IDF Fall 2006 - Day 2: Opteron vs. Xeon, SSE4 & Intel's Torrenza competitor Announced
by Anand Shimpi & Virginia Lee on September 27, 2006 1:12 PM EST- Posted in
- Trade Shows
Intel Announces new Instructions for 45nm, SSE4 and more
Unexpectedly, Gelsinger announced the 50 new instructions that will be present in its upcoming 45nm processors (Penryn and Nehalem) - it's "SSE4 and more".
These are some of the workloads the new instructions will help accelerate
Today Intel will be publishing a white paper detailing the new instructions and as they get implemented in silicon, Intel will disclose more information about the instructions.
Intel's approach to SSE4 disclosure is a significant departure from how previous x86 instruction extensions have been disclosed by the company. It's another example of Intel's more open and friendly approach to its upcoming roadmaps; we're hoping Intel will only continue to be more open with its future plans, and we do have AMD to thank for keeping the pressure on Intel to bring about such a dramatic change in behavior.
Facebook
Twitter
RSS
18 Comments
View All Comments
jiulemoigt - Wednesday, September 27, 2006 - link
Actully it's even funnier than that as all of the 1207 mobo's are dual or multi socket mobos at this point so it is most likely to be a AM2 940 pin 1000 chip or the Op 285 which is a socket 939. Any way you look at that it is questionible which is interesting since I was under the impresion that the core2 were faster than anything amd had even if the mobo chipsets have issues.JarredWalton - Wednesday, September 27, 2006 - link
I'm not sure if Anand already updated this, but both systems were dual socket and it was a "Socket F 2.80 GHz" processor config in the AMD unit, not AM2, Opteron 285, or some other hypothetical config to show Intel in a better light. The benchmark, on the other hand, is a different story as it could be just about anything. :)theteamaqua - Wednesday, September 27, 2006 - link
this company is in big troubleAmpedSilence - Wednesday, September 27, 2006 - link
Based on what?The Core 2 Duo is doing well and trumps AMD64 for the time being. They are releasing a quad-core almost a year before AMD.
What are you using as a basis for this conclusion?
btw, i have three AMD64 machines (one X2 and two AMD64's).
Viditor - Wednesday, September 27, 2006 - link
closer to a half year...at the very most it will be 7 months.
He might be referring to Torrenza. It's probably the most underestimated and misunderstood advancement AMD has announced. Remember that there are already coprocessers that can just drop into an Opteron ssocket, and IBM is already shipping servers with HTX connections. Intel will be another 1-1.5 years before they are ready to do this...and there isn't any idea what kind of support they will end up with at that point.
psychobriggsy - Wednesday, September 27, 2006 - link
Quad-core on two dies, utilising their experience from the Smithfield panic reaction to AMD's X2. However it shows that once Intel gets woken up, they don't go back to sleep after doing 'enough', they'll continue until they have the lead.AMD fell asleep after dual-core, indeed after K8 considering dual-core was meant to be a possibility from day one. It is costing them now, apart from their platform work which is still ahead. I think AMD should have investigated dual-die MCMs for quad-core, instead 4x4 is a reactionary hack solution (that'll work nevertheless, and it'll have good memory bandwidth).
Calin - Thursday, September 28, 2006 - link
AMD would better ramp the 65nm technology as soon as possible - only after that, quad cores will become a real/profitable possibilityPirks - Wednesday, September 27, 2006 - link
he's using some green smelly stuff as his basis, isn't that obvious ;))