backtop


Print 73 comment(s) - last by Calin.. on Jul 27 at 4:05 AM

Native quad-core en route

Yesterday during AMD's Q2'06 earnings conference call, AMD's President and Chief Operating Officer Dirk Meyer recapped the long term plans for the company.  Although the bulk of his comments were already stated in during the June AMD Analyst's Day, Meyer also added the tidbit that the company plans "to demonstration our next-generation processor core, in a native quad-core implementation, before the end of the year."  Earlier this year, AMD's Executive Vice President Henri Richard claimed this native-quad core processor would be called K8L.

Earlier AMD roadmaps have revealed that quad-core production CPUs would not utilize a native quad-core design until late 2007 or 2008. To put that into perspective AMD demonstrated the first dual-core Opteron samples in August 2004, with the processor tape out in June 2004.  The official launch of dual-core Opteron occurred on April 21, 2005.  On the same call Meyer announced that that the native quad-core would launch in the middle of 2007 -- suggesting the non-native quad-core Deerhound designs may come earlier than expected or not at all.

Just this past Wednesday, Intel one-upped K8L plans by announcing quad-core Kentsfield and Clovertown will ship this year, as opposed to Q1'07 originally slated by the company. 


Comments     Threshold


This article is over a month old, voting and posting comments is disabled

RE: Trying to step out of the shadows?
By DallasTexas on 7/22/2006 9:17:41 AM , Rating: -1
Yes, AMD did pull this out of their ass. The new "native quad" is a exactly that - a marketing stunt as if anybody cares how the quad sausage is made. Most people don;t care if it's 4 sausages linked together or one big turd like AMD is not professing to call "native".

The same marketing hype on intergrated mem controller and hypertransport. Too much rhetoric. The fact is, Core 2 Duo has neither an integrated mem controller NOR HT and it kicks AMD's finest and most expensive in the ass.


RE: Trying to step out of the shadows?
By stmok on 7/22/2006 10:21:44 AM , Rating: 1
You really have to get off your high horse.

What's this nonsense about "marketing hype" of intergrated memory controller and HyperTransport?

Intel is gonna be adopting a similar implementation to HyperTransport. Its called CSI.

And what about the integrated memory controller? Intel has done that before, they screwed up. The project/CPU was called "Timna", and it used a integrated memory controller for RDRAM. ("Timna" was the cancelled predecessor to "Banias"). Both designs were done by the same Israeli Intel team. With the "Banias", they used the PIII-S as the basis, and expanded on that. If you really think about it, Conroe is a descendent of the P6. (Pentium Pro, PII, PIII, Pentium-M, Core Solo/Duo).

Intel has improved the instructions per clock and prefetching such that Conroe is the "delaying mechanism" for future Intel CPUs that will integrate CSI and possibly a memory controller. (As in they've done enough to keep AMD back for now, while their R&D work on something new and experiment with new ideas).

Even an Intel engineer admitted an integrated controller is a good idea. (They aren't sure if they're really gonna need it yet...But they are considering it in a future design).

AMD's Hypertransport or Intel's CSI is a good idea as we increase the number of cores into the future. If you've bothered to look in the 4-core or CPU market, you'll see Intel's current bus design has a bottleneck in this scenario. (And you'll see a situation where Conroe-based solutions will be held back by the bus design).

That's why the Xeon version of the Conroe will have dual-independent buses! (one for each dual-core CPU). And its also why AMD is very likely to hold onto the 4-core/CPU market until K8L comes.

You really should get a good feel of CPU history and upcoming ideas before blabbering nonsense to make yourself feel better about Intel. Because you currently sound like a no clue fool.


By masher2 (blog) on 7/22/2006 1:29:02 PM , Rating: 2
> "Even an Intel engineer admitted an integrated controller is a good idea"

Actually, several Intel engineers have 'admitted' what anyone in the industry already knew. That an integrated memory controll is a classic tradeoff scenario...with good points and bad. Intel has toyed with them several times, but decided the benefits weren't worth the disadvantages. AMD decided the opposite. And-- given the differences in the architecture of each-- it can very well be that BOTH companies made the right decision.


RE: Trying to step out of the shadows?
By masher2 (blog) on 7/22/2006 1:26:12 PM , Rating: 1
> "The new "native quad" is a exactly that - a marketing stunt as if anybody cares how the quad sausage is made. Most people don;t care if it's 4 sausages linked together or one big turd"

Colorful languge...but your metaphor is essentially correct. Intel announces plans to deliver a quad core this year, so AMD has to respond in some fashion.


RE: Trying to step out of the shadows?
By Viditor on 7/23/2006 12:26:29 AM , Rating: 4
I think you guys need to understand what it takes to bring a processor into being...
AMD's Native Quad Core has been in development for almost 4 years now!
It was even talked about in the press before Conroe was in Intel's roadmap at all...


RE: Trying to step out of the shadows?
By masher2 (blog) on 7/23/2006 10:17:36 AM , Rating: 3
> "AMD's Native Quad Core has been in development for almost 4 years now! It was even talked about in the press before Conroe was in Intel's roadmap..."

I know that, but action is what counts in the market, not talk. AMD's 'demonstration' this year is being prompted by Intel moving up its quad core release to Q406.


By Viditor on 7/23/2006 7:52:22 PM , Rating: 2
quote:
AMD's 'demonstration' this year is being prompted by Intel moving up its quad core release to Q406

Why do you say that? The timing seems fairly normal to me...


"It seems as though my state-funded math degree has failed me. Let the lashings commence." -- DailyTech Editor-in-Chief Kristopher Kubicki

Related Articles













botimage
Copyright 2014 DailyTech LLC. - RSS Feed | Advertise | About Us | Ethics | FAQ | Terms, Conditions & Privacy Information | Kristopher Kubicki