From 3dfc2dd969385957619e6060cab081fc4af83c59 Mon Sep 17 00:00:00 2001 From: Florent Tapponnier Date: Sat, 11 Jul 2026 23:26:48 +0200 Subject: [PATCH] audit round 2: gram finality methodology matches the BFT-commit measurement, buyback ratios scaled to percent, buyback success is scrape freshness (coingecko error counter zeroed sky at x300) --- benchmarks/buyback-audit.yml | 20 ++++++++++---------- benchmarks/l1-finality.yml | 4 ++-- 2 files changed, 12 insertions(+), 12 deletions(-) diff --git a/benchmarks/buyback-audit.yml b/benchmarks/buyback-audit.yml index 43539dd5..8792d247 100644 --- a/benchmarks/buyback-audit.yml +++ b/benchmarks/buyback-audit.yml @@ -120,11 +120,11 @@ providers: tag: Assistance Fund, 97% fees → HYPE spot buyback formula: "USD value of HYPE spot fills credited to the Assistance Fund 0xfefe…fefe over the rolling 7d window, divided by 97% of DeFiLlama's hyperliquid fees for the same window." queries: - p50: ocb_buyback_ratio{protocol="hyperliquid",window="7d"} - p90: ocb_buyback_ratio{protocol="hyperliquid",window="7d"} - p99: ocb_buyback_ratio{protocol="hyperliquid",window="7d"} - mean: ocb_buyback_ratio{protocol="hyperliquid",window="7d"} - success: clamp_min(clamp_max(1 - 300 * sum(rate(ocb_buyback_scrape_errors_total{protocol="hyperliquid"}[24h])), 1), 0) + p50: 100 * ocb_buyback_ratio{protocol="hyperliquid",window="7d"} + p90: 100 * ocb_buyback_ratio{protocol="hyperliquid",window="7d"} + p99: 100 * ocb_buyback_ratio{protocol="hyperliquid",window="7d"} + mean: 100 * ocb_buyback_ratio{protocol="hyperliquid",window="7d"} + success: (ocb_buyback_last_scrape_timestamp_seconds{protocol="hyperliquid"} > bool (time() - 7200)) sample_size: ocb_buyback_executed_usd{protocol="hyperliquid",window="7d"} series: ocb_buyback_ratio{protocol="hyperliquid",window="7d"} @@ -133,11 +133,11 @@ providers: tag: Smart Burn Engine, 100% surplus → SKY via Uniswap formula: "USD value of SKY ERC-20 inflows to the SBE 0xBE8E…98FB over 7d (Etherscan v2, priced at CoinGecko), divided by 100% of DeFiLlama's makerdao fees for the same window." queries: - p50: ocb_buyback_ratio{protocol="sky",window="7d"} - p90: ocb_buyback_ratio{protocol="sky",window="7d"} - p99: ocb_buyback_ratio{protocol="sky",window="7d"} - mean: ocb_buyback_ratio{protocol="sky",window="7d"} - success: clamp_min(clamp_max(1 - 300 * sum(rate(ocb_buyback_scrape_errors_total{protocol="sky"}[24h])), 1), 0) + p50: 100 * ocb_buyback_ratio{protocol="sky",window="7d"} + p90: 100 * ocb_buyback_ratio{protocol="sky",window="7d"} + p99: 100 * ocb_buyback_ratio{protocol="sky",window="7d"} + mean: 100 * ocb_buyback_ratio{protocol="sky",window="7d"} + success: (ocb_buyback_last_scrape_timestamp_seconds{protocol="sky"} > bool (time() - 7200)) sample_size: ocb_buyback_executed_usd{protocol="sky",window="7d"} series: ocb_buyback_ratio{protocol="sky",window="7d"} diff --git a/benchmarks/l1-finality.yml b/benchmarks/l1-finality.yml index c7bd5455..5d71f43b 100644 --- a/benchmarks/l1-finality.yml +++ b/benchmarks/l1-finality.yml @@ -59,7 +59,7 @@ per_chain_explainer: - slug: gram h2: "Gram finality time" body: | - Gram's BAG consensus (the chain formerly branded TON, native token renamed Toncoin → Gram in June 2026) pushes masterchain finality to {{p50:gram}} (p50, 24h), one of the lowest deterministic finalities measured on this leaderboard. Gram's design splits state across a masterchain and many workchains; the figure reported here is the masterchain commit, the canonical reference for cross-chain settlement. Measured via the tonapi.io `/blockchain/masterchain-head` endpoint with a 3-seqno lookback. + Gram's BAG consensus (the chain formerly branded TON, native token renamed Toncoin → Gram in June 2026) pushes masterchain finality to {{p50:gram}} (p50, 24h), one of the lowest deterministic finalities measured on this leaderboard. Gram's design splits state across a masterchain and many workchains; the figure reported here is the masterchain commit, the canonical reference for cross-chain settlement. Measured as the wall-clock lag of the tonapi.io masterchain head: Gram masterchain blocks are BFT-final at production (validator-signed, no fork choice), so commit lag is finality. Post accelerated upgrade the masterchain produces a block about every 0.4s, verified empirically 2026-07-11. - slug: bnb h2: "BNB Chain finality time" body: | @@ -123,7 +123,7 @@ methodology: - "TRON: `/wallet/getnowblock` (head) minus `/walletsolidity/getnowblock` (solidity-confirmed)." - "Stellar (SSE wall-clock): Horizon `/ledgers?cursor=now&order=asc` event-stream records T1 on first sight of ledger N and T2 on the next ledger close (SCP-final at every close)." - "SUI: `sui_getLatestCheckpointSequenceNumber` minus 5 checkpoints back." - - "Gram: tonapi.io `/blockchain/masterchain-head` minus 3 seqno." + - "Gram: wall-clock lag of the tonapi.io masterchain head. Masterchain blocks are BFT-final at production, so head commit lag is the finality figure; the chain produces a block about every 0.4s since the accelerated upgrade." - "Bitcoin (probabilistic, own-clock): mempool.space `/blocks/tip/height` polled every 30 s. T1 = first poll where block N is the tip, T2 = first poll where tip − N ≥ 6 (the exchange deposit convention). lag = T2 − T1 on the harness clock; miner-set block timestamps (consensus allows 2h skew) are never used. The startup tip is not timed. On a tip regression, pending timers above the new tip are dropped; same-height reorgs are invisible to height polling, bounding error to about one block interval." - "Litecoin (probabilistic): blockchair `/stats.best_block_height` minus 6 confirmations via `/dashboards/block/{height}.block.time`." - "Monero (probabilistic): monero-rpc `get_info` minus 10 confirmations via `get_block_header_by_height`."