frol.near
4.5K posts

frol.near
@Freol
Rust and NEAR Protocol, NEAR AI, IronClaw technical evangelist building @TrezuApp

HUGE. Someone just made a single $635,266 purchase through the EtherFi card And earned roughly $3,177 back in WETH cashback Thoughts?

@mert step 3) switch to near(.)com to never have this issue again and pay in any asset anywhere





🚀 Good news from @NEARProtocol A new bounty target has been added to their bug bounty program: Sputnik DAO Contract You will be rewarded based on these tiers: Low: up to $100 Medium: up to $10,000 High: up to $100,000 Critical: up to $300,000 Start the #bugbounty hunt right now 👉 hackenproof.com/programs/near-…

❗️🚨 BREAKING: Researchers used Mythos Preview to find the first public macOS kernel memory corruption exploit on Apple's M5 silicon, they give a glimpse into Mythos say it’s really powerful. Apple spent five years and an estimated several billion dollars building Memory Integrity Enforcement (MIE), the hardware-assisted memory safety system built around ARM's MTE. It was the flagship security feature of the M5 and A19, designed specifically to kill the entire memory corruption bug class. Researchers from Calif built a working exploit in five days. According to Apple's own research, MIE disrupts every public exploit chain against modern iOS, including the recently leaked Coruna and Darksword kits. Calif walked into Apple Park this week and handed over the report in person. Full 55-page technical report drops after Apple patches the vulnerability.

today we are releasing a qemu escape


42% of all activity on near.com in the last 30 days was confidential. Six weeks ago, that number was zero. The curve is bending.

Crypto hacks, supply chain attacks, major web framework hacks, Linux root privilege escalation, and now we are at VM escapes... I announce the era of offline desktop/CLI apps running on Internet-less devices!



Crypto hacks, supply chain attacks, major web framework hacks, Linux root privilege escalation, and now we are at VM escapes... I announce the era of offline desktop/CLI apps running on Internet-less devices!




0e11c4aa285dffe95d2d7e90d974ad0e72336549b0dd2161dec606ba4955e2e1 qemu.c

Your Trezu treasury can now earn yield on @rhea_finance! Rhea is NEAR's most powerful DeFi hub, bringing together swaps, lending, and yield under one roof with over $130M in TVL. Just connect your treasury as a wallet directly in their UI and start earning on assets you're already holding. Live now on our new Earn page 😎

What you want is to compare web frameworks based on how many tweets they can post. Such benchmarks are only relevant to Twitter team, not to Django / Flask / PHP devs. I invite you to learn about web framework benchmarks: techempower.com/benchmarks/ Benchmarks are for developers who can understand how to interpret the results and it gives the baseline. Good benchmarks are those that provide all the assumptions, conditions, and that try to add as much of real-world context in the setup as possible, but not more than necessary. Comparing how many NEAR Intents swaps a chain can handle is not a completely helpful benchmark. What does NEAR's 1M TPS benchmark mean? x.com/NEARProtocol/s…. It means that @NEARProtocol can handle 1M basic operations per second, it also shows that horizontal scaling works as expected (not like multithreading in Python, and does not hit the ceiling like any other blockchain), and so the chain can actually scale even beyond that.

I said no chain can handle 1M real TPS and got a lot of pushback But most people don't even understand how TPS benchmarks actually work They’re run in local environments with a handful of nodes Zero real-world latency. Using simple native token transfers (or even minimal/no-op txs) Occasionally they go global…but still with basic transactions only Here’s the math ⬇️ A basic transaction is cheap Sending native ETH (or SOL, etc.) from wallet A to B is ~21,000 gas on Ethereum Minimal compute and tiny data That’s what most TPS benchmarks are measuring Now try a real DeFi swap: ETH → token ABC on Uniswap Now you’re at 100k-150k+ gas 5-10x the resources Run that swap through an aggregator (1inch, Jupiter, etc.) with multiple hops, approvals, routing? Easily 300k–1M+ gas/compute. Often 10-50x and sometimes 100x+ the resources of a simple transfer Now look at flash loans, MEV bundles, leveraged positions, NFT bundles? Those can all get even larger. One complex tx can consume the resources of dozens of basic ones And that’s just one category and how all chains operate We also have liquidations, cross-chain bridges with messages, restaking loops, governance executions, account abstraction batches… No chain in the world has ever been stress-tested at scale with a real mix of these transactions Because the demand simply isn’t here yet This is why many chains that claimed “tens or hundreds of thousands of TPS” ran into congestion, failed txs, and degraded performance at just a fraction of those numbers We’ve seen it countless times when surges expose the gap between lab tests and reality A handful of chains have solid infrastructure and could scale up faster than others to meet real demand Maybe by adding more nodes, shards, hardware, or with various tweaks. Some that could happen relatively quickly. But not a single live chain today could handle sudden real-world demand hitting 1 million TPS TPS without transaction complexity is just marketing. Every chain knows it. Most just hope you don't.

Wow, the ethereum ecosystem is currently processing 6,088 TPS across the entire stack (L1 & L2S), 3-4x that of Visa, in real time, on permissionless rails. Visa averages is ~1,700 TPS fyi Thanks @growthepie_eth and @LeonWaidmann for the data!



NEAR recently achieved 1M TPS in a publicly verifiable benchmark. In this technical deep dive, the NEAR One R&D team unpacks the most significant optimizations they worked on across the stack to make this monumental scalability milestone possible: blog.nearone.org/announcement/2…



