History

igneum-labs 696f8d2330 Live page: the ?window=300 capture with a paid shard cell and its tooltip; header counts and the ingest keep-window follow the target scale (the 8 px/s floor was the adaptive scale's); counts right on the first poll; lock label backed; record rotation 80 blocks per tick Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>		2026-10-04 17:43:26 +00:00
..
devnet-before-activation-desktop.png	Live page: shards per block (proving v0), three strips on one time axis (blocks, finality bar, proving), observer proof feed, hero glint	2026-10-04 17:39:28 +00:00
devnet-before-activation-mobile.png	Live page: shards per block (proving v0), three strips on one time axis (blocks, finality bar, proving), observer proof feed, hero glint	2026-10-04 17:39:28 +00:00
README.md	Live page: shards per block (proving v0), three strips on one time axis (blocks, finality bar, proving), observer proof feed, hero glint	2026-10-04 17:39:28 +00:00
testnet-active-desktop.png	Live page: shards per block (proving v0), three strips on one time axis (blocks, finality bar, proving), observer proof feed, hero glint	2026-10-04 17:39:28 +00:00
testnet-active-window300-desktop.png	Live page: the ?window=300 capture with a paid shard cell and its tooltip; header counts and the ingest keep-window follow the target scale (the 8 px/s floor was the adaptive scale's); counts right on the first poll; lock label backed; record rotation 80 blocks per tick	2026-10-04 17:43:26 +00:00
testnet-active-window300-mobile.png	Live page: shards per block (proving v0), three strips on one time axis (blocks, finality bar, proving), observer proof feed, hero glint	2026-10-04 17:39:28 +00:00

README.md

Live page: blocks, finality and proving strips (4 October 2026)

Screenshots of site/live.html after the three-strip layout (one time axis, newest at the right: BLOCKS lanes, a FINALITY bar, a PROVING strip with one cell per shard of every chain block). Captured from the page served locally against two observers' tables, with the installed Chromium test build (never the owner's Chrome), device scale 2.

File	What it shows
`devnet-before-activation-desktop.png`, `devnet-before-activation-mobile.png`	The real devnet (a read-only second observer on the observer node 28640, proving data from the Mac app's node on the proving build, 26800): `proving_v0_activation_daa` unset, so the proving strip carries the one honest line "Proving layer: not yet activated on this devnet; activation not set". 1280 px and 375 px.
`testnet-active-desktop.png`	The private 3-node proving network (`tools/proving-v0/run.mjs --network-only`, activation at DAA 60) at the default adaptive scale: every chain block's planned shard cell, the lag label at the left when proofs land beyond the visible window.
`testnet-active-window300-desktop.png`, `testnet-active-window300-mobile.png`	The same network through the diagnostic `?window=300` view (300 s on the axis; the API returns the same span): paid cells in the prover's colour with the tick, the dashed "proofs land N behind the tip" line, the tooltip of a paid cell (block, shard, prover, lag, payout).

Frame budget (playwright, 6 s samples, requestAnimationFrame wrapped): 1280 px with 110 blocks on screen avg 0.63 ms, max 1.0 ms; 375 px avg 0.49 ms, max 0.9 ms. Budget 10 ms.

What the CPU test prover showed (observer rows, provtest_live_proofs): block 155 shard 0 paid by chain block 405, lag 259 DAA (the compressed proof took 252 s on the loaded Mac), payout 634,116,040,000,000,000 wei; block 154 in proving while node 0's verifier queued behind the same CPU. Nothing here is a proving-time number.