By: someone (someone.delete@this.somewhere.com), November 17, 2010 9:31 am
Room: Moderated Discussions
Linus Torvalds (torvalds@linux-foundation.org) on 11/17/10 wrote:
---------------------------
>
>And sanity has long been lacking in the Itanium world.
>
>Linus
The McKinley core was a perfectly sane design for its
era - large die 180 nm single core FSB based MPU.
I doubt any of its contemporaries (Willamette, Power4,
EV6, US-III) would have held up as well being kept on
life support all the way to a 65 nm quad core.
The problem IPF has had is that Montecito and Tukwila
were each ~2 years late to market. Had they arrived
on schedule they would have largely ruled the non-x86
server world. That was mainly a project management
issue which came mostly from problems integrating 3
rather different groups of MPU designers into a unified
team and Santa Clara letting them run independently
of Intel's best practices for too long. If the spectre of
Beckton and its descendents isn't enough incentive
for Fort Collins to finally get its shit together and deliver
to POR I don't know what will.
---------------------------
>
>And sanity has long been lacking in the Itanium world.
>
>Linus
The McKinley core was a perfectly sane design for its
era - large die 180 nm single core FSB based MPU.
I doubt any of its contemporaries (Willamette, Power4,
EV6, US-III) would have held up as well being kept on
life support all the way to a 65 nm quad core.
The problem IPF has had is that Montecito and Tukwila
were each ~2 years late to market. Had they arrived
on schedule they would have largely ruled the non-x86
server world. That was mainly a project management
issue which came mostly from problems integrating 3
rather different groups of MPU designers into a unified
team and Santa Clara letting them run independently
of Intel's best practices for too long. If the spectre of
Beckton and its descendents isn't enough incentive
for Fort Collins to finally get its shit together and deliver
to POR I don't know what will.