Mining, validators and rewards

Short version: you can mine a local proof-of-work chain today, but it pays nothing and it is not how the public network will run. The launch network is secured by validators, not miners. This page says exactly what works and what does not, so nobody spends electricity on a misunderstanding.

Where things stand

	Today
Proof-of-work mining (ArgonBlake)	Works on a local devnet. Blocks are empty and there is no block reward. Disabled in the production build.
DAG-BFT validators	Works: blocks, finality, transfers, fees and staking bonds at genesis. This is the launch consensus (ADR-016).
Block reward / coin issuance	None. No coin is minted after genesis. The base fee is burned; validators are paid from transaction tips each staking epoch (ADR-029).
Joining the validator set after genesis	Built and tested: a registration with a bond enters the committee at the next staking epoch (crates/node/tests/bft_staking_tests.rs). Not usable publicly until a public testnet exists.

Mining a local proof-of-work chain

Useful for exercising the miner, the difficulty adjustment and the GPU kernels; not for earning anything.

cargo build -p maya2c-node -p maya2c-genesis
maya2c-genesis --chain-id maya-pow-local --difficulty-bits 8 --out genesis.json
maya2c-node --genesis genesis.json --data-dir data \
    --rpc-addr 127.0.0.1:8546 --mine --threads 2
# mined height 2 nonce 12 tip 00677864d4f1f928
# mined height 3 nonce 6 tip c8c0ec10640cfe33

Measured on 2026-09-29: four blocks in twenty seconds at 8 difficulty bits on two threads of a laptop CPU, dev build. Raise --difficulty-bits and it slows accordingly.

What it does not do, and why:

• Blocks are empty. The built-in miner does not yet assemble transactions from the mempool (bins/maya2c-node/src/main.rs, mining_loop). Send a transaction to a proof-of-work devnet and it waits in the mempool.
• No reward. There is no coinbase transaction and the header names no miner. Hash power is recorded only through an optional claim transaction that feeds governance weight, not balances.
• --mine is refused by a production build, and a production node refuses a proof-of-work genesis. That is deliberate: proof of work is a devnet mode.

Why validators and not miners

Proof of work gives probabilistic finality: a payment is “probably final” after some number of blocks. DAG-BFT gives a block final in about a second on a healthy network and never reorganises it, which is what an application or a bank needs to treat a transfer as settled. The trade-off is that a validator set has to exist and be trustworthy, which is what staking and slashing are for.

What is needed before anyone can earn

These are open decisions, not tasks waiting for time:

1. A decision on issuance. Today validators earn only tips, and emission is nil. Whether that stays true, or new coins are minted for validators or a treasury, is an economic decision for the project owner.
2. A public testnet with seed validators, so outside validators have something to join and register against.
3. At least four independent validators. A DAG-BFT committee of n tolerates ⌊(n−1)/3⌋ faults: one validator halts if it stops, two halt if either stops, four survive one failure.

Testnet coins, when a testnet exists, will come from a faucet and will have no value. Anyone offering to sell you Maya2C coins today is not selling anything this project issued.