TCLUG Archive
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [TCLUG:3420] MIPS is dead



> This economies-of-scale thing is depressing, but inexorable. I
> always considered the PPC architecture (and RISC in general) to be
> technologically superior. I watch with amazement as an inferior
> CISC architecture wins out in the marketplace.

I hate to burst your bubble, but PPC is almost as CISC as Intel..  I'm
not certain about the specifics (I wish I was), but RISC (as defined by
IBM for PPC) doesn't have the traditional meaning.  IBM uses RISC to
stand for Reduced Instruction Set Cycles -- meaning that more work can
get done in a sincle clock tick.  PPC still has quite a large set of
instructions (over 200, I believe) where most traditional RISC systems
have less than 100.

Some more info can be found at
http://www.cs.uregina.ca/~bayko/cpu.html#POWER (be warned, it's quite a
large document -- ~250kB.  All text, though..)

Still, I like PPC.  It's still cleaner than Intel, plus the actual
architecture seems to be more flexible (PPC is based on a specification,
not a chip, IIRC)
-- 
.------ ----- ---- ---- --- --- -- -- - -  -   -    -
|               Mike Hicks | Linux User Since: 1.2.13
: http://umn.edu/~hick0088 | mailto:hick0088@tc.umn.edu
`              icq:6883760 | Current Kernel: 2.2.0-pre6