Hardware

Intel told Ars today that its long-delayed Larrabee discrete graphics product has suffered yet another delay, so the company has had to "reset" its overall GPU strategy and reposition plans and its expectations for the first version of the Larrabee product. 

Specifically, Larrabee v1 is so delayed that, at the time it eventually launches, it just won't be competitive as a discrete graphics part, so Intel plans to wring some value out of it by putting it out as a test-bed for doing multicore graphics and supercomputing development. Intel will eventually put out a GPU, but may not be the one we've been calling "Larrabee" for the past few years.

Intel held a press event in downtown San Francisco on Wednesday. It was another Terascale event—Terascale being the company's test-bed for many-core technologies, and not an actual product. But the 48-core, Single-chip Cloud Computer (SCC) that Intel demonstrated is a lot closer to being a viable product than the 80-core "Polaris" chip that the chipmaker first unveiled in 2007.

Multi-hop mesh networks, confined to university labs at the start of this decade, are now widely available from commercial vendors. These vendors tout a number of advantages for mesh technologies: lower costs of deployment, easier administration, better coverage, and lower power consumption. Mesh networking is now being used in an impressive range of applications, from large-scale institutional deployments to networks of tiny sensors.

Mesh networking is sometimes mentioned as a solution to the much-discussed "last mile" problem in US telecommunications policy. Unfortunately, the inherent capacity limits of the wireless medium means that mesh networks are unlikely to provide a serious alternative to fiber or coax broadband connections in this market. Mesh is a reasonable way to provide broadband to consumers in developing countries who might not otherwise be able to afford access at all. But in the developed world, mesh technologies are best viewed as a supplement to wired Internet connections and traditional single-hop access points.

On the day before US companies took off for the long Thanksgiving weekend, Nokia's lawyers wished a number of major electronics manufacturers a very unhappy holiday. The cell phone maker filed suit in a California court, alleging that a price-fixing scheme the companies have already admitted to raised the prices it paid for the LCD screens used in its phones and other handheld devices. The LCD makers could be in for a major payout, as the suit alleges violations of both federal laws and those of over 20 states, and seeks triple damages; a similar suit was filed in the UK on the same day.

The charges cover a decade of misbehavior, a lot of which was uncovered by regulators in Europe, Japan, Korea, and the US. Major manufacturers of LCDs, like Sharp, LG, and Chungwa, apparently reached agreements that allowed them to divide the market for the hardware, keeping manufacturing artificially low, and prices high. Late last year continuing into this one, the US Department of Justice has been settling the cases through fines and, for some executives, actual jail sentences. Most significantly, from Nokia's perspective, the settlements included admissions of guilt.

The word "physicalization" is ten months old. This January, Rackable Systems launched a strange line of servers which defied all the conventional wisdom of server design by disaggregating larger servers into many smaller ones based on consumer parts, and in the process lowering power and performance density. Ars, among others, expressed skepticism over the strange design decisions, but launches from other major vendors suggest that, for some market segments, the server space is taking a turn in a novel new direction. Ars has covered this trend before, but we now take a more detailed look at it, with a survey of available physicalization offerings, analysis of the reasons for their adoption, and some predictions about the future.

An interdisciplinary team of researchers at IBM have presented at paper at the SC09 supercomputing conference describing a milestone in cognitive computing: the group's massively parallel cortical simulator, C2, now has the ability to simulate a brain with about 4.5 percent the cerebral cortex capacity of a human brain, and significantly more brain capacity than a cat.

You might think that Cray, the company whose name has practically been synonymous with "supercomputer" three decades, would be a regular fixture at the top of the Top 500 Supercomputer List. But you'd be mistaken. Today's first-place win by Cray marks the company's debut in the top slot—at long last, after 34 lists, Cray won.

This past Friday, Intel and AMD announced a settlement in their acrimonious antitrust dispute, with AMD clearly coming out on top to the tune of $1.25 billion in cash and a host of concessions. Both companies hosted conference calls that Friday morning, and later in the day the released excerpts from the agreement that makes up the non-cash portion of the deal, in which Intel agrees to a number of conditions that should make life much easier for AMD and its fab spinoff, GlobalFoundries.