So freaking what! Itanium has to be behind in process because it can't fund process development, equipment purchase or anything like that.
Hello? It is 2008, not 2001. Intel announced 18 months ago that IPF would get process access parity with Xeon at 32 nm and had two full design teams working on new ia64 uarches. A few months ago Otellini said Intel had over 1000 MPU engineers working on next generation IPF products. This high level of effort is the result of sustained strong market acceptance.
Kilroy and Boyd Davis, general manager of Intel's Server and Platform Group Marketing, who also attended the meeting, reiterated Intel's commitment to Itanium, a non-x86 server chip that competes with SPARC from Sun Microsystems and POWER-architecture based processors from IBM, both RISC-based chips. A major Itanium customer is computer maker Hewlett-Packard.
While acknowledging that the Itanium market is not as high volume as Xeon, it's still a profitable business. "We want to bring volume economics to this market," Davis said.
IPF has been profitable for nearly two years now and it continues to grow sales rapidly. That's a record AMD can only look on with envy. ;-)
That extra time (2-3 years) allows it to get more out of the older process by throwing lots of transistors into its huge dies.
The relative die size, cache, and transistor count lead held by IPF over Xeon has steadily decreased over the past few years. Beckton Nehalem is almost as big as Tukwila. The funny thing is IPF bashers's prediction that once x86 chips got big caches too they would crush IPF was proven just as erroneous and illusionary as all the rest of their ideologically driven BS.
At 180 nm, 130 nm, and 90 nm IPF chips blew away the best x86 server chips made in the same feature size and in a few months Tukwila will do the same for 65 nm. In about two years or less both IPF and Xeon MP will ship in 32 nm around the same time. End game for bashers, the architectural advantages of EPIC will be clear for all to see.
News 
Market Data 
Discover 
AI
