The Consequence of Waking Up a Sleeping Giant: Intel Roadmaps Inside
by Kristopher Kubicki on January 25, 2005 7:44 PM EST- Posted in
- CPUs
Dual Core on the Horizon
So we lied originally - we have even better news. Dual core Smithfield processors, which are really nothing more than two Prescotts slapped together with independent caches, are scheduled to launch a little earlier than we originally claimed in previous roadmap articles. In fact, part of the push to launch so early seems to be to coincide with the 945/955 launch as those chipsets are the only ones to support the multiple core processors. Recall AMD's dual core launch strategy is to enable existing hardware (nForce4, K8T890, 8xxx) to run multiple cores. So while you can't plug a Smithfield into your existing 925X motherboard, it may be for the better. DDR2 has plenty of bandwidth to offer, but as we have seen in server benchmarks, multiple Pentium 4's competing on the memory bus can be quite slow. Dual core Pentium 4's might be horribly inefficient without DDR2-667, however that is another theory we can put to the test on launch day. If you look carefully, you'll see the Smithfields launching only at 800FSB. We find it slightly unusual that the entire 945/955 platform supports a front side bus speed that two $1000 SKUs utilize.
| Intel Dual Core PerformanceDesktop Lineup LGA775 | ||||
| Processor | Speed | L2 Cache | FSB | Launch |
| Pentium 4 840 | 3.20GHz | 2x1MB | 800MHz | Q2'05 |
| Pentium 4 830 | 3.00GHz | 2x1MB | 800MHz | Q2'05 |
| Pentium 4 820 | 2.80GHz | 2x1MB | 800MHz | Q2'05 |
Also note that the dual core processors on the desktop do not support HyperThreading. The server implementation of Smithfield, "Dempsey," has HyperThreading enabled. For database applications, this makes sense - although we have known for a long time that single threaded applications take a performance hit when a HyperThreading processor exclusively runs that program. Interestingly enough the Smithfield lineup has some very competitive price points according to the launch data. The 820, 830 and 840 models will launch at $241, $316 and $530 respectively - compare that to the Pentium 4 lineup today [RTPE: Pentium 4 775]. At today's prices that's only an $80 premium on the second core.
Facebook
Twitter
RSS
74 Comments
View All Comments
phantom505 - Thursday, January 27, 2005 - link
Whoppee, the paper launch king launches a lot more paper.I'll believe it when I see it.
BTW, where is that 4GHz CPU? SOI anyone? They need 65nm to keep from cooking. AMD has PLENTY of headroom on frequancy, something Intel doesn't (obviously). Now how about more cache, it will fix it, right?
Any word on Intel figuring out how to make a good FPU yet? Who needs that? The rest of world minus the internet... because it takes a monster machine to crunch full screen video?
How about a bus that can handle the data? 1066FSB? Isn't there something about AMD going 1.4GHz plus with HyperTransport II?
How about that slapped together dual P4? Wasn't the K9 being considered even during the design of the K8?
How about that DDR2 junk? Rambus part deux? At least AMD is going to watch Intel sink or float before you jump on that boat. I'd bet they would prefer to go to DDR3 directly.
Folks, I see a desperate company trying to fend off from something that is 1/8th of its size. And you think that's happens by chance? It's called screwing up.
AMD fanboy am I? Sure, at least they are going somewhere, other than down. (look at the Intel stock prices for 5 years).
Peter - Thursday, January 27, 2005 - link
EIST maturity? *laughs to tears*If one of the two contenders does have maturity and experience with processors capable of scaling their clock speeds and supply voltage on the fly, then it's AMD. Remember K6-2+? First one to do this on the x86 stage, even before Transmeta iirc. Mobile Athlons had it, all Athlon-64 have it, Opterons will get it soon.
Intel? They had nothing but the clunky original SpeedStep (which required a high latency sleep-wake cycle to change speed, and had only two steps, slow and fast) before the Pentium-M showed up. Now that technology is getting retrofitted to P4 core. Good move, but the innovation was invented elsewhere ...
Live - Thursday, January 27, 2005 - link
As for the release date and availability I don’t know, but you could always speculate a bit :)The article mentions an NDA coming up and launch next month. So launch in February it is, which coincides with CeBIT Hanover 10-16 March. So we can expect launch in conjunction with that. Then we know Intel will release its new 64 bit ready CPUs to be ready for the launch of Windows 64 that is supposed to be in April. So at the latest by May we should have it all in retail.
So if February is launch then I guess March would be the optimistic and April the pessimistic and May if Intel or M$ stumbles and summer if they screw up.
danidentity - Thursday, January 27, 2005 - link
Is there any more detail on a release time frame for the 945/955 chipsets, other than Q2?footballfan - Wednesday, January 26, 2005 - link
I forget...how big of a gap was there between the official announcement of 915/925 chipsets and being able to actually buy a motherboard based on those chipsets?I'm putting together an Intel system and I don't know if I should get an 925XE motherboard or wait it out for a 955.
Hmmmm
JarredWalton - Wednesday, January 26, 2005 - link
28 - The comment in regards to AMD vs. Intel is that AMD is keeping dual-core backwards compatible while Intel is not. From a performance standpoint, that means Intel can potentially improve aspects of the chipset. AMD is in a sense more limited in that they're constrained to the original S939 specifications. Neither approach is *better*, per se, although a lot of people like the AMD approach simply because it doesn't require a new motherboard. As far as we're aware, *all* S939 motherboards will be capable of running Toledo.When will we see all the new hardware? That's the big question. :) AMD and Intel both tend to be a bit better about avoiding the "paper launch syndrome", but there have been instances in the past where availability lagged far behind the official launch.
Live - Wednesday, January 26, 2005 - link
Always a real treat to read these top insider stories, good work!This makes buying/upgrading decisions at least a little bit easier. I want more ;D
One question tough. You mention that AMD plans "to enable existing hardware (nForce4, K8T890, 8xxx) to run multiple cores."
Those this exclude earlier s939 chipsets? I was under the impression that both nForce3 and K8T800 Pro would be compatible with dual core Toledo.
If you read here (Registration required): http://www2.amd.com/us-en/protected/Weblets/1,,783...
"AMD’s dual-core processors are being designed with today’s infrastructure in mind. System integrators will have no problem incorporating AMD Opteron processors into existing platforms and any desktop motherboard supporting a 90nm AMD Athlon 64 processor will accommodate dual-core descendants of the chip as well."
Toledo is a dual-core descendant of the current crop of 90nm AMD CPUs is it not?
I’m contemplating either nForce3 or nForce4 and I believe future Toledo support would tip the scale.
RockHydra11 - Wednesday, January 26, 2005 - link
I'm quite interested in what kind of response nVIDIA will have to the barrage of new chipsets.footballfan - Wednesday, January 26, 2005 - link
How long from launch to being able to actually buy a motherboard with one of those new chipsets will it be?JarredWalton - Wednesday, January 26, 2005 - link
945 and 955 are scheduled to launch in Q2'05, right along with Smithfield. Smithfield will *not* work on 915 or 925 chipsets - it may also have a new socket, although that wasn't indicated on the roadmaps.As far as what the Prescott 2M will bring, I expect more than a 5% performance increase in most applications, but probably less than 15%. It's difficult to say where it will actually land relative to the P4EE Gallatin cores, since it uses L2 instead of L3 cache, but it also has the longer pipeline. Certain applications perform better on Prescott than Northwood already, so in those instances the lead will increase. Will it catch up to AMD in gaming? Not likely, but it will close the gap. 2MB of cache might also improve HyperThreading performance a bit - that will be interesting to see.