OT: was: Re: 32 or 64??
ubuntu at tigershaunt.com
Tue Feb 2 16:55:46 UTC 2010
Smoot Carl-Mitchell wrote:
> On Tue, 2010-02-02 at 22:58 +0800, Chan Chung Hang Christopher wrote:
>> Oh? I thought the Itanium was slow as molasses? Or maybe I missed
>> something about the Itanium II?
> The Itanium is not slow, but it took a long time for Intel and HP to
> work the kinks out and get the performance up. Also keep in mind the
> Itanium is a very pipelined architecture. That kind of design requires
> very sophisticated compilers to get decent performance.
> The Itanium's market penetration problem is due to the relatively low
> yields of usable chips and the lack of penetration into the low end cpu
> market. In order to get per chip costs down, the number of chips sold
> needs to be very high. The only way to do that is to penetrate the low
> end market with slower chips of the same design. That never happened.
> Intel's original goal with Itanium was to replace the x86 architecture
> and dominate all market segments from notebooks to high end servers.
> You can make the argument the Itanium's slow development cycle allowed
> the 64 bit extensions to the x86 architecture to "catch up" to the
> Itanium and perform just as well for most application loads and eat away
> from the bottom the market intended for the Itanium.
I couldn't help but notice recently, much to my cost, that Intel is
removing Virtualization extensions on all their current socket 775
CPU's.. (how are the i5 and i7 faring?). The only reason I can think
for this move is to prevent the low end cpus from competing with the
workstation/server market, since these cpu's are now more than fast
enough for many workloads to be simply io bound.
More information about the ubuntu-users