tag:blogger.com,1999:blog-11788780.post114403222492114119..comments2023-12-29T13:22:33.104-08:00Comments on JJinuxLand: Hardware: Smarter Memoryjjinuxhttp://www.blogger.com/profile/03270879497119114175noreply@blogger.comBlogger2125tag:blogger.com,1999:blog-11788780.post-1144203054961328522006-04-04T19:10:00.000-07:002006-04-04T19:10:00.000-07:00Wow! You mean I'm not a quack for thinking along ...Wow! You mean I'm not a quack for thinking along these lines? Cool!jjinuxhttps://www.blogger.com/profile/03270879497119114175noreply@blogger.comtag:blogger.com,1999:blog-11788780.post-1144199015673799042006-04-04T18:03:00.000-07:002006-04-04T18:03:00.000-07:00Why stop there? While you're at it, add SIMD inst...Why stop there? While you're at it, add SIMD instructions, which are also easy to do.<BR/><BR/>Of course, if you go down this line of thinking too far, you end up at something like <A HREF="http://iram.cs.berkeley.edu/" REL="nofollow">iRAM</A> -- solve the whole Neumann bottleneck at once. (I'm all for it. The current CPU/RAM architecture is looking sillier all the time.)<BR/><BR/>- timAnonymousnoreply@blogger.com