RayCore Lite IP is designed based on MIMD architecture. This approach enables more efficient execution of independent parallel tasks – each ray is regarded as an independent instruction. Due to this advantage, RayCore Lite with its ray tracing pipeline maintains ray tracing performance, regardless of ray coherence and scene characteristics and consequently supports real-time ray tracing. Accounting for most of the computations in Ray tracing graphics is part of the T&I (Traversal & Intersection test). Customers can achieve significant performance gain by replacing the tracking of the T&I portion with RayCore Lite, which already has a ray tracking solution.
- Powerful Path Tracing Acceleration GPU IP Solution
- ‘Unified’ Traversal and Intersection Test
- Based on High Performance MIMD Architecture
- Highly Applicable for Servers and High End GPU Chips
- H/W (T&I) Performance : 760 Mray/s (16 cores @ 190 MHz) = The Performance is an ideal performance based on the clock speed and the required clock cycles to complete the logic pipeline
- Effective Performance: 160 Mray/s = Effective Performance is bachmarked based on Intel Arria 10 PAC Card with Cornell Box.
Block Diagram of the Path tracing/Ray tracing accelerator