Bitcoin Knots Plans BLAKE2b Fork Test After BIP‑110 Collapse

Bitcoin Knots is scheduling a test on Sunday, Aug. 30, to replace the SHA‑256d proof‑of‑work used by Bitcoin with BLAKE2b on a separate chain. The rehearsal follows the collapse of the earlier BIP‑110 fork, which produced only two blocks before stalling.

Planned test and release schedule

Developer Luke Dashjr asked SHA‑2 miners to stop mining ahead of the test. The upcoming release candidate 29.4.1rc4 is intended to create the final SHA‑2 block before the switch. If the test succeeds, a final 29.4.1 release could lock in the new chain on Sept. 1; otherwise another release candidate would be issued and the chain would revert to the last SHA‑2 block.

Why BLAKE2b?

The new proposal aims to avoid reliance on existing Bitcoin miners by using hardware built for BLAKE2b mining. Devices originally designed for Sia, such as Bitmain’s Antminer A3 and Goldshell SC5, are cited as compatible. Testnet4 mining instructions and a DATUM Gateway fork have been published.

Unresolved technical details

• Activation height for the mainnet fork has not been fixed.
• Block‑weight limit differs between sources (700,000 WU vs 800,000 WU).
• Reduced‑data restrictions are slated to expire in 2027.
• Public release pages did not list rc4 or a final 29.4.1 build as of Aug. 29.

Mining capacity concerns

Preliminary calculations suggest the initial difficulty would need roughly 870 TH/s to keep 10‑minute blocks, while measured testnet4 capacity is only 50–70 TH/s. The test will reveal whether enough BLAKE2b‑compatible hash power is actually pledged.

Economic and ecosystem challenges

Even if blocks are produced, the fork must gain support from exchanges, wallets, custodians, explorers and Lightning implementations. Light‑client compatibility is outside Bitcoin Knots’ scope, and replay protection is limited to an optional SIGHASH_UNIFIED mode.

The upcoming rehearsal will therefore serve as a first indicator of whether the BLAKE2b design can sustain a viable chain after the BIP‑110 failure.