Ethereum founder Vitalik Buterin supports the controversial switch from Casper FFG to Minimmit and believes that censorship is more difficult than maintaining textbook error tolerance as ETH trades around $2,000.
Conclusion
- Vitalik suggests replacing Casper FFG’s two-round Ethereum final gadget with Minimmit, which completes blocks in one round.
- Trade-off: fault tolerance decreases from 33% to 17%, but censorship resistance and recovery from errors or attacks are likely improved.
- The debate comes as ETH hovers around $2,000 and markets are weighing whether a faster, more stable end could justify a premium on the macro-side tape.
Vitalik Buterin has thrown his weight behind one of the most sensitive changes to the Ethereum (ETH) core: ditching the Casper FFG ultimate gadget and replacing it with Minimmit, a single-cycle Byzantine fault-tolerant scheme that deliberately relaxes the pure framework in exchange for an otherworldly theory.
Casper today requires validators to confirm twice – once to validate a block and again to complete it – and tolerates up to 33% stake before breaking the system’s guarantees. Reduce it to a single round: faster and simpler, but with a formal tolerance for errors in the proposed parameters that is reduced to 17%.
On paper, this looks like a letdown. But Buterin’s thread makes a crude point: the world’s worst real attack isn’t ultimate reversal, it’s censorship. The ultimate reversal creates irrefutable cryptographic proofs and leads to massive reductions – millions of ETH or billions of dollars – evaporated on-chain, making such attacks economically senseless for any rational actor with such capital. Censorship, by contrast, is messy: it forces users and developers into social coordination, soft wars, and political struggles. In both the “ideal” three-slot-final (3SF) model and the Minimmit, the attacker requires a 50% stake to censor, but the Minimmit changes the thresholds at which an attacker can unilaterally complete a bad history, raising the threshold from 67% to 83%. According to Buterin, this maximizes scenarios where the network ends up with “two chains of two chains” instead of “something done wrong”—an outcome that is messy but fixable.
The background is a market that no longer only pays for stories. ETH is around $2,000, down from previous session highs around $4,900, with high volatility and macro headwinds still in play. Traders have already seen the outline of Ethereum’s “L1 speed” map, which aims to reduce slot times from 12 seconds to 2 seconds and final completion to single-digit seconds using Minimmit. If this redesign works, Ethereum will stop competing solely in the DeFi aggregation and payment ecosystem and start competing in something much simpler: how quickly and reliably your transaction becomes irreversible. In a market where ETH is still re-pricing its role against rival L2s and L1s, Minimmit isn’t just a deal; it’s an attempt to re-anchor asset value in a raw, observable user experience: click, confirm, complete.
