Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
Register for free to join our community of investors and share your ideas. You will also get access to streaming quotes, interactive charts, trades, portfolio, live options flow and more tools.
No, there's one where Alan and I were talking about it, specifically about EE and loose power stats.
Then DRAM Exchange is to blame.
Shipping to OEMs April 30, channel May 15, Retail avail = June 6.
Interesting about the lowish power Live! desktop parts.
More nonsense about "AMD CPU postponed". As of last October, roadmaps indicated June for nVidia chipset, and hence AM2, per Anandtech article of Nov 1, 2005. Inq repeats this twice now in the last month as if it just occurred.
Nah, I think you're overweighting Dempsey here, and underweighting all the other factors involved.
You didn't exclude Extreme Edition parts in your prediction.
Actually, I believe I did. Look for an exchange with Alan81.
Thanks. So, 1/2 of it is waterfall, and then there's the transition to DC in server, so DC needs to approach old SC prices, now that both companies will have something competitive, plus AMD is growing server volumes.
Not sure it's anything Dempsey specific, really, except inasmuch as Dempsey allows Intel to sell a higher % of DC server parts.
Missed that. I wouldn't call 2.93 vs 2.8 "WAY off". We'll see if they ship it in December, or if it slips to Jan. 14 stage pipeline... could be the IPC isn't all it's been speculated to be.
(EDIT) Keith, the fact that xbit thinks late Q3 is "much earlier than expected" is a little odd, don't you think?
Ok, I'll look.
3GHz. Fair enough. Will they launch that grade by the end of the year? I guess we'll see.
14-stage pipeline. Hmmm. Despite the 4-width, I wonder if Intel's NGA will have a LOWER IPC than the K8?
Anyhow, in the server segment, AMD is holding the QC ace. I'm quite convinced they can offer that whenever they feel the need.
EDIT: regarding launching-- I see no indication of a higher Conroe non-EE in 2006 than the 2.67GHz grade. Do you? There isn't that much time left btw a September launch and the end of the year.
I'll be happy to remind people of your credibility when your bearishness of Intel / bullishness of AMD ends up completely unfounded by the end of the year.
Unfortunately for you, you're already well in the hole on your claim that INTC stock would outperform AMD stock this year.
With Intel's NGA launch now in late Q3 / early Q4, there isn't much catalyst for Intel stock this year. Vista might trigger all-around good news in 2007. Hence my 2008 Intel call purchases. I was a bit early on those buys however. After Intel's terrible earnings, but not after enough.
And indeed, Conroe is launching at 2.66GHz top grade. Now, with the EE, depends how high they want to go with TDP. I doubt xbit's 95W is correct, here.
Also, these xbit articles have a second-hand feel to them, as if they cobbled them together from the Inq, Reg, Tom's.
Keith, so you think introducing a new speedgrade has nothing to do with the price cuts?!
No, you earned that lack of credibility over the past few years on the various AMD iHub boards.
You have no credibility left regarding your buy/sell recs on AMD or Intel.
Dempsey ain't here until May, so I doubt that's the reason. The actual reason is the coming x85 series in a month. Standard pricing waterfall.
"Much earlier than expected" (???) -- I guess I don't need to read any more of that article.
Yeah, the last gasp of NetBust.
I always post full texts, because you just don't know if it is going to go away, and it makes it impossible to search for later if you don't.
AMD's Q3 in the bag, and maybe Q4 too... Intel NGA is too late for the back to school build, and pushing it for the holiday build, based on this latest late Q3 / early Q4 timing.
Conroe not until late Q3, Merom in early Q4. Looks like the attempt to pull products in to early Q3 failed.
-------------------------------------------
Full text from your link:
Chicago (IL) - Intel is counting down to the launch of its next-gen desktop CPU: Conroe will begin to ship to system builders in Q2 of this year with the official introduction to follow in late Q3. Details are already leaking out of Santa Clara - the chip will carry the sequence numbers E4000 and E6000 and arrive in six versions with clock speeds of up to 2.66 GHz, TG Daily has learned.
This year, Intel will have to prove if it can keep the bold promises it made last year - to surpass the processing performance of AMD processors while cutting power consumption in half. Latest roadmaps seen by TG Daily suggest that the introduction of the new micro-architecture is well on track: While Intel will not meet an initially expected launch date in early H2, the company may be in the final stages of prepping the mass production of "Conroe."
Intel's next-generation microprocessor architecture is developed around the mobile processor core "Merom." Conroe is the desktop version of Merom, while "Woodcrest" will be the server-targeted derivate. Conroe will be the first chip of the family to launch: The processor's introduction is expected to coincide with the unveiling of Windows Vista in September of this year.
According to sources, Merom core now appears to be delayed until Q4 of this year, most likely targeting an October announcement. The new architecture will debut in Q3 in the server segment as Xeon DP 5100 (Woodcrest core); a quad-core DP variant (Clovertown core) will follow in Q1 2007.
Conroe to debut as "Core E4200" and "Core E6000"?
Intel's latest roadmaps are listing Conroe's sequence number for the very first time. Sources suggest that Conroe will be introduced as a member of the "Core" processor family. The sequence numbers - E4200 and E6000 series - are final, but we aren't completely sure which brand name Intel will choose for the desktop processor. Considering that it is a part of the Core family and that the "E" in the sequence number indicates a power consumption of 50 watts or higher, it only would make sense to release the chip under the "Core" umbrella.
In such a scenario, the mobile processors would be assigned the letters "U" (14 watts or less), "L" (15 - 24 watts) and "T" (25-49 watts), while the "E" would be reserved for desktop processors. For the first time in several years, Intel's processor portfolio could gain transparency and reduce the confusion that surrounds its current chip lineup.
Averill and Bridge Creek Platforms
Consistent with its platform strategy, Conroe is just one part of several platforms that Intel will offer to its consumers. For example, Viiv entertainment computers as well as consumer computer systems will be based on the "Bridge Creek" platform which will include the chipsets 975X (DDR2-667, FSB 1066) and Q965 (DDR2-800, FSB 1066) on the high end and G965 (DDR2-800, FSB 1066, Clear Video Technology) in the mainstream segment. Bridge Creek will also integrate a Wi-Fi chipset (3945ABG / 3965ABG).
However, Intel will put a strong focus on its "Averill Pro" and "Averill Fundamental" platforms that are aiming especially at business systems. The "Pro" version will come with an E6000 processor, the Q965 chipset in combination with an extended ICH8 Southbridge, integrated Virtualization and Active Management Technology (VT and AMT). The "fundamental" edition includes Pentium 4 and D CPUs as well as E4000 and E6000 CPUs and 946, 975X, Q963 or Q965 chipsets.
Six processors, up to 2.66 GHz clock speed, mainstream pricing
Intel reminded us repeatedly that the new architecture will show a departure from ever increasing clock speeds and that performance per watt will be the future focus. So it is no surprise that Conroe checks in at significantly less Gigahertz than its predecessor Presler - which currently tops out at 3.4 GHz and will receive one more update before it will begin to be phased out in the third quarter of this year: A 3.6 GHz version (Pentium D 960) as well as a 3.73 GHz Pentium EE 965 will debut on 30 April 2006.
Intel will be offering six versions of the Conroe processor at launch:
* E6700: 2.66 GHz / FSB 1066/ 4 MB shared L2 cache
* E6600: 2.40 GHz / FSB 1066/ 4 MB shared L2 cache
* E6400: 2.13 GHz / FSB 1066/ 2 MB shared L2 cache
* E6300: 1.86 GHz / FSB 1066/ 2 MB shared L2 cache
* E4200: 1.60 GHz / FSB 800/ 2 MB shared L2 cache
* Conroe Extreme Edition (XE): Specifications unknown
All Conroes will keep the current LGA775 packaging designation of the current Pentium D processor; however, there is no indication at this time that users will be able to simply swap a Pentium D 800/900 processor with a Conroe chip. According to sources, mainstream Conroes will post a thermal design power (TDP) of 65 watts, which is half of the maximum power consumption of the Pentium D 800/900. On the high end, that TDP will be exceeded by the variants E6600, E6700 and especially Conroe XE.
Pricing of the new architecture will be in line with previous processor introductions. The E6700 model will cost $530 at launch, the E6600 will be priced at $316, the E6400 at $241 and the E6300 at $209, sources said. We were told that Intel plans to quickly ramp the production to have more processors available at launch than it was the case with previous products. While the Pentium D 900 will post the highest production numbers of Intel dual-core desktop processors in 2006, the company plans to ramp down the production of the CPU beginning in the third quarter and shift production capacities in favor of Conroe.
The latest roadmap is a clear indication that Intel will be able to deliver its new architecture close to the time frame it was expected. While we do not know yet how the new architecture will perform, the lineup looks promising on paper. With AMD introducing Socket (A)M2 processors in Q2, the stage is set for the most interesting year in the microprocessor industry since the Gigahertz race back in 2000.
Anyone who followed my advice would have either made money, or they would still be holding onto their shares.
yeah, RIGHT. Only by the new "unless I mark a post as OFFICIAL, any recommendation isn't REALLY a recommendation" wbmw rule, announced last week.
Please spare us this BS.
Agreed.
What do you mean, "later than expected"?
putting
intend
waiting
Say no more.
Wow, I guess the hubby's options got killed in the latest price fall, huh?
My AMD leaps are doing great. They're all listed options, of course, as I am in no way affiliated with AMD, other than as an investor.
Must be hard to be an employee (or employee spouse) and be objective enough to know when to bet on the competition.
BTW, why don't you define "multi multi millionaire" for us, and explain what instruments you hold that would realize that with INTC at $37/sh.
Kate, how's that $37 Intel target looking?
Fair enough. If it is "next steppings" DC, that would mean Rev F, being the stepping after Rev E, at least for 90nm, I presume. And I guess the first 65nm parts will be Rev G.
Buggi, are you referring to better clocking Rev E parts, coming Rev F 90nm parts, the 65nm Rev F parts, or all of the above?
If you want, PM my Silicon Investor account.
When Intel starts kicking AMD's butt in all the benchmarks starting with Woodcrest and continuing through Clovertown (which will ship this year, according to the article), I'll return to let you know how absurd and silly your comments are.
Excellent! So we won't be hearing from you until and unless that happens?
BTW, why do we care when Intel is "shipping" a part, as opposed to launching it, or better, OEMs having it available for sale?
"Launch" is Q107, per Intel.
I think they won't. As has been true over and over again for the past couple years.
That's nice that you said that, but his point is that Intel has a history of finer-grained PR announcements, especially concerning process stuff, and more broadly, when they are behind.
You're exactly right, as any close observer of the two companies over the past several years knows. You just have to laugh at these folks. It's like all they have to hang onto is a delusion that Intel is now "ahead" of AMD in terms of QC development, when in reality, the situation is completely the opposite.
Also amusing is that there are no benchmarks or photos of any kind of Intel's MCM QC Frankenstein. What kind of 'demo' is that, anyway? How do the journalists even know what was actually in the machine?
I guess INTC investors haven't had their remote controlled silence chips installed yet to control pain.
If it takes ranting to make you and Kate and wbmw feel better about your shrinking investment accounts, I guess we'll all have to bear the burden.
No, I know: they took chipdesigner's advice and fired the engineers so they could afford more lawyers.
Still talking up your irrational, illogical fantasies, eh?
That was already dismissed here: http://www.investorshub.com/boards/read_msg.asp?message_id=9649466
How come those idiots down in Sunnyvale can't seem to get a quad core demo ahead of Intel?
They've demoed privately for months already. They've announced they'll be giving a public demo along with Socket F. Which do you think is more important? Demos for OEMs or PR stunts?
AMD simply isn't as desperate as Intel. They can roll out QC from the moment Socket F launches, if they choose. Intel is stuck with their 2 x DC "quad core" hack job, putting on another benchmark-free blackbox "demo" for the press.
And, generally speaking, Intel is always making a big PR fuss over every little step they accomplish... "We've designed a 45nm SRAM!" "We've started building one!" "We've finished it!" "We took a picture of one!" etc., etc.
Could you keep the quote spam to a minimum, please? There are other boards for that.
Well, the article was written by longtime Intel fan Michael Kanellos, so he'd happily oblige Intel in the (B) case.
Now, now... the Duke has "advanced degrees" and was a professor "at a law school".
3GHz SC Opteron (256) coming soon!
http://www.theinquirer.net/?article=29629
BTW, considering this and the previous, looks like BUGGI was right about AMD's improved clocking overhead with recent parts.
Rev E Opti 285 (95W, 2.6GHz DC) in early March.
http://www.theinquirer.net/?article=29629
And that means the 290 SE's (2.8GHz DC) can't be far behind...
We shall.