New research by Big Blue could see CPUs running up to 40% cooler in the near future, without needing prices to rocket. In a paper released at the IEEE Semi-Therm conference, IBM published ideas designed to make thermal paste more efficient, therefore allowing for an improved cooling capacity leading to cooler processors. According to IBM, current thermal paste techniques can lead to 40% of heat given off by the CPU being absorbed by the particles in the paste, largely caused by the fact that the particles don't spread evenly, leading to the "Magic Cross" shown in the image below and to the left. IBM's new technology involves integrating micro-meter length trenches into the copper cap that sits above the CPU core (below on the right), which will allow thermal paste to be more evenly distributed and will lead to a third less thermal paste being required, as well as halving the pressure required to fit a CPU cooler. IBM still needs to finalise the research, but this relatively in-expensive method could be integrated into new CPUs and coolers before too long. For more detailed information you can read IBM's press release.

At the 2007 Optical Fiber Conference, IBM scientists will reveal a prototype optical transceiver chipset capable of reaching speeds at least eight times faster than optical components available today. The breakthrough could transform how data is accessed, shared and used across the Web for corporate and consumer networks. The transceiver is fast enough to reduce the download time for a typical high definition feature-length film to a single second compared to 30 minutes or more.

IBM Previews New Accessibility Tool for People With Low or No Vision to Access Multimedia Web Content

TOKYO & ARMONK, NY - 13 Mar 2007: IBM (NYSE: IBM) today announced a new emerging technology that helps blind and visually impaired people experience streaming video and animation on the Internet. Designed at IBM's Tokyo Research Laboratory, the new multimedia browsing accessibility tool potentially opens a world of rich content to visually impaired people around the world, who number more than 161 million.

Computing giant IBM is planning yet another supercomputer, with this one set to take the performance crown away from its very own BlueGene. The computer is being built for the Department of Energy and will be housed at the Los Alamos National Laboratory in New Mexico, costing $110 million to construct. Named Roadrunner, this new hybrid will use a unique design combining 16,000 conventional AMD Opteron Cores alongside 16,000 Cell processors (yes, the ones in the PS3), with calculations being shared between the two. According to IBM, the Cell processors will act as the workhorse, completing the major floating point calculations, whilst the Opterons will act as the system interface processors and the transactional backbones between the nodes. Once the machine is finished it ought to run at one petaflop - that's one trillion calculations each second, more than capable of belittling BlueGene, which is capable of a mere 280 teraflops.

Harnessing the power of the Cell processor, developed by IBM, Sony and Toshiba, is no easy task - IBM has therefore decided to get together with Vivendi and offer hands on programming sessions on the processor to game developers.

"The Cell Broadband Engine is complex high-performance architecture that processes millions of pieces of information per second to deliver highly detailed graphics," said Hal Lasky, vice president of consumer, media and entertainment for IBM Global Engineering Solutions in a statement. "Our focus is also on enabling the broader eco-system of game developers to fully utilize the power of the Cell Broadband Engine. We hope this takes us to a plateau that has never been approached before."

Clinton Keith, chief technology officer for High Moon, admitted it "would have been great to get it before launch," but the information wasn't available at the time, so they are seeking it out now.

The performance of the Cell is such that IBM has discussed using it in supercomputers and has already released one blade server based on the Cell, the QS20, which provided 205 GFLOPS of floating point performance. With a rackmount chassis, performance could reach 2.8TFLOPs, which would easily put such a system in the Top 500 supercomputer list.

SAN JOSE, Calif. - IBM has devised a way to triple the amount of memory stored on computer chips and double the performance of data-hungry processors by replacing a problematic type of memory with a variety that uses much less space on the slice of silicon.

International Business Machines Corp. said Wednesday that its new memory technology will help unclog crippling bottlenecks that build up as increasingly powerful microprocessors attempt to retrieve data from a separate memory chip faster than it can be delivered.

Armonk, N.Y.-based IBM said its solution entails swapping out most of the static random access memory, or SRAM, used to store information directly on computer chips and integrating onto the chip another kind of memory, known as dynamic random access memory, or DRAM.

IBM said it has been able to speed up the DRAM to the point where it's nearly as fast as SRAM, and that the result is a type of memory known as embedded DRAM, or eDRAM, that helps boost the performance of chips with multiple core calculating engines and is particularly suited for enabling the movement of graphics in gaming and other multimedia applications.FoxNews

The Inquirer reports that IBM has sold off more than a quarter of its shares in the Chinese computer maker Lenovo.

Lenovo gave IBM the shares as part payment when it bought IBM's PC outfit last year.

IBM sold 300 million shares at a fairly low price, triggering a seven per cent share slide before there was a suspension in sales.

IBM made $123 million by selling the shares which made up about 3.5 percent of Lenovo's holdings.

Lenovo had seen its stock increase in value this week, unfortunately, IBM's sale has wiped out most those gains.

At the International Electron Device Meeting (IEDM) today, IBM and AMD presented papers describing the use of immersion lithography, ultra-low-K interconnect dielectrics, and multiple enhanced transistor strain techniques for application to the 45nm microprocessor process generation. AMD and IBM expect the first 45nm products using immersion lithography and ultra-low-K interconnect dielectrics to be available in mid-2008.