the time of this writing is $54,000. No small feat, of course, but there are projects like RocketPool and Lido that help smaller token holders participate.
Because the barrier of entry is reduced, it opens the door to reduced centralization. As mentioned above in the Beacon Chain Stats, there are over 400,000 active validators and according to Glassnode, there are 118,150 addresses with over 32 ETH in them.
Ben’s sidebar: I highly suggest you read Vitalik’s Endgame and his standards on what is sufficient trustlessness and censorship resistance. In short, he believes that users are the decentralization (I mention this in the tweet thread above) to which I agree. Empowering the everyday user is the decentralization. One of the biggest things that I always think about as we try to scale this space to the next billion users is the amount of data that will be generated and consumed, especially related to block production/building. Block building is not going to be accessible for the everyday user. It requires specialized hardware, which is kind of what we’re trying to get away from, right? Well, yes and no. The question is how do we empower everyday users that can’t participate in block building because of the high barrier of entry? We give them the power to validate the block building. We will talk more about this in future editions.
Reduces ETH Issuance
The supply of ETH can be broken down into issuance and burn.
The issuance of ETH is the process of creating ETH that did not previously exist. The burning of ETH is when existing ETH gets destroyed, removing it from circulation. The rate of issuance and burning gets calculated on several parameters, and the balance between them determines the resulting inflation/deflation rate of ether.
Ethereum is currently incentivizing both miners (under PoW) and validators (under PoS). For mining, it’s around 13,000 ETH per day pre-Merge. For staking, it’s around 1,600 per day pre-Merge (source: Ethereum.org). So issuance is coming from two places.
After the Merge, since miners will not exist anymore, rewards to miners will also cease which will reduce ETH’s issuance rate by around 90%
The reason issuance can be reduced is that PoS makes it easier to secure a network than PoW. With POW, Ethereum needs to issue enough to cover miner costs such as electricity and mining rigs. With PoS, it’s just the opportunity cost of capital.
Demand for ETH staking is expected to increase because:
Staking rewards for validators are expected to increase since they will receive transaction tips that are currently earned by PoW
Validators may earn MEV via Flashbots/mev-boost
But remember from earlier: Staking APR is a function of how much percentage of the current total supply is being staked. The general expectation is that the more that is staked, the less the APR. High rates of APR will not be sustainable as more and more stakers come into the network. It’s up to the market of stakers to stake and unstake as they feel aligns with their incentives.
Burn + EIP-1559
Now, let’s talk about burn and EIP-1559. For a transaction to execute on Ethereum, a minimum fee (known as a base fee) must be paid, which fluctuates continuously depending on network activity. The fee is paid in ETH and is required for the transaction to be considered valid. This fee gets burned during the transaction process, removing it from circulation.
Before EIP-1559, the fee market mechanism was a first-price auction where people bid a set amount to pay for their transactions to be processed where the highest bidder wins. With EIP-1559, it introduced a “base fee” for transactions to be included in the next block. For users or applications that want to prioritize their transaction, they can add a tip or a “priority fee.”
Combining all the above, depending upon network activity, the rate of ETH being burned may become higher than the rate of ETH being issued making the supply of ETH decrease effectively making ETH deflationary.
For a fun resource, check out ultrasound.money
When mev-boost is available, validators can choose to run it alongside their validator software. Mev-boost is an implementation of Proposer-Builder Separation (see above resources, such as the tweet or Delphi Digital’s guide) where block production is outsourced to specialized block builders that bid for validator blockspace. According to Flasbots research, validator yields could increase up to 60% (this assumes 8M ETH staked – we’re at 13M+ right now).
Creates division in logic division in the protocol
Instead of one client that runs both consensus and execution, there will be two clients dedicated to each one where the execution layer processes transactions in a block and maintains world state of contract and account balances and the consensus layer chooses which blocks should and should not be included in the canonical chain.
However, since there is a division in logic, there needs to be a way to facilitate communication between the two layers. That’s what the Engine API is for and it requires a secret that is shared only between two clients to authenticate one another. This is a great segue into the “Ingredients of the Merge."
The Ingredients of the Merge
Playing the Merge out
First – the Bellatrix Hard Fork
This is a Consensus Layer upgrade to start to accommodate blocks containing transactions from the Execution Layer. However, unless the TTD is reached on the Execution Layer, it will not let the Consensus Layer start filling up Beacon blocks with transactions from the Execution Layer. Once the hard fork is completed, the Consensus Layer nodes will begin listening for a TTD via the Engine API. The Bellatrix Hard Fork is tentatively scheduled around Tuesday, September 6th, at epoch 144896 at around 7:34 PM EST.
Second – reaching Total Terminal Difficulty (TTD)
TTD is the accumulated mining difficulty of all blocks that have been processed by the network. All the testnets have forked using this instead of using block number. Why TTD? A predefined block number for the fork is unsafe. A potential attack vector may use minority hashpower to build a malicious fork chain that would satisfy the block height requirement and the first Proof of Stake block may be proposed from the malicious fork. For more information on this, check out EIP-3675. The tentative TTD is 58750000000000000000000, which as mentioned a few times above, targets September 15/16.
Third – the Paris Hard Fork
This is an upgrade for the Execution Layer nodes. This upgrade contains a few major elements:
EIP-3675, which removes the dependency on ethash, the Proof of Work mining algorithm Ethereum has used.
EIP-4399, which changes how randomness is generated on-chain by replacing the `DIFFICULTY` opcode with the `PREVRANDO` opcode. This would be useful for Dapps to generate random integers since it will be accessible in the EVM.
All of these things plus more will be activated once the TTD is reached. And once it’s reached, it’s party time! Don’t forget to check out some of these resources for Merge!
Finally – the Difficulty Bomb
Since Ethereum is moving to Proof of Stake, a potential scenario that might pop up is that some miners might reject it. The difficulty bomb was introduced to make it increasingly hard and eventually impossible for miners to try to keep Proof of Work alive after the Mainnet Merge happens. We don’t want that to happen until the Mainnet Merge actually happens.
Back in June, it was decided that to delay the difficulty bomb until mid September 2022 and the Gray Glacier Upgrade on June 29 put that into play. With the tentative announcement of the mainnet Merge happening around September 15-16, it lines up perfectly.
It’s all coming up aces y’all! However, I do want to point out that until client upgrades get released, everything is still tentative. But folks were confident enough to put actual dates on things which is always a great sign. We’re witnessing history in the making and I’m happy to be able to spend it together with y’all. By the time Devcon rolls around (which I will be at, so if you see me, say hi!), we will be living in an Ethereum Proof of Stake world.
I’ll talk to y’all soon!