The long read

A network for everyone

What a blockchain actually is, what went wrong with the existing ones, and what Paritr does differently. No prior knowledge assumed - the expandable boxes hold the detail if you want it.

Chapter 1 - The basics

What is a blockchain, really?

Strip away the jargon and a blockchain is a shared account book. Everyone holds a copy, everyone can check it, and no single person can quietly change a past entry. That is the entire trick - everything else is engineering in service of it.

receipt_long

A ledger nobody owns

Every transfer ever made is written into a public list of blocks, each one referring to the one before it. Change an old entry and every later block stops matching - which is why the past is effectively frozen.

key

Keys instead of accounts

You do not have an account with a company; you have a pair of keys. The public one is your address, which you hand out to receive money. The private one signs your payments and never leaves your device.

construction

Mining is bookkeeping

Someone has to bundle new transfers into the next block. Miners compete for that job by solving a hard puzzle, and the winner is paid in new coins. That reward is what makes the system run without an operator.

In more detail: why the puzzle matters
The puzzle has no shortcut - the only way to solve it is to try enormous numbers of guesses. That makes writing a block expensive, and rewriting history means redoing every block since. An attacker would need more computing power than everyone else combined, continuously. Security is therefore not a promise anyone makes; it is the cost of the electricity an attacker would have to burn. That is what “proof of work” means.
Chapter 2 - The problem

What went wrong

Bitcoin’s original paper described a network where “one CPU, one vote” decided the outcome. That held for about two years. What happened next was not a conspiracy - it was economics doing exactly what economics does.

  1. 1

    The puzzle rewarded specialised silicon

    Bitcoin’s puzzle, SHA-256, is simple and repetitive - ideal for a purpose-built chip. Once those existed, a general-purpose processor was thousands of times too slow to matter. Ordinary computers were not outcompeted; they were made irrelevant.

  2. 2

    Difficulty rises to match the strongest

    The network automatically makes the puzzle harder as more power joins, keeping block times steady. So the industrial machines did not merely win more often - they raised the bar until everyone else won essentially never.

  3. 3

    Which forced everyone into pools

    A small miner might find a block once a century, so miners joined pools and shared the payouts. That works, but it hands coordination to a handful of pool operators - the exact concentration the design set out to avoid.

The promise was that anyone could take part. Paritr is an attempt to keep that promise workable.

Chapter 3 - The answer

Two decisions, and what follows from them

Paritr follows Bitcoin’s architecture closely - the same 80-byte header, the same Merkle tree, the same most-work-chain rule. Two things are deliberately different, and between them they undo both halves of the problem above.

memory

A puzzle bound to memory, not to silicon

Instead of SHA-256, the proof of work is scrypt with a 2 MiB scratchpad per attempt. Each guess has to shuffle real memory, so speed is limited by memory latency rather than by how many logic gates you can afford. A custom chip gains very little - the bottleneck is something it cannot buy its way out of.

In more detail: why not RandomX?
Monero’s RandomX is stronger still: it compiles a random program per block and effectively demands a full CPU. But it needs a native library that every operator has to build. That would end one-command installation on Windows and on a Raspberry Pi - and a defence nobody can install is not a defence. scrypt at 2 MiB is the same scratchpad size that kept ASICs away from CryptoNight for years, and it needs nothing beyond the Python standard library. Block identity stays SHA256d, exactly as in Litecoin, so lookups and Merkle proofs remain cheap.
groups

Pool payouts, without the pool

Every miner keeps finding partial solutions - “shares” - that are far easier than a full block. Paritr counts those directly in consensus and splits 95 % of each block reward across everyone whose shares landed in the last 120 blocks. You are paid steadily for the work you actually did, with no operator, no fee and no account to open.

In more detail: why can nobody cheat this?
A share costs real computation, so it cannot be faked - unlike a self-reported hash rate. The payout address is baked into the hashed header via a Merkle path, so a relay cannot swap in its own address without destroying the proof. And the finder has no reason to drop foreign shares: without them the pool falls to him anyway, while including them additionally earns him the guaranteed 5 % plus every transaction fee. Every node recomputes the same distribution from the chain alone and rejects a block whose coinbase does not match.
Chapter 4 - In context

How Paritr compares

Bitcoin and Monero are the honest reference points: one defined the field, the other has fought hardest to keep mining on general-purpose processors.

Bitcoin Monero Paritr
Proof of work SHA-256d RandomX scrypt (2 MiB)
Realistic hardware ASIC only CPU CPU
Payout for small miners External pool External pool Built into consensus
Block time ~10 min ~2 min 60 s
Supply 21 Mio. Tail emission Tail emission
Solo setup effort Not viable Moderate One command

Bitcoin and Monero are mature networks with enormous security budgets and years of adversarial testing. Paritr is young and makes no claim to match that - the table compares design choices, not track records.

Chapter 5 - Plainly

What Paritr is not

Paritr is not an investment product and makes no promise about value. It is infrastructure: a ledger, a currency, and the tools to use them. Nobody is selling anything here.

It is also deliberately unfinished in the sense that any young network is: it has not been attacked at scale, and its security budget is a fraction of an established chain’s. Those are facts, not marketing footnotes.

warning

Before you put anything in

Coins may lose all value. Transactions cannot be reversed - a payment to a wrong address is gone. If you lose your private key or your password, nobody can recover your funds, including us. Back up your key the moment you create a wallet.

Chapter 6 - Reference

Key figures

Algorithm scrypt (N=2048, r=8, p=1) · 2 MiB
Block time 60 s
Difficulty retarget every 60 blocks (~1 h)
Block reward 10 PAR
Halving every 1,600,000 blocks (~3 years)
Maximum supply None - tail emission
Reward window 120 blocks (~2 h)
Reward maturity 10 blocks
Addresses Base58Check · start with "P"

The name

Paritr comes from the Latin pariter - equally, on equal footing.

Full protocol details in the documentation arrow_forward

The network only works if people take part

A wallet takes under a minute. A mining node takes one command - and from then on your machine helps secure the ledger and earns its share.