[cap-talk] Butler Lampson's upcoming talk

Jack Lloyd lloyd at randombit.net
Wed Apr 9 12:32:30 CDT 2008


On Wed, Apr 09, 2008 at 06:16:27PM +0100, Sam Mason wrote:

> I was under the impression that RISC was well and truly alive.  Most of
> the newer (desktop/server) processor designs (i.e. PowerPC and Cell)
> seem to be basically RISC designs.  I also thought that most embedded
> processors were RISC as well, except maybe the really tiny ones.

Even x86/x86-64, the last suriving CISC (?), is basically a RISC these
days - everything [*] since the Pentium Pro era works by dynamically
translating x86 instructions into a series of fixed-length micro-ops
which are a very RISC-like load/store architecture with a large
register window and few complex instructions. But then, a modern
PowerPC like the PPC970 is arguably CISC - certainly it is much more
complex ISA than the SPARCv7 or early MIPS had.

I feel the CISC/RISC distinction was a lot more clear cut (and more
interesting) when you were comparing the VAX or PDP with the SPARC or
Alpha, rather than, say, the STI Cell with Intel's Core2.

[*] From Intel and AMD, I'm not sure about VIA's x86 CPUs

-Jack


More information about the cap-talk mailing list