The eCash Alpha, Days 16 to 28: Drivechains Come Alive & 150k pECX Gets Burned Before Beta Net

September 20, 2026 · 9 min read

Two weeks ago, in The eCash Alpha, Days 4 to 15, the alpha was a network learning to breathe. Hashrate hangovers, orphan blocks, the first withdrawal bundle, and a $320 million warning shot from Liquid.

Since then the alpha finished its job, the community finished its homework, and on September 19 the beta fork fired on schedule. This is the story of the final fortnight, and of the odd fortnight that came right after it.

The Blukies' Grand September Race

While eCashers argued about where to point the hash, Luke Dash Jr.'s BIP-110 fork, the Blake2b chain the chat promptly renamed Lukecoin, staged the loudest liquidity theatre of the month. One illiquid exchange, a handful of cult bidders, and a price that screamed $1,000, then $1,300, while five coins could crater it.

Freedom Monk asked the obvious question: "has to be wash trading and exchange manipulation right?" Don Wonton supplied the verdict of the age: "B2B will probably have its own contentious hardfork within a year. Purity spiral." Within days he was right in spirit. Internal discord erupted, a blacklist PR against the biggest pool appeared, and the community started debating 45 day, then 360 day, coinbase maturity to stop the dreaded shitcoin miners from dumping. kac summarized the schism beautifully: "there already is a civil war... layer 2 labs and layer tree labs... l3l."

Some of our own got paid. Felix Ha split and sold at $400, owning the classic seller's curse: "it'll probably go 100x." Coinholio pulled a 9.5x return and held a little back, "in case it does take over BTC." Then the truth arrived. Renters dumped, buyers went home, and Lukecoin slid back from #8-on-coinmarketcap-air to a few hundred dollars on the two sketch venues that would host it. Louis's orderbook autopsy was surgical: one USDC bid of 105 coins at $1,025, and everything else thin enough to vanish on contact. "24H Volume (USDT) 2.303K. Illiquid POS."

The useful part, beyond the entertainment, is what it normalized. Paul put it plainly in chat: "They kind of trailblaze it for us, a little. Re-normalize forks!" A second Bitcoin fork existing, trading, arguing with itself on Discord, and surviving is market education for the fork people will actually care about on October 31. It also taught the entire sector what thin books produce: fake prices, fake confidence, and a permanent reminder that an exchange listing is not liquidity. For the full comparison of the two forks and their philosophies, see Two Bitcoin Forks, Two Philosophies.

Liquid Already Taught the Lesson Twice

The previous article ended as Liquid's 4,000 BTC pegout disaster was still burning through the megaphones. In the fortnight after, the aftermath turned into a full media cycle. Paul's X thread hit 1.5 million views in twelve hours. The "white hat" and Blockstream negotiated in OP_RETURN, the attacker kept 598.5 BTC as a 15% tip, and the federation resumed block production with pegouts disabled.

Serious Cat floated the theoreme that it was an inside job to sunset Liquid, then watched the onchain chat logs make less sense of that, then watched Blockstream refuse a "ransom" on stage as if the room hadn't already been robbed. Marcus found the whole thing easy to summarize when he was called the person who soaked Liquid to build pegs: "brother, this is extremely bullish for ecash."

The substance, which Vlad's article covered properly, is worth repeating because October 31 will not give us a second chance to learn it. A federation fails in secret, all at once. A BIP-300 peg fails only in public, over months, with hashpower voting visibly on each block. The same week that proved this to millions of impressions, Lukecoin was proving something else to the same people: forks are normal again, and custody questions matter even for coins that look small.

Six Sidechains Walk into a Rally

On the development front, Bjørn Olav and the sidechain crew ended the alpha with more activated work than anyone had a right to expect. Thunder picked up a built-in block explorer, and light clients for Thunder, BitAssets, BitNames, Photon, and CoinShift all landed inside BitWindow a week before beta day.

Wiredancer's pool added Thunder to its setup so withdrawal requests could actually be checked and processed. Felix Ha deposited to every chain that would take it and reported, in the most casual voice in the world, "I'm pretty sure light clients for all the rust chains are live now! Thunder, BitAssets, BitNames, Photon and CoinShift, at your pleasure."

The neighborhood watches grew teeth. Michael Blowes announced FreeBank, an admitted experiment in Scottish free banking lineage, credit creation, bills of exchange, settlement in ECX with a gold unit of account, and no central banking, at slot 130. It is not polite, and it knows it: "At the moment I see this project more as tcp/ip. Plumbing for after the apocalypse."

He also made the sidechain's bid while noting the AI workforce in his own words: "I had about 5 clanker sessions running in parallel for beta launch... and mowing the lawn." JK, meanwhile, pushed Elements+ toward slot 24 with an Esplora and a growing set of Simplicity experiments, and Joe started writing the RPC endpoints the prediction market UIs will need on Truthcoin.

BitNames quietly became the most social thing on the chain. Decentralized handles, IP commitments hashed into registrations, and PGP-style encrypted chat included in the same wallet. BTCapsule squatted "ecash" in the first day of registration and passed a little encrypted note back and forth with Felix, which is how the feature got its first user review: lmao, it's a good site though, from JK to Coinholio's faucet. The joke about DNM adoption came right after, because the obvious UI for reputation and escrow is sitting in plain view.

The Great Burn

The alpha's retirement account had a simple rule. To claim mainnet ECX from alpha pECX, you had to burn at least 1,000 of them to the Eater address, with an OP_RETURN carrying a return address, for a 1000:10 ration in November. BitWindow got a burn button a week before the deadline. Wallets got consolidated. UTXO sprawl met the consolidation tool, sometimes choked on it, and then still burned. Don Wonton, reporting from the deep end of the UTXO pool: "I have too many UTXOs to even consolidate."

The bookkeeping closed with roughly 150,000 ECX burned at the end, with the largest single bundle coming in around 12,000. Juan Libertad's rental dynasty closed the books he opened, and Don Wonton accidentally found the alpha's final block while typing pool info. Even the mempool cleared its throat: 18,000 stuck transactions had been worrying Vlad a day earlier. On the 19th, Wiredancer's explorer confirmed the fork watched by everyone, and the beta explorer lit up on schedule.

The Small Change Problem

Here is where the fortnight got genuinely good. The 1,000 pECX floor excluded exactly the kind of person this launch needs tomorrow: the small miner who ran a node through orphan blocks and dust storms, the investor whose paperwork said 167 coins, the tester with 0.3 ECX of goodwill and no way to redeem it.

Coinholio saw it early: "Too late to mine 1000 now." Laurent Colman put a formal plea in for a lower threshold. No lowering came, because the floor exists to save Paul's time, not to hurt anyone. What came instead was better. The community built two offramps itself.

Marcus stitched waECX onto Solana, with token ID BZEmay...JncmhZ, concentrated liquidity at Orca, Jupiter already routing properly, and a custodian guarantee you could exit against immediately for USDC or USDT, or hold to redeem 100:1 into mainnet ECX. It traded around $0.92 per wrapped alpha coin and even stayed tradeable in the six week bridge window.

Vlad ran the other rail, the human one: send him your small balance, he burns it all into one transaction, and he pays you your 1% in November. More than 2,100 pECX came in that way, a spreadsheet of names and percent promises attached. "banker santa claus," he said. The effect was real. People who would have gotten nothing will get approximately 21 ECX split between them at launch. The last coin, sent 166 pECx too late, turned into the proof that some of us always arrive five minutes after the last class.

Beta Day

September 19, block 967,680. BitWindow offered the migration prompt, the betanet config updated with the Esplora URLs, and the fork happened on time with pools queued. eCPool.tech renamed itself on schedule. Louis's ePool.cash automation stalled and finished a few hours late, still same day.

Juan Libertad pointed 140 petahash at a 1G network inside the first hour ("oh boy Juans at it again"). Braiins blacklists arrived for the high‑frequency template pushes, and eCPool opened its Braiins-only stratum on port 3337 to solve it professionally. The beta explorer went down with two abnormal hardware failures at once around midday, and was back by mid-afternoon. The temporary Cloudflare explorer Wiredancer spun up even got proper pool tags.

The miners, in their awkward way, validated the whole thing on schedule. Six activation proposals landed within an hour of launch and sat at 1,009 acks each. Pool share stayed roughly collegial while Juan's 800 PH turbominer took the headlines, and the retarget to over 4G happened in about a day, where the alpha had needed a week.

The szats.world faucet refilled for beta before night fell. By the end of Sunday, six slot proposals were being bmm'd, FreeBank's slot 130 was active, Elements+ waited on its slot 24 to clear, and the network had gone from proposal to retarget in one epoch instead of a weekend.

Why October 31 Is a Different Beast

No hopium, only the receipts. The alpha ended with working BMM, slot proposals, a working burn, and honestly bad blocks to complain about. The beta began with activation proposals inside the hour, governance voters weighting their hashrate, withdrawal bundles being processed, and a retarget inside a day. Liquidity is not hope: CoinShift's OTC rails from Rob, Elements+' AMM-adjacent line from JK, the Solana wrap from Marcus, and Vlad's burn pool form four market paths before launch.

Price discovery, which Lukecoin is teaching everyone to mis-handle, is planned on-purpose this time. Paul's public framing to Peter Rizzo, the year one target of roughly one percent of Bitcoin's market cap, "beating Bitcoin Cash," has an anchoring investor baseline around $30 discussed openly in chat, and the Oct 31 fork has forty days of beta proofs packed into it already.

The creator is ranked only by the receipts he collects per decade: the misinformation archive, the fee economics paper, the thread that won September, and now crypto.news running the conclusion as fact: "All soft forks since Taproot have failed to activate, and BIP-110 was no exception."

Paul Sztorc's most useful skill here is not invention, it is patience. Ten years of the same argument, in public, receiving the same dismissal, with no deviation. We watched the market catch up inside two weeks: Lukecoin's fork failing to ban profit, Liquid's federation failing in public, and six drivechain proposals going bmm'd within an hour of the beta fork. That is the actual thesis, which is the same thesis, at last, arriving on time.