Skip to content

BOB is a hybrid L2 that combines the security of Bitcoin with the versatility of Ethereum.

Notifications You must be signed in to change notification settings

bob-collective/bob

 
 

Repository files navigation

Build on Bitcoin

BOB is a Bitcoin-augmented rollup bringing experimentation and freedom of choice to builders to make a real-world impact. BOBs vision is to onboard the next billion users to Bitcoin.

Learn more

Contribution

BOB is an open-source project. We welcome contributions of all sorts. There are many ways to help, from reporting issues, contributing code, and helping us improve our community.

Getting Started

We use foundry extensively for maintaining and testing this contract suite:

Build

forge build

# production
FOUNDRY_PROFILE=optimized forge build

Test

forge test

Format

forge fmt

Local Testnet

To deploy the relay contract to a local environment for testing use our convenience script here:

# start local ethereum testnet node
docker-compose up anvil
# run script to deploy the relay contract
cd sdk
npm run deploy-relay -- --init-height=latest --dev --network=mainnet --testnet
npm run update-relay -- --dev --network=mainnet --relay-address=0x<relay-address>

# sepolia
FOUNDRY_PROFILE=optimized npm run deploy-relay -- --network=testnet --testnet --rpc-url=testnet --init-height=latest --proof-length=20 --private-key=0x<exported-privatekey>

# mainnet
FOUNDRY_PROFILE=optimized npm run deploy-relay -- --network=mainnet --rpc-url=mainnet --init-height=latest --proof-length=20 --private-key=0x<exported-privatekey>
FOUNDRY_PROFILE=optimized forge verify-contract --verifier blockscout --verifier-url https://explorer.gobob.xyz/api 0x9fe7ef727da3d79e0308ff43f31ea1d077ee9f41 src/relay/LightRelay.sol:LightRelay
FOUNDRY_PROFILE=optimized npm run update-relay -- --network=mainnet --rpc-url=mainnet --relay-address 0x9fe7ef727da3d79e0308ff43f31ea1d077ee9f41 --private-key=0x<exported-privatekey>

The initialization height should be a multiple of 2016 with at least one subsequent retarget (i.e. if using 2016 as the starting height, blocks 4031-4032 must exist).

Proof length is the number of headers needed before and after a retarget to update the relay - tBTC-v2 uses 20 on ETH mainnet.

Contracts

Network Name Address
BOB Sepolia (Testnet) TestLightRelay 0x377d6993b848013991e3d8b3028db74ec6fdf03a
BOB Mainnet LightRelay 0x9fe7ef727da3d79e0308ff43f31ea1d077ee9f41

Security (Light Relay)

Let's review the calculation given in the Bitcoin Wiki 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)

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).

hashrate * 6/(2016*2)
595 * 6/(2016*2) = 0.885 EH/s (~885712 TH/s)

So we need ~0.148% 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. 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".