239632
bertcmiller

@bertcmiller #239632

⚡️🤖
251 Follower 43 Following
Since it is a slow news day... excited to share a new building architecture: parallel block building

https://writings.flashbots.net/parallel-builder
Brief technical writeup of a proof-of-concept builder implementing synchronous interop between multiple OP-Stack rollups 🦾


https://collective.flashbots.net/t/crucible-hackathon-opstack-interop-builder/3572
The rollup fee dilemma: lower your fees and get spammed by bots, raise them to deal with bots and price out users
Flashbots Protect has leveled up:
- Fast mode: wicked fast inclusion speeds
- Upgraded backend for best in class reliability and speed
- Etherscan integration
- Beautiful and simple UI to get started

Add to your wallet on mainnet:

https://protect.flashbots.net
Been years since I've made an MEV searcher but got to scratch that itch lately with... a Resy reservation bot. Surprisingly competitive?
One of the nice things that you get from using a TEE builder is fast, credible pre-confs. Could bring OP-Stack "block times" as felt by users down a lot.
Was on a podcast recently talking MEV, rollups, and what isn't being paid enough attention to in Ethereum research! Fun conversation, give it a listen:
https://open.spotify.com/episode/0expIwKw9P23RmS6tXwNE9
What is the best way to level up from being "good enough" to "great" at Rust? Have written a bunch of code through a variety of projects and touched most parts of Rust by now. Obviously past the tutorial phase.
Couple of books gifted by the parents for my birthday. Excited for this stack, going to start with Natural Goodness - a “reply to Nietzsche.”
Enjoyed "Children of Time" - a long ranging, imaginative sci-fi novel. I was looking for a followup after reading The Three Body Problem trilogy and settled on this after also reading Hail Mary. I liked Three Body more, that was more "philosophy" and less "fantasy." But it was still enjoyable nonetheless.
Wrote a review of The End of History, lamenting our loss of History itself: https://bertcmiller.com/the_end_of_history.html
https://gist.github.com/karalabe/4313ece67b528113ea72388ac2df25d8

Intrigued by this new CR proposal from Péter Szilágyi. First takes are:
1. I am worried that tying the fork choice rule to the mempool will not be a good idea. Mempools are brittle.
2. This lowers the value of blocks by adding constraints for builders
Two big trends for this cycle:
1. Scaling: increasing the overall economic and compute throughput of chains (see: rollups, reth, parallel EVM, etc)
2. Expressivity: increasing the kinds of applications that can be built without trust (see: SUAVE, TEE/ZKP coprocessors, restaking, etc)
Tinkered with the Base p2p layer and found:
* One bot is crawling the network hard. Several fake nodes all named "nodecrawler" from the same IP but different ports that repeatedly connect and disconnect to my node.
* It is pretty hard to find real nodes
* A few nodes are sharing their mempool but it's all old txs
Two surprising things about MEV on rollups:
1. Just how much $ is being left on the table by sequencers today by not implementing some kind of MEV auction
2. Just how much of their txs are actually bots spamming the network trying to capture arbs
Been trying to sync a Base op-reth node for awhile. Here's the config that, after much tinkering, works:
1. standard L1 reth on beta v6 release
2. standard L1 lighthouse
3. op-reth pointed to base with discv5 discovery syncing with --debug.tip
Low fees and MEV = spam

Spam = wasted block space, higher fees, and higher costs to run a node

You need to address MEV to scale a chain

/eth /base /mev