The Ethereum Name Service has just delivered the biggest upgrade in its history. Notably, he did this on the Ethereum mainnet rather than the L2 he spent years planning.
On Tuesday October 27, ENSv2 has been put online on L1, about nine months after ENS abandoned Namechain, which was the rollup that was originally supposed to host the redesign.
— (@)
The ENS is not alone either. Over the past year, a growing number of applications have left L2s, canceled their own projects or shut down chains they had already built, and then moved their main contracts back to Ethereum L1.
Return to mainnetin other words.
The backdrop to this trend is Ethereum’s gas limit (i.e. the amount of computation contained in each block). double from 30M to 60M during the year 2025, so L1 fees fell at the same time.
Meanwhile, running your own channel has proven costly. Two L2s, Blast and Abstract, announced closures within four days of each other this month alone.
Let’s look at the big picture here, starting with the last domino…
As many of you know, ENS transforms wallet addresses into readable names like william.eth.
ENSv2 arrived as a total rebuild of the protocol. In short, each ENS name can now manage its own ledger for its subnames, while owners benefit from fine-grained permission controls and fees are paid in stablecoins like USDC.
The most interesting thing for me is where v2 is. In May 2024, I wrote ENSv2 101 about ENS Labs’ plan to bring the protocol to an L2, which then took the form of its own rollup, Namechain. At the time, I said that this upgrade felt like “the beginning of the mainstream era of ENS.”
ENSv2 has updated and welcomed its new era, but not on an L2. On February 6, ENS founder Nick Johnson explained the U-turn: “Ethereum L1 is evolving, and it’s evolving faster than almost anyone predicted two years ago.”
Indeed, ENS saw its gas registration costs drop 99% from the previous year as Ethereum grew. Additionally, Nick argued that an L2 would add trust assumptions that L1 does not add.
For his part, Vitalik Buterin sign the same day: “It’s a good decision!”
ENS is the newest name on this growing list of “comebacks,” but it is far from the first.
For his part, Synthetix was one of the first to move. Although it is one of the first protocols to launch on an Ethereum rollup in 2021, it published a paper in June 2025 titled “L2s are dead, long live Ethereum“, citing, among other problems, fragmented liquidity.
As a result, Synthetix began winding down its L2 deployments that summer, and its new perps exchange open in L1 in December. As founder Kain Warwick later put it on: “I spent two years of my life memorizing L2’s narrative. It ultimately didn’t work out well for Synthetix, so I returned to the mainnet.”
Then, in February, came THE job. Vitalik Buterin, creator of Ethereum argued that L2’s progress toward full decentralization had been “much slower and more difficult than initially anticipated” while L1 itself evolved. His conclusion was that “the original vision of L2s and their role in Ethereum no longer makes sense.”
I covered up the post at the time and ENS pulled the plug on Namechain three days later.
— (@)
Do you like this article?
Subscribe to Bankless Or log in
This summer, Super seed became another case study. Its entire design used its own L2’s fees to repay users’ loans. Instead, he deprecated the channel on August 15 and is now in the process of “bringing its entire ecosystem back to the Ethereum Mainnet,” trading on-chain fees for yield strategies at the application layer.
Then, in September, Truethe prediction market protocol behind the DefiLlama markets that I wrote about last week, announced that it was “returning to Ethereum” from Base. Vitalik acclaimed the plan.
This month of October was also marked by a series of related movements:
-
Shape change moved its rFOX staking program from Arbitrum to L1 on October 1st since most FOX already lives on the main network.
-
Blast announcement on October 2, she ceased operations because her costs exceeded her channel’s revenues.
-
Millions took his Blacklight network live on L1 on October 5, and its road map retires his L2.
-
Abstract said on Oct. 6 that it will close its doors on Dec. 15, calling a consumer-only chain “unsustainable as a standalone model.”
Some of Abstract’s NFT projects, like Gigaverse ROMs And BEARISHalso announced their move to L1.
There is a more subtle pattern that characterizes these migrations, namely that applications are increasingly separating where their State is from where their calculations are happening.
Take Synthetix’s new criminals. Orders receive off-chain correspondentbut user funds are kept on Ethereum L1, where transactions are also settled.
Or take Derive V3, which I broke earlier this month. Its order matching remains off-chain for speed, its risk calculations are proven with zero-knowledge proofs, and user funds reside in Ethereum L1 contracts.
ENSv2 also works in this vein. NFT names themselves live on L1, while cheap, gasless subnames can still be resolved from L2s or off-chain via CCIP-Read, a standard that allows a name to point to data stored elsewhere.
In other words, the valuable elements (custody, ownership and settlement) are transferred to L1, while the bulk of the work is directed where it is cheapest and quickest. This is different from the classic rollup model, which contains your state And your calculation, and therefore needs an output when you want to get your “stuff” back.
However, note that there are also technical avenues that will increasingly flatten the distinction between L1 and non-L1.
For example, the Ethereum Economic Zone (EEZ), which I explained in March, is an effort to give Ethereum and its rollups synchronous composability, that is, allowing contracts on different chains to interact within a single transaction.
Rather than removing applications from L2s, the ZEE aims to make rollups appear to be part of L1. Plans are also continuing. GnosisDAO vote in August to transform Gnosis Chain from a standalone L1 into a cumulative ZEE, as I explained during its proposal.
There are also newer, less proven ideas. This week, a researcher going through Namnc launched a “ephemeral accumulation,” Or “L1.5“, on the Eth Research forum, which takes the state-computing split to its logical end.
In this proposed design, L1 holds commitments to users’ state and verifies each proof, users’ wallets hold their own private state, and calculations run as short-lived tasks that retain nothing afterwards. Since an L1.5 contains no state, there is nothing to exit from.
Of course, this migration dynamic is based on Ethereum stay spacious. If applications arrive faster than L1 capacity increases, fees could increase again and change the calculations.
Glamsterdam, Ethereum’s upcoming big upgrade, had its next testnet activation postponed last week, and according to the Ethereum Foundationno mainnet date has been decided.
Yet Glamsterdam is already operating at a gas limit of 200 million on testnet, more than 3 times the current 60 million. Considering what the last doubling did for fees, this bodes well for L1’s margin if the mainnet follows.
Whatever happens, it’s clear that more and more applications are moving their most valuable state back to L1, and ENS is just the latest to make this journey.
Now keep an eye out for the next dominoes to drop. And in the meantime, if you own a .eth name, you can upgrade it to ENSv2 at: app.ens.domains when you want to. It cost me about $1.40 in gas for three names!
About the Author
