First impression on unpacking the Q702 test unit was the solid feel and clean, minimalist styling.
Nvidia brings console gaming to ultrabooks with new GPU
- — 23 March, 2012 00:06
Nvidia hopes to bring console-style gaming to ultrabooks with its latest GeForce GT 640M graphics processing unit (GPU) for laptops, which was officially announced by the company on Thursday.
Ultrabooks are a new class of thin and light laptops that resemble Apple's MacBook Air, but most models come with integrated graphics processors that can handle only casual games. Nvidia's GT 640M GPU fits into the thin profile of ultrabooks and can handle the most demanding games, said Chris Daniel, product manager at Nvidia.
The GPU is two times faster than its predecessor, the GeForce GT 540M, which was used in weightier laptops. One ultrabook using the latest GPU is Acer's Timeline Ultra M3, which is 20 millimeters (0.79 inches) thick and weighs 5 pounds (2.27 kilograms). That is lighter than a comparable MSI GE620 laptop with a GT 540M, which weighs 5.7 pounds and is 37 mm thick.
However, the Timeline Ultra M3 feels slightly bigger and heavier than other ultrabooks like Hewlett-Packard's Folio and Dell's XPS 13, which don't have discrete graphics units. But the laptop fits within the design criteria set by Intel, which states that ultrabooks cannot be more than 21 millimeters thick (0.8 inches).
Daniel said the new GPU can handle all the games available in the market. Nvidia has design wins with top PC makers for the GPU, including HP, Lenovo, Dell, Sony and Toshiba. Laptops with the new GPU could become available in April, Daniel said.
The ultramobile GPU is a boon for laptop users editing video or playing sophisticated games, said Dean McCarron, principal analyst at Mercury Research.
"It opens the door for them to migrate to ultrabooks as well, or they'd have to stay with conventional notebooks that are heavier, thicker and bigger," McCarron said, adding that the GPU could possibly go in ultrabooks even thinner than Acer's Timeline Ultra M3.
The new chip is based on Nvidia's new Kepler architecture, McCarron said. The company focused on performance-per-watt with Kepler after teething problems with its previous Fermi architecture, which was criticized for being too power-hungry.
In addition to the size breakthrough, the graphics chips help preserve battery life by working with Intel's latest Core chips to share video processing, Daniel said. The discrete GPU, which is superior to integrated graphics, activates only when needed, helping preserve battery life.
Nvidia also launched the high-end GTX 680 GPU for desktops. The GPU has 1536 processing cores and draws 195 watts of power. Depending on the game, the GPU is 20 percent to 40 percent faster than its predecessor, the GTX 580. The GPU is priced at US$499 and is available worldwide immediately.
Nvidia is also expected to introduce new Tesla graphics chips based on Kepler for workstations, servers and supercomputers later this year. The Oak Ridge National Laboratory is building a supercomputer named Titan that will combine Advanced Micro Devices' Opteron 6200 series server processors with Tesla graphics cards based on Kepler to deliver 10 petaflops of performance, which could make it one of the fastest supercomputers in the world.