Author: Andreas M. Antonopoulos 2013-11-08 19:49:11
Published on: 2013-11-08T19:49:11+00:00
Nicholas Weaver reported on November 3rd that some pools have started delaying blocks, suggesting selfish mining tactics. Other reasons for delayed block propagation were dismissed. It was unclear whether this report is accurate or due to other network issues. Daniel Lidstrom presented an analysis showing that miners would withhold their blocks to increase expected revenue rather than for greater probability of winning the next one. If a cabal has fraction q of total hashing power and is n blocks ahead on a secret branch with r_tot coins mined and r_next as its next block's reward, then it will withhold its chain if q > 1 - (1 + r_tot / r_next )^(-1/n). This means that if the cabal has mined its first block off the public chain and the block reward is stable, it needs q > 50% to profitably withhold. Mike Hearn suggested that once the ASIC race calms down, people will dissipate from large pools to eliminate their fees for extra profit. He also stated that a few hundred pools or lots of miners on p2pool would make many of these theoretical attacks not very relevant. Peter Todd mentioned that GHash.IO has physical control of a bunch of hashing power, and they are a bit short of 30%, but behind-the-scenes deals could fix that. He also presented the idea of a witch hunt against large pools. Finally, Peter Todd and Gavin Andresen agreed that miners should stop mining on pools with more than 1% hashing power and mine on smaller pools or p2pool instead.
Updated on: 2023-06-07T19:36:31.017542+00:00