Talk:Scoreboarding

Fi[FU] ← dst; Fj[FU] ← src1; Fk[FU] ← src2; Qj[FU] ← Result[src1]; Qk[FU] ← Result[src2]; Rj[FU] ← not Qj; Rk[FU] ← not Qk;

etc. I don't understand what's meant here. Can someone explain scoreboarding in more detail? Thanks! --Abdull (talk) 17:40, 12 August 2008 (UTC)


 * it's directly out of the patent (1960s) which would have been deliberately written in the most bare summary obtuse fashion as possible. i could not in the least understand it either.  the academic literature cannot properly understand it either, and debases its value and importance as a result.  FIVE MONTHS later, with help from Mitch Alsup on comp.arch i had a full in-depth understanding of the algorithm, and had augmented it to be precise-capable. see https://libre-soc.org/3d_gpu/architecture/6600scoreboard/ where i documented it as much as i could. the absolutely critical thing to understand is that Q-Tables are a SMALL PART (about 15%) of the overall algorithm. User:lkcl

= Updating page from Libre-SOC resources =

i'm going to start updating this page using images from libre-soc.org's wiki, particularly the images. such as https://libre-soc.org/3d_gpu/dependence_cell_pending.jpg however they need redrawing as some of the originals are copyright Mitch Alsup reproduced with his permission and require crediting him as the source. i am perfectly fine with this on libre-soc.org but for wikipedia we need to only upload material that has the appropriate license. hence redrawing them. User:lkcl — Preceding unsigned comment added by 92.40.200.146 (talk) 01:52, 20 May 2021 (UTC)