From fabrice.drouin at acinq.fr Mon Mar 7 20:01:58 2016 From: fabrice.drouin at acinq.fr (Fabrice Drouin) Date: Mon, 7 Mar 2016 21:01:58 +0100 Subject: [Lightning-dev] We don't need R-Value, how OP_CODESEPARATOR saves the day In-Reply-To: References: Message-ID: Ok, got my answer from the thread you started on deterministic R values and which I had missed :) On 7 March 2016 at 20:09, Fabrice Drouin wrote: > Hi, > I may have misunderstood something, but with this scheme instead of R > + shachain/elkrem, then how do nodes react when old commit tx are > published ? It seems that they would have to store lots of signatures > ? > > > On 7 March 2016 at 08:28, Nicolas Dorier wrote: >> I managed to adapt the payment channel script to use less space. >> >> This version is winning around 32 bytes in the scriptpubkey (for the R >> value) as well as 70 bytes in the signature when Alice close the channel. >> >> Alice closing the channel: >> http://n.bitcoin.ninja/checkscript?savedScript=51225750-f245-45b5-a86c-6ca1e87dcafb >> >> Bob using revocation: >> http://n.bitcoin.ninja/checkscript?savedScript=c4d7ebaa-5a79-4c03-ab55-c499854f1e94 >> >> This amount to 100 bytes saved between the anchor transaction + commitment >> broadcasted. >> >> _______________________________________________ >> Lightning-dev mailing list >> Lightning-dev at lists.linuxfoundation.org >> https://lists.linuxfoundation.org/mailman/listinfo/lightning-dev >>