P4 piece hertz reader's feelings
An apprentice boffin lets rip
Posted in Letters, 26th January 2001 13:47 GMT
Free whitepaper – Dell/EMC CX4 and Dell PowerEdge blades
Math/CS major James Perry has got a bee in his bonnet about Mike Magee's Pentium 4 Foster may sink the Itanic.
I'm writing this since I think someone needs to correct the general misconception surrounding the Itanium that many people have - particularly Register journalists.
In particular, in your article you state: "Will they [Itaniums] ever reach the giggle hurtz speeds that Foster and the Pentium 4 seem capable of now?"
The IA64's architecture is completely different from that of the IA32's, particularly when it comes to what happens each clock-cycle. The Itanium uses a technique called VLIW(Very Long Instruction Word) where each instruction package is actually 3 instructions that the CPU procedes to execute /in parallel/. Hence, during each clock-cycle an Itanium executes at least 3 instructions whereas a P4 runs barely more than 1.
Now if the Itanium were to run only a single instruction per clock, like the P4 does, it'd be clocking in at around _2ghz_. So excuse me, but what was that about 'giggle hurtz' speeds? (And yes, I know I'm simplifying things a lot. The IA64 architecture specifies even more goodies such as speculative execution.)
You guys should bother checking your facts before writing up this BS.
Right on brother. Anyone want to reply to this one?

Enabling The Agile Data Center
Analyst Keynote: The Register Agile Data Center Summit
Analyst Keynote: The Register Agile Data Center Summit

Dirty, dirty PCs: The X-rated picture guide
Top 500 supers - rise of the Linux quad-cores
Early adopters bloodied by Ubuntu's Karmic Koala
Sign up, sign up for The Register IT security newsletter