Quinvarium network architecture
Quinvarium is designed as an independent UTXO Proof-of-Work Layer 1 with one linear accepted block history and four equal target-share PoW lanes.
Current design
The four lanes are not four separate chains. Every accepted block advances one global height. Equal target shares define allocation, not guaranteed decentralization, ASIC resistance or throughput.
Quinvarium network architecture
| Parameter | Current value | Status |
|---|---|---|
| Ledger | UTXO model | CURRENT DESIGN |
| Accepted history | One linear chain and one global accepted height | CURRENT DESIGN |
| Consensus direction | Nakamoto-style Proof of Work | CURRENT DESIGN |
| PoW lanes | RandomX-M, RandomX-T, SHA3x, Cuckaroo29 | 4 × 25% TARGET SHARE |
| Aggregate block target | 120 seconds; derived 480 seconds per lane | TARGET · NOT LIVE MEASUREMENT |
| DAA and fork score | Final algorithms and wire contracts | OPEN |
Technical documentation
One accepted sequence
Lane identity travels with a block, but accepted blocks share one global sequence. Monetary transitions, maturity and other height-based rules use the global accepted height rather than a separate height for each lane.
Four target-share lanes
RandomX-M is the Monero-compatible merge-mining direction. RandomX-T is a separate QNV-specific RandomX profile. SHA3x and Cuckaroo29 are independent directions. Each has a 25% target share in the current baseline.
What remains open
The final per-lane difficulty adjustment, work normalization, fork-score composition, tie-breaks, final targets, payload wrappers and wire bytes are not complete. Local easy-target tests do not prove the public 120-second target.
What the model does not claim
Four equal shares do not by themselves prove decentralization, hardware fairness, ASIC resistance, security under public adversarial traffic or a particular transaction throughput.
Related documentation
Questions and answers
Are the four PoW lanes four blockchains?
No. The current design has one accepted linear history with a lane identifier on each accepted block.
Is the 120-second target already measured on a public network?
No. It is the aggregate design target. The public network and final DAA are not live.