At the end of September, Intel was joined by AMD GPU maker Vineet Goel. The company has confirmed to CRN that Goel has taken over as vice president and general manager of the Xe Architecture and IP Engineering Group. Goel reports directly to Raja Koduri, an AMD veteran who spearheaded Intel’s discrete GPU development.
An Intel spokesperson told CRN that Goel has led a team to architect, design, and verify the Intel Xe roadmap, which includes graphics, multimedia, and display functionality for future graphics products ( GPU) of the company.
Goel was previously corporate vice president of GPU architecture at AMD, where he led the GPU architecture team from concept to turnkey solution for all of the company’s GPU families. This includes products in segments such as mobile devices, consoles and PCs, as well as HPC servers. His team also worked on GPUs for AMD APUs.
Prior to AMD, Goel served as Director of GPU Architecture for Qualcomm, where he led the architecture development for the mobile GPUs in the Adreno family. Goel earned his PhD in Computer Science from the University of Central Florida and an MS in Systems Science and Automation from the Indian Institute of Science in Bangalore and an MBA from Rollins College, Florida. Goel is the author of 57 registered patents (nine more are pending at the patent office). His works on the topic of “computer graphics” have been published in 15 scientific journals and collections.
The addition of a new employee of this level at Intel is due to the fact that the company’s CEO Pat Gelsinger (Pat Gelsinger) pledged to challenge NVIDIA, the market leader in GPUs. Intel plans to introduce discrete GPUs for the PC gaming market next year. They will be followed by the highly anticipated Ponte Vecchio GPU for high performance computing and artificial intelligence.
Microway’s vice president of strategic clients, Eliot Eshelman, an Intel partner, told CRN that Intel Ponte Vecchio GPUs are likely to have the same features as NVIDIA and AMD GPUs for high performance computing. The big question, he said, is whether Intel can overcome the challenges it previously faced in accelerators with the already discontinued Xeon Phi chip, and that would require the chipmaker to excel in software and user support.