Solana (SOL) Alpenglow Migration Orders Explorers to Reset TPS Baselines at Epoch 1064

Update 05:57 UTC: New details from the handover show how quickly the switch itself ran: the Solana (SOL) Foundation put the transition between old and new consensus at about 900 milliseconds, with votes from 446 of the network's 654 participating validators certifying the block used as the starting point.

Vote Transactions Leave Solana’s Blocks

Explorers, analytics services and node operators are the parties still facing work before Solana (SOL) completes its Alpenglow consensus migration, scheduled for Oct. 21. The Solana Foundation’s guidance tells every service that counts activity on the Solana network to reset its transaction-count and TPS baselines, because validator vote transactions, the byproduct of staking , will vanish from blocks once the handoff completes. A total that includes votes will fall even if payments, trades and DeFi activity stay unchanged. Services that track validator participation face a second shift: they must stop reading Vote program instructions and read certificate data from block footers instead, and the Foundation notes that Yellowstone’s gRPC footer message does not yet forward those certificates. Anza’s operator documentation adds that Agave 4.3 ships an optional block-footer notification, which a service has to enable and implement if it needs the data. An earlier developer note also told explorers to keep competing candidate blocks separate until confirmation, retaining the selected block rather than merging their transactions. Where the Solana price trades this week is secondary to the mechanics: the Foundation identifies epoch 1064 as the trigger and estimates the start at roughly 13:20 UTC on Oct. 21. Anza separates feature activation from the consensus handoff, with the migration boundary landing 5,000 slots after activation, a window in which blocks temporarily carry only votes while validators finish the transition. Solana (SOL) co-founder Anatoly Yakovenko used the moment to argue the upgrade took longer than necessary, writing in an Oct. 21 post on X that “Alpenglow would have shipped in half the time if they just changed the vote program to the ag state machine.” His remark quoted a comment celebrating the removal of on-chain votes and mocking the long-running criticism that most of the network’s transactions per second are validator votes rather than user traffic. Developers saw the changed behavior when Alpenglow reached devnet in September, after the earlier testnet transition.

@toly · X post

Post on X.

View on X

A Burned 0.8 SOL Fee Per Epoch

The fee side is set out in the Foundation’s documentation: once on-chain votes disappear, the fees paid on individual vote transactions end with them. Validators instead pay a flat 0.8 Solana (SOL) admission charge per epoch, and the entire amount is burned. The payment comes from the vote account, which replaces the identity account as the source of voting-related costs, and operators must keep the charge above the account’s rent-exempt minimum. Billing runs one epoch ahead: the first charge, due at the start of epoch 1064, covers participation in epoch 1065, and a validator that cannot pay loses permission to vote or produce blocks in the following epoch. Preparatory pieces are already live, with BLS public-key registration available on mainnet since July 8 and the admission feature activating July 22, although neither switched on Alpenglow consensus itself. The Foundation’s checklist requires version 4.3.2 of Agave, Jito-Solana or Harmonic Salsa, and states that Firedancer and Frankendancer are not supported during the handoff. Under the hood, Anza’s published design replaces TowerBFT, the system validators have used to agree on blocks, with Votor, which runs two voting paths at once: one finalizes a block in a single round once 80% of stake responds, while the other uses two rounds at 60% thresholds. Votes become direct messages signed with BLS cryptography and combine into compact certificates, and validators use their own timers to flag an unresponsive leader and vote to skip it. The published SIMD-0326 proposal caps the admitted set at 2,000 validators, prioritizing the highest-staked operators, so certificates stay small enough for a single UDP message. Governance cleared the way in September 2025, when the proposal passed with 98.27% support among votes cast and participating stake represented 52% of the total. Anza’s operator documentation also qualifies the roughly 150-millisecond finality target: nodes outside the admitted validator set learn finality from certificates included in a subsequent block, which can add up to one produced-block interval, with longer waits when blocks are skipped or delayed. An admitted validator can send consensus messages directly to a partner node, leaving latency between the partners as the main added delay.

The mechanical effect is the part to price correctly. Removing votes cuts the altcoin’s headline throughput figure without any change in what users actually do; our records put non-vote activity at 8.92 billion transactions in Q3 2026, a baseline that will now sit beside a visibly lower gross number. Cross-chain comparisons will need re-basing once the migration settles, and analytics firms that have not enabled Agave 4.3’s footer notifications will under-report validator participation. The party that has not moved yet is the Firedancer and Frankendancer cohort, which cannot join the handoff at all; until support ships, the transition runs on Agave, Jito-Solana and Harmonic Salsa only.

The migration crossed its threshold on schedule: Alpenglow is now live on Solana mainnet, with the official Solana account announcing the switch at 13:59 UTC on Oct. 21 and Anza confirming it five minutes earlier. On-chain records show the last TowerBFT-era block at slot 459,652,999 around 13:54 UTC, followed by eight empty slots before block production resumed at slot 459,653,008, with no rollback. The handover landed ahead of both Anza's original estimate of roughly 14:50 UTC and the revised projection of about 14:00 UTC. Both teams put finality at roughly 150 milliseconds, down from a stated 12.8 seconds, though the Alpenglow dashboard showed 6.6-second finality and 4,721 TPS at launch. Wednesday's activation covers Votor only; Rotor, which would replace Turbine, has no date. Solana traded near $108.40, up about 3.11% in the 24 hours before the switch.

New details from the handover show how quickly the switch itself ran: the Solana Foundation put the transition between old and new consensus at about 900 milliseconds, with votes from 446 of the network's 654 participating validators certifying the block used as the starting point. Anza also noted that some computers receive finality information later than the validators voting directly, meaning the speed a user sees depends partly on the service handling the payment. The roughly 150-millisecond settlement figure stands as a target under normal network conditions rather than a guaranteed time for every payment. For users, wallets, fees and application behavior stay unchanged, and existing wallets continue to work; the visible difference comes from apps being able to confirm settled payments, such as crediting an exchange deposit or releasing goods, with stronger assurance.