GpuTechnologyConference

Latest

  • Caption Contest: NVIDIA CEO flaunts tattoo on stage, still serving cans of whoop-ass

    by 
    Richard Lai
    Richard Lai
    09.25.2010

    Oh sure, NVIDIA might not have had any physical chips to show off at the GPU Technology Conference earlier this week, but CEO Jen-Hsun Huang did have one very, very special exhibit up his sleeve. Paul: "And below the logo is a picture of all the shipping devices running Tegra 2." Don: "Here are the new logo designs we've been working on, Mr. Huang." Ross: "Take a cue from Peter Moore and go with temporary. You never know where you'll be in even just a few years' time." Chris: "Huang's Boxee Box tramp stamp, of course, would remain a secret." Joanna: "And this is why I hate gun, er GPU shows." Darren: "Whatever. At least it's not a Zune tattoo." Richard Lai: "Yo Intel, this is how embedded is done." Vlad: "Jen-Hsun's displays of support for Notion Ink are starting to get out of hand..."

  • Malata's SMB-A1011 Tegra 2 tablet spotted at GTC 2010, we go hands-on

    by 
    Sean Hollister
    Sean Hollister
    09.22.2010

    Despite perennial delays that have mysteriously kept every slate of note from formally shipping out, there are actually a surprising number of Tegra 2 tablets floating around. Like this 10-inch Malata SMB-A1011, for instance, which seems to be very closely related to Hannspree tablet and the Interpad -- all 1GHz, 1024 x 600 tablets with theoretically fabulous spec sheets including Froyo and HDMI-out. We found the device sitting alone and unloved at Allegorithmic's texture compression booth, but after spending a few minutes of quality time we figured out why -- though the Tegra 2 T20 was plenty powerful enough to tilt windmills in real time, build quality was severely lacking. Though not light by any stretch of the imagination, the chassis nonetheless flexed when we picked it up, and we nearly dropped it more than once due to the lack of a good place to grip. There's no stand of any sort to prop it up on a table, but the screen's viewing angles are so poor (especially in the vertical directions) as to completely wash out or darken the screen when we set it down... and at least a half-dozen times the capacitive multitouch digitizer totally failed to respond to our finger. Though Froyo looked clean and had plenty of real estate to work with, the official Google suite of apps (and the Android Market) were lacking, often a major concern even on quality Android devices, and possibly the nail in the coffin for this one at the €399 (about $528) that Hannspree and E-Noa figure their versions are worth. Take a gander at the half-baked hardware yourself in our gallery immediately below, and for Tegra's sake pray that things improve before the Malata hits the market. %Gallery-102961%

  • NVIDIA CEO: Tegra 3 almost done, Tegra 4 on the way, expect a new Tegra annually

    by 
    Sean Hollister
    Sean Hollister
    09.21.2010

    Though NVIDIA's GPU Technology Conference is primarily about the applications of GPU computing, CEO Jen-Hsun Huang hasn't shied away from revealing new silicon, and he just promised something quite unexpected to attending press: new Tegra chips. Though the Tegra 2 has yet to leave a single dent on the consumer marketplace -- the Boxee Box famously tossed it out -- Huang told us that not only is a Tegra 3 almost done and a Tegra 4 currently underway, but that we should expect a new Tegra SOC "every single year." Forgive us for being a mite skeptical of the company's ability to attract customers, but the only notches we see on Tegra's belt are the Zune HD... and Microsoft's failed Kin.

  • NVIDIA teams with PGI for CUDA-x86, gifts its brand of parallelism to the world

    by 
    Sean Hollister
    Sean Hollister
    09.21.2010

    NVIDIA's GPU Technology Conference 2010 just kicked off in San Jose, and CEO Jen-Hsun Huang has shared something interesting with us on stage -- thanks to a partnership with The Portland Group, it's bringing the CUDA parallel computing framework to x86. Previously limited to NVIDIA GPUs -- and the lynchpin of NVIDIA's argument for GPGPU computing -- CUDA applications will now run on "any computer, or any server in the world." Except those based on ARM, we suppose. Still no word on NVIDIA's x86 CPU.