Roughnecks Pulls the Plug as the Fork Freezes
The group announced the decision shortly before 4 a.m. UTC on Aug. 9, following a team meeting around 3:40 a.m. Roughnecks stressed that the move was difficult but insisted it did not view the retreat as a defeat for BIP-110, instead calling it an “escalation to the next step.” It also advised anyone still mining the BIP-110 chain under Bitcoin’s current proof-of-work (PoW) algorithm to stop until further notice.
The timing leaves little ambiguity about the pressure behind the decision. Roughnecks had mined BIP-110 blocks 961632 and 961633 after Bitcoin fractured at the mandatory signaling point Saturday, but no third block followed. Meanwhile, Bitcoin’s dominant chain continued stacking blocks normally and quickly left the minority fork dozens of blocks behind.
Adding another warning light, Ocean’s displayed hashrate has collapsed. The pool reached roughly 36 exahash per second (EH/s) on Aug. 8 before falling to about 1.25 EH/s by Aug. 9, a decline of roughly 96.5%.
Ocean had become the central mining venue associated with BIP-110 signaling, particularly through its DATUM system, which lets individual miners construct their own block templates and choose whether to signal the proposal.
Bitcoin’s Difficulty Became the Minority Chain’s Trap
The problem facing the breakaway chain is brutally mechanical. When BIP-110 nodes rejected Bitcoin’s non-signaling block 961632 and accepted Roughnecks’ competing block instead, their new chain inherited Bitcoin’s existing mining difficulty. That difficulty had just climbed to roughly 127 trillion, a level calibrated for the enormous amount of computing power securing the dominant Bitcoin network.
Difficulty determines how hard miners must work, on average, to find a valid block. Bitcoin normally adjusts that difficulty every 2,016 blocks to keep block production near one block every 10 minutes. A minority chain carrying only a tiny sliver of Bitcoin’s former hashrate does not receive an immediate discount. It must keep wrestling with the same difficulty until it reaches another adjustment point.
That created a punishing feedback loop for BIP-110 miners. Less hashrate meant blocks arrived dramatically slower, but slower blocks also meant reaching the next 2,016-block difficulty adjustment became increasingly distant. Monitors projected that completing the necessary period could take hundreds of days under severely reduced mining power.
BIP-110 Never Won the Hashrate Battle
BIP-110, formally called the Reduced Data Temporary Softfork, was designed to temporarily restrict several methods used to place non-financial data inside Bitcoin transactions. Supporters argued those techniques, including Ordinals-related activity and large data payloads, increased storage and validation burdens while pulling Bitcoin away from its primary monetary purpose.
The proposal’s activation design became the decisive fault line. BIP-110 sought 55% miner signaling during a 2,016-block difficulty period. When that threshold failed, nodes enforcing the proposal were programmed to begin rejecting non-signaling blocks at height 961,632. Miner support never came close, topping out around 2.5% to 2.6% before the mandatory window began.
That meant the split arrived with the overwhelming majority of Bitcoin’s mining power on the other side. Antpool produced a non-signaling block 961632 that Bitcoin’s main network accepted, while BIP-110 nodes rejected it and followed Roughnecks’ competing Ocean-linked block. Roughnecks found one more block at 961,633, but the expected migration of additional mining power never materialized into anything meaningful.
BIP-110 proponent and Club Orange founder and CEO Matteo Pellegrini, along with several other supporters, appeared stunned by the result. “The lesson from BIP-110 is that 15-20% of the nodes is not enough to change Bitcoin consensus,” Pellegrini wrote on X. Pellegrini’s admission drew immediate pushback, with several respondents suggesting Pellegrini revisit Satoshi Nakamoto’s white paper, in which Bitcoin’s inventor states:
“They vote with their CPU power, expressing their acceptance of valid blocks by working on extending them and rejecting invalid blocks by refusing to work on them. Any needed rules and incentives can be enforced with this consensus mechanism.”
Two Blocks Expose the Economics of a Minority Fork
The result offers a stark demonstration of what happens when node rules and mining economics diverge. Software can reject blocks according to a different rule set, but those nodes still need miners willing to spend electricity and computing resources building the alternative chain. Without enough hashrate, confirmations slow dramatically and the chain’s usefulness deteriorates.
Roughnecks acknowledged that reality by describing continued mining under current conditions as largely wasteful while maintaining that the broader BIP-110 campaign is not finished. Possible next steps discussed around the effort include different coordination strategies or even a change to the proof-of-work (PoW) algorithm, although no replacement strategy had been established this weekend.
For users running BIP-110-enforcing software, the immediate problem is practical rather than philosophical. Their nodes are following a chain that has produced only two blocks since the split. Because both chains recognize many of the same transactions, users also face potential replay complications if coins are spent without deliberately separating activity between the two networks.
The next question is whether anyone replaces Roughnecks as a meaningful source of mining power. As of the group’s announcement, BIP-110 had not locked in or activated its transaction restrictions on Bitcoin’s dominant chain, Roughnecks had stopped operating under its name, Ocean’s displayed hashrate had fallen sharply and the minority fork remained stranded at two blocks.
Any revival now depends on fresh hashrate, a revised technical strategy or another coordinated attempt to keep the breakaway chain alive. Despite Roughnecks’ decision to stop pointing hashpower at the minority chain, at 8 a.m. EDT on Aug. 9, statistics still show a small fraction of SHA256 hashrate is still chasing BIP-110 block 961634.
