🚀 go-pugleaf

In article <DOCONNOR.92Nov20100231@potato.sedona.intel.com>, doconnor@sedona.intel.com (Dennis O'Connor) writes:
> 
> mu@terapin.com (Mu Hong Lin) writes:
> ] Will someone explain ( technially ) why RISC chips are so much faster
> ] than CISC chips??
> 
> RISC chips don't need to support obsolete mis-features from 10 years ago.
> CISC chips generally do, because they are all old architectures.
> --
> Dennis O'Connor			doconnor@sedona.intel.com

  To be fair, that answer may fit a particular part, but has nothing to do with
the RISC/CISC question.

  The most simplistic answer is that RISC, being relatively simpler to build,
can be made to clock faster. That way it may take five RISC instructions to
duplicate one CISC instruction that takes five cycles, but it will finish first
because it is running faster. For example, as I recall the first HP-PA
(Precision Architecture) chip set was actually bonded to large copper
"Finstrates" to dissapate heat because it ran so fast.

  In a more general sense, there is no clear winner in the RISC/CISC wars,
although RISC seems to be gaining the edge, based on announcements from the last
couple of years.

    Cheers,

    Tom McConnell

 Tom McConnell          |     Internet: tmcconne@sedona.intel.com
 Intel, Corp. C3-91     |     Phone: (602)-554-8229
 5000 W. Chandler Blvd. | The opinions expressed are my own. No one in 
 Chandler, AZ  85226    | their right mind would claim them.
-- 
 Tom McConnell          |     Internet: tmcconne@sedona.intel.com
 Intel, Corp. C3-91     |     Phone: (602)-554-8229
 5000 W. Chandler Blvd. | The opinions expressed are my own. No one in 
 Chandler, AZ  85226    | their right mind would claim them.