Ethereum writing and news
20 items match, 0 of them low signal.
Top of this filter
Items with a reason of their own, ranked by urgency, weight and how often the project publishesWe run 84 full-custody PeerDAS supernodes. Under BPO2's 14/21 blob ceiling mainnet fills about 6, so the hardware bill is the client you run, not the blobs.StereumLabs
An independent measurement of Optimum's mump2p gateway on 36 Ethereum mainnet nodes: blocks arrived earlier on three of six consensus clients, none significant.StereumLabs
Seven weeks, six instruments, three Ethrex generations. Three results turned up every time, one dissolved under a finer cut, and the middle never settled.StereumLabs
Six consensus clients timed the same engine call. Their stopwatches disagree with the execution client's own by 6 to 208 ms, before any comparison starts.StereumLabs
Lodestar times the slot end to end in milliseconds. Its engine wall clock also runs 200 ms longer than the execution client's own stopwatch.StereumLabs
Teku splits the slot into eleven named stages. Its own counters also rated a node that never committed a block the healthiest of six.StereumLabs
Grandine hides its engine-call timing behind a metric name no keyword search reaches. Besu's slow head update is confirmed on a fourth instrument.StereumLabs
The stream
Newest firstEverything here already leads the page above.