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
Mobility
Although we get real annoyed with paper launches and PR vaporware, the Centrino lineup never struck us as a set of hardware that didn't show up when promised (at least in the OEM sector). In fact, Sonoma based notebooks with Dothan 2.1GHz processors are already shipping from at least two of the Tier 1 notebook manufacturers although the launch was just last week. This gets us particularly excited since dual core, 65nm Yonah processors are scheduled to launch with the third generation Centrino platform, Napa. And given how reliable Centrino's track record has been, it could be very possible for us to review a 65nm, dual core notebook 365 days from now.
| Intel Dual Core Performance Desktop Lineup LGA775 | ||||
| Processor | Speed | L2 Cache | FSB | Launch |
| Pentium M 780 | 2.26GHz | 2MB | 533MHz | Q3'05 |
| Pentium M 770 | 2.13GHz | 2MB | 533MHz | Now |
| Pentium M X50 | ??? | 2MB (Shared) | ??? | Q1'06 |
| Pentium M X40 | ??? | 2MB (Shared) | ??? | Q1'06 |
| Pentium M X30 | ??? | 2MB (Shared) | ??? | Q1'06 |
| Pentium M X20 | ??? | 2MB (Shared) | ??? | Q1'06 |
| Pentium M X48 | ??? | 2MB (Shared) | ??? | Q1'06 |
| Pentium M X38 | ??? | 2MB (Shared) | ??? | Q1'06 |
To quote Bender from Futurama, "... the X makes it sound cool." When Yonah hits the retail market it seems fairly likely that Intel will drop the X in favor of a number - although it seems like all the good ones thus far are used up. Well, X is the Roman numeral for 10...
One of the first things we knew about Yonah when we first saw the preliminary work in Taipei for Computex 2004 was that it uses a shared L2 cache between the two cores. While the Smithfield dual core processors separate their cache per core, Yonah is unique as arbitrating the cache for both processors. We also know a bit about the further enhancements on EIST utilized in the processor, including a technology that dynamically throttles power from unused portions of the cache. Feel free to check out or IDF 2004 coverage for that announcement. Another interesting tidbit on the roadmap reveals that the Napa platform will utilize both the 945 and the 955 chipsets in production models. Why a notebook would need ECC support is beyond me, but perhaps there are some features in 955X that haven't been fully leaked yet.
Unfortunately thermals are a huge piece of the pie, and we don't have any confirmed behind the scenes details on that yet. It looks like you'll just have to wait until launch dates for Prescott 2M, Dempsey and Yonah for that scoop.
Though Intel's roadmap reveals a gung-ho attitude for EM64T, the mobility platform looks completely devoid of any such notion. Even the much famed Yonah doesn't explicitly mention any 64-bit support, which may give AMD the upper hand in that match up. AMD's Turion platform will have 64-bit support - will their competition?
Unannounced Secret Stuff
The very forward looking stuff on Intel's roadmap, Q1'06, shows more promise than we had also originally anticipated. First off - get used to the names Presler and Cedar Mill. We had mentioned Cedar Mill before as a single core Pentium 4 evolution. While probably not a direct NetBurst revision, don't be surprised if some of those wonderful projects scrapped with Tejas show up in Cedar Mill instead. Cedar Mill utilizes 2MB of L2 cache, Socket 775 architecture and a 65nm process. On the enterprise portion of the roadmaps Intel is very careful to separate Cedar Mill from the rest of the Prescott 2M SKUs so perhaps there is more to meets the eye for this little processor.
Presler is a whole different animal. On the roadmaps Intel marks Presler as the eventual dual core replacement for Smithfield albeit with an extra megabyte of cache per core. Since this is the first we have heard of the processor in official circles, details were pretty light.
We talked real casually in the past about Dempsey - the dual core Xeon. From the roadmaps Dempsey doesn't look similar to any Pentium 4 or Xeon processor we know about. For starters, expect 1066FSB, dual core, and HyperThreading. If four logical processors per socket didn't seem to catch your attention the addition of Fully Buffered DIMM (FB-DIMM) and iAMT will also show up on the chip platform. Intel also refers quite often to its "Diamond Peak" technology - which they loosely describe as:
Platform level LAN acceleration based on improvements in processor, MCH and ESB-2
Like the desktop platforms, the next generation server core logic will also feature Vanderpool Technology. This leads to a real lot of promise for those who rely on User Mode Linux or VMWare for their enterprise solutions. Rather than placing separate operating systems in different machines, VT opens the door to putting different operating systems on the same processor. The roadmap stresses this sort of configuration makes sense for high availability; if one OS crashes, its OK because we have 2 running.
"Blackford," the E7520 replacement for Xeon, will utilize ESB-2 and will become Dempsey's chariot in the server market. A cheaper, stripped down version of Blackford dubbed Bensley will perform the task of Dempsey's value platform, Greencreek is the workstation variant.
Final Thoughts
Generally we do not get to write this much about the Intel world, so today's opportunity to look at everything Intel has to offer is quite refreshing. While a lot of people have looked at Intel's recent reorganization as a sign of weakness, we feel the company is just trimming itself back to the main arteries it does best. We are at least optimistic that 2005 will be a better year for consumers than the last one - now we just need to see if AMD has enough up their sleeves to make things interesting.
Facebook
Twitter
RSS
74 Comments
View All Comments
jiulemoigt - Friday, January 28, 2005 - link
#42 the problem with Rambus was the company tried to make the entire industry pay them for DDR2 because of tech they suggested to the Industry standards comision without telling them they patented it already when the Standards comminty was tring to find open standards. That and it is like P4 a long pipe when forced to branch wastes clock cycles{way to often}.Zebo - Friday, January 28, 2005 - link
Unless they get 2-2-2 DDR2 out AMD is wasting it's time (and performance) with DDR2. Negitivty twards Intel? Probably because they keep pimping that marketing gimmick called netburst. And heaters called prescott, but worse this time with two. EVERY, I mean every CPU in history has worked twards more effecientcy. Not intel.You can read about it all here.
http://arstechnica.com/articles/paedia/cpu/pentium...
That article includes the excellent Dothan too, which they should have done in the first place instead of raising our power bills and room temps.
JarredWalton - Thursday, January 27, 2005 - link
#34 and #35 - Why all the negativity towards Intel? I certainly don't think this is all the greatest thing ever, but it's a welcome change from the last 3 months of Intel roadmaps where chips were canceled and release dates were postponed. AMD is still way back in terms of revenue, and that isn't going to change over night. I hope they continue to make improvements in their design, but anyone that thinks Intel is just sitting still is loco, plain and simple.As for the technologies you "poo-poo" above, DDR2 is an industry standard. AMD is avoiding it initially because they don't really need it yet, so when they do need it they can just join the club. That's fine, but at the same time it's good to have one company pushing things forward. AMD pushed 64-bit and forced Intel to join them, and Intel is pushing DDR2 and FB-DIMM technologies, which will benefit everyone in the long run.
In retrospect, do you REALLY think Rambus memory was that bad? It wasn't necessary on the Pentium 3, and it was more expensive than DDR at the time, but economies of scale come into play. If the public had not had a huge backlash against RDRAM, it would probably still outperform equivalent DDR on Intel platforms. The only real problem with RDRAM was that it was a closed standard, so you had to pay royalties.
If you look at the big picture, none of these companies are really out there trying to make the world a better place just for altruistic purposes. They all want to make money. If AMD gets bigger, it will be because they're making more money, and generally speaking that means that they'll be acting more like Intel. To #39, I would say that it *IS* a competition, and we want it to stay that way. If it just becomes another ho hum update each year, we'll end up just like the car industry.
danidentity - Thursday, January 27, 2005 - link
One more question, is there any confirmation to the rumors that the 6xx series of P4's will have downward unlocked multipliers because of EIST?Anemone - Thursday, January 27, 2005 - link
From what I"ve seen a 600 series @ 3.6-3.8ghz should keep up with the FX55 in a lot more games than you'd think. That's based partly on results of EE's clocked @ 3.7-3.8... Of course we will see soon, but of course AMD might easily find another 200 speed bump somewhere too.If you all remember the performance bump from the 845 to the 875, I think you might want to give some thought as to just what "could" be provided by the 955.
I'm also rather sad the 925XE 'may' not accept the dual core. Oh well. If they'd get rid of that idiotic oc lock, I'd spring for the 955. Would be nice to get back to 875 days or better imo.
I'm mostly glad the two are so tightly in competition. I bet that doesn't make Intel happy but it is sure doing nice things for customers. Again, with the exception of the OC lock which was a stupid mistake...
Quanticles - Thursday, January 27, 2005 - link
#8, AMD partners with IBM for their fabricationAll of this is the same old inefficient core slapped together with a little bit more cache, a faster FSB, blah blah blah. Their dual-core setup with have an amazing 10% performance increase I'm sure, maybe that extra cache will make up the last 5%. Now if only this stuff wasnt being released a year from now, cause I'm sure AMD will have better than this in the next 6 months - they just keep their cards to themselves.
Anandtech - tnx for hyping this for Intel. Every page had even "better" news, although I'm not sure who the news was better for. You make it sound almost like a competition.
KristopherKubicki - Thursday, January 27, 2005 - link
Peter: Surely you jest. The predecessor to PowerNow! technology showed up in spring 2000. The first real production processor with PowerNow was the mobile Athlon4 line in May 2001 - which had 5 frequency stages and 6 core voltage stages.Not to split hairs or anything, but the first P3 to show up with speedstep was the 600MHz variant which showed up 3 months before K6-2+ in Jan 2000. In Q2'01 SpeedStep improved and which allowed dynamic clocking, but also allowed voltage adjustment with deepsleep.
Concerning latency: EIST today requires a 30microsecond delay to transition frequency, 100microseconds for voltage. Last I checked the AMD driver for CNQ had a 0.03 second hard delay on frequency/voltage adjustment.
OK fine - things are even up until 2003. Then something called Pentium M showed up in Q1 with EIST. EIST goes beyond changing the clock speed and voltage and will actually switch processor logic on and off when it isnt needed. On the Yonah processor EIST will actually disable portions of the cache it isnt using.
CNQ just isn't doing this yet, but Intel already has 2 years of experience doing that with Banias/Dothan. If we want to talk about innovation, what has AMD been doing for the last 2 years with PowerNow other than renaming it to Cool N Quiet for the K8?
Kristopher
KristopherKubicki - Thursday, January 27, 2005 - link
Live: there are two launches - one of the 6xx line, another a little bit later for 945/955. I dont know if the dual core launch falls at the same time as the 945/955 launch but there will be more data available then at least.Kristopher
Live - Thursday, January 27, 2005 - link
O I whish we could edit our comments. CeBIT is of course in February not March. My whole guessing game kind of falls a part if it would be in March.“So launch in February it is, which coincides with CeBIT Hanover 10-16 February.”
miketheidiot - Thursday, January 27, 2005 - link
First of all I see nothing on here that will allow intel to take back the performance crown. Of course I already see that intel probably will win the dual core publicity/prestige/etc, even though I think their design will probably be garbage.#8, amd has had 90nm out for 3 months and it has worked much better than intel's 90 IMHO. Second they are well into the construction of another fab, so I highly doubt they will go fabless any time soon. On top of that intel was, and still is, getting its ass handed to it by AMD 130nm
#34 there was some news about AMD unfortunately going to ddr2 in '06. This will require a new socket, so if there is to be a Hypertranport2, which I do remember hearing, it probably accompany that change.
I do hope that Intels accelerated dualcore will get AMD to start moving a bit quicker. They demoed dual core quite a while ago, I would suspect that it could go into production very soon if they really needed it to.