News
 

Bookmark and Share

(18) 

Intel Corp. may not only be making central processing units (CPUs) with four processing engines for high-performance workstation, desktop and server machines, but may also target powerful laptops with its quad-core chips, according to an analyst.

Jim McGregor, principal analyst of Microprocessor Report from In-Stat, said in his recent report that Intel is also planning a quad-core mobile processor. TG Daily web-site claims that the observer believed that Intel’s quad-core mobile processor could be compatible with “Santa Rosa” platform, “which will be launched in Q2 of 2007 as a refresh for the Merom processor”.

The code-named Santa Rosa platform and Crestine chipset will boost the processor system bus speed of the processor known under Merom name – expected to be introduced later this year – to 800MHz, which should significantly increase performance of this dual-core chip in multimedia tasks that require high bandwidth, according to earlier reports. It is yet uncertain whether Crestine chipset will support higher-speed dual-channel DDR2 memory, but in the past years memory bus clock-speed was usually the same as processor system bus speed.

According to the new report, Intel asserted its technical prowess and squashed concerns about losing market share to its rival by opening the doors to the new Core micro-architecture and the future-product roadmaps at the spring 2006 Intel Developer Forum (IDF). The report by Mr. McGregor claims that the result would be an aggressive transition to the Core micro-architecture for all mobile PC, desktop PC, and volume server dual- and quad-core processors within the following 12 months. The transition would begin with the Merom mobile PC, Conroe desktop PC, and Sossaman server processors, beginning in Q3 2006.

Earlier Intel officially showcased its first quad-core processor code-named Clovertown, which is a product that features two separate chips on a single piece of substrate. Additionally, Intel plans to introduce quad-core Kentsfield, next-generation “extreme” desktop processor. It is highly likely that Kentsfield features two Conroe dual-core chips on the same piece of substrate, thus, providing four execution engines. The Kentsfield is expected to have four cores, 4MB or 8MB cache, 1066MHz or 1333MHz processor system bus and other characteristics similar to Conroe.

Intel did not comment on the news-story.

Discussion

Comments currently: 18
Discussion started: 04/14/06 04:08:51 PM
Latest comment: 08/25/06 09:38:51 AM
Expand all threads | Collapse all threads

[1-2]

1. 
Is there really that big of a market for quad core mobile chips? Considering the extra power they would consume, and the nature of mobile computing, it seems like a relatively small market to me, and that's putting it mildly.

With regards to transitioning to the P8, it is hard to imagine Intel being in a worse situation when those CPUs come out. I have to believe they will delay the P8s for a while to allow the transition to take place, because who will buy a P7 once the P8 is out? They have a difficult time competing with the mediocre Athlon, coming from a knock-off brand company. How will they compete against a vastly superior processor like the P8, from the leading company (themselves) when it becomes available? How will they continue to sell the wretched P7s they are making while the transition is still in progress? Why would anyone buy these chips when the P8 is out? Actually, why would anyone buy them even now with the Athlon 64 out? But, it will get even worse.

Has anyone else noticed that Intel "odd" numbered processors suck big time, whereas the even numbered ones are excellent? 8086 kind of sucked, but it was a big improvement over the 8085, actually, an enormous one. So, call it mediocre. 80186 sucked so bad it was barely used except in embedded. 80286 was shockingly fast, and opened up multitasking operating systems and massive amounts of memory. I still remember using an PC AT and being stunned by the performance. The 386 was a miserable processor, and could not even perform as well as a 286 on 16-bit code clock normalized. It added 32-bit processing, but was so slow it really didn't matter all that much, except to run multiple real mode applications. The 486 ran like a raped ape too, being roughly twice as fast as the 386 clock normalized. It also allowed multiplying the internal bus by the memory bus, because it had a L1 cache (a whopping 8K on the original 486). The Pentium kind of sucked too, particularly since one could get a 486 running at 100 MHz at the same time Intel was selling their buggy, very hot 60 and 66 MHz Pentiums. The Pentium wasn't that bad though, but still was disappointing. The P6 core is legendary and put an end to the RISC era for workstations, essentially. It needs no further comment. The P7 is also legendary, at sucking. It's a fascinatingly bad design that I still find difficult to believe. How did Intel let this piece of crap see the light of day? A huge mystery I'll never understand. Now we have the P8, which I haven't seen yet, but by all indications seems utterly fantastic, well, within relative terms. The days of the 286 type performance increases are over, and even the 486 type improvement will probably never happen again. But, considering the crap we've been getting for the past few years from Intel and AMD, this is a huge leap forward. I knew it would be better than the Athlon 64, but I had no idea it would that much better. The bad part is, the P9 is probably going to be another Intel dog. Weird how that pattern is so hard for them to deviate from.
0 0 [Posted by:  | Date: 04/14/06 04:08:51 PM]
Reply
- collapse thread

 
dude I don't know what planet you're from but the original pentium (P5) certainly did not suck at the time.

Yea the 486 was running at 100MHz while the P5 ran at 66MHz, but the Pentium was also the first CISC/RISC combination bringing superscalar and out of order excecution to X86. For 10 years leading up to that moment, EVERYONE said RISC would replace the CISC type CPUs that comprised the x86 lineup...and intel shut the entire argument down with the pentium- a RISC-CISC hybrid.

The original Pentium might have had a slower clock...but it was more that twice as fast clock for clock, especially in floating point. That and a year later you could throw in a Pentium 133 or 150/166 in the same socket. The 486's socket was dead.
0 0 [Posted by:  | Date: 04/15/06 10:54:41 AM]
Reply

2. 
Once again, Intel has to release ANOTHER chipset for a new cpu.

Knowing Intel lately, they'll just take 2 conroes, shrink to 45nm, super glue them together, and call it quad core.
0 0 [Posted by:  | Date: 04/15/06 10:55:38 AM]
Reply
- collapse thread

 
They would be Meroms, but if it works identically to pumping 4 cores on one die, why would you care how they do it? As I mentioned in a previous message, you can probably get better performance by doing it the way Intel does, because if have even one bad core when they are all on the same die, you have a defective die. So, you have to lower the clock speeds to increase acceptable yields. If you use 2-core per die, if one core can not run at the desired clock speed, you don't have to throw out, or de-clock, the other three, just the one is it stuck with.

I guess I don't understand why people complain about Intel doing this. If you can explain why you dislike it, it might help me understand. To me, if it performs the same, why would you care?
0 0 [Posted by:  | Date: 04/15/06 04:55:34 PM]
Reply
 
It still uses FSB!
0 0 [Posted by:  | Date: 04/16/06 06:40:49 PM]
Reply
 
That's irrelevant to this point. Whether they were four cores on the same die, or a pair of two cores stuck together, they'd share a FSB. I am asking why people have a problem with Intel slapping together two single cores onto one package as they do now, rather than two cores on a single die, or in this case a pair of dual cores rather than four processors on a single die.

People seem to hate this arrangment, although so far no one has given a lucid answer as to why.

I am guessing people simply are looking for a way to take a shot at Intel and really do not understand a heck of a lot, and this seems like a negative with their fuzzy understanding. It's all I can come up with until someone explains it a little better.

I think a lot of people are scared that Intel is going to crush AMD now, since AMD's assault on marketshare has already taken a downturn, and it is widely accepted that the Athlon 64 is not remotely competitive with the Conroe. It's a valid concern, I am as well, particularly since AMD is being badly mismanaged and are spending money for capacity they will not need. AMD already can meet demand, and they are just starting production in their new 300 mm fabrication facility, yet they are outsourcing to Charter and want to build yet another facility. By all indications, they are already not gaining market share and Intel is dumping their garbage chips at bargain basement prices to facilitate this. Conroe is coming out, and everyone knows it is vastly superior to the Athlon 64. So why is AMD continuing to create so much capacity when they obviously don't need it? Their 300 mm Fab running full speed will offer a huge increase in capacity, well over double what they have. They need another, why?

When the Conroe comes out and becomes the processor everyone wants, what will AMD do with all these expenses and an inferior processor? If they kept lean and mean, they could offer them inexpensively and make money, and at the same time keep Intel from demanding so much. But, the way they are set up now with their high cost, they need to keep high margins and good market share, both of which are almost certain to reverse when Intel has the superior product. They could be headed for disaster with their stupidity, and Intel can break them easily with a price war, particularly with their superior chip. Hector needs to go, and go soon. They are going to be in the red soon with this jackass at the helm.
0 0 [Posted by:  | Date: 04/17/06 01:20:56 AM]
Reply
 
Go on carry that Intel cross. No one else is going to help you.
0 0 [Posted by:  | Date: 04/17/06 04:50:13 AM]
Reply

[1-2]

Add your Comment




Related news

Latest News

Monday, April 14, 2014

8:23 am | Microsoft Vows to Release Xbox 360 Emulator for Xbox One. Microsoft Xbox One May Gain Compatibility with Xbox 360 Games

Tuesday, April 1, 2014

10:39 am | Microsoft Reveals Kinect for Windows v2 Hardware. Launch of New Kinect for Windows Approaches

Tuesday, March 25, 2014

1:57 pm | Facebook to Acquire Virtual Reality Pioneer, Oculus VR. Facebook Considers Virtual Reality as Next-Gen Social Platform

1:35 pm | Intel Acquires Maker of Wearable Computing Devices. Basis Science Becomes Fully-Owned Subsidiary of Intel

Monday, March 24, 2014

10:53 pm | Global UHD TV Shipments Total 1.6 Million Units in 2013 – Analysts. China Ahead of the Whole World with 4K TV Adoption

10:40 pm | Crytek to Adopt AMD Mantle Mantle API for CryEngine. Leading Game Developer Adopts AMD Mantle

9:08 pm | Microsoft Unleashes DirectX 12: One API for PCs, Mobile Gadgets and Xbox One. Microsoft Promises Increased Performance, New Features with DirectX 12

3:33 pm | PowerVR Wizard: Imagination Reveals World’s First Ray-Tracing GPU IP for Mobile Devices. Imagination Technologies Brings Ray-Tracing, Hybrid Rendering Modes to Smartphones and Tablets

2:00 pm | Nokia Now Expects to Close Deal with Microsoft in Q2. Sale of Nokia’s Division to Close Next Month