Intel gave a briefing today on what the public should expect to see in its upcoming Core i7 processors and the X58 chipset coming with it.

Thomas from Intel’s Platform Applications Engineering division gave a complete overview of the Core i7 processors as well as the specifications for the X58 chipset. The Core i7 will change several things over the Core 2 processors, including an integrated memory controller as well as the new Intel QuickPath Interconnect, which will eliminate the Front Side Bus (FSB). Beyond that, we can expect the first wave of i7 Processors to come with 12 MB of unified L3 cache, and also have the OverSpeed Protection removed, good news for all you overclockers out there.


Another bonus coming with the new Nehalem architecture is how the processor will handle heat. With current processors, if a given heat is reached, the processor will automatically go to its lowest available frequency. While the processors will also lower the frequency, it will not automatically go down to the lowest possible point, instead utilizing a step process, going down one small step at a time until a desirable temperature is reached.


One of the most innovative features being found in the new Nehalem chips is the Turbo Mode, which will automatically overclock one or two cores in given situations. If only two of four available cores are needed, Turbo mode will kick in, shutting down the two unused cores as well as boosting the frequency of the other cores (add one performance bin to each). If only one core is needed, three will shut down and the frequency of the one active core will be boosted even more two performance bins).


The biggest change we will see with the new processors and chipset is a new socket, LGA1366. Because of the increased size and thus increased tension, Intel is changing how the processor attaches to the motherboard by separating the retention mechanism from the board. Instead of being soldered to the board, the retention mechanism will screw into the board after the manufacturer or enthusiast places the processor into the bracket. Also, because everything is bigger, you’ll need a new Heat_sink or cooling solution, so if you plan on investing in a new i7/X58 setup, be prepared to buy a new aftermarket heatsink or liquid cooling system.

The new X58 chipset will bring a lot of welcome additions and changes, including three memory channels, which will allow for up to 24 GB of DDR3 memory. The new chipset will also use ICH10, so look for the expected eSATA as well as 12 USB ports.


As far as discrete graphics support is concerned, you should expect what is already available – two x16 or four x8. It’s also noteworthy that the new ICH10 will be Halogen free, making the new chipset a little more environmentally friendly. Also look for the inclusion of Intel Turbo Memory, allowing for up to 4GB of memory to be attached, which should help with performance across a wide array of applications.

Intel http://en.wikipedia.org/wiki/Intel_Corporation announced that the first ’Nehalem http://en.wikipedia.org/wiki/Intel_Nehalem_(microarchitecture) ’ family of processors to be made available will be desktop processors (Core i7) and high-performance server products code-named Nehalem-EP.

Intel said it is planning to manufacture a second server derivative designed for the expandable server market ("Nehalem-EX"), as well as desktop (Havendale and Lynnfield) and mobile (Auburndale and Clarksfield) client versions in the second half of 2009.

All Nehalem processors will be quad-core http://en.wikipedia.org/wiki/Multi-core versions initially, processing up to eight threads simultaneously, thanks to the use of a revived and updated Hyperthreading technology.

Intel’s 6-core processor code-named Dunnington will be called Xeon X7460 and is expected to become available in servers beginning next month. Intel claims that servers based on the chip already have broken performance records, including an 8-socket 48-core IBM System x3950 M2 server, which became the first platform to break the 1 million tpmC barrier on the TPC-C benchmark, Intel said.

Intel also demonstrated the company’s first mobile quad-core processor and, for the first time, the next-generation Calpella mobile platform, as well as Intel’s upcoming solid state disk drives, which will debut in 32 GB and 80 GB flavors http://en.wikipedia.org/wiki/Flavor .

Yorktown Heights (NY) - IBM http://en.wikipedia.org/wiki/IBM and its chip development partners, including AMD http://en.wikipedia.org/wiki/Advanced_Micro_Devices , made a stunning announcement today, apparently beating Intel in the successful production of the first functional 22 nm http://en.wikipedia.org/wiki/22_nanometer SRAM cell. 22 nm processors are still three years out in the future, but IBM’s news is a good sign that chip manufacturer will be able to easily scale to this new level by the end of 2011. It appears that, for the first time in several decades, Intel may have to put some extra time into its research and development efforts to make sure it can keep its manufacturing lead at 22 nm and beyond.

SRAM chips are typically the first semiconductor devices to test a new manufacturing process as a precursor to actual microprocessors. The devices developed and manufactured by AMD, Freescale http://en.wikipedia.org/wiki/Freescale_Semiconductor , IBM STMicroelectronics, Toshiba and the College of Nanoscale Science and Engineering (CNSE) were built in a traditional six-transistor http://en.wikipedia.org/wiki/Transistor design on a 300 mm wafer and had a memory cell size of just 0.1 μm2, which compares to Intel’s 45 nm SRAM cell size (the test chip that was used for today’s 45 nm processors) of 0.346 μm2.

A 22 nm chip is two generations out in the future and AMD even has to catch up with Intel’s 45 nm. Intel presented the first 32 nm SRAM cell wafer in September of last year and in fact is not expected to show 22 nm SRAM cells for at least another year, while first 32 nm CPU prototypes could be shown at IDF this week.

IBM said that it is on track with its 32 nm process and promises that it will use a "leading 32 nm high-K metal gate technology that no other company or consortium can match." IBM did not provide further details to substantiate this claim, however, Intel has been using its high-K metal gate technology since the introduction of the 45 nm Penryn processors in late 2007.

While we are far from actual 22 nm and 32 nm products, it is clear that IBM and its partners are turning up the heat on Intel. For the first time in decades, there could actually be an interesting race towards a new production node.

NVIDIA's really pushing the GPU-as-CPU angle at SIGGRAPH this year -- we've already seen the PhysX and CUDA-powered GeForce Power Pack for consumers, and the company is also updating the Quadro Plex series of visual co-processors for workstation customers. The new Quadro Plex 2200 D2, designed for large datasets and models, crunches data through two Quadro FX 5800 GPUs (totalling 480 CUDA cores) and 8GB of RAM, while the Quadro Plex 2100 D2 is optimized for large multidisplay rigs with four Quadro FX 4700 GPUs and support for up to eight monitors. Sounds fun -- and we're guessing the people who can justify the $10,500 starting price for these rigs think so too.
Darn, Too bad it can't go any cheaper!:(

As if the whole defective NVIDIA GPU situation couldn't get any more confusing, The Inquirer is now reporting that the previous batch of bad GPUs may be far from the end of NVIDIA's problems. Apparently, four unspecified board partners are now saying that they're seeing G92 and G94 chips going bad at "high rates" as well, and in both desktop and laptop cards no less. That includes 8800GT, 8800GTS, 8800GS graphics cards, "several mobile flavors" of the 8800, "most" of the 9800 suffixes, and a few 9600 variants, all of which are based on the G92. As for the G94, it seems the only card affected is the 9600GT. Of course, none of this is nearly as set in stone as the previous lot of problems, but we have a sneaking suspicion this won't be last we hear about it.