Intel Snags AMD’s Chief Discrete GPU Architect

In what can only be described as a significant loss for AMD, Intel has successfully poached one of the company’s top discrete GPU architects to help the company with its upcoming line of Arc graphics solutions. Rohit Verma, who was formerly Lead SoC Architect for Discrete GPUs at AMD, is now Lead Product Architect at Intel’s Accelerated Computing Group.

Verma was previously a senior fellow at AMD, where he had worked for a bit over three years in the company’s semi-custom business unit, according to analyst Dylan Patel. The timing is notable as it includes the period where AMD rose from the ashes and started releasing competitive products, not to mention its success in securing the hardware wins for both next-gen consoles from Sony and Microsoft. Prior to his stint at AMD he actually worked for Intel, where he was a lead SoC architect from 1999 to 2013. His time at the company overlapped with current CEO Pat Gelsinger, who began working for Intel himself all the way back in 1979. The timing of their employment together certainly makes it seem like Gelsinger likely was the catalyst for the surprising acquisition of Verma. A number of old-time Intel employees have rejoined the company since Gelsinger took over.

On his LinkedIn page Verma announced he was “excited to be back at Intel,” and that his mandate was simple: to “define and build next generation GPU products!” Verma’s arrival coincides with Intel’s ongoing efforts to disrupt the discrete GPU market with its very own Arc line of GPUs, which will be launching in Q2 of this year according to the roadmap Intel recently released. Given the long development times involved in a GPU’s lifecycle, it’s doubtful Verma will have much impact on the company’s go-to-market GPU, which is code-named Alchemist. However Intel recently shared its GPU roadmap through 2024 in its recent investor meeting, and announced it’s already begun the architecture planning for Celestial, which is its “ultra performance” GPU set to arrive in 2024 (or beyond). It seems highly likely Verma will have substantial input on that design, as well as the GPU planned beyond Celestial, code-named Druid.

Verma will likely be working side-by-side with another high profile AMD defector, Raja Koduri, who is currently the VP and general manager of the Accelerated Computing Systems and Graphics group (AXG) at Intel. Koduri famously left AMD immediately following the launch of the company’s ill-fated Vega GPU launch, and has been spearheading Intel’s nascent discrete GPU effort ever since.

This is just the latest in a series of aggressive moves on Intel’s part as it works to reestablish its dominance in industry as a whole. Gelsinger has set a goal to reclaim overall semiconductor leadership globally by 2025. It follows news from January that Intel had poached Apple’s lead designer for its M1 silicon SoC, who was also previously employed by Intel. This is on top of the company’s plans to expand its capacity via a new fab in Ohio, as well as its recent announcement that it would begin licensing x86 CPU designs to customers as part of its Intel Foundry Services division (IFS). It also has begun to invest in an effort to work with partners in the development of new RISC-V CPU designs via IFS as well.

Continue reading

In Massive Shift, Apple Announces New Macs With ARM-Based M1 Chip

Apple saw huge success the last time it switched architectures to Intel, but this time? The jury's still out, but one thing is certain: Apple is about to make a lot more money.

Microsoft: Pluton Chip Will Bring Xbox-Like Security to Windows PCs

Intel, AMD, and Qualcomm are working to make Pluton part of their upcoming designs, which should make PCs more difficult to hack, but it also bakes Microsoft technology into your hardware.

Why Apple’s M1 Chip Threatens Intel and AMD

Intel's own history suggests it and AMD should take Apple's new M1 SoC very seriously.

How L1 and L2 CPU Caches Work, and Why They’re an Essential Part of Modern Chips

Ever been curious how L1 and L2 cache work? We're glad you asked. Here, we deep dive into the structure and nature of one of computing's most fundamental designs and innovations.