yes run it once

toly 🇺🇸 dr cavey phd ⏳ yes. Alessandro Decina promised this is the final XDP fix before we can all retire github.com/anza-xyz/agave…

most ibrl shit we could by just agreeing to flip some constants instead of being autistic: - cut rent 100x to 69 lamports per byte - cut epoch length to something like ~4-8 hours - "pseudo-remove" CU cap on blocks by making it 200m - lower block times

"we are not poor - we are on low-liquidity diet."

So a guy becomes a drug dealer, goes to prison for 10 years, then (while in prison) somehow becomes a “Rust fanatic”, lands a full time job re-writing SQLite in Rust, and buys a house. While still in prison. For being a drug dealer. Huh. There’s… I mean… huh.

Purrteil.hl Because they have something none of those have. Liquidation protection mechanism against scam wicks. It's ultimately on us, as both participants and builders to fix our own problems. That's the beauty of crypto. We don't need regulators to tell us what we need to do. Drift

"cpu affinity casino" = when you skip pinning and your threads go gambling across ccds. jackpots? only for cache misses. dont turn your app into perf roulette. pin it or spin it.

one chain, one plan.

discord.com/invite/epochto…

Suddenly, you’re 80. Markets open, joints don’t. You still wake up at 6am, out of habit more than purpose. The screen lights up your wrinkled face, same charts, same patterns - different generation making the same mistakes. Your wife’s gone, the grandkids visit when they

to summarize: - jito copied Flashbot, kept themselves as sole blockbuilder - temporal copied what Flashbot is doing now - jito working on BAM, copying Flashbots buildernet all while FB system on ETH is pretty bad, uses my trusted relays, and devs try to figure how to fix it

me when people want to do "cleanups" to bpf lkml.org/lkml/2012/12/2…

generally, i removed some latency parts, such as pinning threads to cores on the same ccd to take advantage of the shared l3 cache on amd (improve inter-thread communication). alessandro told me that agave isnt ready for this

4kb coming soon 👀👀

We’ve been building this tool for months to study sub-millisecond behavior on Triton One 🌊🌋. Now it’s open-source, use it to see how fast your RPC really is and learn things no simple benchmark will show. Link’s in the comments 👇