By Peter Debenham, Senior Consultant, Signal Processing
Fundamentally, power dissipation requirements are limiting single-core CPU speeds but we selfishly want yet faster computing.
In recent years, Graphics card Processing Units (GPUs) have moved from being highly specialised for graphics processing to being high-speed general computing systems.
This process was started by NVIDIA but has been followed by AMD and Intel. Of particular advantage is that Graphics cards have a large number (100s) of individual processors compared to the standard CPU (with 4 to 12 cores). With GPUs having a similar speed clock to modern CPUs, this, in theory, leads to processing speed-ups when looking at similar cost/power consumption.