{"id":12434,"date":"2026-08-27T02:56:15","date_gmt":"2026-08-27T02:56:15","guid":{"rendered":"https:\/\/cryptonews.uk.com\/?p=12434"},"modified":"2026-08-27T02:56:15","modified_gmt":"2026-08-27T02:56:15","slug":"solana-takes-its-first-step-toward-sub-second-speed-by-cutting-block-confirmation-times-across-the-network","status":"publish","type":"post","link":"https:\/\/cryptonews.uk.com\/?p=12434","title":{"rendered":"Solana takes its first step toward sub-second speed by cutting block confirmation times across the network"},"content":{"rendered":"

<\/p>\n

\n

Solana\u2019s mainnet is producing blocks faster after its first staged slot-time reduction moved the network\u2019s target from 400 milliseconds to 350 milliseconds.<\/p>\n

Faster slots shorten block-level feedback and confirmation thresholds measured in slots, while throughput depends on a separate set of limits.<\/p>\n

The Solana Foundation confirmed the mainnet change after the feature gate activated at slot 440,208,000, the first slot of epoch 1019. A feature gate is the switch validators use to coordinate a protocol change, and under the one-epoch delay required by SIMD-0525, the new timing applied when epoch 1020 began on Aug. 21.<\/p>\n

A slot is the window in which a designated validator can produce a block, while an epoch is a fixed period of 432,000 slots.<\/p>\n

Trillium, a Solana validator-telemetry provider, measured a slot-weighted mean of 365.4ms across 431,505 timed slots in post-change epoch 1021. Its view of pre-change epoch 1015 recorded a 420.7ms mean.<\/p>\n

The same dataset recorded 331 skipped slots in epoch 1021, or 0.077%, compared with 1,890 skips and 0.438% in epoch 1015. The lower post-change reading offers an early stability signal across those two epochs, though the comparison cannot establish that the timing cut caused the change.<\/p>\n

What faster slots change<\/h2>\n

Shortening the slot window reduces the wall-clock time for confirmation thresholds measured in slots. It also cuts the four-slot leader window from a nominal 1.6 seconds at 400ms to 1.4 seconds at 350ms, narrowing the period one block producer controls.<\/p>\n

The proposal keeps four slots per leader and 432,000 slots per epoch. It scales per-slot compute, account-write, vote, data, and shred limits down with each shorter target. Blocks arrive more frequently and carry smaller budgets, leaving approximate work capacity per second broadly unchanged.<\/p>\n