On Thu, Apr 6, 2017 at 4:39 AM, Bram Cohen via bitcoin-dev <email@example.com> wrote: > > Asicboost also has the problem that it isn't treating the hashing as a black > box, and thus has impacts on what gets mined. In particular it creates an > incentive to make blocks smaller. That's a very unwanted effect, and > anything like it should be engineered out on principle.
This is an interesting point. If you have a precise description why it makes an incentive to make blocks smaller I would love to read it. Somebody asked and I didn't have an answer. I imagine you try several reorderings sometimes excluding certain branches of the merkle tree, permuting the branches you exclude or something similar, but I really don't know the algorithm in detail and I didn't want to say something inaccurate. _______________________________________________ bitcoin-dev mailing list firstname.lastname@example.org https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev