Author: Luke Dashjr 2017-04-08 14:59:12
Published on: 2017-04-08T14:59:12+00:00
In response to a proposal by Gregory Maxwell, Luke Dashjr has suggested that the mandatory commitment should expire. This would simplify the process of hardforking when required as the commitment in the "old block" would no longer be necessary. However, Dashjr is opposed to Jimmy Song's proposal regarding ASICBoost. He argues that ASIC-resistance elevates the barrier of entry for new mining chip manufacturers and gives an advantage to miners who can make use of it. Instead, Dashjr suggests fixing the vulnerability exploited by ASICBoost entirely to keep SHA2 ASIC-friendly or changing the PoW algorithm to one that is more ASIC-friendly. Nevertheless, he may not oppose Song's proposal if it gains better support than Segwit currently has, and concerns are addressed, such as Bitfury and Canaan stating they can compete using ASICBoost and the patents being licensed freely to everyone. Jimmy Song modified his proposal on April 8, 2017, after receiving feedback from Adam Back. Of the 32-bit version field, bits 16 to 23 are reserved for miners, and the witness commitment remains as defined in BIP-141, except that it is now required. As per Gregory Maxwell's proposal, BIP9 is modified so that bits 16 to 23 are no longer usable. Song's original proposal aimed to combine Gregory Maxwell's proposal with BIP-141 (Segwit) and add a slight modification that, in theory, makes overt ASICBoost more attractive to miners. The modification involves moving the 32 bits over to the Coinbase transaction as part of the witness commitment. The witness commitment becomes mandatory. This change brings ASICBoost out into the open, making covert ASICBoost much more costly and encouraging overt ASICBoost. These modifications can be deployed relatively quickly without imposing an extra burden on upgraded light clients.
Updated on: 2023-06-11T23:57:56.236305+00:00