Could Samsung's In-House GPU Design Be A Credible Threat To Qualcomm, NVIDIA And AMD?

Dr. Chien-Ping Lu, known as CP, has been brought in by Samsung to help put the finishing touches on the product and to ensure that it has a successful launch. CP is a known quantity in the tech space, having been involved with the development of NVIDIA's ahead-of-its-time nForce IGP along with his subsequent work at MediaTek.

According to Graphic Speak, CP is leading Samsung engineering teams based in Austin and San Jose dedicated to "hardware, software, test, verification, and implementation." Although not many details have been given regarding the hardware, we do know that it is a SIMD design that forgoes VLIW (to reduce overhead), which has resulted in a treasure chest of patents for Samsung.

"The simulations have indicated an even better performance than predicted, and the batch instruction processing holds the promise of significantly reducing the latency in the motion-to-photon path, and bringing low-power (consumption) VR, with fast recovery very high-resolution dynamic display capability," writes said Jon Peddie of Jon Peddie Research.

It should be interesting to see how this plays out; Qualcomm's GPUs have been the gold standard on the Android side of the smartphone market, and Apple has been pushing its own designs for iPhones. On the desktop/notebook side of things, Intel has the integrated market covered, while AMD and NVIDIA vie for domination with discrete GPUs. In the supercomputing realm, NVIDIA has thrown down the gauntlet with its Tesla GPUs. 

Given its command of the high-end smartphone market, Samsung could make its mark here with the SGPU, and it stands to reason that this scalable architecture could have big ambitions down the road.

“Put it on par with Apple. The only question is where and when Samsung’s GPU will show up," adds Peddie. "This design is so good they could deploy it in every platform — it’s a function of their ambition. If I owned it, it would be in everything including cockpits and supercomputers.

“This design is so good, they could deploy it in every platform — it’s a function of their ambition. If I owned it, it would be in everything including cockpits and supercomputers.”

The first stop for this new SGPU is likely to be in a future Samsung Exynos SoC to be released next year.