Head of research @blockspace_eth | @blockspaceforum | fmr. head of research @ChorusOne | block building enthusiast
Switzerland
Joined September 2022
- Tweets637
- Following215
- Followers815
- Likes2K
Pinned Tweet
Excited to announce that I've been leading research at Blockspace for multiple months now.
More than 90% of blocks are constructed out-of-protocol. Improving this process will benefit Ethereum, originators, validators, builders...
When Ethereum and ETH win, Blockspace wins.
Expect tangible improvements within weeks.
A little data puzzle from MPBC.
While total gas added to blocks appears roughly log-normally distributed with a fat right tail, blobs do not appear to follow a clear distribution.
Why is that?
Let's examine blobs in more detail:
The bucket containing 5-6 blobs is the largest, followed by the buckets containing 3 blobs and 1 blob.
The likely solution:
L2s typically post batches of blobs in fixed sizes that fit these large buckets.
These are then included through MPBC in one go.
For example, the standard OP batcher configuration recommends 6 blobs per TX, corresponding to the largest bucket in our data.
Michael retweeted
When I wrote "Sharding the Builder" in 2022, my naive solution to builder centralization was multiple builders competing to each win a chunk (Near nightshade style) of the execution block. There would be three competitive MEV auctions per block enforcing a gas cap on each, along with one non-competitive censorship resistance chunk. The proposer would aggregate the winning chunks. This was a multi builder design with an inclusion list in it, written before the Merge.
Seeing the concept come to fruition over the years with contributions from many brilliant researchers, developers, operators, and teams is pretty sweet.
It was even sweeter to get to talk to @mostlyblocks about @Blockspace_ETH MPBC design and its growing adoption (Shout out to TOOL and BuilderNet as well!) We spent a good chunk of the episode going over the mechanism and dashboard including a few thought experiments. There is a ton of alpha in this episode if you listen carefully.
Enjoyed the conversation with aPriori!
Wide-ranging from block construction basics, to data on MPBC, down to API details.
Improving Ethereum for users is a total win<>win.
Hope the episode sheds light on how MPBC contributes to this, and where it may go.
What FOCIL does is provide hard guarantees for the transactions the Ethereum protocol sees.
What Multi-Party Block Construction (MPBC) provides is softer fast inclusion guarantees for transactions that are known by any block builder.
Now you have multiple builders contributing to blocks. In particular, the base builder and the contributing builder work with the MPBC operator and proposer to deliver the most valuable block to the Ethereum network.
In the future, both FOCIL and MPBC can continue to scale up and get bigger together, making Ethereum more robust.
In the endgame it maybe that FOCIL includers construct most of the block, with the MPBC pipeline doing the rest. The builder role then becomes an "off-to-the-side function", simply adding a few transactions to capture MEV.
@mostlyblocks from @Blockspace_ETH touches on these topics and more as he eloquently provides a comprehensive first principles overview of MPBC on Ethereum in this quintessential episode of Credible Commitments.
Same reason Quick Slots are S-tier. Faster actions, more information equals better outcomes, larger pie.
MPBC is fast approaching 100k included user transactions.
Inclusion Tracing allows users to easily check how their transactions were handled.
Knowing a transaction's hash and the time it was sent is enough.
Outputs are specific without obscure language.
Making UX better and transparent is really key for Ethereum to succeed!
Stellar weekend for MPBC:
- On Saturday 5.35% of all Ethereum blocks were multi-party blocks.
- New all-time record blocks for highest value uplift , highest gas uplift, and blobs added.
More improvements coming shortly that will improve performance and UX further.
Bid streams allow proposers to receive multiple fast bids per block.
This is a notable improvement on the round-trip request proposers currently send to get bids.
As a result bids are delivered more reliably, which means fewer missed slots & better blocks. It's "Kaizen"!
Today 3.5% of all Ethereum blocks are multi-party blocks.
This is up from around 2% when MPBC was announced.
The average number of builders per multi-party block is 2.63 today.
This is the highest to date, and up from the 2.42 previous high I mentioned last Friday.
There are also three new blocks by gas added on the all-time leaderboard.
This morning, Aestus delivered blocks adding up to 25.8m gas or an additional 1190%.
This came from contributions by the Bombora, BTCS, and Quasar builders.
The number of builders per multi-party block has been trending up nicely this week.
It's now reached its high to date, with an average of 2.42 builders contributing to multi-party blocks.
This is a good start, but it will go higher.
First multi-party block with contributions from 5 builders landed.
I expect blocks to hit 5+ contributing builders regularly, very soon.
More builders means less subjective discretion and more inclusion lanes.
The net result for users is even faster, more predictable inclusion
@VitalikButerin calls it a "party", we call it a "participant." Either way, we recently saw MPBC produce a block with 5 participants contributing. We like how this party looks! 🎉🍾🥳
Highly recommended!
Agree that a property of the future Ethereum will be:
> "multiple forms of multi-participant block construction[...]"
FOCIL and MPBC v1 are the first step towards this. They are complementary.
Agree that there will be a strong non-chain layer between users and Ethereum.
There are many interests that go into a block to make it valuable.
This layer will rest on coordinating these efficiently (remove all friction). Maximum "information" in every block.
This is one ambition for MPBC I personally have.
Separately, great to see "4-8s slot times" in the post.
Now is an opportune time to start building this with Quick Slots.
The cryptographic world computer:
vitalik.eth.limo/general/202…
My attempt to express in somewhat concise terms the true meaning of basically everything planned to happen to Ethereum starting from the fork after Hegota. It's really not just a blockchain anymore. It's a hybrid architecture that combines together blockchains and modern cryptography, to enable much more powerful properties.
FOCIL, EIP-8288, Lean consensus, state management, formal verification, advanced mempool improvements (including privacy), and the longer-term specter of obfuscation all mentioned.
Many such reports from searchers landing faster through multi-party blocks.
Most transactions added to blocks in MPBC v1 are normal user activity like transfers and trades.
But self-contained bundles (e.g. backruns) can also land.
Two out of the three top multi-party blocks by value uplift landed in the last 2 days.
Don't expect these to stay on the leaderboard for long.
New significant performance improvements are already in testing.
These will allow MPBC to support even larger blocks.
Here's the multi-party block with the highest gas uplift to date.
MPBC increased gas usage by 1346%, adding 24.5M gas.
It's Slot: 15213880
The highest number of transactions added to date is 810.
Assuming users would have waited ~12s for the next slot, it saved them 2.7h.
You can find it on the dashboard (Slot: 15180396).
All top blocks by gas added are by newer and smaller base builders.
This includes 24.5m gas added to one block (+1346%)!
MPBC rewards providing faster and more predictable inclusion to users.
Blocks from smaller builders typically have comparatively more space available.
Check out the MPBC dashboard to view every single MPBC block.
Multi-party block construction could make smaller builders more competitive.
Builders earn when they contribute to other's blocks, or when others contribute to their blocks.
-> The less flow a builder has, the higher the relative value that can be added to its blocks.
-> The less likely a builder is to win a slot, the higher the relative revenue from adding TXs to others' blocks.
Curious how this will reflect on bids.
Quick Slots are S-Tier.
We contributed to this because the best way to improve Ethereum is to make transactions cheap, fast, and reliable.
This depends on block size, slot time, and the block construction pipeline.
With Quick Slots, MPBC, and gas limit increases we're taking strides towards a better Ethereum.
Our more detailed take: xitter.cf/Blockspace_ETH/status/…
We want Ethereum to produce the highest grade blockspace. One requirement is providing end consumers predictable inclusion. Users should know when their transaction will be included in a block and what it will cost.
Quick Slots are a significant leap in this direction and something we strongly support. With this upgrade, your transaction is either included faster or given another opportunity faster.
Since blocks will have less time to reach the network, we also think Quick Slots should arrive with the performance engineering EIPs such as 8131, 8279, and 8334 provide to help keep the network at peak performance.
We are excited to support @ethlabs_org and others in this push! Read more on their post below.
Multi-Party Block Construction (MPBC) is live!
Over the past 24h, 3% of Ethereum blocks were multi-party blocks.
Since August 28th, >20k transactions have been included faster through MPBC.
MPBC means larger & more robust blocks, faster and more predictable inclusion.
Highlights to date include blocks that gained 810 additional transactions, or almost 14x the gas.
MPBC comes with comprehensive docs and metrics. You can check every multi-party block out on our dashboard.
MPBC will continue to improve and scale.
Looking forward to taking the next step together with the many teams that make up the Blockspace Forum.
Let's make transaction inclusion fast and predictable, and Ethereum the best it can be.
