PPC is better better better!!!!

[Hammer]

@HyperionMP

Note that Intel has IPC bias “Pentium M” for “thin and light” and embedded markets. It's available from 400Mhz to 1.6Ghz. From tech news sites (e.g. ZDNET),“Pentium M” @ 1.6Ghz is said to be equal with Intel's Pentium 4 @ ~2.2Ghz-2.5Ghz on integer performance (benefits office type applications).

"Pentium M" @1.1Ghz, it consumes ~12 watts.

Just keeping things into perspective in relation to Motorola’s G4 targeted market.

Reference;
http://www.intel.com/design/intarch/prodbref/25269601.pdf

[Hammer]

 

It does however have Altivec which is far superior to SSE2.

Which version of SSE2 implementation (i.e. Opteron’s SSE2 or Pentium 4’s SSE2)?

The 970 is still somewhat slower that the very fastest x86 CPU's from Intel and AMD, at least in integer performance.

In SPECfp rate context, 2P Opteron @1.8Ghz was ahead** of 2P PPC 970 @2.0Ghz.  

[Hammer]

strobe wrote:
Apple's prices are competative with dual Xeon systems.

I don’t recall current Xeon prices being static (e.g. June 2003 -> August/September 2003) …

Comparing the current Xeon prices to August/September release of Apple’s PowerPC 970 is inherently flawed.

[Hammer]

Jaruzel wrote:

Hammer said:

Note that Intel has IPC bias “Pentium M” for “thin and light” and embedded markets. It's available from 400Mhz to 1.6Ghz. From tech news sites (e.g. ZDNET),“Pentium M” @ 1.6Ghz is said to be equal with Intel's Pentium 4 @ ~2.2Ghz-2.5Ghz on integer performance (benefits office type applications).

I have a brand spanking new centrino 1.6Ghz laptop, and I can say quite definately that 1.6Ghz on these new Pentium Ms match a P4 2.2Ghz.

Runs bloody hot tho...

Note that, @ 1.6Ghz it jumps to 24.5 watts from 1.1Ghz's 12 watts. Other factors may contribute with the heat generation.

[Hammer]

PPC vs x86 is pretty much a no brainer. PPC is a better architecture, overall.

Refer to http://ars-technica.com/cpu/4q99/risc-cisc/rvc-1.html

Modern X86 processors are merely post-RISC like CPU with a hardware decoders/translators/emulators.  You will also notice that PPC 970 has a crush/decode stage.

Referring to http://www.arstechnica.com/cpu/03q1/x86-64/x86-64-4.html

In addition to fattening up the x86 ISA by increasing the number and sizes of its registers, x86-64 also slims it down by kicking out some of the older and less frequently used features that have been kept thus far in the name of backward compatibility.

When AMD's engineers started looking for legacy x86 features to jettison, the first thing to go was the segmented memory model. Programs written to the x86-64 ISA will use a flat, 64-bit virtual address space. Furthermore, legacy x86 applications running in long mode's compatibility sub-mode must run in protected mode. Support for real mode and virtual-8086 mode are absent in long mode and available only in legacy mode. This isn't too much of a hassle, though, since, except for a few fairly old legacy applications, modern x86 apps use protected mode.

//AMD's clean up of X86 ISA. It will not affect most WinNT/Win32/Linux based applications. **It can still run at least from DOS 3.30 to DOS 7.