Solana Transaction V1 is live on Testnet and Devnet. It raises the transaction size limit to 4,096 bytes but drops Address Lookup Tables. Here’s why and when to use V1 vs V0 + ALTs. cc @solana_devs

If you want it done right, build it yourself. 5M downloads later @zbr0wn and @anza_xyz keep @solana and the broader Rust ecosystem winning. 💪

What Transaction V1 unlocks: a 4096 byte tx limit, up from 1232. 🔹 Confidential-balance ZK proofs 🔹 Untruncated Winternitz signatures 🔹Nested Squads multisigs 🔹BLS verification without precompiles All in one tx, with protocol-level atomicity no Jito bundle workaround.

Solana slot time: 400 → 350 → 300 → 250 → 200ms. The blockhash validity window is ~150 slots: 60s at 400ms, 45s at 300ms, 30s at 200ms. Audit hardcoded timeouts, retries, “wait N slots,” cache TTLs and expiry logic. Anything else to add @solana_devs?

In my view, the SGP-0003 vote to increase fees for compute-heavy transactions failed for a few key reasons: 🔴 Why it failed: - Fear of higher transaction costs - Major impact on aggregators and other CU-heavy DeFi apps - Not enough data showing the real cost 🟢 What could help it pass: - Real-world fee benchmarks - Before-and-after CU comparisons - Guidance on optimizing apps to match or beat today’s fees, showing that higher costs aren’t inevitable just a reason to optimize

SGP-0002 Looks like we did it, congratulations to everyone on this milestone! 🎉 @SolanaFndn

I’m disappointed by the current SGP-0002 (reduction of SOL emissions) vote. It’s a shame that @krakenfx, @everstake_pool, and @Figment_io voted at their own discretion without publicly seeking input from their delegators or at least explaining how they reached their decisions.

Solana is entering a new governance era. 🔹SGP-0001 formalizes on-chain decision-making 🔹SGP-0002 could reduce SOL issuance faster, with ≈18.9M fewer SOL created over six years 🔹SGP-0003 proposes raising daily burns from≈650 to 7,500–9,000 SOL Governance is becoming a key part of Solana’s tokenomics.

SGP-0003 would replace Solana’s flat fee with an inclusion fee + a burned fee based on requested compute. Efficient txs may get cheaper but compute-heavy apps could pay much more. Stake leans YES, many app builders say NO. The real question: what should @solana optimize for?

Agave v4.3 reverted the banking-stage change that accounted for actual compute consumed. It’s back to reserve-then-refund: your requested CU limit counts against block capacity again. Padding SetComputeUnitLimit to 1.4M can hurt inclusion. Measure usage and set a tight limit.

Why is Solana targeting 200ms slots? Not because someone wanted a round number. The real constraint discussed by validators was global leader handoff latency. Tokyo → Europe propagation can already approach ~200ms. So the target was not chosen around CPU limits. It was chosen around geography, networking and physics. @solana_devs

I went through the latest Solana validator call. One thing surprised me: the path to 100M CU blocks / 200ms slots is not just consensus or execution. It also depends on validator networking. More shreds, less time, less kernel overhead. That is why XDP matters.