A people-mined, ASIC-resistant proof-of-work cryptocurrency built to spend, not hoard.
Mine EMBER on the computer you already own — no farms, no pools, no premine.
-blue){kind=icon}
-ff4d00){kind=icon}
🌐 Website & live explorer: savvaniss.github.io/hearth
proof of work · CPU mining · ASIC-resistant · fair launch · digital cash
Proof-of-work was supposed to be one CPU, one vote. Instead Bitcoin became a game for warehouse farms and a few pools, and almost everything since became a speculative chip you hoard rather than money you spend.
Hearth gives the original promise back to ordinary people and fixes the specific things that pushed them out:
| Big crypto problem | What Hearth does |
|---|---|
| Mining centralized into ASIC farms & pools | Homefire PoW: RandomX-class, memory-hard, non-outsourceable. A farm earns no more per dollar than your laptop, and pools can't form. |
| Coins behave like "gold" you hoard | Money-first coinnomics: disinflationary emission → perpetual tail, offset by a fee burn. Net inflation trends toward ~0%. Built for velocity. |
| Slow & expensive to pay with | 15s blocks, sub-cent fees, instant retail settlement over Tab channels. |
| The "fee cliff" (security dies when rewards end) | Perpetual tail emission funds security forever. |
| Development captured by VCs / premines | Fair launch + on-chain Commons treasury (10% of each block), community-governed. |
| Too hard for normal humans | One app (node + wallet + miner), polite mining, a web wallet, a block explorer, and a two-line "Accept EMBER" SDK. |
Full argument: WHITEPAPER.md.
- Network: Hearth · Coin: Ember · Ticker:
EMBER - Smallest unit: 1 EMBER = 100,000,000 sparks
- Block time: 15 seconds · PoW: Homefire (CPU-only, non-outsourceable)
- Emission: smooth halving-free decay → perpetual tail; 10% to the Commons; base fee burned
- Supply: uncapped but disinflationary — modeled net inflation ~0.37% at year 30
yr reward issued/yr burned/yr supply gross% net%
1 4.24 10,667,873 1,133,462 9,534,412 100.00 100.00
5 1.06 2,666,968 1,416,827 22,527,372 11.14 5.55
10 0.30 631,152 536,479 23,890,811 2.58 0.40
30 0.30 631,152 536,479 25,784,267 2.40 0.37 (node proto/emission.js)
Money, not gold. Full breakdown: docs/coinnomics.md.
Everything below runs:
- ✅ a real blockchain node (
node/) — signed UTXO ledger, Homefire PoW, mempool, emission, fee burn, LWMA difficulty, disk persistence - ✅ most-work fork choice with chain reorganization — not a toy single-chain accept
- ✅ P2P networking — nodes gossip and sync over TCP (network-id handshake, DoS caps)
- ✅ wallet + CLI — keys, checksummed addresses, balances, signed payments
- ✅ HTTP/JSON-RPC/SSE API + a live block explorer, web wallet, and Hearth Pay merchant SDK
- ✅ a Rust production core (
rust/hearthd) — Homefire PoW, Ed25519-signed ledger, mempool, and Tab payment channels;fmt/clippy/29 testsclean - ✅ Docker Compose to boot a multi-node network on your Mac
- ✅ CI running Node tests + Rust build on every push
Hardened against an adversarial audit — unlimited-mint, fork-choice, coinbase maturity, timestamp, DoS, address-checksum, deterministic-emission, and XSS issues are fixed with tests (see docs/security-review.md).
Verified locally: 50 node checks + 29 Rust tests pass, two nodes sync over P2P, a reorg replaces the active chain, and the Rust core mines ~8× faster than the JS prototype on the same machine.
One implementation is the goal. The Rust node is the target; the JS node is a transitional reference we're porting away from (emission is already byte-identical across both). See docs/why-two-implementations.md.
1 — Boot a real multi-node network (seed + 2 miners + web/explorer):
docker compose up --build
# open http://localhost:8080 · explorer reads the live chain from :8645No Docker? ./scripts/run-local-network.sh. Details: docs/network.md.
2 — Run a node + wallet directly (needs Node 18+, no install):
cd node
node bin/hearthd.js --mine # a mining node + API on :8645
node bin/hearth-cli.js supply # query it
node bin/hearth-cli.js send <toAddress> 5 # send 5 EMBER
node test/unit.js && node test/e2e.js # 40 checks3 — Build the Rust core (needs stable Rust):
cd rust/hearthd
cargo run --release -- 20 # mine a demo block
cargo fmt --check && cargo clippy -- -D warnings && cargo test4 — Open the web experience (works offline on demo data, or live against a node):
npx serve web # or: open web/index.htmlindex.html (site) · explorer.html (live block explorer) · wallet.html
(web wallet + in-browser mining) · pay-demo.html (Accept EMBER checkout).
hearth/
├── README.md · WHITEPAPER.md the pitch and the full design
├── docs/ architecture · coinnomics · mining · network · roadmap · faq
├── node/ hearthd — runnable reference node/wallet/miner (JS) + tests
│ ├── src/ bin/ test/ Dockerfile
├── rust/hearthd/ production core (Rust): SHA-256 + Homefire PoW + ledger + P2P, zero deps
├── app-desktop/ Tauri desktop app scaffold (one-click node + wallet + miner)
├── web/ website + web wallet + explorer + Hearth Pay SDK (SEO-ready)
├── proto/ coinnomics simulator + standalone PoW demo
├── scripts/run-local-network.sh local network without Docker
├── docker-compose.yml seed + 2 miners + web
└── .github/workflows/ci.yml CI: node tests + rust build + web lint
A complete design with a working reference network, a compiling Rust core, a full web layer, tests and CI — but pre-mainnet: the production Homefire VM, Tab channels, and consensus still need audits and a public testnet before launch. See docs/roadmap.md.
Hearth is a commons. See CONTRIBUTING.md and docs/why-two-implementations.md. Highest-leverage areas: the Rust Homefire VM, the Tab payment layer, and the Tauri desktop app.
Please report vulnerabilities privately — see SECURITY.md.
MIT — money should be free to fork.
Keywords: Hearth, EMBER, cryptocurrency, proof of work, CPU mining, ASIC resistant, RandomX, non-outsourceable puzzle, decentralized digital cash, mine crypto at home, no premine, fair launch, crypto wallet, block explorer, blockchain node, Rust blockchain, tail emission, fee burn.