Post AS96K3L9pVfnsnbncW by mulad@mastodon.sdf.org
 (DIR) More posts by mulad@mastodon.sdf.org
 (DIR) Post #AS96K2A8D1G8EJRTnM by mulad@mastodon.sdf.org
       2023-01-29T20:02:08Z
       
       0 likes, 1 repeats
       
       RetroBytes' recent video on ARM is pretty good, and suggests one of their strong suits is the testing/verification that they do -- something that may be a hangup for RISC-V as companies seek to compete using that.(this channel uses stock footage loops a bit too heavily, but I otherwise find pretty enjoyable)https://www.youtube.com/watch?v=t59EtDxpYmM
       
 (DIR) Post #AS96K3L9pVfnsnbncW by mulad@mastodon.sdf.org
       2023-01-29T20:04:20Z
       
       0 likes, 0 repeats
       
       It's fascinating to me that the 32-bit ARM1 had around 25,000 transistors, which is actually on par with the 16-bit Intel 8086 from seven years earlier. Ken Shirriff recently counted 19,618 actual transistors, but space for potentially 29,277 in ROMs and programmable logic arrays that aid in decoding instructions and controlling chip behavior. Intel's 32-bit 80386 started with around 275,000 transistors, by comparison http://www.righto.com/2023/01/counting-transistors-in-8086-processor.html