eWatts is built on a strict separation of execution, interpretation, and adversarial validation. These layers do not mix. Each has a distinct role, rule set, and falsifiability condition.
Objective: A single deterministic execution system that any node can independently verify.
State transition: St+1 = f(St, Bt)
A block is valid if: PoW satisfies the difficulty threshold • prior state exists • UTXO transitions are consistent • commitments are verifiable • emission invariants hold.
E_t = g(height, state)smoke.rs, block.rs, state.rs, commitment.rs, reward.rs, reorg.rsObjective: Define the economic meaning of the system without interfering with the protocol.
Core hypothesis: Monetary systems anchored in physical computation substrates exhibit different long-term stability properties depending on the improvement rate of their underlying physical constraints.
Key variable: Δ = rate(substrate improvement). ASIC: high Δ (~25-30%). DRAM latency: low Δ (~1-2%). Lower Δ predicts more stable emission regimes.
The hypothesis is false if: (1) substrate advantage converges quickly, (2) system becomes economically dominated by a secondary abstraction layer, or (3) cost of attack decouples from physical constraint.
Objective: Model system behavior under strategic, noisy, and adversarial conditions.
Unlike Layer 1's single-chain assumption, Layer 3 models fork reality: multiple competing valid chains can exist simultaneously. Each node selects the chain maximizing S(C) = W(C) - C_attack(C), where W is accumulated work and C_attack is the cost-adjusted adversarial penalty.
| Layer 1 | YES | Can influence protocol — execution truth |
| Layer 2 | NO | Interpretation only — must not affect protocol |
| Layer 3 | NO | Testing environment only — must not affect protocol |
smoke.rs / reorg.rs | Layer 1 | Pipeline correctness, fork resolution, round-robin mining |
tests.rs (integration) | Layer 1 | Private tx, founder lock, double-spend, reorg |
commitment.rs | Layer 1 | Bandwidth commitment validation |
reward.rs | Layer 1+2 | Emission math (L1) + economic interpretation (L2) |
shuffle.rs | Layer 3 | Block propagation under network noise |
(future) adversarial.rs | Layer 3 | Strategic miner competition simulation |
eWatts has no fundraising, no investor allocation, and no entity controlling its treasury.
Mining is the only way to acquire coins. Founder outputs are time-locked on-chain
(spendable_after = max(50000, block + 40000)). Mining public key published before block 1.
| Consensus | MBPoW (Memory-Bound Proof of Work) |
| Target block time | 600 seconds |
| Signature verification | <3 ms per MLSAG signature (ring of 11) |
| Ring size | 11 |
| Signature scheme | MLSAG over Ristretto255 |
| Address format | Stealth (spend key + view key), 32 bytes |
| Amount privacy | Pedersen commitments + range proofs |
| Emission formula | R = BASE_EMISSION x (total_eff / historical_avg), clamped [0.05x, 20x] |
| BASE_EMISSION | 100 eWatt/block |
| Historical window | 4,300 blocks (30 days) |
| Ramp-up (blocks 1-10k) | 80% cap per miner, excess burned |
| Founder lock | On-chain: max(50000, block + 40000) |
| Privacy change threshold | 95% miners + 95% nodes |
| Unit | 1 eWatt = 100 ecents |
| Implementation | Rust, ~3,300 LOC |