Breathless reporting makes news of Nvidia's and Arm's collaboration on Xavier sound like earth shattering news. Nvidia and Arm recently announced that they would be working together to make Nvidia’s Deep learning Accelerator (NVDLA) accessible to programmers using Arm’s Trillium platform. Project Trillium was announced by Arm in February this year as a way for developers to get direct access from … Read more
Technology matters in March at SXSW, GDC and GTC
It’s ten years since Nvidia launched the GPU Technology Conference (albeit under a different name) to promote the use of their technology. This year, they are claiming an attendance of more than 8000 delegates along with one of the largest press and analyst delegations you will see outside of behemoths like CES and MWC. What started as a bold statement … Read more
The Machine Learning for Games panel (GDC, 21st March), hosted by Imagination Technologies and moderated by our very own Dr. Peddie, gave us a pretty good sense of where we are on some parts of that topic, as well as an indication that we need to be very careful to define terms. Panel mambers represented mobile, desktop, and console gaming. … Read more
Nvidia demonstrated real-time raytracing at GDC and again at GTC using their DGX station super computer through the DirectX Raytracing (DXR) API. That system provides a theoretical compute capability of over 63 TFLOPS for the total system, of which about 62 TFLOPS were used for the ray tracing. So a threshold or benchmark of sorts has been established for what … Read more
Nvidia demonstrated real-time raytracing at GDC and again at GTC using their DGX station super computer through the DirectX Raytracing (DXR) API. That system provides a theoretical compute capability of over 63 TFLOPS for the total system, of which about 62 TFLOPS were used for the ray tracing. So a threshold or benchmark of sorts has been established for what … Read more
It will take a little longer for smartphones Nvidia demonstrated Microsoft’s approach to raytracing for games at GDC and again at GTC using their DGX station super computer through the DXR API. The $50,000 DGX contains four NVLink 4-Way interconnected Nvidia Tesla V100 add-in boards (AIBs). Each AIB has a 5,120-shader Volta GPU running at 1.46 GHz with 16 GB … Read more
After alerting the media at the Supercomputing Conference in 2016, Nvidia introduced the DGX-1 supercomputer at GTC 2016, as the world's first deep learning supercomputer to meet the unlimited computing demands of artificial intelligence. The price was $149,000 and came equipped with eight Tesla V100s GPUs, linked together via next-gen Nvidia NVLink interconnect technology that ups the bandwidth per GPU … Read more
At GDC, Nvidia and Microsoft made a joint announcement about the possibility of doing real time ray tracing, something the CG community has striven for since ray tracing was popularized by Turner Whitted in 1979. Whitted did it on big computers at Bell Labs, just before the PC was introduced. Since then, scores of papers and books have been written … Read more
Using AI techniques, and powerful processor
Major versions of SPECviewperf and SPECwpc (to be renamed SPECworkstation) are on tap for Q2 of this year. SPECviewperf 13 will feature new volume visualization viewsets for energy and medical applications, radically new shaders, and updated viewsets to match more recent versions of the applications on which they are based. SPECworkstation 3.0 will include new GPU performance measurement, completely redesigned … Read more
Expanded SPECworkstation benchmark slated for Q2
