Peer-to-peer network · Self-certifying certificates · Zero premine

Verifying a transaction shouldn't require the whole network.

Zycord is a peer-to-peer network where every transaction carries its own proof of validity. No node has to replay the history to confirm one part — just the transaction's own bytes, checkable by any machine, in parallel. Proof-of-work mining with RandomX, no premine, no founder allocation, no investor round.

0% premined, allocated to founders,
or sold before launch
883K state operations per second,
per CPU core
~2s to validate and apply a whole
block of certificates
RandomX CPU mining

Zycord (ZCD) — how the network works

Assembled off-chain.
Verified in parallel.
Committed in order.

Every large blockchain makes every node re-execute every transaction against a global state. Zycord takes execution off the chain: the transaction is assembled between servers outside it, arrives complete with everything it read and wrote, any machine checks it from the bytes alone, and a single loop commits the result.

Three phases: construction between servers, parallel verification, sequential fold Construction — off-chain The transaction bounces between servers until it closes Wallet Sequencer contract A Sequencer contract B Underwriter co-signs Verification — parallel, stateless Many nodes, each checking its slice Fold — sequential One loop, compares and adds 9.9 ms per 2,900-certificate block

// 01 — The problem

Every fast blockchain gets heavier as the network grows

Most blockchains make every computer on the network replay every transaction against the same global state. It works — until the network really grows: then every node has to become a bigger machine, and fewer and fewer people can keep up. Zycord exists to solve that specific problem, not to promise being generically "faster".

The core idea: every transaction — called a certificate here — already carries what it read and what it wrote of the network's state. Checking whether it's valid becomes a matter of checking its own bytes, without consulting the whole history. Any machine can do that check on its own, in parallel, using several processor cores at once.

The only step that has to happen in order, one at a time, is applying the final result to the ledger — and that step is simple enough to run almost for free: comparing and writing values, with no signature checks, no code execution, no heavy cryptography along the way.

No premine

No coin exists before block 0. No founder, investor, or early adopter gets a special allocation.

Accessible mining

RandomX runs well on ordinary CPUs — no ASIC farm to compete with you.

Parallel processing

Validation depends only on the transaction's own bytes — never on the whole network history. That's what opens room to run in parallel, with no queue.

// 02 — Launch

Four eras, two heights, zero premine

The network is born by proof of work and evolves toward proof of stake through thresholds measured on-chain — block height and sustained stake — never by calendar date or by vote.

Public testnet · genesis

15 September 2026 — 00:00 UTC

The testnet carries mainnet's consensus rules under its own chain id, so what it measures is mainnet's behaviour and not its own. It can be reset from genesis; coins on it are worth nothing and are meant to be.

Mainnet · genesis

1 October 2026 — 00:00 UTC

Block 0, and the first block anyone can mine. No premine, no founder allocation, no investor round, no admin keys — the absence of privilege is checkable line by line in genesis rather than promised.

Era 0 genesis, RandomX

Mine & Tip

Proof-of-work mining with RandomX — CPU, no ASIC. Native assets (issue, mint, transfer, tip) already work, and every certificate is self-insured: you can use the network with no counterparty from the very first block.

Height 40,320 ~day 14

Bond

Stake can be bonded, so underwriters exist before anything needs them. The height is a consensus parameter fixed at genesis, not a release: the binary that mines block 0 already knows it.

Height 80,640 ~day 28, same binary

Parallel cEVM

The Solidity-compatible machine turns on, with a delta-first standard library already deployed — alongside bonded underwriters, application sequencers and the forced-inclusion queue. About four weeks after block 0, with no fork and nothing new to install.

Era 1 stake ≥ 1% sustained

Finality

Staked validators start finalizing checkpoints alongside miners, and the block subsidy splits between the producer and the attesters. Stake cannot be measured before the operation that creates it exists, so this threshold waits on the network rather than on a height.

Era 2 stake ≥ 10% for 30 epochs

Platform

A committee of validators starts proposing blocks; the proof-of-work mining reward retires gradually, on a ramp, never all at once. The treasury vault opens under public control of 5 parties, 3 signatures required.

Era S after 2+ public audits

Shielded

Payments with hidden value and disposable addresses go live — only after independent audits are published and a public adversarial testing period. It doesn't switch on before it's ready.

Issuance with no cliffs

The block reward drops gradually, in small steps each epoch — no sudden cut all at once, like Bitcoin's halving. After about 12 years, it settles into a small, permanent tail that pays for security forever.

A treasury with no key and no owner

3% of every block reward goes into a vault — but it stays locked until Era 2, with no key, no address, no quorum defined at genesis. If nobody shows up publicly to claim it with transparency, that money simply never gets spent.

// 03 — Differentiators

What already runs from block 0 — and what comes later

Each card shows which era of the protocol turns that feature on. Nothing here is sold as ready before it's ready.

Era 0

Genuinely parallel processing

Every certificate already arrives proving on its own that it's valid — no need to consult the network's history. Hundreds of payments to the same wallet process at the same time, with no queue and nobody blocking anybody.

Era 0

Capacity that grows with real usage

The TPS ceiling isn't fixed or decided by vote: it doubles every year of full, sustained usage. The design's structural limit — 1.1 million TPS — is reachable over the next decade, as real usage calls for it.

Era 0

Fair fees

Two separate markets — one for state changes, one for verification — adjust on their own as usage rises or falls, with no fixed auction. A certificate fee that fails is burned, never kept by whoever assembles the block.

Era 0

No-surprise launch

Zero premine, zero founder allocation, zero investor round, zero admin key. The genesis block is public and reproducible before the network even exists.

Era 0

Miner today, operator tomorrow

Whoever mines and bonds their reward from Era 0 gets priority in the first sequencer rotations, which open at height 80,640. Miners don't get discarded as the network evolves — they become part of the operation.

Height 80,640

Censorship resistance

If an application sequencer tries to ignore you, there's a forced-inclusion queue: your transaction gets in regardless, guaranteed. Arrives at height 80,640, about four weeks after block 0, alongside the Solidity-compatible machine.

Era 0

Mine with what you already have

Mining uses RandomX, built to run well on an ordinary processor — the same one already in your computer. No specialized hardware needed to start participating from block 0.

Height 80,640

Solidity-compatible

Anyone who already writes contracts on Ethereum can port them to the cEVM with few changes — same language, same tools, same auditors. It ships inside the genesis binary and turns on at height 80,640, with no fork and nothing new to install.

Era S

Privacy in payments

Each payment's value stays hidden, and whoever receives it gets a new address every transaction — nothing to trace by repetition. This layer only takes effect after independent audits are published: privacy that arrives ready, not privacy that arrives fast.

Height 80,640

Room for heavy cryptography

A heavy proof — zero-knowledge, post-quantum signature — never competes for space with a simple payment: it pays in a market of its own. What doesn't fit in a contract on an ordinary chain today fits here.

Era 0

294K TPS measured. Only 90 released at genesis.

The system's one sequential step — the one that decides what actually counts — already runs at 294,000 TPS on a single core of an ordinary desktop. At launch, the network releases only 90: the ceiling grows with measured trust in the network, not with hardware limits. The bottleneck was never the machine.

Era 0

99.5% of block time runs in parallel

Out of roughly 2 seconds to validate and apply a block of 2,900 certificates, only 9.9 milliseconds is the part that can't be parallelized. The system's real bottleneck measures under 1% of total time — the rest is verification, and verification scales with hardware.

// 04 — Why

Trust that doesn't ask you to trust anyone

Four reasons to take this launch seriously — and none of them ask you to take a promise on faith.

  • 01

    Why trust without trusting anyone

    The whitepaper doesn't hide its open questions — it lists them, gives measured numbers from a real implementation, and signs with a public key. You don't need to take a promise on faith: you can check the code, and run the test vectors against it yourself.

  • 02

    Why now, and not later

    The network runs on testnet before any mainnet date is announced — and when it is, it's announced in the open, well ahead of time, so anyone can reach block 0. There's no silent presale and no privileged waitlist.

  • 03

    Why mine here

    RandomX was designed to keep ordinary hardware competitive — CPU, not an ASIC farm. Whoever mines and bonds the reward from the start gets priority when the network opens its first operator slots, in Era 1.

  • 04

    Why this solves a real problem

    Every fast blockchain today makes every computer on the network replay every transaction against the same state. The more people use it, the bigger the machine every node has to be — and the fewer people can run one. Here, verifying a transaction doesn't require the whole history — just the transaction's own bytes.

// 05 — See for yourself

Don't take our word for it. Check it.

Documentation, and soon the complete source code — all published for anyone to audit before trusting it.

Read the whitepaper

Self-certifying certificates, deterministic fold, two fee markets — the full specification, with the academic references the design leans on.

Document · v1.0

Check the protocol against its vectors

The parameter files and the golden vectors are the protocol: every fold, block and validity rule has its case, positive and negative. Run them against this implementation — or against one you write yourself, which is what they are for.

Spec · vectors

The community lives on Bitcointalk

No official Discord and no official Telegram — no channel with an admin who could silence someone or erase the history later. Community-run channels are welcome and explicitly unofficial. Official announcements are signed and posted in one public forum thread, with dated, ownerless posts, the way anonymous projects have done it since 2009.

Forum · official

Watch mainnet produce blocks

The block explorer for mainnet, from block 0. Block 1 carries the launch message in plain text; every block after it, every certificate, and whether each was applied or skipped and billed.

Mainnet · live

Watch the testnet produce blocks

The block explorer for the running testnet. Every block, every certificate, and — the page no other chain's explorer can render — whether each certificate was applied, or skipped and billed, and which declared read stopped holding.

Testnet · live

Source code & testnet

Reference implementation, golden test vectors, and instructions to run your own node — published before any mainnet date, under the MIT licence.

Public · MIT

// 06 — Join

Run your own node, no permission needed

Reference client, RandomX mining binary, and a light wallet — published together with the testnet, before any mainnet date is set.

Download it, start it, and leave it running. A node started before the network's first block waits for it and begins mining on its own — there is nothing to schedule, restart or reconfigure when the hour arrives.

🟢Testnet

Each release ships two sets of archives and they are not interchangeable. The -randomx builds are the ones that join a network, and they are published for Linux, macOS and Windows on x86-64. The plain builds are byte-for-byte reproducible so that anyone can rebuild the tag and compare — they carry no mining engine and run only a local devnet, which is all that Windows on ARM has. The downloads page names which is which.

// 07 — Community

Designed from day one not to depend on me

I am an anonymous developer. I will launch this network and maintain it for as long as it takes — but the goal is for the community to take over every part of it.

Donations

Donations fund development. Eras 0 and 1 are donation-funded only: proposals public before payment, milestones and budget up front, release against delivered work, and unspent amounts returned to the fund.

I take no allocation and hold no privilege over any other user. There is no premine, no founder allocation and no investor round — so these addresses are the only way I am paid for this work.

I am looking for volunteers

Every part of this is work the community can own, and most of it is work I cannot do alone or should not do at all:

  • Community channels — Telegram, Discord, Reddit. Community-run and explicitly unofficial: a chat has an admin, and an admin can silence people and erase history, which is why no channel is ever official here.
  • Translations of the documentation and the whitepaper.
  • Documentation — guides, corrections, and anything that makes the docs easier to arrive at cold.
  • Pools. None exist and none are official; if you build one, it is a third party and should say so.
  • Above all, people willing to attack the design and publish what they find. Post it in the thread or open an issue — neither can be deleted, and a defect found before mainnet costs nobody anything.

Post in the announcement thread. It is not self-moderated, so nothing in it can be deleted — including criticism of me.

Follow the project by e-mail

Releases, testnet resets and the mainnet date, sent by hand and not often. One address, no tracking, no mailing service.

Get Updates by E-mail