GlobalFoundries offers a little preview at one of the upcoming Fusion processors from AMD, the Orochi based on the Bulldozer architecture as Softpedia reports.

Bobcat’s firstborn will be the 40nm Ontario mobile chip and comes with embedded DirectX 11 graphics. is intended for desktops and servers with Orochi being one of the flagship chips.

No real information was provided on this 8-core product besides that it will be the second 32nm-based product and will be built immediately after the Llano. The Orochi will feature a set of 4 Bulldozer modules, which provide 8 cores and 8 threads, and will probably have an AM3+ package with an integrated DDR3 memory controller.

On the other hand, the individual L2 caches will be of 1MB and will support dual-channel DDR3 and will, of course, have its own built-in graphics with support for DirectX 11.The APU should also feature an integrated northbridge, integrated PCI Express root complex, a HyperTransport interface to the chipset and an integrated SIMD array with 480 stream processors.

The Llano will have to compete with the Sandy Bridge processors from Intel, which also have integrated graphics and will feature better performance than the current Core series [via Softpedia / PCPer].

The Germany based tech site Computer Base seems to have their hands on a Intel’s 32nm Sandy Bridge roadmap for the desktop and notebook market.

It seems Intel plans to launch Sandy Bridge as the Core i7 2000 series. Several processors will have a suffix to identify special features of the processor, such as the “K” for unlocked chips, “S” for energy-efficient versions, “T” for dual-core 35W TDP models, “M” for laptop processors, “QM” for quad-core laptop CPUs and “XM” for the fastest laptop chips.

Incoming search terms:

• intel sandy bridge
• Sandy Bridge laptop
• sandy bridge processor
• sandy Bridge
• intel sandy bridge roadmap
• technology roadmap intel processor
• sandybridge
• Notebook CPU Lists for reducing
• intel technology roadmap
• intel laptop roadmap

AMDs Fusion processor will be a merge between a GPU and a CPU, there’s a new ETA. Currently, Fusion 32nm is scheduled to debut sometime during 2011.

According to AMD, the Barzos platform that will be paired with the Fusion processors will be targeted at ultra-thin notebooks. They will supposedly operate on less power than current solutions, which means that the chip and platform might be very suited for tablet PCs. This fact comes even though AMD is “currently not painting a strategy” around such devices. The Fusion APU (accelerated processing unit) will be codenamed Ontario and will integrate the CPU and GPU on a single die, similar to what Intel’s 32nm CPUs do.

All APUs will be made based on the 32nm manufacturing process. They will target not just ultra-thin and ultra-mobile systems, but the line will also feature desktop PC offerings. The company will still monitor the “exciting” tablet market and will watch out for an opportune moment to enter it.

Most speculations around the APU deal with the actual graphics capabilities. Considering the extensive graphical know-how from ATI, there is little doubt that AMD should be capable of making APUs capable of running even the most modern games, even with all the difficulties of combining the two technologies into a single die. However, making such a super powered chip might actually threaten AMD’s graphics card sales. As such, while the Fusion is expected to surpass Intel’s 32nm chips, a high-end graphics performance may be out of the question.

Alongside a 6-core launch for consumers, in the enterprise segment Intel will start shipping eight-core Nehalem-EX processors this month already. The pending 8-core Xeons will be able to run in four-socket servers and, by leveraging the two threads of each core, the logical CPU count can reach 64 cores on a single system.

This high scalability is complemented by 256KB L2 dedicated cache memory for each core, a shared L3 cache of 24MB and even Turbo Boost technology, which can disable unused cores in order to safely and effectively boost the heavily used ones. The 8-core also has a memory controller with four DDR3 memory channels and is built on the 45nm HKMG process (it is based on the Nehalem architecture).

The 8-cores will make their official debut later this month. In addition, Intel is preparing to formally add its Westmere-EP processors around the same time. These will be based on the 32nm Westmere process and will likely have six cores (aimed at two-socket servers), a 12MB L3 cache and three DDR3 channels.

In this video, Intel engineers from the Technology and Manufacturing Group and computer chip design and test teams show and tell how they helped create Westmere, the codename for the world’s first 32nm (Westmere) microprocessor.

This multichip processor will be available in 2010, and will feature CPU, graphics, memory controller, Intel Hyper-Threading and Turbo Boost all on one package.

It’s taken many years (and a $7 billion manufacturing investment) to get these Westmere chips onto silicon, with the Clarkdale (desktop) and Arrandale (notebook) lines expected to start showing up in over-sized cardboard boxes before the end of the year. What’s next for 45nm chips, like those new Lynnfields? A slow fade to obsolescence, as they’re likely to be the last of the breed.

Click here to view the embedded video.

Chinese website Expreview published some screenshots of Intel’s Gulftown, this 32nm six-core LGA1366 part is expected to be launched in the first half of 2010 under the Core i9 brand. The sample operated at a frequency of 2.4GHz (18x133MHz) and features a massive 12MB L3 cache.

Gulftown will be socket 1366 compatible and work with all X58 motherboards after a BIOS upgrade. Check out the pics.

Incoming search terms:

• intel core i8
• intel i8
• core i8
• backlit usb socket pcb
• intel i9
• intel core 9
• after intel core i8
• easypix digi scope sample
• dell precision intel (R) xeon (TM) 3 2
• intel core 32

Information about Intel’s next generation 32nm CPU has found its way onto the internet. The new chunk of silicon is being called SandyBridge and represents a new architecture all together.

Intel has always followed a tick-tock schedule. In this way each CPU design gets to run on two separate Processes while each Process will see a new architectural design. SandyBridge is the new design for 32nm.

The new CPU will feature more integration including an integrated graphics chip. So far SandyBridge has only been taped out, but they still expect it to reach production grade by 2011. SandyBridge should maintain the 1156 Socket used by the upcoming i5 CPU.

Read the whole story here.

An exclusive look at DFI’s upcoming codenamed “Ibex Peak” single chip design Intel P55 Express based motherboards.

DFI’s UT P55-T3eH8 and DK P55-T3eH6 motherboards that we saw today are not mass production level samples and will see some changes – you will see some hand wiring and so on that obviously won’t be there in the final shipping product. You’ll probably be able to buy them sometime in September this year, once the compatible Socket LGA-1156 Core i5 quad-core “Lynnfield” processors go on sale.

We think Intel pushed back the P55 launch again and again to allow the market to shift more of its P45 and Core 2 Duo parts before the new platform comes in and takes over the mainstream level.

P55 also supports the codenamed “Havendale” Core i5 dual-core processor which includes an on-die GPU, but we were told this will be phased out and never actually reach mass production (just for testing purposes). Instead, a newer 32nm version of it will be released sometime in Q1 next year, according to our sources.

Back onto the DFI boards – UT is DFI’s ultra high-end board and DK is a slightly cut-down version, coming in a little cheaper. The difference between the two is that the UT features a more advanced power system using a digital power management (PWM) system. UT also gets a more advanced audio add-on card as well as a better cooling system which exhausts air outside of the case – we were not shown this, DFI has to keep some surprises until Computex in a couple weeks time.

One feature you will probably notice straight away on the UT board is a the SO-DIMM like looking slot between the DDR3 DIMM slots and the IDE connector. This is an Open NAND Flash Interface (ONFI) slot that allows for an Intel made and sold 4GB or 8GB flash module to be installed to boost system performance. You can think of it as a roughly similar feature to Microsoft’s ReadyBoost in Windows Vista, but much quicker than plugging in a USB 2.0 pen drive. It acts as an extra type of cache (some are calling it “L4″ cache) and apparently you’ll even be able to access this flash memory as a hard drive and even install your main operating system on it.

Some common features on both boards – you’ll notice that the CPU socket is different from what you get on X58 and P45 boards since these boards only support Socket LGA-1156 processors – namely Core i5, as we spoke about above. Core i5 is designed to be the mainstream and cheaper version CPU of the high-end enthusiast Core i7 processor. One of the big differences is that Core i5 will only have dual-channel DDR3 support whereas Core i7 has triple-channel.

At the time of writing, DFI was unable to provide any details on pricing for either motherboard model. However, they did state a little dissatisfaction over Intel charging the same price for P55 chipsets as the P45 north and southbridge combo. Keep in mind that P55 is a single chip solution and P45 was a dual chip solution – something is clearly not quite adding up.

DFI UT P55-T3eH8 motherboard

DFI DK P55-T3eH6 motherboard

Incoming search terms:

• DFI P45-T2RS Plus
• dfi p45 hybrid
• dfi special force
• sf dfi
• dfi x58
• DFI P45
• toshıba satellite L635 BIOS CHIP
• smart pwm dfi x58
• sf dfi video
• sf dfi pictures

Intel CEO Paul Otellini unveiled his company has shipped thousands of 32nm Westmere processor samples to over 30 EOM customers:

“Our 32nm process is very healthy and shows great promise for us. This is the reason we recently announced a $7.0 billion investment in plants and equipment to accelerate high volume deployment of this technology,” noted Otellini. “We have pulled in Westmere, our fist 32nm product family, and will now be shipping those products later this year. We have shipped thousands of Westmere samples to over 30 EOM customers already.”

Although fab utilization rates fell in Q1, executives noted in the conference call that utilization rates would trend upwards as demand returned and the 32nm ramp gather pace. Expenses for the ramp are expected to peak in 2Q, Intel said.

In this sentiment, here’s a picture of Clarkdale, the first 32nm desktop part.