Home Tags CUDA


AMD’s game plan to become a machine-learning giant

Right now, the market for GPUs for use in machine learning is essentially a market of one: Nvidia.AMD, the only other major discrete GPU vendor of consequence, holds around 30 percent of the market for total GPU sales compared to Nvidiarsquo;s 70 percent.

For machine-learning work, though, Nvidiarsquo;s lead is near-total. Not just because all the major clouds with GPU support are overwhelmingly Nvidia-powered, but because the GPU middleware used in machine learning is by and large Nvidiarsquo;s own CUDA.[ Roundup: TensorFlow, Spark MLlib, Scikit-learn, MXNet, Microsoft Cognitive Toolkit, and Caffe machine learning and deep learning frameworks. | Get a digest of the dayrsquo;s top tech stories in the InfoWorld Daily newsletter. ]AMD has long had plans to fight back.
Itrsquo;s been prepping hardware that can compete with Nividia on performance and price, but itrsquo;s also ginning up a platform for vendor-neutral GPU programming resourcesnbsp;— a way for developers to freely choose AMD when putting together a GPU-powered solution without worrying about software support.To read this article in full or to leave a comment, please click here

Nvidia delivers new and improved Titan Xp—3,840 cores, 550GB/s memory bandwidth

Full-fat GP102 Pascal in a consumer card; same cost as previous Titan X. On sale today.

Nvidia GTX 1080 Ti review: The fastest graphics card, again

Feel free to save us from the monotony and Nvidia's high prices any time, AMD.

Nvidia GeForce GTX 1080Ti will launch for $699 “next week”

Cores, RAM confirmed; normal 1080 gets price cut.

Nvidia Quadro GP100: Big Pascal, HBM2, and NVLink comes to workstations

Big Pascal on a PCIe interface is coming for those that need 64-bit double precision math.