Nvidia Debuts New Graphics Processors For Artificial Intelligence To Compete With Intel

Nvidia Corp. debuted two new graphics processors for systems and computers that use artificial intelligence putting them directly in competition with chips giant Intel.

Nvidia hopes that its two new graphic processors would be utilized by tech companies in their products that run artificial intelligence software like smartphones, driverless cars and smart homes, according to a report by Bloomberg.

The chipmaker, which is the leading company in video gaming graphics, released a new set of graphics chips for running software that "makes split-second decisions needed when everything from phones to cars to internet search engines respond to inputs such as speech, images and moving objects," according to Bloomberg.

Nvidia calls the new chip "Tesla P4 chip" and is designed for servers used in massive data centers that process big data like Facebook, "The Division" and "World of Warcraft."

Based on Nvidia's Pascal design, the P4 is more than three times as efficient at processing images than its predecessor and 40 times more efficient than Intel server chips, as stated in Nvidia's report. Another new chip, called the P40, is designed for more-powerful single computers like supercomputers.

Nvidia is seeking to go head to head with Intel, which last month announced its own artificial intelligence chips. Both tech titans also want to make sure big data operators like Google not only rely on their technology but are not swayed to design their own custom solutions.

Nvidia told Bloomberg that its graphics chips, which "perform multiple small manipulations of data simultaneously," are apparently the perfect answer for computers and software that use artificial intelligence, stating that the P4 and P40 will make the use of AI easier.

The chipmaker announced the two new chips in its GPU conference in Beijing on Tuesday, as reported by Forbes. Along with the Tesla P4 and Tesla P40, Nvidia also revealed two new pieces of software - TensorRT and DeepStream designed to make inferencing in computing machines faster and resource efficient than existing systems on the processor market.

Forbes describes the two new chip as based on Nvidia's "newest Pascal architecture with the Tesla P4 is smaller in size and about the length of a pencil."

Intel countered saying that although its chips have less ability to work in parallel, they are more capable in general purpose computing and can offer the right solutions for everyday needs.

Bloomberg opined that data center owners, whose biggest cost is energy, are focused on components that can get the job done using less power.

The site speculates that the market for AI chips is still in its infancy. In its latest quarter, Nvidia's data center business more than doubled its sales to $151 million and most of this came from AI tasks. Intel's data center unit had sales of $4 billion, but very little of that revenue came from AI projects.

© 2017 iTech Post All rights reserved. Do not reproduce without permission.

More from iTechPost