backtop


Print 20 comment(s) - last by ADDAvenger.. on Sep 22 at 9:17 PM

Roadmaps reveal launch dates for upcoming Intel products

DailyTech has come across a roadmap outline Intel’s launch plans for the next couple of months. Intel next week will launch five new products. At the beginning of the week Intel plans on silently releasing its Pentium D 925 processor. The Presler based Pentium D 925 is clocked at 3 GHz on a 1066 MHz front-side bus with 4MB L2 cache. Unlike the previous Pentium D 920, the 925 lacks Intel Virtualization Technology. Intel will finally release its dual-core Xeon 3000 series processors towards the middle of the week. Accompanying the Xeon 3000 series processor launch is Intel’s Kaylo server platform and Intel 3000 Mukilteo 2 and 3010 Mukilteo 2P chipsets.

While the Xeon 3000 series and Intel 3000/3010 chipsets will have a silent launch, Intel plans a high profile launch for its new storage products. Launching the same day as the Intel Xeon 3000 series and 3000/3010 chipsets are three new storage products. Intel will launch Sunrise Lake, Val Vista and Chevelon products. Sunrise Lake is a single chip SAS and SATA controller with an integrated storage processor while Val Vista is an SAS/SATA controller. Chevelon will be a high performance system-on-chip processor that’s based on a 90nm Xscale core, possibly replacing Intel’s previous i960. This completes Intel product launches for September.

October will be a quiet month for Intel, with the only product releasing being a Celeron D. Nevertheless the Celeron D 347 is expected to silently launch the fourth week of October. It will be based on Cedar Mill with a 3.06 GHz clock speed on a 533 MHz front-side bus. L2 cache is bumped up to 512KB from 256K found on current Prescott based Celeron D’s. Intel Enhanced Memory 64 Technology and Execute Disable Bit are supported with the Celeron D 347.

November brings the official launch of Intel’s quad-core processors for desktop, server and workstation environments. In mid-November Intel will launch its quad-core Clovertown Xeon 5300 series and quad-core Kentsfield Intel Core 2 Extreme processors. Quad-core Clovertown Xeons will launch with models X5355, E5345, E5320 and E5310 as previously reported. Kentsfield based quad-core Intel Core 2 Extreme will launch with a single QX6700 model. The Intel Core 2 Extreme QX6700 is expected to launch at 2.66 GHz and target entry-level workstation and enthusiast markets. Intel is planning a high profile launch for Clovertown Xeon 5300 series and Kentsfield Core 2 Extreme QX6700 processors.

Intel has nothing planned for the month of December, though January brings plenty of launches. On the desktop side of things Intel will introduce four new processors. Two new Core 2 architecture processors will arrive in the form of the Kentsfield Core 2 Quadro Q6600 and Core 2 Duo E4300. The Kentsfield Core 2 Quadro Q6600 will be clocked at 2.4 GHz while the E4300 will be clocked at 1.8 GHz. Intel’s Core 2 Duo E4300 is based on Conroe with 2MB of L2 cache and an 800 MHz front-side bus. The Kentsfield Core 2 Quadro Q6600 and Core 2 Duo E4300 launches will be a medium profile launch. Also launching in January is the previously reported quad-core Kentsfield Xeon 3200 series for single-processor servers and workstations.

Intel will also release two new Netburst based processors in the form of the Pentium D 935 and Celeron D 365. The Pentium D 935 will be similar to the 925 launching next week with a 3.2 GHz clock speed. Intel’s Celeron D 365 will be based on Cedar Mill like the 347 launching next month. Clock speed of the Celeron D 365 will be 3.60 GHz. No exact date has been given for the launch of the Core 2 Quadro Q6600, Core 2 Duo E4300, Xeon 3200 series, Pentium D 935 and Celeron D 365 except sometime between the second and last week of January.


Comments     Threshold


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

Wtf is with all the technology advances?
By Nightmare225 on 9/21/2006 9:09:42 PM , Rating: 1
I just built an e6600 system and in January there will be a replacement for this processor, why? Does Intel want to piss off its customers?




By Pirks on 9/21/2006 9:25:31 PM , Rating: 1
this is life in the world of fashion, man. follow the latest trendy fad or join the women that constantly lament about buying new clothes 'cause yesterday's green is less fashionable than today's red ;))


By Capt Caveman on 9/21/2006 9:27:04 PM , Rating: 4
What are you talking about?

The Q6600 is Quad-core and not a replacement for the E6600. It's the second Kentsfield Quadro CPU that's coming out.


RE: Wtf is with all the technology advances?
By D1abolus on 9/21/2006 11:12:46 PM , Rating: 2
So the whole tech world should stagnate in order for your purchase to not become obsolete? ROFL

It's not as if your PC will stop running the latest software. If it bothers you that your rig isn't the latest and greatest anymore, just stop reading tech news; go read Cosmo or something for the next 2 years.

I'm still waiting to put my money down for a E6600 to replace the 3.46EE I currently have. Looks like I'll be waiting a bit longer... :)


RE: Wtf is with all the technology advances?
By Capt Caveman on 9/21/2006 11:30:53 PM , Rating: 2
The Q6600 I believe will be price over $600. So, unless you do a whole lot of encoding, a E6600 is the way to go.


By PlasmaBomb on 9/22/2006 9:59:07 AM , Rating: 2
Yeah a quad core at $600+ doesn't really do a whole lot more than a similarly clocked E6600 at $300 for games and other apps. You really need the right software to get the juice outta it.


By AzureKevin on 9/21/2006 11:26:48 PM , Rating: 2
Same here dude.


But what about the front side bus??
By daschneider on 9/22/2006 11:45:49 AM , Rating: 1
Is anyone else wandering when Intel is going to do something about the front side bus to handle the memory access needs of multiple cores? Today we have dual cores sharing a 1066MHz bus. Soon we'll have twice the processing horsepower (4 cores) in the package but still a 1066MHz or maybe a 1333MHz bus. How are we going to keep all those cores feed with data from memory? As a gross approximation you need to double the memory bandwidth if you double the number of cores. The consequence of sharing the same slow bus with more and more cores is that we're going to see less and less improvement in performance as cores are added because they're going to be starved for data.




By s12033722 on 9/22/2006 12:20:02 PM , Rating: 2
What was that about "possibly replacing Intel's previous i960"? The 80303 was the last i960 chip, and it was over 4 years ago. Xscale (based on an ARM core) replaced it AGES ago.


RE: But what about the front side bus??
By ADDAvenger on 9/22/2006 3:16:34 PM , Rating: 2
I'm wondering the same thing. They can probably get by with a 1066 for quadcore, but they'll have to do something different when octcore comes out, as is seeming inevitable. Already they're down to 266mhz per core with quad-core (333mhz if they go up to a 1333FSB), granted they have a lot of shared cache so the bandwidth comparision isn't quite apples to apples, but it's close enough for me.
Now is this quadcore a native quad or two dual dies on the same chip? That will make a pretty big difference in the effectiveness of the FSB.


RE: But what about the front side bus??
By daschneider on 9/22/2006 3:47:13 PM , Rating: 2
It's two dual cores in one package, not four cores on one die. I'm not sure how the two dual cores are tied together or if they use the FSB to communicate with each other. I actually think packing two dual cores into one packages makes a lot of sense. A native quad core would be big and have a lot of surface area. That in turn would lower yield rates since the odds of a flaw on a die is pretty much a function of die size. Being able to discard two cores instead of all four due to a manufacturing flaw will help to keep prices down.


RE: But what about the front side bus??
By ADDAvenger on 9/22/2006 9:14:44 PM , Rating: 2
Yeah, it will improve yields for Intel, but at the expense of FSB crosstalk. I was pretty sure it was a two-die setup, since it is easier to do that than to do a redesign for native quad-core. Not that they won't do native quad-core, it'll come in a few months or so (I have no dates for that, I'm pretty much pulling stuff out of my butt on that little tidbit).

Then again, just about the only people buying this in the near future with be OCing it anyway, and AFAIK the only way to do that is by upping the FSB, so this may be a moot point.


By ADDAvenger on 9/22/2006 9:17:06 PM , Rating: 2
Before the IT guys yell at me, I was thinking of the desktop parts being OC'd, not the servers.


All for Intel and AMD
By Ftedla on 9/22/2006 10:34:43 AM , Rating: 2
Okay, so what's the issue here? Intel is advancing the cpu threshold and I for one am excited. I'm sure that some of my programs (for rendering and such) would highly benefit from a quad-core upgrade.

Now my only concern comes from the fact that ALMOST everytime Intel releases a new CPU, you have to buy a new motherboard. AMD is currently behind but with thier reputation for maintaning socket type I would highly recommend waiting to see thier next move.

I'll buy the $600 quad core cpu if I don't have to buy another $200 motherboard.




RE: All for Intel and AMD
By cochy on 9/22/2006 11:54:33 AM , Rating: 3
I don't agree. Aren't most current motherboards compatible with P4, PD, C2D, and quad core? LGA775, just like AM2.

What I am confused with however, is why Intel is continuing to release Netburst products? I thought Core was to be a complete replacement. At what market is Intel targeting with new Pentiums?


No X6900
By Webgod on 9/22/2006 11:03:45 AM , Rating: 2
Interesting, no X6900. I was thinking dual core 3.2G would be a better buy than the X6800 at 2.93G, but we're going straight to QUAD CORE. Let's get ready to rumble. Bring on UT2007, Crysis, and Quake Wars.




Intel will exploit every customer they can
By cornfedone on 9/22/06, Rating: -1
By Devious David on 9/22/2006 4:20:25 AM , Rating: 3
Oh yes, God help us! Intel is already coming out with Quad Core chips! What a horrible development! I'm being taken advantage of! And God forbid that AMD follow suit to compete! LOL.

I guess you would just as soon insist that Intel never should have come out with the 486, because all those poor poor people who were "FORCED" to buy a 386 shouldn't have their machines no longer top of the heap. It would be bad for their egos. Of course those who had 286s, well what about them? Should the 386 have never been developed due to the same logic?

Making newer/better/faster chips available and lowering prices on slightly older ones is taking advantage of the consumer!? What a crock.

Anti-trust laws? Smoke and mirrors. AMD is doing fine, thank you. Monopoly can only happen via government grant.


RE: Intel will exploit every customer they can
By ZmaxDP on 9/22/2006 2:09:32 PM , Rating: 2
In general I agree with you, however, I'd add one thing:

"Monopoly can only happen via government grant, or illegal manipulation of the market."

If a company violates the free market by somehow restricting competition or consumers, then they can create a monopoly without government interference. In fact, in these cases the government interference is a good thing. (Rare, I know) This is a minor concern in a stabilized market, like perhaps bed sheets, but in markets where there is a dominant player (like Intel is in mocroprocessors) then there is some concern for the latter happening. Intel has enough money and influence to possible buy out all good competitors, or to limit customers from buying other products (as they did with Dell for a long time).


By Devious David on 9/22/2006 3:48:28 PM , Rating: 3
Companies can't violate the free market - not without using The State. Only coercive State intervention violates the free market. The free market is composed of mutually agreed upon exchange. By logical extension, everything The State does is an inhibition of the free market and thus every action destroys wealth in some fashion.

This is why we have such an unequitable distribution of wealth in the world. The State picks winners and losers and through various means of intervention causes a GROSS misallocation of capital and resources. More intervention, more misallocation. And needless to say, "people" like Chevron and Pfizer have just a little more clout in DC than your or I.

If Intel were to buy AMD, there are opportunities for others. Just as there is now. Of course, government regulation/intervention of labor law, property zoning, taxation, environmental permitting, anti-dumping laws, etc etc etc. raise the bar of entry for new entrants into the market. So, it would take an artificially high threshold of private capital accumulation for a competitor to enter the market than it would in a true free market.

The United States, economically, is a quasi-feudalistic fascism. Plain and simple.


"We shipped it on Saturday. Then on Sunday, we rested." -- Steve Jobs on the iPad launch

Related Articles
Quad-core Xeon Details Unveiled
September 21, 2006, 9:02 AM
Intel Core 2 Quad Announced Internally
September 19, 2006, 1:56 AM
"Kentsfield" to Debut at 2.66GHz
August 16, 2006, 10:43 PM
Intel Adds Low End Xeons to Roadmap
July 19, 2006, 5:13 AM













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