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
Single Core Desktop
Now for desktop processing; we have good news and better news. The good news is almost all desktop Prescotts (including the Celerons) will get a 64-bit makeover real soon. Unfortunately, you'll still need to buy a new processor but the roadmaps indicate there will be virtually no price premium on the 64-bit versions. With Windows XP 64-bit release less than a few months away, it makes sense that Intel's 64-bit push comes strong and hard in the 9th inning. We are particularly interested in how fully committed the roadmap details EM64T; even the puny Celerons get the instructions. You may recall that the Socket 754 Sempron processors are nearly identical to AMD's Athlon 64 processors with half the cache and the 64-bit instructions removed. It will be interesting to see how the two companies play this against each other as AMD will be the 32-bit SKU on the desktop when WinXP Pro x64 launches.
Just to detail the whole outline for 64-bit Socket 775 processors, here is a quick roadmap of what we have to look forward to:
| Intel Single Core Mid Range Desktop Lineup LGA775 | ||||
| Processor | Speed | L2 Cache | FSB | Launch |
| Pentium 4 XE 3.73GHz | 3.73GHz | 2MB | 1066MHz | Soon |
| Pentium 4 XE 3.46GHz | 3.43GHz | 512KB | 1066MHz | Nov 2004 |
| Pentium 4 571 | 3.80GHz | 1MB | 800MHz | Q2'05 |
| Pentium 4 561 | 3.60GHz | 1MB | 800MHz | Q2'05 |
| Pentium 4 551 | 3.40GHz | 1MB | 800MHz | Q2'05 |
| Pentium 4 541 | 3.20GHz | 1MB | 800MHz | Q2'05 |
| Pentium 4 531 | 3.00GHz | 1MB | 800MHz | Q2'05 |
| Pentium 4 521 | 2.80GHz | 1MB | 800MHz | Q2'05 |
| Celeron D 355 | 3.33GHz | 256KB | 533MHz | Q4'05 |
| Celeron D 351 | 3.20GHz | 256KB | 533MHz | Q2'05 |
| Celeron D 346 | 3.06GHz | 256KB | 533MHz | Q2'05 |
| Celeron D 341 | 2.93GHz | 256KB | 533MHz | Q2'05 |
| Celeron D 336 | 2.80GHz | 256KB | 533MHz | Q2'05 |
| Celeron D 331 | 2.66GHz | 256KB | 533MHz | Q2'05 |
| Celeron D 326 | 2.53GHz | 256KB | 533MHz | Q2'05 |
As you can see, the EM64T enabled CPUs have incremented their model numbers by 1 relative to their non-EM64T counterparts. We're glad that Intel is making a clear distinction between the two variants, rather than simply adding a new suffix. Where there is no earlier part, like the 3.33GHz Celeron D, the model numbers do not have the +1.
The 3.73GHz Pentium 4 EE will show up real soon, if that's your thing, and it will become the second processor to support 1066FSB. We haven't been real impressed with the 1066FSB launch thus far, and a 300MHz bump in clock speed doesn't strike us as something that will revolutionize the performance desktop anytime soon either. However, keep in mind this new P4EE is very different from the previous 3.46GHz revision, and with a different core we may see a very different performance curve on the 1066MHz front side bus. The rest of the Intel roadmap neglects to mention any other 1066FSB processors, including the dual core behemoths, so the technology is either a little bit ahead of its time or simply a temporary dead end.
Next we have the great news. Not only will we see the launch of four Prescott 2M/Iriwindale processors next month, but soon after we will also get our first taste of Smithfield - several quarters ahead of what the previous roadmap had anticipated! Prescott 2M will launch with four SKUs listed below, along with a "670" model clocked at 3.8GHz sometime shortly after.
| Intel Single Core Performance Desktop Lineup LGA775 | ||||
| Processor | Speed | L2 Cache | FSB | Launch | Pentium 4 670 | 3.80GHz | 2MB | 800MHz | Q2'05 |
| Pentium 4 660 | 3.60GHz | 2MB | 800MHz | Q1'05 |
| Pentium 4 650 | 3.40GHz | 2MB | 800MHz | Q1'05 |
| Pentium 4 640 | 3.20GHz | 2MB | 800MHz | Q1'05 |
Like the other Prescott processors, Prescott 2M will launch with EM64T and XD, but it adds Enhanced Speed Step (EIST) as well. EIST is very similar to AMD's Cool n' Quiet as it dynamically ramps the clock speed of the processor to conserve thermals and power. However, the big difference between CnQ and EIST is maturity - EIST has existed in some form or another since the earliest days of the P6 architecture. How EIST will affect performance on everyday desktop processing - particularly on a processor with such a high clock speed - we leave for the actual launch date sometime next month.
Facebook
Twitter
RSS
74 Comments
View All Comments
Ozenmacher - Thursday, February 3, 2005 - link
Naw, I have nothing wrong with writing a positive article for Intel or AMD. I mean, I understand both companies have great chips, so I don't care if a positive article is written. But yeah, I still don't understand where sleazes came from, lol.JarredWalton - Wednesday, February 2, 2005 - link
Learn to type very fast (80 WPM or more) and write a lot of articles. You get all sorts of interesting slips. I would correct the "sleazes" for Kris, but unfortunately I can't. (Not enough access.)As for Kris being an Intel "fanboy", give it a rest. So he writes a relatively positive article on the latest Intel roadmap, what's the big deal? Does a relatively postitive AMD roadmap article make me an AMD fanboy? (Obviously not, since I get accused of being an Intel supporter just as often when I provide my own take on the market.)
We're all just interested in performance - price/performance for many of us. Competition from Intel is great, because AMD needs it just as much as Intel does. Given that the last few Intel roadmaps had little information on Smithfield and it looked like it would slip to late 2005 or 2006 instead of launching earlier, how can this revised roadmap be anything but good news? We'll still give a critical look at the final performance when all the products launch, and hopefully XP-64 will even become a factor some time this decade.
Ozenmacher - Wednesday, February 2, 2005 - link
Very good point #61, how could you "accidentally" type sleazes instead of sleeves? lolRoosterKooster - Tuesday, February 1, 2005 - link
Kris, "sleazes" can't be a typo. Even it it was, how did it escape the proofing?Perhaps you've been in a daze, but Intel has hardly been napping - just a bit on the back side of the power curve. Put all the vaporware aside and let's see what spring has in the air. Just tell the folks here, you are 100% pro-Intel.
EglsFly - Tuesday, February 1, 2005 - link
Wow, that article had Intel fanboy written all over it. To sum up: "Intel Better, and it gets better, need to see if AMD has enough up their sleazes ...."Give me a break!
Regs - Tuesday, February 1, 2005 - link
Good article Zebo. I agree with it.Intel has to slowly adapt the Pentium-M into the market. If they came out with a desk top CPU based on the PM, what would they say? "Look at this amazing CPU with twice the power with less the clock cycle!" While AMD just sits back and goes, "Hey, we've been doing this for years" Wouldn't look good.
Regs - Tuesday, February 1, 2005 - link
The reason why I think Intel won't scrap their Netburst Northwood's is because they sold the market on High Clock Speeds while also selling the mobile market with less power hungry yet more clock efficient processors. If they came out with something like AMD's processor, what would that make them look like? It seems like Intel's marketing department is running the course for Intel in the future other than their engineering department.When 939's come out with strained silicon that could possibly push through the 3.0 GHz barrier with SEE3, Intel's NetBurst processors will be looking pretty desolate.
Intel needs a knight-in-shinning armor and a 4.GHz Prescott with 2MB L2 cache is not going to do it. They would need to bring out a entire new line of CPU's to match performance with all A64's instead of pushing out one new CPU every 6 months that costs over 500 dollars.
Quanticles - Monday, January 31, 2005 - link
I liked Anandtech's point that Microsoft was delaying Windows64 until Intel was ready.Peter - Monday, January 31, 2005 - link
#38, I wouldn't say things were "even" before the Pentium-M showed up. Before that, Intel's speed step always required a deep-sleep transition and had only max and min states, while AMD's were quicker and more versatile, even the original implementation on K6-2+ (where transition time was even fine tuneable to the voltage regulator's needs).What happened at AMD for the last two years? They kept the power down in general, something Intel quite miserably failed to do on the desktop.
Note that I'm completely with you in that the Pentium-M is a very fine part, and that AMD has some catching up to do in the mobile arena. On the desktop, it's the other way around, and it's exactly this catching up that we see documented in your article.
We're both wrong on who did it first anyway. Guess what, it was Cyrix. Their 5x86 could do live transitions from its native (2x or 3x) multiplier down to 1x and even 1/2x and back up. And it actually worked. In 1995. (Separate voltage regulators for the CPU core were nonexistant back then.)
regards,
Peter
johnsonx - Monday, January 31, 2005 - link
Anytime the socket-754 Sempron has been discussed, I've said it makes little sense for AMD to purposely cripple a 64-bit processor down to 32-bits, as it leaves them no competitive advantage over Intel.With Intel extending EM64T all the way down the Celeron D line, now we see that AMD will now be at a competitive DISADVANTAGE vs. Intel because of their foolish crippling of the 754 Sempron. Dumb, AMD, dumb...
And before anyone comments, no I don't think Semprons are just a way for AMD to sell bad A64 cores that won't do 64-bit, but run 32-bit fine... that just isn't going to be a common enough failure mode. Cores with some bad cache? Sure. Cores with a malfunctioning HT link? Maybe. Cores with one memory channel on the fritz? Perhaps. 64-bit extensions not working? Nope.