• 60 Seconds — The Heartbeat
  • Why faster
  • What stays the same
  • The hidden consequence
• 137 Days — The Memory
  • Why a hard limit, and not “optional pruning”
• 120 Seconds — The Open Hand
  • What it actually does
  • Why it is not a security hole
  • The hand-up for small miners
  • The barbell
  • The energy ceiling
• The Synthesis — Why the Three Numbers Need Each Other
• 51% Attacks — What This Architecture Changes
  • What is harder
  • What remains the same
  • What is structurally improved
• The Gold-Panner

Three numbers shape almost everything about how Elektron Net feels, behaves, and stays secure:

• 60 seconds — the heartbeat of the chain (block interval)

• 120 seconds — the open hand for small miners (Stoic Awakening window)

• 137 days — the limit of memory (mandatory prune depth, 197,280 blocks)

Each number on its own looks like a tuning parameter. Together they form a coherent design that trades the archival, industrial Bitcoin of today for something closer to Satoshi’s original vision: a peer-to-peer cash system secured by many small, individual participants, with privacy built in by the simple act of forgetting.

This document explains the why. The how lives in BITCOIN_CORE_DIFF.md, the philosophy in WHITEPAPER.md, and the audit in AUDIT_PRUNING_SNAPSHOT.md.

§60 Seconds — The Heartbeat

Bitcoin targets a block every 600 seconds (10 minutes). Elektron Net targets one every 60. Code reference: nPowTargetSpacing = 60 in src/kernel/chainparams.cpp.

§Why faster

A 10-minute block was a reasonable choice for a system that had to prove itself in 2009. Today, when people pay for coffee with their phones, waiting ten minutes for the first confirmation is a museum piece. At 60 seconds:

• First confirmation in a minute, six confirmations in six.

• 1,440 blocks per day (vs. Bitcoin’s 144).

• The chain feels alive — block explorers update like a pulse.

§What stays the same

• Total supply is unchanged: MAX_MONEY = 21,000,000 * COIN.

• Halving cadence is rescaled to nSubsidyHalvingInterval = 2,102,400 — same ~4 calendar years, just expressed in more blocks.

• Smallest unit is still 10⁸ per coin (called lepton here, instead of satoshi).

§The hidden consequence

Faster blocks create more frequent mining opportunities. That alone wouldn’t matter — Litecoin and Dogecoin already proved that. But combined with the other two numbers, 60 seconds becomes the foundation of something bigger.

§137 Days — The Memory

Constant: MANDATORY_PRUNE_DEPTH = 197280 in src/validation.h:79. At 60-second blocks, that’s almost exactly 137 days.

§Why a hard limit, and not “optional pruning”

Bitcoin lets the user choose: keep the full archive, or prune to save space. Elektron Net makes the choice for everyone: all nodes prune to the last 137 days. There is no full-archive mode. The user’s -prune=<GB> setting is silently ignored (src/node/blockmanager_args.cpp).

This is not a storage optimization. It is a statement about what a public ledger should be.

§The “Pocket” model

A wallet should be like a pocket — light enough to carry, holding what’s useful, forgetting what isn’t. The right to be forgotten is not an afterthought; it’s the first principle. If the network cannot serve old transactions, no third party can dig them up either. Forensic de-anonymization tools depend on archival history. Without it, the chain itself protects its users.

§What is preserved, what is lost

Crucially: balances are not lost when blocks are pruned. The UTXO set in chainstate/ is a cumulative database. An unspent output created in block 100 stays in the UTXO set until it is spent, even if block 100 itself was deleted a year ago. The snapshot mechanism (see below) captures this complete state at every checkpoint, so newcomers can join the network with full and current knowledge of every balance.

§Implicit finality

A 51% attacker on Bitcoin can, in theory, rewrite an arbitrarily deep section of history given enough hashrate and time. On Elektron Net, after 137 days the blocks are simply gone. A reorg that deep is physically impossible because no honest node has the data to participate. The chain becomes finalized by erasure, not by checkpoint consensus.

§120 Seconds — The Open Hand

If 60 seconds is the heartbeat and 137 days is the memory, 120 seconds is the gesture of welcome.

The rule (Stoic Awakening, mainnet, active from height 1): if more than 120 seconds have passed since the last block, the next block may be mined at the minimum difficulty.

Code: MinDifficultyActivationHeight = 1 in src/kernel/chainparams.cpp, implementation in src/pow.cpp.

§What it actually does

It is one block of relief, then the timer resets. The next block after that is back to normal difficulty unless 120 seconds pass again. This is not a sustained low-difficulty period — it is a single open window, and it only opens when the network slows down.

§Why it is not a security hole

A 51% attacker cannot use this to “chain-rush” — every min-difficulty block needs another 120-second gap, which means waiting in real time. The chainwork gained is marginal.

But for a Bitaxe in someone’s kitchen, those 120 seconds are everything.

§The hand-up for small miners

Modern small mining devices — Bitaxe, NerdMiner, hobby ASICs in the 500 GH/s to 5 TH/s range — have essentially zero chance of finding a Bitcoin block. The math is brutal: a single device against a global exahash network is a lottery with no winners.

On Elektron Net, when the 120-second window opens, the difficulty drops to powLimit. A solo Bitaxe can realistically find that block. Not often — but realistically. The lottery has winners now.

§The barbell

This produces a deliberately asymmetric miner population:

• Professional pools and data centres mine the ~95% of blocks that appear within 120 seconds, at full difficulty. They carry the cost and carry the security.

• Solo adventurers with Bitaxes, NerdMiners, and basement rigs catch the blocks that fall through the 120-second window.

Nobody is displaced. Professionals are not undercut — they still win the vast majority of blocks and the steady income. Hobbyists are not excluded — they have a real, non-zero chance every time the network breathes.

§The energy ceiling

Bitcoin’s energy use grows because difficulty grows because hashrate grows. The arms race has no built-in brake. On Elektron Net, if difficulty rises so high that even professional miners regularly take more than 120 seconds per block, the overflow falls to solo miners at minimum difficulty. This acts as an implicit safety valve against unbounded hashrate escalation. Over long horizons, the system self-regulates toward a hashrate level where the arms race is no longer economically attractive.

This is not proof-of-work versus proof-of-stake. It is proof-of-work with self-limitation built in.

§The Synthesis — Why the Three Numbers Need Each Other

Each constant in isolation is interesting. Together they form an architecture.

• 60 seconds makes mining frequent enough that solo participation is meaningful. At 600-second blocks, a Bitaxe owner would wait years between meaningful chances even with min-difficulty windows. At 60 seconds, the chances arrive constantly.

• 120 seconds opens the door for small miners without taking anything from large miners. The big operators still win when blocks come fast (most of the time). The small operators win when blocks come slow.

• 137 days keeps the chain light enough that a Raspberry Pi, a phone, or a low-end laptop can be a full node. Light chains attract more nodes. More nodes mean more verification, more diversity, more resilience.

• Per-block UTXO attestation stitches it all together cryptographically. Every block commits, in its coinbase, to the hash of the UTXO set after that block. A new node bootstrapping from a snapshot does not have to trust the snapshot’s author — it can verify the snapshot against the on-chain commitment, which is secured by all the proof-of-work backing the chain. (Code: ValidateUTXOCheckpoint, src/validation.cpp:2920; WriteAutomaticSnapshot, src/validation.cpp:2439.)

Pull on any one of these threads and the others come along.

§51% Attacks — What This Architecture Changes

A 51% attack is still possible in principle — proof-of-work makes no other promise. But the shape of the threat changes in ways that matter.

§What is harder

• Renting hashrate. Solo miners are not on NiceHash. They are in kitchens, garages, workshops. An attacker cannot simply purchase enough hashrate to overpower the chain — much of the honest hashrate is geographically and operationally distributed in places no marketplace reaches.

• Geographic concentration. Industrial mining tends toward a handful of regulatory regimes and a handful of cheap-electricity locations. Hobbyist mining lives everywhere — across power grids, ISPs, and jurisdictions. Confiscating or coercing that hashrate would mean knocking on a million doors.

• Deep reorgs. Past 137 days, no honest node holds the data needed to validate an alternative chain. Reorgs are bounded by erasure.

• Snapshot poisoning. A malicious peer cannot feed a new node a fabricated UTXO set, because the snapshot must hash-match the on-chain coinbase attestation. To forge a snapshot, the attacker would have to re-mine the checkpoint block and every block after it.

§What remains the same

• Short-range double-spends within the recent window are as feasible as on any proof-of-work chain.

• A young chain with low total hashrate is vulnerable, regardless of pruning model. This is a launch-phase risk, not a design flaw.

§What is structurally improved

A 51% adversary on Bitcoin must out-spend a few large pools. A 51% adversary on a mature Elektron Net must out-spend a global swarm of individuals who, by selection, are running their devices for reasons other than pure profit. Ideologically motivated hashrate is sticky hashrate. It is the hardest kind of hashrate to buy.

§The Gold-Panner

There is an older image that captures what this design is trying to do.

Before industrial mining came to the American West, the early gold rushes were prospectors with pans, sluices, and stubborn optimism. The Colorado rivers were full of strangers who built small towns, traded news, watched each other’s claims, and occasionally — not often, but occasionally — found a nugget. The economy of those rivers was not concentrated. It spread itself across the continent precisely because the work was small enough for individuals.

Industrial mining made that economy obsolete. The same has happened to Bitcoin: the romance of running a node and finding a block is now mostly just romance.

Elektron Net puts the panners back at the river.

A Bitaxe under a desk is a pan in a stream. Most days it finds nothing. But the stream is moving, the 120-second window keeps opening, and every so often someone in Stuttgart or São Paulo or Sapporo will check their device in the morning and find a block.

Everyone, eventually, finds a nugget.

That is not just a feature. It is the security model.