On June 13, Intel had the GPU world in a flurry when it tweeted “Intel’s first GPU coming in 2020”. The media were quick to post stories of this incoming new GPU, which would add interesting competition to a market dominated by NVIDIA with AMD a distant second.
The two-horse race between NVIDIA and AMD (when it acquired ATI) has been going on for years though NVIDIA has been pulling further and further ahead with its innovation in using the GPU not just for consumer but in the enterprise space in specific areas such as supercomputing and deep learning.
ATI provided more of a competition when it was a standalone company prior to its acquisition by AMD in 2006. Somehow, being part of a bigger company that was fighting Intel on the CPU front, weakened the focus on GPU development, giving NVIDIA the opportunity to stride ahead.
This explains why the media and industry were all abuzz about Intel’s venture into this space. A new player in the form of Intel is formidable in any sphere. The dominant player in the CPU arena has a wealth of technological know-how as well as the financial clout to develop a worthy competitive product.
Interestingly, the tweet was linked to the appointment of Raja Kodurias in November 2017 as the Chief Architect, Senior Vice President of Intel’s newly-formed Core and Visual Computing Group, and General Manager of a new initiative to drive edge computing solutions.
According to that post, “in this position, Koduri will expand Intel’s leading position in integrated graphics for the PC market with high-end discrete graphics solutions for a broad range of computing segments”.
This confidence steps from his previous job as AMD, where Koduri’s last appointment was Senior Vice President and Chief Architect of the Radeon Technologies Group, which is responsible for graphics technologies.
While Koduri’s appointment is definitely a game changer in the GPU race, it takes more than just one person to build a GPU architecture from scratch. Both NVIDIA and AMD have invested heavily in research and technology (R&D) over the years to bring GPUs to where they are today — used in both gaming and driving high-end applications. And it’s going to take a lot of work for Intel to get a foot in.
Nevertheless, it would be interesting to see what Intel can come up with by 2020. If nothing else, it’s a win for a market that’s been longing for more competition to drive more innovations and lower prices. Ironically, that is what organisations and consumers want too in the CPU space where Intel has such a commanding lead. Perhaps NVIDIA should consider bringing competition into Intel’s backyard.