Re: Emulator, Inc

[mjmahon@aol.com (Michael J. Mahon)]

Sheldon Simms wrote:

>On Sun, 12 Oct 2003 23:17:14 -0700, Eric Smith wrote:
>
>> "Bryan Parkoff" <nospam@nospam.com> writes:
>>>     I know that Pentium 4 has some problems, but I stay on Xeon Pentium
>III
>>> until Pentium 4 will be fixed.
>> 
>> What problems are these?
>
>Cycle for cycle, It executes code a lot more slowly than a Pentium III,
>Athlon, or Athlon 64. That's all.
>
>It's obvious that the Pentium 4 is microarchitecturally cripped with
>respect to these other processors when you look at the benchmarks and
>see that a Pentium 4 Extreme Edition running at 3.6 Ghz with 2580kb
>of onboard cache is essentially the same speed as an Athlon 64 FX-51
>processor running at 2.2Ghz with 1088k onboard cache.
>
>All that, however, it hardly matters since the Pentium 4 is capable
>of being run at much higher core frequencies, which makes up for the
>cycle for cycle disadvantage.

In any processor implementation, there is a tradeoff between the
amount of work done in one clock cycle versus the time between
cycles.  Less work per cycle makes for faster clock speed,
more work, slower clock speed--and both designs may have
similar total time to accomplish a task requiring many instructions.

This tradeoff was characterized in the mid-90s in Microprocessor
Report as the "Brainiacs" (lots of logic per cycle, but slower clocks)
and the "Speed Demons" (little logic per cycle, but fast clocks).

Which approach is superior in net cost-performance is a dynamic
function of the state of semiconductor technology and microprocessor
design, but in the _marketplace_, higher clock speed has always
been an apparent advantage (whether it delivers the goods or not).

-michael

Check out amazing quality sound for 8-bit Apples on my
Home page:  http://members.aol.com/MJMahon/