AMD Terascale
Realizes: Unified VLIW5 shader compute and Eyefinity multi-display pipeline
The HD 5000-era Terascale architecture couples 40nm VLIW5 shader arrays with Eyefinity-aware raster and memory subsystems, providing a general-purpose GPU fabric where GPGPU shader arrays power multi-display graphics and OpenCL-style compute domains.
Examples
🔗
Radeon HD 5870 / FirePro V8800
The reference HD 5870 and its workstation sibling FirePro V8800 sport the Terascale 2 die with 1600 VLIW5 stream processors, enabling Eyefinity across three displays while exposing OpenCL compute to workloads.
OpenCL-style dispatch across shader arrays and vector units while driving Eyefinity display pipes.
high
single-chip
tens of picojoules per vector operation (varies by clock).