Tracy R Reed wrote:
Dexter Filmore wrote:
Intel did that on Itanium. Performed like a dead nun in a closet
filled with concrete.
But they didn't just throw out the old cruft, they went with a whole new
CPU design philosophy. VLIW and other stuff IIRC.
Yes, and VLIW is what kills it. I still remember the DEC architecture
guys cheering when they found out that Intel chose VLIW. DEC had been
trying for 15 years to make VLIW perform and couldn't; they (rightfully)
believed that Intel couldn't do it either.
Of course, then you had the hostile giveaway from DEC to Intel that made
it all moot.
That is completely
different and does not suggest that breaking backwards compatibility
every now and then is a bad thing.
Agreed.
Itanium has been out for years now and I have still never actually laid
eyes on one or seen an ad for one. I wouldn't know where to buy one if I
wanted it. No wonder it's dead.
Well, it served its purpose. It wiped out MIPS, SGI, HP, and DEC as
microprocessor competitors and left only IBM. I don't know how the
Itanium funding is going, but there really isn't any point to doing much
more with it. The x86 line performs better, IBM isn't going away, and
AMD is x86 compatible and is chewing at Intel's market share.
Taking Itanium money and dumping it at the mainline x86 is going to be
the most profitable choice, at this point.
(I'm not ignoring the backward compatibility post, but I don't have time
to craft a useful answer until I get past my Japanese test today.)
-a
--
[email protected]
http://www.kernel-panic.org/cgi-bin/mailman/listinfo/kplug-list
