It is the first company to open-source its unified RISC-V Vector CPU-with-GPU Instruction Set Architecture (ISA) and provides register-level access to its hardware via a Hardware Abstraction Layer (HAL). This allows OEMs and content providers to leverage a RISC-V Vector Standard ISA with GPU extensions to customise their drivers and applications to a greater level than is currently possible.
X-Silicon is also the first company to offer Vulkan on RISC-V with fused GPU acceleration.
GPUs were originally designed to run games and are predominantly triangle-processors. As new workloads for parallel computing are defined, these legacy architectures are struggling and patched to perform tasks that were never intended for them.
Consequently, memory latencies and compute inefficiencies for dynamically changing content are not being well handled, so X-Silicon's solution has been designed to optimally render and manage this new type of dynamic content using a single C-GPU processor with multiple NanoTiles.
As such, this new architecture is suited to implementing more efficient graphics, AI and compute algorithms without being bound by the limitations of today's GPUs.
The NanoTile architecture is supported by 14 patents. The patent portfolio is focused on bringing a new generation of AI/ML Compute and GPU rendering using a single processor and claims to deal with ways in which data is shuttled among the processing cores and memory, reducing latencies, and improving compute efficiency. The patented IP also enables the NanoTile architecture to be deployed within edge and cloud configurations, enabling a federated model for GPU compute.
"We want to enable enhanced visual experiences directly on industrial and consumer devices where users have become accustomed to using their mobile phone displays for UI, while enhancing the intelligence on these edge devices. We also want to enable low power visual computing in the cloud that supports novel methods, leapfrogging the capabilities that are available with closed systems. Our Open-Standard, Patented C-GPU Nano Tile architecture is suited for processor companies targeting SoC solutions for low-power markets," said Dan Nilsson, CEO of X-Silicon. "By providing register level access through a Hardware Abstraction Layer, we not only enable novel and more efficient methods for rendering but also ensure compatibility across multiple hardware generations – a feature highly desired by consumer, industrial IoT and automotive OEMs with longer product life cycles."
