London: Here’s what’ll happen at Block #12965000
I’m not going into details about the EIPs, but just wanted to summarize what will happen in 8 hours’ time. I had no plans to write this post, as London has been covered very well, but browsing through the Daily I see there are still some unanswered questions.
Firstly, a common misconception is that miners run the network. They do not — they just provide a service. It’s the users running full nodes — exchanges, wallets, infra providers, end users transacting, developers etc. that are in charge. Given there’s a near universal consensus among these users, there’s zero need to worry about miners. Even if there was a majority (>51%) of miners declining London and choosing to fork, it doesn’t matter — they’ll be stuck on a fork that Ethereum users will simply ignore.
At block #12965000, the difficulty bomb will be delayed till December 2021. Some may have noticed an
mild uptick in block times lately — this will reset back to targeting low 13s seconds. Given the same target gas limits, we should see a minor (~5%) bump in TPS. Gas tokens like CHI will be deemed worthless. Contracts starting with the 0xEF byte will be rejected.
As for EIP-1559, once again, I’m assuming you understand the base fee and priority fee model.
At block #12965000, base fee will reset to 1 gwei. The block gas limits will temporarily double for the next few minutes, till base fee finds an equilibrium. Given recent gas prices, it’d be somewhere in the 20–50 gwei region. By my estimation, it’ll take somewhere between 30 and 35 blocks, or 6 to 8 minutes. During this time, the network will be running at its highest throughput ever — 110 TPS for ETH transfers; 30 TPS for more complex transactions. This does not mean you can lowball 1 or 2 gwei and hope to get included — it’ll just mean that the priority fees for the first few minutes will be very high.
At launch, MyCrypto.com will support 1559-style transaction. MetaMask has made incredible progress in the last couple of days, but it’s unlikely they’ll have a new release supporting 1559 in the next 8 hours. You can still make transactions as usual on wallets that do not support 1559 transactions — they’ll just be legacy transactions. The downside with legacy transactions is that if you overbid, you’ll still be overpaying. However, you can mitigate this by following what the base fee is, offering a minimal overhead on top to account for priority fees (usually 1–2 gwei should suffice, but we’ll need to wait and see what the typical priority fee is like) and manually entering this as your bid. For example, if base fee is 30 gwei and the typical priority fee is 3 gwei, you can just manually enter 33 gwei and there’s a very good chance you’ll get included. Of course, we should see all wallets support 1559-style transactions over the coming days and weeks — once that happens the wallet will do all of this for you, and you just need to accept their given estimate (advanced users can edit in their custom priority fee and max fee). In most cases, your transactions will confirm within an average of ~7 seconds. Irrespective of transactions being 1559 or legacy, base fees will always be burned.
Etherscan will support 1559, and there are other sites that’ll let you follow the burn. Miners will need to double their block gas limit variable, which we’ll assume they will, so users will not be affected.
The target gas price for deflationary ETH is approximately 190 gwei. Given current average gas price has been hovering around 30 gwei, deflation is very unlikely any time soon, though there’ll be a not insignificant reduction in net inflation. Post-Merge the target is ~20 gwei, so that’s when we might see deflation kick in. By the way, the assumption around these estimates is that the base fee : priority fee ratio will be 75:25, but with 1559 on mainnet we’ll finally get more representative figures. I’ll make an updated post around gas price targets once we have a better data.
Feel free to ask if you have any further doubts, I’ll answer them to the best of my knowledge.