I should have stated that we're assuming the actual total hashrate remains constant. Obviously this is not what would actually happen - the rest of the post discusses ways to counter the economic forces at play pushing total hashrate down using only soft forks. The increased variance is still unaccounted for (pool operators would have to deal with this somehow)...and we still have larger block intervals even with compensation. And the practicality of deployment and usability are clearly problematic, to understate things. It's merely an exercise seeking the theoretical limit of what's actually possible to do with a soft fork. On December 26, 2015 8:09:18 AM PST, Tier Nolan via bitcoin-dev wrote: >On Sat, Dec 26, 2015 at 8:23 AM, Eric Lombrozo via bitcoin-dev < >bitcoin-dev@lists.linuxfoundation.org> wrote: > >> Unfortunately, this also means longer confirmation times, lower >> throughput, and lower miner revenue. Note, however, that >confirmations >> would (on average) represent more PoW, so fewer confirmations would >be >> required to achieve the same level of security. >> > > >No, the re-target compensates so that the number of blocks in the last >two >weeks is 2016. If a soft fork forces miners to throw away 25% of their >blocks, then the difficulty will drop by 75% to keep things balanced. >Throwing away 75% of blocks has the same effect on difficulty as >destroying >75% of mining hardware. > >The block interval will only increase until the next re-target. > >Slowly increasing the fraction of blocks which are thrown away gives >the >re-target algorithm time to adjust, so it is another advantage. > >If the rule was instantly changed so that 95% of blocks were thrown >away, >then there could be up to 40 weeks until the next retarget and that >would >give 200 minute block times until the adjustment. > > >------------------------------------------------------------------------ > >_______________________________________________ >bitcoin-dev mailing list >bitcoin-dev@lists.linuxfoundation.org >https://lists.linuxfoundation.org/mailman/listinfo/bitcoin-dev