diff --git a/README.md b/README.md index 19c33805..477c6cd2 100644 --- a/README.md +++ b/README.md @@ -70,27 +70,27 @@ Proof length is the number of headers needed before and after a retarget to upda ## Security (Light Relay) -Let's review the calculation given in the [Bitcoin Wiki](https://en.bitcoin.it/wiki/Difficulty) to compute the hashrate: +Let's review the calculation given in the [Bitcoin Wiki](https://en.bitcoin.it/wiki/Difficulty) to compute the hashrate: ``` hashrate = difficulty * 2**32 / 600 (60 * 10 = 10 minutes) hashrate = ~157 (GH/s) = (22012.4941572 * 2**32 / 600) / 10**9 (example) -hashrate = ~595 (EH/s) (83148355189239.77 * 2**32 / 600) / 10**18 (current) +hashrate = ~743 (EH/s) = (103919634711492.2 * 2**32 / 600) / 10**18 (time of writing, Dec 2024) ``` -The `LightRelay` requires that the proof is included at the *current* or *previous* difficulty so we can assume the attacker has 2016 * 2 blocks to brute-force a valid chain of `proofLength`. -This is possible since due to SPV assumptions we can not verify the transactions references by the block header are valid, only that sufficient PoW has accumulated on the chain. +The `LightRelay` requires that the proof is included at the _current_ or _previous_ difficulty so we can assume the attacker has 2016 \* 2 blocks to brute-force a valid chain of `proofLength`. +This is possible since, due to SPV assumptions, we can not verify the transactions references by the block header are valid, only that sufficient PoW has accumulated on the chain. -Let's assume the attacker can generate 6 blocks (with some invalid transactions) within two difficulty adjustment period, 2016 * 2 blocks (four weeks). +Let's assume the attacker can generate 6 blocks (with some invalid transactions) within two difficulty adjustment period, 2016 \* 2 blocks (four weeks). ``` hashrate * 6/(2016*2) -595 * 6/(2016*2) = 0.885 EH/s (~885712 TH/s) +743 (EH/s) * 6/(2016*2) = 1.105 EH/s (~1,105,654 TH/s) ``` -So we need ~0.148% of the current hashrate to build six blocks in two weeks. +So we need ~0.142% of the current hashrate to build six blocks in two weeks. -If we estimate it would cost $11 per TH/s (excluding electricity and other setup costs) then the total cost of that hashrate would be $9,742,832. +If we estimate it would cost $11 per TH/s (excluding electricity and other setup costs) then the total cost of that hashrate would be $12,162,194. This excludes the opportunity cost from actually mining on Bitcoin mainnet, receiving fees and block rewards. -Therefore, provided the value protected by the relay is less than $10m protocols secured by the relay are "economically safe". +Therefore, provided the value protected by the relay is less than $12m protocols secured by the relay are "economically safe".