Accelerated processing teaches autonomous cars to drive in minutes - SiliconANGLE

Speed is the name of the game in the processor world, and the latest competitive sprint down the innovation track involves field programmable gate arrays, known as FPGAs. Since Intel Corp. announced FPGA acceleration platforms operating with Xeon CPUs early last month, several companies have been showcasing a number of use cases for the lightning-fast technology. At a computing conference last month, one firm created buzz among attendees with a machine learning-based demonstration where simulated cars were taught to drive in less time than it takes to eat a sandwich. "Within a few minutes and about 15 million simulations, the cars start driving better than humans," said John Lockwood (pictured), chief executive officer at Algo-Logic Systems Inc. "You can give [machines] man-years of experience in a few minutes with these scale-out computer systems." Lockwood visited theCUBE, SiliconANGLE Media's mobile livestreaming studio, and spoke with host Jeff Frick (@JeffFrick) during the recent Supercomputing event in Denver, Colorado.